{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 7.46909161641704,
  "eval_steps": 3000,
  "global_step": 96000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 10.691896057128906,
      "epoch": 0.00038902937171756465,
      "grad_norm": 13.375,
      "learning_rate": 2e-06,
      "loss": 10.7682,
      "mean_token_accuracy": 0.0,
      "num_tokens": 8095.0,
      "step": 5
    },
    {
      "entropy": 10.691933250427246,
      "epoch": 0.0007780587434351293,
      "grad_norm": 12.6875,
      "learning_rate": 4.5e-06,
      "loss": 10.7403,
      "mean_token_accuracy": 0.0006377055658958852,
      "num_tokens": 17329.0,
      "step": 10
    },
    {
      "entropy": 10.69080991744995,
      "epoch": 0.001167088115152694,
      "grad_norm": 10.125,
      "learning_rate": 7e-06,
      "loss": 10.4717,
      "mean_token_accuracy": 0.03278036494739354,
      "num_tokens": 25729.0,
      "step": 15
    },
    {
      "entropy": 10.676905536651612,
      "epoch": 0.0015561174868702586,
      "grad_norm": 6.5625,
      "learning_rate": 9.5e-06,
      "loss": 10.1506,
      "mean_token_accuracy": 0.04596309997141361,
      "num_tokens": 33980.0,
      "step": 20
    },
    {
      "entropy": 10.59173231124878,
      "epoch": 0.0019451468585878235,
      "grad_norm": 4.28125,
      "learning_rate": 1.2e-05,
      "loss": 9.8699,
      "mean_token_accuracy": 0.04366149380803108,
      "num_tokens": 41675.0,
      "step": 25
    },
    {
      "entropy": 10.51305980682373,
      "epoch": 0.002334176230305388,
      "grad_norm": 3.34375,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 9.7305,
      "mean_token_accuracy": 0.046745364740490916,
      "num_tokens": 50277.0,
      "step": 30
    },
    {
      "entropy": 10.490173530578613,
      "epoch": 0.0027232056020229526,
      "grad_norm": 3.046875,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 9.6493,
      "mean_token_accuracy": 0.0512831661850214,
      "num_tokens": 58790.0,
      "step": 35
    },
    {
      "entropy": 10.494702529907226,
      "epoch": 0.0031122349737405172,
      "grad_norm": 2.96875,
      "learning_rate": 1.95e-05,
      "loss": 9.6088,
      "mean_token_accuracy": 0.05120518207550049,
      "num_tokens": 67247.0,
      "step": 40
    },
    {
      "entropy": 10.47821397781372,
      "epoch": 0.003501264345458082,
      "grad_norm": 2.828125,
      "learning_rate": 2.2e-05,
      "loss": 9.5376,
      "mean_token_accuracy": 0.054223894327878955,
      "num_tokens": 75174.0,
      "step": 45
    },
    {
      "entropy": 10.43958568572998,
      "epoch": 0.003890293717175647,
      "grad_norm": 2.46875,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.4437,
      "mean_token_accuracy": 0.06763513237237931,
      "num_tokens": 83203.0,
      "step": 50
    },
    {
      "entropy": 10.308233833312988,
      "epoch": 0.004279323088893211,
      "grad_norm": 2.3125,
      "learning_rate": 2.7e-05,
      "loss": 9.4079,
      "mean_token_accuracy": 0.06488834656774997,
      "num_tokens": 91798.0,
      "step": 55
    },
    {
      "entropy": 10.269000720977782,
      "epoch": 0.004668352460610776,
      "grad_norm": 2.09375,
      "learning_rate": 2.95e-05,
      "loss": 9.4289,
      "mean_token_accuracy": 0.06329003609716892,
      "num_tokens": 100869.0,
      "step": 60
    },
    {
      "entropy": 10.361470031738282,
      "epoch": 0.0050573818323283405,
      "grad_norm": 2.296875,
      "learning_rate": 3.2e-05,
      "loss": 9.3296,
      "mean_token_accuracy": 0.061626766622066495,
      "num_tokens": 109697.0,
      "step": 65
    },
    {
      "entropy": 10.3142409324646,
      "epoch": 0.005446411204045905,
      "grad_norm": 2.0625,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.2921,
      "mean_token_accuracy": 0.06110430583357811,
      "num_tokens": 118656.0,
      "step": 70
    },
    {
      "entropy": 10.19054298400879,
      "epoch": 0.00583544057576347,
      "grad_norm": 1.859375,
      "learning_rate": 3.7e-05,
      "loss": 9.2586,
      "mean_token_accuracy": 0.060313938185572624,
      "num_tokens": 127748.0,
      "step": 75
    },
    {
      "entropy": 10.34864444732666,
      "epoch": 0.0062244699474810344,
      "grad_norm": 1.984375,
      "learning_rate": 3.95e-05,
      "loss": 9.1176,
      "mean_token_accuracy": 0.0627338420599699,
      "num_tokens": 136621.0,
      "step": 80
    },
    {
      "entropy": 10.175674057006836,
      "epoch": 0.006613499319198599,
      "grad_norm": 2.015625,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 8.9517,
      "mean_token_accuracy": 0.06631159856915474,
      "num_tokens": 145793.0,
      "step": 85
    },
    {
      "entropy": 10.162372493743897,
      "epoch": 0.007002528690916164,
      "grad_norm": 1.796875,
      "learning_rate": 4.45e-05,
      "loss": 8.8629,
      "mean_token_accuracy": 0.06395913921296596,
      "num_tokens": 154466.0,
      "step": 90
    },
    {
      "entropy": 10.15381736755371,
      "epoch": 0.007391558062633729,
      "grad_norm": 2.09375,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 8.8443,
      "mean_token_accuracy": 0.06343522258102893,
      "num_tokens": 163297.0,
      "step": 95
    },
    {
      "entropy": 10.048249244689941,
      "epoch": 0.007780587434351294,
      "grad_norm": 1.84375,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 8.696,
      "mean_token_accuracy": 0.06365865133702756,
      "num_tokens": 171747.0,
      "step": 100
    },
    {
      "entropy": 10.09149627685547,
      "epoch": 0.008169616806068859,
      "grad_norm": 1.6953125,
      "learning_rate": 5.2e-05,
      "loss": 8.6166,
      "mean_token_accuracy": 0.06534692756831646,
      "num_tokens": 180499.0,
      "step": 105
    },
    {
      "entropy": 9.920382881164551,
      "epoch": 0.008558646177786422,
      "grad_norm": 1.71875,
      "learning_rate": 5.45e-05,
      "loss": 8.4788,
      "mean_token_accuracy": 0.06633736491203308,
      "num_tokens": 189227.0,
      "step": 110
    },
    {
      "entropy": 9.86361379623413,
      "epoch": 0.008947675549503988,
      "grad_norm": 1.609375,
      "learning_rate": 5.7e-05,
      "loss": 8.3439,
      "mean_token_accuracy": 0.07009124867618084,
      "num_tokens": 198388.0,
      "step": 115
    },
    {
      "entropy": 9.680712413787841,
      "epoch": 0.009336704921221552,
      "grad_norm": 1.421875,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.2396,
      "mean_token_accuracy": 0.06940936222672463,
      "num_tokens": 206854.0,
      "step": 120
    },
    {
      "entropy": 9.601908302307129,
      "epoch": 0.009725734292939117,
      "grad_norm": 1.6015625,
      "learning_rate": 6.2e-05,
      "loss": 8.119,
      "mean_token_accuracy": 0.0706846684217453,
      "num_tokens": 215506.0,
      "step": 125
    },
    {
      "entropy": 9.40778636932373,
      "epoch": 0.010114763664656681,
      "grad_norm": 1.40625,
      "learning_rate": 6.450000000000001e-05,
      "loss": 8.027,
      "mean_token_accuracy": 0.07141325175762177,
      "num_tokens": 223717.0,
      "step": 130
    },
    {
      "entropy": 9.245368766784669,
      "epoch": 0.010503793036374246,
      "grad_norm": 1.25,
      "learning_rate": 6.7e-05,
      "loss": 7.9189,
      "mean_token_accuracy": 0.07270697467029094,
      "num_tokens": 232480.0,
      "step": 135
    },
    {
      "entropy": 9.090145778656005,
      "epoch": 0.01089282240809181,
      "grad_norm": 1.390625,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.8471,
      "mean_token_accuracy": 0.07338932827115059,
      "num_tokens": 241350.0,
      "step": 140
    },
    {
      "entropy": 8.89556589126587,
      "epoch": 0.011281851779809376,
      "grad_norm": 1.234375,
      "learning_rate": 7.2e-05,
      "loss": 7.7957,
      "mean_token_accuracy": 0.07191371396183968,
      "num_tokens": 250610.0,
      "step": 145
    },
    {
      "entropy": 8.702104473114014,
      "epoch": 0.01167088115152694,
      "grad_norm": 1.2109375,
      "learning_rate": 7.45e-05,
      "loss": 7.7138,
      "mean_token_accuracy": 0.06948468647897243,
      "num_tokens": 259672.0,
      "step": 150
    },
    {
      "entropy": 8.603777408599854,
      "epoch": 0.012059910523244505,
      "grad_norm": 1.3046875,
      "learning_rate": 7.7e-05,
      "loss": 7.6923,
      "mean_token_accuracy": 0.0712867472320795,
      "num_tokens": 267826.0,
      "step": 155
    },
    {
      "entropy": 8.510001087188721,
      "epoch": 0.012448939894962069,
      "grad_norm": 1.2890625,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.6847,
      "mean_token_accuracy": 0.06826631203293801,
      "num_tokens": 277324.0,
      "step": 160
    },
    {
      "entropy": 8.372300243377685,
      "epoch": 0.012837969266679634,
      "grad_norm": 1.28125,
      "learning_rate": 8.2e-05,
      "loss": 7.5139,
      "mean_token_accuracy": 0.07538944780826569,
      "num_tokens": 285482.0,
      "step": 165
    },
    {
      "entropy": 8.333262157440185,
      "epoch": 0.013226998638397198,
      "grad_norm": 1.3671875,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.5443,
      "mean_token_accuracy": 0.07613336965441704,
      "num_tokens": 293684.0,
      "step": 170
    },
    {
      "entropy": 8.189146041870117,
      "epoch": 0.013616028010114764,
      "grad_norm": 1.359375,
      "learning_rate": 8.7e-05,
      "loss": 7.4992,
      "mean_token_accuracy": 0.0754831738770008,
      "num_tokens": 302605.0,
      "step": 175
    },
    {
      "entropy": 8.120375633239746,
      "epoch": 0.014005057381832327,
      "grad_norm": 1.4453125,
      "learning_rate": 8.95e-05,
      "loss": 7.5501,
      "mean_token_accuracy": 0.07501808553934097,
      "num_tokens": 310849.0,
      "step": 180
    },
    {
      "entropy": 8.180096626281738,
      "epoch": 0.014394086753549893,
      "grad_norm": 1.3984375,
      "learning_rate": 9.2e-05,
      "loss": 7.5376,
      "mean_token_accuracy": 0.07213864214718342,
      "num_tokens": 319752.0,
      "step": 185
    },
    {
      "entropy": 8.150297737121582,
      "epoch": 0.014783116125267459,
      "grad_norm": 1.609375,
      "learning_rate": 9.45e-05,
      "loss": 7.5715,
      "mean_token_accuracy": 0.07293041869997978,
      "num_tokens": 328009.0,
      "step": 190
    },
    {
      "entropy": 8.01708698272705,
      "epoch": 0.015172145496985022,
      "grad_norm": 1.1953125,
      "learning_rate": 9.7e-05,
      "loss": 7.4454,
      "mean_token_accuracy": 0.07922360971570015,
      "num_tokens": 337414.0,
      "step": 195
    },
    {
      "entropy": 8.00553069114685,
      "epoch": 0.015561174868702588,
      "grad_norm": 1.234375,
      "learning_rate": 9.95e-05,
      "loss": 7.4438,
      "mean_token_accuracy": 0.08163341209292412,
      "num_tokens": 345419.0,
      "step": 200
    },
    {
      "entropy": 8.010188388824464,
      "epoch": 0.01595020424042015,
      "grad_norm": 1.28125,
      "learning_rate": 0.000102,
      "loss": 7.4176,
      "mean_token_accuracy": 0.08399317860603332,
      "num_tokens": 353435.0,
      "step": 205
    },
    {
      "entropy": 7.945941305160522,
      "epoch": 0.016339233612137717,
      "grad_norm": 1.34375,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.4398,
      "mean_token_accuracy": 0.07745959497988224,
      "num_tokens": 361840.0,
      "step": 210
    },
    {
      "entropy": 8.018308210372926,
      "epoch": 0.016728262983855283,
      "grad_norm": 1.1875,
      "learning_rate": 0.000107,
      "loss": 7.5097,
      "mean_token_accuracy": 0.07523712664842605,
      "num_tokens": 370873.0,
      "step": 215
    },
    {
      "entropy": 7.983619546890258,
      "epoch": 0.017117292355572845,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0001095,
      "loss": 7.4308,
      "mean_token_accuracy": 0.07907425090670586,
      "num_tokens": 379558.0,
      "step": 220
    },
    {
      "entropy": 7.867334222793579,
      "epoch": 0.01750632172729041,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000112,
      "loss": 7.4184,
      "mean_token_accuracy": 0.08053626418113709,
      "num_tokens": 387870.0,
      "step": 225
    },
    {
      "entropy": 7.964390230178833,
      "epoch": 0.017895351099007976,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001145,
      "loss": 7.4773,
      "mean_token_accuracy": 0.08220710828900338,
      "num_tokens": 396442.0,
      "step": 230
    },
    {
      "entropy": 7.887078332901001,
      "epoch": 0.01828438047072554,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.4199,
      "mean_token_accuracy": 0.07810803577303886,
      "num_tokens": 404764.0,
      "step": 235
    },
    {
      "entropy": 7.896487426757813,
      "epoch": 0.018673409842443103,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.3561,
      "mean_token_accuracy": 0.08208074867725372,
      "num_tokens": 413355.0,
      "step": 240
    },
    {
      "entropy": 7.845146036148071,
      "epoch": 0.01906243921416067,
      "grad_norm": 1.203125,
      "learning_rate": 0.000122,
      "loss": 7.25,
      "mean_token_accuracy": 0.08628582656383514,
      "num_tokens": 421868.0,
      "step": 245
    },
    {
      "entropy": 7.847237968444825,
      "epoch": 0.019451468585878234,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001245,
      "loss": 7.2885,
      "mean_token_accuracy": 0.08563556969165802,
      "num_tokens": 431106.0,
      "step": 250
    },
    {
      "entropy": 7.879620504379273,
      "epoch": 0.0198404979575958,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000127,
      "loss": 7.4271,
      "mean_token_accuracy": 0.0771660901606083,
      "num_tokens": 439459.0,
      "step": 255
    },
    {
      "entropy": 7.715308475494385,
      "epoch": 0.020229527329313362,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001295,
      "loss": 7.2523,
      "mean_token_accuracy": 0.08652433380484581,
      "num_tokens": 447943.0,
      "step": 260
    },
    {
      "entropy": 7.791458606719971,
      "epoch": 0.020618556701030927,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000132,
      "loss": 7.3749,
      "mean_token_accuracy": 0.08243331611156464,
      "num_tokens": 456691.0,
      "step": 265
    },
    {
      "entropy": 7.827027750015259,
      "epoch": 0.021007586072748493,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.2679,
      "mean_token_accuracy": 0.08309322819113732,
      "num_tokens": 466388.0,
      "step": 270
    },
    {
      "entropy": 7.677990436553955,
      "epoch": 0.02139661544446606,
      "grad_norm": 1.265625,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.3463,
      "mean_token_accuracy": 0.084238101541996,
      "num_tokens": 474544.0,
      "step": 275
    },
    {
      "entropy": 7.855551147460938,
      "epoch": 0.02178564481618362,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001395,
      "loss": 7.3638,
      "mean_token_accuracy": 0.08487658277153969,
      "num_tokens": 483561.0,
      "step": 280
    },
    {
      "entropy": 7.682491683959961,
      "epoch": 0.022174674187901186,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.1825,
      "mean_token_accuracy": 0.08649473935365677,
      "num_tokens": 492036.0,
      "step": 285
    },
    {
      "entropy": 7.736964845657349,
      "epoch": 0.02256370355961875,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001445,
      "loss": 7.3058,
      "mean_token_accuracy": 0.08307258263230324,
      "num_tokens": 500978.0,
      "step": 290
    },
    {
      "entropy": 7.750108766555786,
      "epoch": 0.022952732931336317,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000147,
      "loss": 7.331,
      "mean_token_accuracy": 0.08541621565818787,
      "num_tokens": 509719.0,
      "step": 295
    },
    {
      "entropy": 7.716811800003052,
      "epoch": 0.02334176230305388,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001495,
      "loss": 7.2418,
      "mean_token_accuracy": 0.0862745113670826,
      "num_tokens": 518422.0,
      "step": 300
    },
    {
      "entropy": 7.7041215896606445,
      "epoch": 0.023730791674771445,
      "grad_norm": 1.375,
      "learning_rate": 0.000152,
      "loss": 7.2726,
      "mean_token_accuracy": 0.08733444213867188,
      "num_tokens": 527172.0,
      "step": 305
    },
    {
      "entropy": 7.661469459533691,
      "epoch": 0.02411982104648901,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.1871,
      "mean_token_accuracy": 0.08940150961279869,
      "num_tokens": 536070.0,
      "step": 310
    },
    {
      "entropy": 7.650560522079468,
      "epoch": 0.024508850418206576,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000157,
      "loss": 7.1815,
      "mean_token_accuracy": 0.08751222342252732,
      "num_tokens": 544556.0,
      "step": 315
    },
    {
      "entropy": 7.71123743057251,
      "epoch": 0.024897879789924138,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001595,
      "loss": 7.1717,
      "mean_token_accuracy": 0.08722432479262351,
      "num_tokens": 553122.0,
      "step": 320
    },
    {
      "entropy": 7.609269285202027,
      "epoch": 0.025286909161641703,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000162,
      "loss": 7.235,
      "mean_token_accuracy": 0.07931372076272965,
      "num_tokens": 562265.0,
      "step": 325
    },
    {
      "entropy": 7.594164276123047,
      "epoch": 0.02567593853335927,
      "grad_norm": 1.1875,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.1853,
      "mean_token_accuracy": 0.09104064255952835,
      "num_tokens": 570938.0,
      "step": 330
    },
    {
      "entropy": 7.591356801986694,
      "epoch": 0.026064967905076834,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.1202,
      "mean_token_accuracy": 0.0924045629799366,
      "num_tokens": 578568.0,
      "step": 335
    },
    {
      "entropy": 7.654570388793945,
      "epoch": 0.026453997276794396,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.192,
      "mean_token_accuracy": 0.09244821965694427,
      "num_tokens": 587013.0,
      "step": 340
    },
    {
      "entropy": 7.647093391418457,
      "epoch": 0.026843026648511962,
      "grad_norm": 1.21875,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.2278,
      "mean_token_accuracy": 0.08628092929720879,
      "num_tokens": 595354.0,
      "step": 345
    },
    {
      "entropy": 7.5597922801971436,
      "epoch": 0.027232056020229527,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.1994,
      "mean_token_accuracy": 0.086347945779562,
      "num_tokens": 603971.0,
      "step": 350
    },
    {
      "entropy": 7.513363790512085,
      "epoch": 0.027621085391947093,
      "grad_norm": 1.25,
      "learning_rate": 0.000177,
      "loss": 7.1077,
      "mean_token_accuracy": 0.0958691768348217,
      "num_tokens": 612097.0,
      "step": 355
    },
    {
      "entropy": 7.607126188278198,
      "epoch": 0.028010114763664655,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001795,
      "loss": 7.1708,
      "mean_token_accuracy": 0.09145483076572418,
      "num_tokens": 621445.0,
      "step": 360
    },
    {
      "entropy": 7.514036798477173,
      "epoch": 0.02839914413538222,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000182,
      "loss": 7.2337,
      "mean_token_accuracy": 0.08651573956012726,
      "num_tokens": 630041.0,
      "step": 365
    },
    {
      "entropy": 7.50714807510376,
      "epoch": 0.028788173507099786,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001845,
      "loss": 7.0418,
      "mean_token_accuracy": 0.09683061838150024,
      "num_tokens": 638317.0,
      "step": 370
    },
    {
      "entropy": 7.585668706893921,
      "epoch": 0.02917720287881735,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000187,
      "loss": 7.1733,
      "mean_token_accuracy": 0.09136604964733124,
      "num_tokens": 647297.0,
      "step": 375
    },
    {
      "entropy": 7.613582992553711,
      "epoch": 0.029566232250534917,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0001895,
      "loss": 7.2214,
      "mean_token_accuracy": 0.09044467359781265,
      "num_tokens": 655309.0,
      "step": 380
    },
    {
      "entropy": 7.432076501846313,
      "epoch": 0.02995526162225248,
      "grad_norm": 1.265625,
      "learning_rate": 0.000192,
      "loss": 7.08,
      "mean_token_accuracy": 0.0861615188419819,
      "num_tokens": 663879.0,
      "step": 385
    },
    {
      "entropy": 7.516607666015625,
      "epoch": 0.030344290993970045,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001945,
      "loss": 6.9856,
      "mean_token_accuracy": 0.09427314698696136,
      "num_tokens": 672913.0,
      "step": 390
    },
    {
      "entropy": 7.394821453094482,
      "epoch": 0.03073332036568761,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.0979,
      "mean_token_accuracy": 0.09137862846255303,
      "num_tokens": 682282.0,
      "step": 395
    },
    {
      "entropy": 7.4547014236450195,
      "epoch": 0.031122349737405176,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019950000000000002,
      "loss": 7.1059,
      "mean_token_accuracy": 0.08515429720282555,
      "num_tokens": 691432.0,
      "step": 400
    },
    {
      "entropy": 7.444170570373535,
      "epoch": 0.03151137910912274,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000202,
      "loss": 7.102,
      "mean_token_accuracy": 0.08980939239263534,
      "num_tokens": 700639.0,
      "step": 405
    },
    {
      "entropy": 7.551699113845825,
      "epoch": 0.0319004084808403,
      "grad_norm": 1.5,
      "learning_rate": 0.00020449999999999998,
      "loss": 7.184,
      "mean_token_accuracy": 0.0844474621117115,
      "num_tokens": 709200.0,
      "step": 410
    },
    {
      "entropy": 7.30176272392273,
      "epoch": 0.03228943785255787,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000207,
      "loss": 7.0024,
      "mean_token_accuracy": 0.09402795061469078,
      "num_tokens": 717266.0,
      "step": 415
    },
    {
      "entropy": 7.484861898422241,
      "epoch": 0.032678467224275434,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002095,
      "loss": 6.9873,
      "mean_token_accuracy": 0.09403469264507294,
      "num_tokens": 725799.0,
      "step": 420
    },
    {
      "entropy": 7.3827354431152346,
      "epoch": 0.033067496595993,
      "grad_norm": 1.46875,
      "learning_rate": 0.000212,
      "loss": 7.1101,
      "mean_token_accuracy": 0.09061254039406777,
      "num_tokens": 734375.0,
      "step": 425
    },
    {
      "entropy": 7.475268840789795,
      "epoch": 0.033456525967710565,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002145,
      "loss": 7.1403,
      "mean_token_accuracy": 0.090643859654665,
      "num_tokens": 743453.0,
      "step": 430
    },
    {
      "entropy": 7.326716709136963,
      "epoch": 0.033845555339428124,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00021700000000000002,
      "loss": 7.0003,
      "mean_token_accuracy": 0.09671235233545303,
      "num_tokens": 752443.0,
      "step": 435
    },
    {
      "entropy": 7.399496269226074,
      "epoch": 0.03423458471114569,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002195,
      "loss": 6.9699,
      "mean_token_accuracy": 0.09153517037630081,
      "num_tokens": 761294.0,
      "step": 440
    },
    {
      "entropy": 7.355255746841431,
      "epoch": 0.034623614082863255,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000222,
      "loss": 6.9395,
      "mean_token_accuracy": 0.0974013090133667,
      "num_tokens": 769489.0,
      "step": 445
    },
    {
      "entropy": 7.445426177978516,
      "epoch": 0.03501264345458082,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002245,
      "loss": 7.0191,
      "mean_token_accuracy": 0.09303558543324471,
      "num_tokens": 777836.0,
      "step": 450
    },
    {
      "entropy": 7.343358373641967,
      "epoch": 0.035401672826298386,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022700000000000002,
      "loss": 7.0442,
      "mean_token_accuracy": 0.08798035010695457,
      "num_tokens": 786543.0,
      "step": 455
    },
    {
      "entropy": 7.382262325286865,
      "epoch": 0.03579070219801595,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022950000000000002,
      "loss": 7.1011,
      "mean_token_accuracy": 0.09493663161993027,
      "num_tokens": 794525.0,
      "step": 460
    },
    {
      "entropy": 7.329567766189575,
      "epoch": 0.03617973156973352,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023200000000000003,
      "loss": 7.0001,
      "mean_token_accuracy": 0.09533165432512761,
      "num_tokens": 803537.0,
      "step": 465
    },
    {
      "entropy": 7.326742553710938,
      "epoch": 0.03656876094145108,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.9507,
      "mean_token_accuracy": 0.09251324906945228,
      "num_tokens": 811706.0,
      "step": 470
    },
    {
      "entropy": 7.262205600738525,
      "epoch": 0.03695779031316864,
      "grad_norm": 1.171875,
      "learning_rate": 0.000237,
      "loss": 7.0091,
      "mean_token_accuracy": 0.09699861034750938,
      "num_tokens": 820182.0,
      "step": 475
    },
    {
      "entropy": 7.3125176429748535,
      "epoch": 0.03734681968488621,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002395,
      "loss": 6.9423,
      "mean_token_accuracy": 0.09304441586136818,
      "num_tokens": 829808.0,
      "step": 480
    },
    {
      "entropy": 7.292161369323731,
      "epoch": 0.03773584905660377,
      "grad_norm": 1.359375,
      "learning_rate": 0.000242,
      "loss": 6.8887,
      "mean_token_accuracy": 0.09506509900093078,
      "num_tokens": 838838.0,
      "step": 485
    },
    {
      "entropy": 7.33060131072998,
      "epoch": 0.03812487842832134,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002445,
      "loss": 6.9833,
      "mean_token_accuracy": 0.09907156899571419,
      "num_tokens": 847800.0,
      "step": 490
    },
    {
      "entropy": 7.394680452346802,
      "epoch": 0.0385139078000389,
      "grad_norm": 1.484375,
      "learning_rate": 0.000247,
      "loss": 7.08,
      "mean_token_accuracy": 0.09246540367603302,
      "num_tokens": 856503.0,
      "step": 495
    },
    {
      "entropy": 7.226006984710693,
      "epoch": 0.03890293717175647,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002495,
      "loss": 6.9683,
      "mean_token_accuracy": 0.09611012637615204,
      "num_tokens": 865245.0,
      "step": 500
    },
    {
      "entropy": 7.303643465042114,
      "epoch": 0.039291966543474034,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000252,
      "loss": 6.9749,
      "mean_token_accuracy": 0.0949372723698616,
      "num_tokens": 874485.0,
      "step": 505
    },
    {
      "entropy": 7.133185148239136,
      "epoch": 0.0396809959151916,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002545,
      "loss": 6.8372,
      "mean_token_accuracy": 0.10009809583425522,
      "num_tokens": 883301.0,
      "step": 510
    },
    {
      "entropy": 7.179492998123169,
      "epoch": 0.04007002528690916,
      "grad_norm": 1.296875,
      "learning_rate": 0.000257,
      "loss": 6.8987,
      "mean_token_accuracy": 0.09437932819128036,
      "num_tokens": 892387.0,
      "step": 515
    },
    {
      "entropy": 7.161367273330688,
      "epoch": 0.040459054658626724,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002595,
      "loss": 6.8902,
      "mean_token_accuracy": 0.101420196890831,
      "num_tokens": 901522.0,
      "step": 520
    },
    {
      "entropy": 7.167306137084961,
      "epoch": 0.04084808403034429,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000262,
      "loss": 6.8411,
      "mean_token_accuracy": 0.09641877040266991,
      "num_tokens": 909848.0,
      "step": 525
    },
    {
      "entropy": 7.3447840213775635,
      "epoch": 0.041237113402061855,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026450000000000003,
      "loss": 7.0797,
      "mean_token_accuracy": 0.09041683003306389,
      "num_tokens": 919079.0,
      "step": 530
    },
    {
      "entropy": 7.150223207473755,
      "epoch": 0.04162614277377942,
      "grad_norm": 1.390625,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.8357,
      "mean_token_accuracy": 0.10363327041268348,
      "num_tokens": 927807.0,
      "step": 535
    },
    {
      "entropy": 7.154421615600586,
      "epoch": 0.042015172145496986,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.9331,
      "mean_token_accuracy": 0.10026705339550972,
      "num_tokens": 936498.0,
      "step": 540
    },
    {
      "entropy": 7.247080707550049,
      "epoch": 0.04240420151721455,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00027200000000000005,
      "loss": 7.0055,
      "mean_token_accuracy": 0.10041748508810996,
      "num_tokens": 945436.0,
      "step": 545
    },
    {
      "entropy": 7.224219846725464,
      "epoch": 0.04279323088893212,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002745,
      "loss": 6.9165,
      "mean_token_accuracy": 0.09951690733432769,
      "num_tokens": 953990.0,
      "step": 550
    },
    {
      "entropy": 7.183721542358398,
      "epoch": 0.04318226026064968,
      "grad_norm": 1.328125,
      "learning_rate": 0.000277,
      "loss": 6.8412,
      "mean_token_accuracy": 0.0982144832611084,
      "num_tokens": 962008.0,
      "step": 555
    },
    {
      "entropy": 7.243313407897949,
      "epoch": 0.04357128963236724,
      "grad_norm": 1.125,
      "learning_rate": 0.0002795,
      "loss": 7.004,
      "mean_token_accuracy": 0.097385935485363,
      "num_tokens": 970935.0,
      "step": 560
    },
    {
      "entropy": 7.262167167663574,
      "epoch": 0.04396031900408481,
      "grad_norm": 1.171875,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.9564,
      "mean_token_accuracy": 0.09126960262656211,
      "num_tokens": 979492.0,
      "step": 565
    },
    {
      "entropy": 7.1115397930145265,
      "epoch": 0.04434934837580237,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002845,
      "loss": 6.8659,
      "mean_token_accuracy": 0.09609195366501808,
      "num_tokens": 987866.0,
      "step": 570
    },
    {
      "entropy": 7.152759647369384,
      "epoch": 0.04473837774751994,
      "grad_norm": 1.140625,
      "learning_rate": 0.000287,
      "loss": 6.9064,
      "mean_token_accuracy": 0.09902488738298416,
      "num_tokens": 996760.0,
      "step": 575
    },
    {
      "entropy": 7.205654001235962,
      "epoch": 0.0451274071192375,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002895,
      "loss": 6.8252,
      "mean_token_accuracy": 0.10003674700856209,
      "num_tokens": 1005432.0,
      "step": 580
    },
    {
      "entropy": 7.1957074165344235,
      "epoch": 0.04551643649095507,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000292,
      "loss": 6.9228,
      "mean_token_accuracy": 0.10067613199353218,
      "num_tokens": 1013729.0,
      "step": 585
    },
    {
      "entropy": 6.997878503799439,
      "epoch": 0.045905465862672634,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002945,
      "loss": 6.7348,
      "mean_token_accuracy": 0.1082767240703106,
      "num_tokens": 1022485.0,
      "step": 590
    },
    {
      "entropy": 7.08715558052063,
      "epoch": 0.0462944952343902,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000297,
      "loss": 6.8166,
      "mean_token_accuracy": 0.10034638941287995,
      "num_tokens": 1030523.0,
      "step": 595
    },
    {
      "entropy": 7.2036830425262455,
      "epoch": 0.04668352460610776,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002995,
      "loss": 6.9339,
      "mean_token_accuracy": 0.09973467364907265,
      "num_tokens": 1039268.0,
      "step": 600
    },
    {
      "entropy": 7.0301995277404785,
      "epoch": 0.047072553977825324,
      "grad_norm": 1.25,
      "learning_rate": 0.000302,
      "loss": 6.8202,
      "mean_token_accuracy": 0.10042936652898789,
      "num_tokens": 1048590.0,
      "step": 605
    },
    {
      "entropy": 7.0609663963317875,
      "epoch": 0.04746158334954289,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003045,
      "loss": 6.7556,
      "mean_token_accuracy": 0.10278418138623238,
      "num_tokens": 1057232.0,
      "step": 610
    },
    {
      "entropy": 7.049582195281983,
      "epoch": 0.047850612721260455,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000307,
      "loss": 6.8216,
      "mean_token_accuracy": 0.09878815785050392,
      "num_tokens": 1066404.0,
      "step": 615
    },
    {
      "entropy": 7.038175964355469,
      "epoch": 0.04823964209297802,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003095,
      "loss": 6.8842,
      "mean_token_accuracy": 0.0939465455710888,
      "num_tokens": 1075093.0,
      "step": 620
    },
    {
      "entropy": 7.106451797485351,
      "epoch": 0.048628671464695586,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000312,
      "loss": 6.7875,
      "mean_token_accuracy": 0.09886922240257263,
      "num_tokens": 1084454.0,
      "step": 625
    },
    {
      "entropy": 6.990231323242187,
      "epoch": 0.04901770083641315,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003145,
      "loss": 6.7069,
      "mean_token_accuracy": 0.10300704166293144,
      "num_tokens": 1093452.0,
      "step": 630
    },
    {
      "entropy": 7.079220771789551,
      "epoch": 0.04940673020813072,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000317,
      "loss": 6.7417,
      "mean_token_accuracy": 0.10723969489336013,
      "num_tokens": 1101814.0,
      "step": 635
    },
    {
      "entropy": 6.995628929138183,
      "epoch": 0.049795759579848276,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003195,
      "loss": 6.8303,
      "mean_token_accuracy": 0.10031761527061463,
      "num_tokens": 1110386.0,
      "step": 640
    },
    {
      "entropy": 7.119993448257446,
      "epoch": 0.05018478895156584,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000322,
      "loss": 6.9206,
      "mean_token_accuracy": 0.09205639511346816,
      "num_tokens": 1120141.0,
      "step": 645
    },
    {
      "entropy": 7.026125097274781,
      "epoch": 0.05057381832328341,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.8553,
      "mean_token_accuracy": 0.09914684519171715,
      "num_tokens": 1129920.0,
      "step": 650
    },
    {
      "entropy": 6.9065900325775145,
      "epoch": 0.05096284769500097,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.7387,
      "mean_token_accuracy": 0.10237120762467385,
      "num_tokens": 1138518.0,
      "step": 655
    },
    {
      "entropy": 7.056087446212769,
      "epoch": 0.05135187706671854,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.8477,
      "mean_token_accuracy": 0.09784239828586579,
      "num_tokens": 1147113.0,
      "step": 660
    },
    {
      "entropy": 6.963723516464233,
      "epoch": 0.0517409064384361,
      "grad_norm": 1.359375,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.6747,
      "mean_token_accuracy": 0.10692296847701073,
      "num_tokens": 1155886.0,
      "step": 665
    },
    {
      "entropy": 6.999239587783814,
      "epoch": 0.05212993581015367,
      "grad_norm": 1.359375,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.8402,
      "mean_token_accuracy": 0.10142037197947502,
      "num_tokens": 1164200.0,
      "step": 670
    },
    {
      "entropy": 6.969135522842407,
      "epoch": 0.052518965181871234,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000337,
      "loss": 6.7113,
      "mean_token_accuracy": 0.10580320581793785,
      "num_tokens": 1172720.0,
      "step": 675
    },
    {
      "entropy": 6.8918328285217285,
      "epoch": 0.05290799455358879,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003395,
      "loss": 6.6591,
      "mean_token_accuracy": 0.09922505840659142,
      "num_tokens": 1181225.0,
      "step": 680
    },
    {
      "entropy": 7.110291481018066,
      "epoch": 0.05329702392530636,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000342,
      "loss": 6.8661,
      "mean_token_accuracy": 0.1002252757549286,
      "num_tokens": 1190070.0,
      "step": 685
    },
    {
      "entropy": 6.889911794662476,
      "epoch": 0.053686053297023924,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.745,
      "mean_token_accuracy": 0.09947578459978104,
      "num_tokens": 1198995.0,
      "step": 690
    },
    {
      "entropy": 7.012227392196655,
      "epoch": 0.05407508266874149,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000347,
      "loss": 6.7964,
      "mean_token_accuracy": 0.10564659833908081,
      "num_tokens": 1207664.0,
      "step": 695
    },
    {
      "entropy": 7.011436891555786,
      "epoch": 0.054464112040459055,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003495,
      "loss": 6.6877,
      "mean_token_accuracy": 0.11616539508104325,
      "num_tokens": 1216232.0,
      "step": 700
    },
    {
      "entropy": 6.915061187744141,
      "epoch": 0.05485314141217662,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000352,
      "loss": 6.7785,
      "mean_token_accuracy": 0.10488367006182671,
      "num_tokens": 1224106.0,
      "step": 705
    },
    {
      "entropy": 6.85550856590271,
      "epoch": 0.055242170783894186,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003545,
      "loss": 6.6638,
      "mean_token_accuracy": 0.1070045679807663,
      "num_tokens": 1232514.0,
      "step": 710
    },
    {
      "entropy": 7.0201925277709964,
      "epoch": 0.05563120015561175,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000357,
      "loss": 6.7192,
      "mean_token_accuracy": 0.1072799600660801,
      "num_tokens": 1240178.0,
      "step": 715
    },
    {
      "entropy": 6.829072189331055,
      "epoch": 0.05602022952732931,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003595,
      "loss": 6.6433,
      "mean_token_accuracy": 0.10768454670906066,
      "num_tokens": 1248635.0,
      "step": 720
    },
    {
      "entropy": 6.899731540679932,
      "epoch": 0.056409258899046875,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000362,
      "loss": 6.7462,
      "mean_token_accuracy": 0.10741591677069665,
      "num_tokens": 1256389.0,
      "step": 725
    },
    {
      "entropy": 6.866089391708374,
      "epoch": 0.05679828827076444,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003645,
      "loss": 6.6941,
      "mean_token_accuracy": 0.1057390108704567,
      "num_tokens": 1264035.0,
      "step": 730
    },
    {
      "entropy": 6.956638479232788,
      "epoch": 0.057187317642482006,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000367,
      "loss": 6.6953,
      "mean_token_accuracy": 0.10426240265369416,
      "num_tokens": 1272941.0,
      "step": 735
    },
    {
      "entropy": 6.688238668441772,
      "epoch": 0.05757634701419957,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003695,
      "loss": 6.676,
      "mean_token_accuracy": 0.106101393699646,
      "num_tokens": 1281767.0,
      "step": 740
    },
    {
      "entropy": 7.095763778686523,
      "epoch": 0.05796537638591714,
      "grad_norm": 1.1875,
      "learning_rate": 0.000372,
      "loss": 6.7623,
      "mean_token_accuracy": 0.10797640457749366,
      "num_tokens": 1290049.0,
      "step": 745
    },
    {
      "entropy": 6.762351226806641,
      "epoch": 0.0583544057576347,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003745,
      "loss": 6.5951,
      "mean_token_accuracy": 0.11266529634594917,
      "num_tokens": 1299032.0,
      "step": 750
    },
    {
      "entropy": 6.872419738769532,
      "epoch": 0.05874343512935227,
      "grad_norm": 1.234375,
      "learning_rate": 0.000377,
      "loss": 6.851,
      "mean_token_accuracy": 0.10019904971122742,
      "num_tokens": 1308508.0,
      "step": 755
    },
    {
      "entropy": 6.993260669708252,
      "epoch": 0.059132464501069834,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003795,
      "loss": 6.6769,
      "mean_token_accuracy": 0.10615287944674492,
      "num_tokens": 1317044.0,
      "step": 760
    },
    {
      "entropy": 6.876901865005493,
      "epoch": 0.05952149387278739,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000382,
      "loss": 6.7072,
      "mean_token_accuracy": 0.10667419582605361,
      "num_tokens": 1325554.0,
      "step": 765
    },
    {
      "entropy": 6.8257323741912845,
      "epoch": 0.05991052324450496,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003845,
      "loss": 6.5916,
      "mean_token_accuracy": 0.10517750829458236,
      "num_tokens": 1334201.0,
      "step": 770
    },
    {
      "entropy": 6.738331365585327,
      "epoch": 0.060299552616222524,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.6132,
      "mean_token_accuracy": 0.11592961698770524,
      "num_tokens": 1343061.0,
      "step": 775
    },
    {
      "entropy": 6.788525867462158,
      "epoch": 0.06068858198794009,
      "grad_norm": 1.25,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.5887,
      "mean_token_accuracy": 0.11328529119491577,
      "num_tokens": 1352262.0,
      "step": 780
    },
    {
      "entropy": 6.766341304779052,
      "epoch": 0.061077611359657655,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.6355,
      "mean_token_accuracy": 0.10724243074655533,
      "num_tokens": 1361400.0,
      "step": 785
    },
    {
      "entropy": 6.8809511184692385,
      "epoch": 0.06146664073137522,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.7431,
      "mean_token_accuracy": 0.10390153974294662,
      "num_tokens": 1369850.0,
      "step": 790
    },
    {
      "entropy": 6.770475769042969,
      "epoch": 0.061855670103092786,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.5993,
      "mean_token_accuracy": 0.11133262068033219,
      "num_tokens": 1378342.0,
      "step": 795
    },
    {
      "entropy": 6.833333206176758,
      "epoch": 0.06224469947481035,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003995,
      "loss": 6.7006,
      "mean_token_accuracy": 0.10400806367397308,
      "num_tokens": 1386981.0,
      "step": 800
    },
    {
      "entropy": 6.896929168701172,
      "epoch": 0.06263372884652792,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000402,
      "loss": 6.7767,
      "mean_token_accuracy": 0.10226056203246117,
      "num_tokens": 1395801.0,
      "step": 805
    },
    {
      "entropy": 6.881122875213623,
      "epoch": 0.06302275821824548,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004045,
      "loss": 6.6733,
      "mean_token_accuracy": 0.11250713765621186,
      "num_tokens": 1403715.0,
      "step": 810
    },
    {
      "entropy": 6.890682744979858,
      "epoch": 0.06341178758996305,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.766,
      "mean_token_accuracy": 0.10130376219749451,
      "num_tokens": 1413199.0,
      "step": 815
    },
    {
      "entropy": 6.845803165435791,
      "epoch": 0.0638008169616806,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004095,
      "loss": 6.7297,
      "mean_token_accuracy": 0.10811256021261215,
      "num_tokens": 1421918.0,
      "step": 820
    },
    {
      "entropy": 6.727932119369507,
      "epoch": 0.06418984633339817,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000412,
      "loss": 6.6275,
      "mean_token_accuracy": 0.10078049674630166,
      "num_tokens": 1430625.0,
      "step": 825
    },
    {
      "entropy": 6.839619207382202,
      "epoch": 0.06457887570511574,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004145,
      "loss": 6.6537,
      "mean_token_accuracy": 0.10472832173109055,
      "num_tokens": 1439158.0,
      "step": 830
    },
    {
      "entropy": 6.7103558540344235,
      "epoch": 0.0649679050768333,
      "grad_norm": 1.15625,
      "learning_rate": 0.000417,
      "loss": 6.6218,
      "mean_token_accuracy": 0.11031484305858612,
      "num_tokens": 1448266.0,
      "step": 835
    },
    {
      "entropy": 6.824801921844482,
      "epoch": 0.06535693444855087,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004195,
      "loss": 6.7042,
      "mean_token_accuracy": 0.10656223520636558,
      "num_tokens": 1457322.0,
      "step": 840
    },
    {
      "entropy": 6.962126445770264,
      "epoch": 0.06574596382026843,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000422,
      "loss": 6.8104,
      "mean_token_accuracy": 0.10119246616959572,
      "num_tokens": 1466120.0,
      "step": 845
    },
    {
      "entropy": 6.734336900711059,
      "epoch": 0.066134993191986,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004245,
      "loss": 6.5668,
      "mean_token_accuracy": 0.10800775364041329,
      "num_tokens": 1474965.0,
      "step": 850
    },
    {
      "entropy": 6.679752445220947,
      "epoch": 0.06652402256370356,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000427,
      "loss": 6.6775,
      "mean_token_accuracy": 0.10365973934531211,
      "num_tokens": 1483483.0,
      "step": 855
    },
    {
      "entropy": 6.906285905838013,
      "epoch": 0.06691305193542113,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004295,
      "loss": 6.6044,
      "mean_token_accuracy": 0.10145875811576843,
      "num_tokens": 1492313.0,
      "step": 860
    },
    {
      "entropy": 6.7070897102355955,
      "epoch": 0.06730208130713869,
      "grad_norm": 1.203125,
      "learning_rate": 0.000432,
      "loss": 6.533,
      "mean_token_accuracy": 0.10664732605218888,
      "num_tokens": 1501081.0,
      "step": 865
    },
    {
      "entropy": 6.653385829925537,
      "epoch": 0.06769111067885625,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004345,
      "loss": 6.5538,
      "mean_token_accuracy": 0.10266396701335907,
      "num_tokens": 1510169.0,
      "step": 870
    },
    {
      "entropy": 6.73896050453186,
      "epoch": 0.06808014005057382,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000437,
      "loss": 6.5099,
      "mean_token_accuracy": 0.10998809859156608,
      "num_tokens": 1519546.0,
      "step": 875
    },
    {
      "entropy": 6.610293054580689,
      "epoch": 0.06846916942229138,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004395,
      "loss": 6.6221,
      "mean_token_accuracy": 0.10572586581110954,
      "num_tokens": 1529104.0,
      "step": 880
    },
    {
      "entropy": 6.671034288406372,
      "epoch": 0.06885819879400895,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000442,
      "loss": 6.5747,
      "mean_token_accuracy": 0.11553266048431396,
      "num_tokens": 1537033.0,
      "step": 885
    },
    {
      "entropy": 6.745483112335205,
      "epoch": 0.06924722816572651,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004445,
      "loss": 6.5984,
      "mean_token_accuracy": 0.10939133316278457,
      "num_tokens": 1545059.0,
      "step": 890
    },
    {
      "entropy": 6.75796480178833,
      "epoch": 0.06963625753744408,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000447,
      "loss": 6.5969,
      "mean_token_accuracy": 0.10726859346032143,
      "num_tokens": 1553784.0,
      "step": 895
    },
    {
      "entropy": 6.6265500545501705,
      "epoch": 0.07002528690916164,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.5325,
      "mean_token_accuracy": 0.1090914063155651,
      "num_tokens": 1562059.0,
      "step": 900
    },
    {
      "entropy": 6.712047672271728,
      "epoch": 0.0704143162808792,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.591,
      "mean_token_accuracy": 0.10841032788157463,
      "num_tokens": 1570061.0,
      "step": 905
    },
    {
      "entropy": 6.632591104507446,
      "epoch": 0.07080334565259677,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.5654,
      "mean_token_accuracy": 0.11026803255081177,
      "num_tokens": 1578261.0,
      "step": 910
    },
    {
      "entropy": 6.621666669845581,
      "epoch": 0.07119237502431433,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.4859,
      "mean_token_accuracy": 0.11358301490545272,
      "num_tokens": 1586260.0,
      "step": 915
    },
    {
      "entropy": 6.711280393600464,
      "epoch": 0.0715814043960319,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.6821,
      "mean_token_accuracy": 0.10423496067523956,
      "num_tokens": 1595234.0,
      "step": 920
    },
    {
      "entropy": 6.685232305526734,
      "epoch": 0.07197043376774946,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000462,
      "loss": 6.5795,
      "mean_token_accuracy": 0.10836847126483917,
      "num_tokens": 1604212.0,
      "step": 925
    },
    {
      "entropy": 6.730190992355347,
      "epoch": 0.07235946313946703,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004645,
      "loss": 6.6145,
      "mean_token_accuracy": 0.11065342202782631,
      "num_tokens": 1612925.0,
      "step": 930
    },
    {
      "entropy": 6.590854501724243,
      "epoch": 0.07274849251118459,
      "grad_norm": 1.21875,
      "learning_rate": 0.000467,
      "loss": 6.5066,
      "mean_token_accuracy": 0.10939636155962944,
      "num_tokens": 1621804.0,
      "step": 935
    },
    {
      "entropy": 6.572239732742309,
      "epoch": 0.07313752188290217,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004695,
      "loss": 6.4856,
      "mean_token_accuracy": 0.11304571852087975,
      "num_tokens": 1630943.0,
      "step": 940
    },
    {
      "entropy": 6.617482995986938,
      "epoch": 0.07352655125461972,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000472,
      "loss": 6.3922,
      "mean_token_accuracy": 0.11573764458298683,
      "num_tokens": 1638899.0,
      "step": 945
    },
    {
      "entropy": 6.57882833480835,
      "epoch": 0.07391558062633728,
      "grad_norm": 1.25,
      "learning_rate": 0.0004745,
      "loss": 6.6125,
      "mean_token_accuracy": 0.11446217894554138,
      "num_tokens": 1647485.0,
      "step": 950
    },
    {
      "entropy": 6.657685661315918,
      "epoch": 0.07430460999805485,
      "grad_norm": 1.171875,
      "learning_rate": 0.000477,
      "loss": 6.5656,
      "mean_token_accuracy": 0.11251236274838447,
      "num_tokens": 1656290.0,
      "step": 955
    },
    {
      "entropy": 6.712019300460815,
      "epoch": 0.07469363936977241,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004795,
      "loss": 6.5924,
      "mean_token_accuracy": 0.11167355254292488,
      "num_tokens": 1665474.0,
      "step": 960
    },
    {
      "entropy": 6.533619403839111,
      "epoch": 0.07508266874148999,
      "grad_norm": 1.3125,
      "learning_rate": 0.000482,
      "loss": 6.464,
      "mean_token_accuracy": 0.1160540908575058,
      "num_tokens": 1673599.0,
      "step": 965
    },
    {
      "entropy": 6.607017183303833,
      "epoch": 0.07547169811320754,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004845,
      "loss": 6.4586,
      "mean_token_accuracy": 0.11052219942212105,
      "num_tokens": 1683282.0,
      "step": 970
    },
    {
      "entropy": 6.630674982070923,
      "epoch": 0.07586072748492512,
      "grad_norm": 1.15625,
      "learning_rate": 0.000487,
      "loss": 6.5825,
      "mean_token_accuracy": 0.1119205579161644,
      "num_tokens": 1691094.0,
      "step": 975
    },
    {
      "entropy": 6.443245840072632,
      "epoch": 0.07624975685664268,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004895,
      "loss": 6.444,
      "mean_token_accuracy": 0.11762246191501617,
      "num_tokens": 1700053.0,
      "step": 980
    },
    {
      "entropy": 6.574719095230103,
      "epoch": 0.07663878622836025,
      "grad_norm": 1.265625,
      "learning_rate": 0.000492,
      "loss": 6.3455,
      "mean_token_accuracy": 0.1151976853609085,
      "num_tokens": 1707732.0,
      "step": 985
    },
    {
      "entropy": 6.575902795791626,
      "epoch": 0.0770278156000778,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004945,
      "loss": 6.6179,
      "mean_token_accuracy": 0.10963503271341324,
      "num_tokens": 1716406.0,
      "step": 990
    },
    {
      "entropy": 6.661358880996704,
      "epoch": 0.07741684497179536,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000497,
      "loss": 6.6325,
      "mean_token_accuracy": 0.10895756259560585,
      "num_tokens": 1725832.0,
      "step": 995
    },
    {
      "entropy": 6.733318710327149,
      "epoch": 0.07780587434351294,
      "grad_norm": 1.125,
      "learning_rate": 0.0004995,
      "loss": 6.5474,
      "mean_token_accuracy": 0.10725341215729714,
      "num_tokens": 1734853.0,
      "step": 1000
    },
    {
      "entropy": 6.559145545959472,
      "epoch": 0.0781949037152305,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004999999989075146,
      "loss": 6.5731,
      "mean_token_accuracy": 0.11213273108005524,
      "num_tokens": 1744240.0,
      "step": 1005
    },
    {
      "entropy": 6.637544441223144,
      "epoch": 0.07858393308694807,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999999944692927,
      "loss": 6.566,
      "mean_token_accuracy": 0.10990296676754951,
      "num_tokens": 1753334.0,
      "step": 1010
    },
    {
      "entropy": 6.47784652709961,
      "epoch": 0.07897296245866563,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499999986617054,
      "loss": 6.4764,
      "mean_token_accuracy": 0.11047938987612724,
      "num_tokens": 1762032.0,
      "step": 1015
    },
    {
      "entropy": 6.572625017166137,
      "epoch": 0.0793619918303832,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999999753507986,
      "loss": 6.4775,
      "mean_token_accuracy": 0.11230112910270691,
      "num_tokens": 1770508.0,
      "step": 1020
    },
    {
      "entropy": 6.491672945022583,
      "epoch": 0.07975102120210076,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999999606705267,
      "loss": 6.3729,
      "mean_token_accuracy": 0.11614649370312691,
      "num_tokens": 1780167.0,
      "step": 1025
    },
    {
      "entropy": 6.619625043869019,
      "epoch": 0.08014005057381832,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999999425762385,
      "loss": 6.5776,
      "mean_token_accuracy": 0.10773374065756798,
      "num_tokens": 1788809.0,
      "step": 1030
    },
    {
      "entropy": 6.548305225372315,
      "epoch": 0.08052907994553589,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004999999210679344,
      "loss": 6.4943,
      "mean_token_accuracy": 0.11628751307725907,
      "num_tokens": 1797445.0,
      "step": 1035
    },
    {
      "entropy": 6.631950330734253,
      "epoch": 0.08091810931725345,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999998961456145,
      "loss": 6.5027,
      "mean_token_accuracy": 0.11015650182962418,
      "num_tokens": 1806488.0,
      "step": 1040
    },
    {
      "entropy": 6.501224613189697,
      "epoch": 0.08130713868897102,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999998678092794,
      "loss": 6.5833,
      "mean_token_accuracy": 0.11565212160348892,
      "num_tokens": 1815692.0,
      "step": 1045
    },
    {
      "entropy": 6.625118780136108,
      "epoch": 0.08169616806068858,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999998360589293,
      "loss": 6.4938,
      "mean_token_accuracy": 0.11357240229845048,
      "num_tokens": 1824599.0,
      "step": 1050
    },
    {
      "entropy": 6.689116191864014,
      "epoch": 0.08208519743240615,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000499999800894565,
      "loss": 6.6448,
      "mean_token_accuracy": 0.11452656835317612,
      "num_tokens": 1833411.0,
      "step": 1055
    },
    {
      "entropy": 6.518345499038697,
      "epoch": 0.08247422680412371,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004999997623161866,
      "loss": 6.4549,
      "mean_token_accuracy": 0.10907695516943931,
      "num_tokens": 1842177.0,
      "step": 1060
    },
    {
      "entropy": 6.470740461349488,
      "epoch": 0.08286325617584128,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999997203237951,
      "loss": 6.4244,
      "mean_token_accuracy": 0.116130381077528,
      "num_tokens": 1850176.0,
      "step": 1065
    },
    {
      "entropy": 6.579597043991089,
      "epoch": 0.08325228554755884,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499999674917391,
      "loss": 6.4737,
      "mean_token_accuracy": 0.10759498700499534,
      "num_tokens": 1859464.0,
      "step": 1070
    },
    {
      "entropy": 6.453550958633423,
      "epoch": 0.0836413149192764,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999996260969748,
      "loss": 6.3705,
      "mean_token_accuracy": 0.11378242447972298,
      "num_tokens": 1867733.0,
      "step": 1075
    },
    {
      "entropy": 6.466144609451294,
      "epoch": 0.08403034429099397,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999995738625475,
      "loss": 6.3725,
      "mean_token_accuracy": 0.11944899260997772,
      "num_tokens": 1876010.0,
      "step": 1080
    },
    {
      "entropy": 6.499561977386475,
      "epoch": 0.08441937366271153,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999995182141099,
      "loss": 6.5391,
      "mean_token_accuracy": 0.1163535013794899,
      "num_tokens": 1884906.0,
      "step": 1085
    },
    {
      "entropy": 6.527617025375366,
      "epoch": 0.0848084030344291,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999994591516626,
      "loss": 6.4293,
      "mean_token_accuracy": 0.11143236458301545,
      "num_tokens": 1893464.0,
      "step": 1090
    },
    {
      "entropy": 6.541065073013305,
      "epoch": 0.08519743240614666,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999993966752066,
      "loss": 6.4873,
      "mean_token_accuracy": 0.11323103830218315,
      "num_tokens": 1902769.0,
      "step": 1095
    },
    {
      "entropy": 6.517040348052978,
      "epoch": 0.08558646177786423,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499999330784743,
      "loss": 6.4036,
      "mean_token_accuracy": 0.11408817544579505,
      "num_tokens": 1911968.0,
      "step": 1100
    },
    {
      "entropy": 6.483025693893433,
      "epoch": 0.08597549114958179,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004999992614802725,
      "loss": 6.3911,
      "mean_token_accuracy": 0.11573496237397193,
      "num_tokens": 1920063.0,
      "step": 1105
    },
    {
      "entropy": 6.351643323898315,
      "epoch": 0.08636452052129936,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999991887617966,
      "loss": 6.3803,
      "mean_token_accuracy": 0.11390967592597008,
      "num_tokens": 1929192.0,
      "step": 1110
    },
    {
      "entropy": 6.570942449569702,
      "epoch": 0.08675354989301692,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999991126293158,
      "loss": 6.5798,
      "mean_token_accuracy": 0.10959510430693627,
      "num_tokens": 1939082.0,
      "step": 1115
    },
    {
      "entropy": 6.373979282379151,
      "epoch": 0.08714257926473448,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999990330828318,
      "loss": 6.43,
      "mean_token_accuracy": 0.1143827423453331,
      "num_tokens": 1948313.0,
      "step": 1120
    },
    {
      "entropy": 6.586628484725952,
      "epoch": 0.08753160863645205,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004999989501223456,
      "loss": 6.4296,
      "mean_token_accuracy": 0.11529109328985214,
      "num_tokens": 1957035.0,
      "step": 1125
    },
    {
      "entropy": 6.405944681167602,
      "epoch": 0.08792063800816961,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999988637478584,
      "loss": 6.4257,
      "mean_token_accuracy": 0.11158483400940895,
      "num_tokens": 1966041.0,
      "step": 1130
    },
    {
      "entropy": 6.511107158660889,
      "epoch": 0.08830966737988719,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999987739593716,
      "loss": 6.4577,
      "mean_token_accuracy": 0.10981602147221566,
      "num_tokens": 1975019.0,
      "step": 1135
    },
    {
      "entropy": 6.464044380187988,
      "epoch": 0.08869869675160474,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999986807568865,
      "loss": 6.3725,
      "mean_token_accuracy": 0.12287635952234269,
      "num_tokens": 1982930.0,
      "step": 1140
    },
    {
      "entropy": 6.514709520339966,
      "epoch": 0.08908772612332232,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004999985841404045,
      "loss": 6.4993,
      "mean_token_accuracy": 0.1092392586171627,
      "num_tokens": 1991888.0,
      "step": 1145
    },
    {
      "entropy": 6.3899940013885494,
      "epoch": 0.08947675549503988,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999984841099271,
      "loss": 6.3541,
      "mean_token_accuracy": 0.11540794596076012,
      "num_tokens": 2000202.0,
      "step": 1150
    },
    {
      "entropy": 6.460285663604736,
      "epoch": 0.08986578486675743,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499998380665456,
      "loss": 6.3698,
      "mean_token_accuracy": 0.11728540211915969,
      "num_tokens": 2009281.0,
      "step": 1155
    },
    {
      "entropy": 6.4477497100830075,
      "epoch": 0.090254814238475,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999982738069924,
      "loss": 6.3702,
      "mean_token_accuracy": 0.11771893873810768,
      "num_tokens": 2017806.0,
      "step": 1160
    },
    {
      "entropy": 6.440757131576538,
      "epoch": 0.09064384361019256,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999981635345382,
      "loss": 6.4672,
      "mean_token_accuracy": 0.1128288336098194,
      "num_tokens": 2027582.0,
      "step": 1165
    },
    {
      "entropy": 6.436817359924317,
      "epoch": 0.09103287298191014,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999980498480949,
      "loss": 6.3218,
      "mean_token_accuracy": 0.12205893769860268,
      "num_tokens": 2035427.0,
      "step": 1170
    },
    {
      "entropy": 6.547386360168457,
      "epoch": 0.0914219023536277,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999979327476644,
      "loss": 6.5285,
      "mean_token_accuracy": 0.11442452073097228,
      "num_tokens": 2044603.0,
      "step": 1175
    },
    {
      "entropy": 6.3849019527435305,
      "epoch": 0.09181093172534527,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999978122332485,
      "loss": 6.3552,
      "mean_token_accuracy": 0.12413827255368233,
      "num_tokens": 2053272.0,
      "step": 1180
    },
    {
      "entropy": 6.4762718200683596,
      "epoch": 0.09219996109706283,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999976883048487,
      "loss": 6.3797,
      "mean_token_accuracy": 0.11840806007385254,
      "num_tokens": 2061677.0,
      "step": 1185
    },
    {
      "entropy": 6.43572359085083,
      "epoch": 0.0925889904687804,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999975609624673,
      "loss": 6.397,
      "mean_token_accuracy": 0.11727287769317626,
      "num_tokens": 2070282.0,
      "step": 1190
    },
    {
      "entropy": 6.5320531845092775,
      "epoch": 0.09297801984049796,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004999974302061059,
      "loss": 6.5035,
      "mean_token_accuracy": 0.111043381690979,
      "num_tokens": 2078812.0,
      "step": 1195
    },
    {
      "entropy": 6.397515392303466,
      "epoch": 0.09336704921221552,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999972960357666,
      "loss": 6.3813,
      "mean_token_accuracy": 0.11424519494175911,
      "num_tokens": 2087020.0,
      "step": 1200
    },
    {
      "entropy": 6.413018131256104,
      "epoch": 0.09375607858393309,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999971584514516,
      "loss": 6.3238,
      "mean_token_accuracy": 0.11890219822525978,
      "num_tokens": 2095933.0,
      "step": 1205
    },
    {
      "entropy": 6.320618343353272,
      "epoch": 0.09414510795565065,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999970174531627,
      "loss": 6.3436,
      "mean_token_accuracy": 0.12122933119535446,
      "num_tokens": 2105218.0,
      "step": 1210
    },
    {
      "entropy": 6.293485689163208,
      "epoch": 0.09453413732736822,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999968730409022,
      "loss": 6.2566,
      "mean_token_accuracy": 0.11917761564254761,
      "num_tokens": 2113808.0,
      "step": 1215
    },
    {
      "entropy": 6.390006256103516,
      "epoch": 0.09492316669908578,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999967252146723,
      "loss": 6.4251,
      "mean_token_accuracy": 0.1174811840057373,
      "num_tokens": 2123014.0,
      "step": 1220
    },
    {
      "entropy": 6.498880815505982,
      "epoch": 0.09531219607080335,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999965739744752,
      "loss": 6.346,
      "mean_token_accuracy": 0.12444588094949723,
      "num_tokens": 2131840.0,
      "step": 1225
    },
    {
      "entropy": 6.458331155776977,
      "epoch": 0.09570122544252091,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999964193203132,
      "loss": 6.4309,
      "mean_token_accuracy": 0.11823861449956893,
      "num_tokens": 2140467.0,
      "step": 1230
    },
    {
      "entropy": 6.428109121322632,
      "epoch": 0.09609025481423847,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999962612521888,
      "loss": 6.3798,
      "mean_token_accuracy": 0.12282692417502403,
      "num_tokens": 2149282.0,
      "step": 1235
    },
    {
      "entropy": 6.429515647888183,
      "epoch": 0.09647928418595604,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999960997701041,
      "loss": 6.3801,
      "mean_token_accuracy": 0.11871640384197235,
      "num_tokens": 2157781.0,
      "step": 1240
    },
    {
      "entropy": 6.419339513778686,
      "epoch": 0.0968683135576736,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999959348740618,
      "loss": 6.3565,
      "mean_token_accuracy": 0.11975458636879921,
      "num_tokens": 2166482.0,
      "step": 1245
    },
    {
      "entropy": 6.3982480525970455,
      "epoch": 0.09725734292939117,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999957665640641,
      "loss": 6.3726,
      "mean_token_accuracy": 0.11583413928747177,
      "num_tokens": 2174559.0,
      "step": 1250
    },
    {
      "entropy": 6.392813014984131,
      "epoch": 0.09764637230110873,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499995594840114,
      "loss": 6.2627,
      "mean_token_accuracy": 0.12039380520582199,
      "num_tokens": 2182374.0,
      "step": 1255
    },
    {
      "entropy": 6.315843868255615,
      "epoch": 0.0980354016728263,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999954197022138,
      "loss": 6.3739,
      "mean_token_accuracy": 0.12241144180297851,
      "num_tokens": 2191427.0,
      "step": 1260
    },
    {
      "entropy": 6.491170120239258,
      "epoch": 0.09842443104454386,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999952411503661,
      "loss": 6.4657,
      "mean_token_accuracy": 0.11634750217199326,
      "num_tokens": 2199698.0,
      "step": 1265
    },
    {
      "entropy": 6.412351083755493,
      "epoch": 0.09881346041626143,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999950591845738,
      "loss": 6.2891,
      "mean_token_accuracy": 0.12139204517006874,
      "num_tokens": 2208519.0,
      "step": 1270
    },
    {
      "entropy": 6.28710150718689,
      "epoch": 0.09920248978797899,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999948738048397,
      "loss": 6.3098,
      "mean_token_accuracy": 0.12150795012712479,
      "num_tokens": 2217453.0,
      "step": 1275
    },
    {
      "entropy": 6.34256329536438,
      "epoch": 0.09959151915969655,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999946850111663,
      "loss": 6.4153,
      "mean_token_accuracy": 0.1209065318107605,
      "num_tokens": 2225623.0,
      "step": 1280
    },
    {
      "entropy": 6.43469386100769,
      "epoch": 0.09998054853141412,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999944928035569,
      "loss": 6.3593,
      "mean_token_accuracy": 0.121299659460783,
      "num_tokens": 2234519.0,
      "step": 1285
    },
    {
      "entropy": 6.3267598152160645,
      "epoch": 0.10036957790313168,
      "grad_norm": 1.296875,
      "learning_rate": 0.000499994297182014,
      "loss": 6.3602,
      "mean_token_accuracy": 0.11970539391040802,
      "num_tokens": 2242994.0,
      "step": 1290
    },
    {
      "entropy": 6.463318681716919,
      "epoch": 0.10075860727484925,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999940981465408,
      "loss": 6.3755,
      "mean_token_accuracy": 0.11988109499216079,
      "num_tokens": 2251687.0,
      "step": 1295
    },
    {
      "entropy": 6.263813257217407,
      "epoch": 0.10114763664656681,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999938956971404,
      "loss": 6.2951,
      "mean_token_accuracy": 0.12231847047805786,
      "num_tokens": 2260751.0,
      "step": 1300
    },
    {
      "entropy": 6.42852816581726,
      "epoch": 0.10153666601828439,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999936898338156,
      "loss": 6.2395,
      "mean_token_accuracy": 0.1288893111050129,
      "num_tokens": 2269125.0,
      "step": 1305
    },
    {
      "entropy": 6.226840353012085,
      "epoch": 0.10192569539000194,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999934805565698,
      "loss": 6.1775,
      "mean_token_accuracy": 0.12845248058438302,
      "num_tokens": 2277098.0,
      "step": 1310
    },
    {
      "entropy": 6.373805332183838,
      "epoch": 0.10231472476171952,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499993267865406,
      "loss": 6.455,
      "mean_token_accuracy": 0.11760153844952584,
      "num_tokens": 2285953.0,
      "step": 1315
    },
    {
      "entropy": 6.406149291992188,
      "epoch": 0.10270375413343708,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999930517603275,
      "loss": 6.3714,
      "mean_token_accuracy": 0.12183143571019173,
      "num_tokens": 2294487.0,
      "step": 1320
    },
    {
      "entropy": 6.39051923751831,
      "epoch": 0.10309278350515463,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999928322413374,
      "loss": 6.3909,
      "mean_token_accuracy": 0.11598639711737632,
      "num_tokens": 2302915.0,
      "step": 1325
    },
    {
      "entropy": 6.3545863151550295,
      "epoch": 0.1034818128768722,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999926093084394,
      "loss": 6.3443,
      "mean_token_accuracy": 0.11570636853575707,
      "num_tokens": 2310769.0,
      "step": 1330
    },
    {
      "entropy": 6.3191876888275145,
      "epoch": 0.10387084224858976,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999923829616365,
      "loss": 6.2362,
      "mean_token_accuracy": 0.11870313882827759,
      "num_tokens": 2319987.0,
      "step": 1335
    },
    {
      "entropy": 6.339432001113892,
      "epoch": 0.10425987162030734,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999921532009323,
      "loss": 6.3474,
      "mean_token_accuracy": 0.11524386182427407,
      "num_tokens": 2328964.0,
      "step": 1340
    },
    {
      "entropy": 6.397040700912475,
      "epoch": 0.1046489009920249,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999919200263304,
      "loss": 6.3133,
      "mean_token_accuracy": 0.11604936346411705,
      "num_tokens": 2337945.0,
      "step": 1345
    },
    {
      "entropy": 6.3518479347229,
      "epoch": 0.10503793036374247,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999916834378342,
      "loss": 6.2137,
      "mean_token_accuracy": 0.1228095144033432,
      "num_tokens": 2345989.0,
      "step": 1350
    },
    {
      "entropy": 6.209484100341797,
      "epoch": 0.10542695973546003,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999914434354472,
      "loss": 6.3234,
      "mean_token_accuracy": 0.12122376561164856,
      "num_tokens": 2353752.0,
      "step": 1355
    },
    {
      "entropy": 6.382920408248902,
      "epoch": 0.10581598910717759,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999912000191733,
      "loss": 6.2872,
      "mean_token_accuracy": 0.11960536614060402,
      "num_tokens": 2361961.0,
      "step": 1360
    },
    {
      "entropy": 6.2765013694763185,
      "epoch": 0.10620501847889516,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499990953189016,
      "loss": 6.2129,
      "mean_token_accuracy": 0.12653318867087365,
      "num_tokens": 2370148.0,
      "step": 1365
    },
    {
      "entropy": 6.378698635101318,
      "epoch": 0.10659404785061272,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999907029449791,
      "loss": 6.2592,
      "mean_token_accuracy": 0.12621283754706383,
      "num_tokens": 2379105.0,
      "step": 1370
    },
    {
      "entropy": 6.351945304870606,
      "epoch": 0.10698307722233029,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999904492870663,
      "loss": 6.3305,
      "mean_token_accuracy": 0.12122908383607864,
      "num_tokens": 2387417.0,
      "step": 1375
    },
    {
      "entropy": 6.3498913764953615,
      "epoch": 0.10737210659404785,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999901922152816,
      "loss": 6.3536,
      "mean_token_accuracy": 0.12595803663134575,
      "num_tokens": 2396163.0,
      "step": 1380
    },
    {
      "entropy": 6.179137945175171,
      "epoch": 0.10776113596576542,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999899317296288,
      "loss": 6.2105,
      "mean_token_accuracy": 0.12522833421826363,
      "num_tokens": 2404661.0,
      "step": 1385
    },
    {
      "entropy": 6.334095621109009,
      "epoch": 0.10815016533748298,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004999896678301121,
      "loss": 6.302,
      "mean_token_accuracy": 0.11695858612656593,
      "num_tokens": 2413459.0,
      "step": 1390
    },
    {
      "entropy": 6.355089330673218,
      "epoch": 0.10853919470920055,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999894005167352,
      "loss": 6.2443,
      "mean_token_accuracy": 0.12691024467349052,
      "num_tokens": 2421401.0,
      "step": 1395
    },
    {
      "entropy": 6.320974922180175,
      "epoch": 0.10892822408091811,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499989129789502,
      "loss": 6.252,
      "mean_token_accuracy": 0.1299786426126957,
      "num_tokens": 2429983.0,
      "step": 1400
    },
    {
      "entropy": 6.366431760787964,
      "epoch": 0.10931725345263567,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999888556484172,
      "loss": 6.3893,
      "mean_token_accuracy": 0.11359420493245125,
      "num_tokens": 2439206.0,
      "step": 1405
    },
    {
      "entropy": 6.271417093276978,
      "epoch": 0.10970628282435324,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999885780934845,
      "loss": 6.3147,
      "mean_token_accuracy": 0.13233309537172316,
      "num_tokens": 2448801.0,
      "step": 1410
    },
    {
      "entropy": 6.330869722366333,
      "epoch": 0.1100953121960708,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999882971247081,
      "loss": 6.3539,
      "mean_token_accuracy": 0.12248845249414445,
      "num_tokens": 2458007.0,
      "step": 1415
    },
    {
      "entropy": 6.32300910949707,
      "epoch": 0.11048434156778837,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999880127420926,
      "loss": 6.2393,
      "mean_token_accuracy": 0.13286312147974969,
      "num_tokens": 2466574.0,
      "step": 1420
    },
    {
      "entropy": 6.4055665016174315,
      "epoch": 0.11087337093950593,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999877249456421,
      "loss": 6.3219,
      "mean_token_accuracy": 0.12253004312515259,
      "num_tokens": 2475263.0,
      "step": 1425
    },
    {
      "entropy": 6.054247617721558,
      "epoch": 0.1112624003112235,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999874337353609,
      "loss": 6.0664,
      "mean_token_accuracy": 0.12846405506134034,
      "num_tokens": 2483738.0,
      "step": 1430
    },
    {
      "entropy": 6.2588417530059814,
      "epoch": 0.11165142968294106,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999871391112535,
      "loss": 6.187,
      "mean_token_accuracy": 0.1216193564236164,
      "num_tokens": 2492937.0,
      "step": 1435
    },
    {
      "entropy": 6.194214534759522,
      "epoch": 0.11204045905465862,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999868410733244,
      "loss": 6.2503,
      "mean_token_accuracy": 0.12336413487792015,
      "num_tokens": 2500716.0,
      "step": 1440
    },
    {
      "entropy": 6.254022169113159,
      "epoch": 0.11242948842637619,
      "grad_norm": 1.15625,
      "learning_rate": 0.000499986539621578,
      "loss": 6.2575,
      "mean_token_accuracy": 0.12012252137064934,
      "num_tokens": 2509385.0,
      "step": 1445
    },
    {
      "entropy": 6.420900154113769,
      "epoch": 0.11281851779809375,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999862347560191,
      "loss": 6.3259,
      "mean_token_accuracy": 0.11690858453512191,
      "num_tokens": 2518179.0,
      "step": 1450
    },
    {
      "entropy": 6.249479198455811,
      "epoch": 0.11320754716981132,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999859264766521,
      "loss": 6.2585,
      "mean_token_accuracy": 0.12355490997433663,
      "num_tokens": 2526644.0,
      "step": 1455
    },
    {
      "entropy": 6.265651607513428,
      "epoch": 0.11359657654152888,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999856147834817,
      "loss": 6.1782,
      "mean_token_accuracy": 0.12229724824428559,
      "num_tokens": 2534998.0,
      "step": 1460
    },
    {
      "entropy": 6.28430061340332,
      "epoch": 0.11398560591324645,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999852996765129,
      "loss": 6.253,
      "mean_token_accuracy": 0.125435870885849,
      "num_tokens": 2543562.0,
      "step": 1465
    },
    {
      "entropy": 6.282908535003662,
      "epoch": 0.11437463528496401,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999849811557502,
      "loss": 6.2506,
      "mean_token_accuracy": 0.1216797761619091,
      "num_tokens": 2551688.0,
      "step": 1470
    },
    {
      "entropy": 6.275887537002563,
      "epoch": 0.11476366465668159,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999846592211984,
      "loss": 6.3275,
      "mean_token_accuracy": 0.12465370073914528,
      "num_tokens": 2560227.0,
      "step": 1475
    },
    {
      "entropy": 6.321870994567871,
      "epoch": 0.11515269402839914,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999843338728626,
      "loss": 6.2034,
      "mean_token_accuracy": 0.12613042816519737,
      "num_tokens": 2569245.0,
      "step": 1480
    },
    {
      "entropy": 6.224797201156616,
      "epoch": 0.1155417234001167,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999840051107476,
      "loss": 6.223,
      "mean_token_accuracy": 0.1265519917011261,
      "num_tokens": 2577706.0,
      "step": 1485
    },
    {
      "entropy": 6.319393301010132,
      "epoch": 0.11593075277183428,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999836729348584,
      "loss": 6.2579,
      "mean_token_accuracy": 0.12350616082549096,
      "num_tokens": 2586075.0,
      "step": 1490
    },
    {
      "entropy": 6.252358913421631,
      "epoch": 0.11631978214355183,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999833373452,
      "loss": 6.2734,
      "mean_token_accuracy": 0.12134807705879211,
      "num_tokens": 2594965.0,
      "step": 1495
    },
    {
      "entropy": 6.257946348190307,
      "epoch": 0.1167088115152694,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999829983417778,
      "loss": 6.1954,
      "mean_token_accuracy": 0.12667486742138861,
      "num_tokens": 2603355.0,
      "step": 1500
    },
    {
      "entropy": 6.317920351028443,
      "epoch": 0.11709784088698696,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999826559245965,
      "loss": 6.3263,
      "mean_token_accuracy": 0.12176407799124718,
      "num_tokens": 2611697.0,
      "step": 1505
    },
    {
      "entropy": 6.24525089263916,
      "epoch": 0.11748687025870454,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999823100936615,
      "loss": 6.1881,
      "mean_token_accuracy": 0.12955067977309226,
      "num_tokens": 2620821.0,
      "step": 1510
    },
    {
      "entropy": 6.2969574451446535,
      "epoch": 0.1178758996304221,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999819608489781,
      "loss": 6.1926,
      "mean_token_accuracy": 0.1269972711801529,
      "num_tokens": 2628891.0,
      "step": 1515
    },
    {
      "entropy": 6.232346153259277,
      "epoch": 0.11826492900213967,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999816081905515,
      "loss": 6.3299,
      "mean_token_accuracy": 0.12064500376582146,
      "num_tokens": 2638641.0,
      "step": 1520
    },
    {
      "entropy": 6.277464914321899,
      "epoch": 0.11865395837385723,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999812521183872,
      "loss": 6.2218,
      "mean_token_accuracy": 0.12372512891888618,
      "num_tokens": 2646898.0,
      "step": 1525
    },
    {
      "entropy": 6.343517303466797,
      "epoch": 0.11904298774557479,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999808926324903,
      "loss": 6.2575,
      "mean_token_accuracy": 0.12216506227850914,
      "num_tokens": 2655404.0,
      "step": 1530
    },
    {
      "entropy": 6.174682712554931,
      "epoch": 0.11943201711729236,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999805297328665,
      "loss": 6.1524,
      "mean_token_accuracy": 0.1293441690504551,
      "num_tokens": 2663851.0,
      "step": 1535
    },
    {
      "entropy": 6.2491882801055905,
      "epoch": 0.11982104648900992,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999801634195214,
      "loss": 6.1957,
      "mean_token_accuracy": 0.12269230708479881,
      "num_tokens": 2672231.0,
      "step": 1540
    },
    {
      "entropy": 6.198257303237915,
      "epoch": 0.12021007586072749,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999797936924603,
      "loss": 6.209,
      "mean_token_accuracy": 0.12302690893411636,
      "num_tokens": 2680198.0,
      "step": 1545
    },
    {
      "entropy": 6.30586805343628,
      "epoch": 0.12059910523244505,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499979420551689,
      "loss": 6.2377,
      "mean_token_accuracy": 0.12162458598613739,
      "num_tokens": 2689208.0,
      "step": 1550
    },
    {
      "entropy": 6.306586265563965,
      "epoch": 0.12098813460416262,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499979043997213,
      "loss": 6.2899,
      "mean_token_accuracy": 0.11871301680803299,
      "num_tokens": 2698116.0,
      "step": 1555
    },
    {
      "entropy": 6.267044591903686,
      "epoch": 0.12137716397588018,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999786640290381,
      "loss": 6.1975,
      "mean_token_accuracy": 0.11922681853175163,
      "num_tokens": 2707046.0,
      "step": 1560
    },
    {
      "entropy": 6.23398265838623,
      "epoch": 0.12176619334759774,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999782806471701,
      "loss": 6.192,
      "mean_token_accuracy": 0.1261439874768257,
      "num_tokens": 2714929.0,
      "step": 1565
    },
    {
      "entropy": 6.2020190238952635,
      "epoch": 0.12215522271931531,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999778938516147,
      "loss": 6.2158,
      "mean_token_accuracy": 0.1232543520629406,
      "num_tokens": 2723436.0,
      "step": 1570
    },
    {
      "entropy": 6.223355484008789,
      "epoch": 0.12254425209103287,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000499977503642378,
      "loss": 6.266,
      "mean_token_accuracy": 0.12410560548305512,
      "num_tokens": 2731935.0,
      "step": 1575
    },
    {
      "entropy": 6.175629901885986,
      "epoch": 0.12293328146275044,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999771100194656,
      "loss": 6.0801,
      "mean_token_accuracy": 0.13136413991451262,
      "num_tokens": 2740307.0,
      "step": 1580
    },
    {
      "entropy": 6.136730718612671,
      "epoch": 0.123322310834468,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999767129828837,
      "loss": 6.0848,
      "mean_token_accuracy": 0.13186190873384476,
      "num_tokens": 2749077.0,
      "step": 1585
    },
    {
      "entropy": 6.22774748802185,
      "epoch": 0.12371134020618557,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999763125326383,
      "loss": 6.2122,
      "mean_token_accuracy": 0.1224086344242096,
      "num_tokens": 2758133.0,
      "step": 1590
    },
    {
      "entropy": 6.179558944702149,
      "epoch": 0.12410036957790313,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999759086687355,
      "loss": 6.1341,
      "mean_token_accuracy": 0.1290878973901272,
      "num_tokens": 2766509.0,
      "step": 1595
    },
    {
      "entropy": 6.1690037727355955,
      "epoch": 0.1244893989496207,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999755013911813,
      "loss": 6.308,
      "mean_token_accuracy": 0.12781446427106857,
      "num_tokens": 2775908.0,
      "step": 1600
    },
    {
      "entropy": 6.2387913227081295,
      "epoch": 0.12487842832133826,
      "grad_norm": 1.125,
      "learning_rate": 0.000499975090699982,
      "loss": 6.0694,
      "mean_token_accuracy": 0.13497328236699105,
      "num_tokens": 2784803.0,
      "step": 1605
    },
    {
      "entropy": 6.248407363891602,
      "epoch": 0.12526745769305583,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999746765951438,
      "loss": 6.2272,
      "mean_token_accuracy": 0.12323559522628784,
      "num_tokens": 2793897.0,
      "step": 1610
    },
    {
      "entropy": 6.144442796707153,
      "epoch": 0.1256564870647734,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999742590766729,
      "loss": 6.0709,
      "mean_token_accuracy": 0.12970903664827346,
      "num_tokens": 2802652.0,
      "step": 1615
    },
    {
      "entropy": 6.195466232299805,
      "epoch": 0.12604551643649095,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999738381445757,
      "loss": 6.2156,
      "mean_token_accuracy": 0.12098970115184784,
      "num_tokens": 2811527.0,
      "step": 1620
    },
    {
      "entropy": 6.168512582778931,
      "epoch": 0.1264345458082085,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999734137988586,
      "loss": 6.0632,
      "mean_token_accuracy": 0.13398998901247977,
      "num_tokens": 2819410.0,
      "step": 1625
    },
    {
      "entropy": 6.040783309936524,
      "epoch": 0.1268235751799261,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499972986039528,
      "loss": 6.0735,
      "mean_token_accuracy": 0.1366802491247654,
      "num_tokens": 2828079.0,
      "step": 1630
    },
    {
      "entropy": 6.200006103515625,
      "epoch": 0.12721260455164365,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999725548665904,
      "loss": 6.1926,
      "mean_token_accuracy": 0.13287901505827904,
      "num_tokens": 2836326.0,
      "step": 1635
    },
    {
      "entropy": 6.217862844467163,
      "epoch": 0.1276016339233612,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999721202800524,
      "loss": 6.2434,
      "mean_token_accuracy": 0.12198355495929718,
      "num_tokens": 2845271.0,
      "step": 1640
    },
    {
      "entropy": 6.14015941619873,
      "epoch": 0.12799066329507877,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999716822799206,
      "loss": 6.1008,
      "mean_token_accuracy": 0.13585591837763786,
      "num_tokens": 2853688.0,
      "step": 1645
    },
    {
      "entropy": 6.2203113555908205,
      "epoch": 0.12837969266679633,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999712408662016,
      "loss": 6.1861,
      "mean_token_accuracy": 0.13381585478782654,
      "num_tokens": 2862261.0,
      "step": 1650
    },
    {
      "entropy": 6.182382297515869,
      "epoch": 0.12876872203851392,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999707960389021,
      "loss": 6.1977,
      "mean_token_accuracy": 0.12676158770918847,
      "num_tokens": 2871324.0,
      "step": 1655
    },
    {
      "entropy": 6.26635627746582,
      "epoch": 0.12915775141023147,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999703477980288,
      "loss": 6.2583,
      "mean_token_accuracy": 0.11872835904359817,
      "num_tokens": 2880268.0,
      "step": 1660
    },
    {
      "entropy": 6.208874273300171,
      "epoch": 0.12954678078194903,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999698961435885,
      "loss": 6.1612,
      "mean_token_accuracy": 0.12779268771409988,
      "num_tokens": 2888510.0,
      "step": 1665
    },
    {
      "entropy": 6.200055122375488,
      "epoch": 0.1299358101536666,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999694410755882,
      "loss": 6.07,
      "mean_token_accuracy": 0.13593013808131219,
      "num_tokens": 2897616.0,
      "step": 1670
    },
    {
      "entropy": 6.106130218505859,
      "epoch": 0.13032483952538418,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999689825940347,
      "loss": 6.166,
      "mean_token_accuracy": 0.13071656748652458,
      "num_tokens": 2905703.0,
      "step": 1675
    },
    {
      "entropy": 6.1497344970703125,
      "epoch": 0.13071386889710174,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999685206989349,
      "loss": 6.0503,
      "mean_token_accuracy": 0.13300523832440375,
      "num_tokens": 2914881.0,
      "step": 1680
    },
    {
      "entropy": 6.064972543716431,
      "epoch": 0.1311028982688193,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999680553902959,
      "loss": 6.111,
      "mean_token_accuracy": 0.13151929527521133,
      "num_tokens": 2923563.0,
      "step": 1685
    },
    {
      "entropy": 6.2831262111663815,
      "epoch": 0.13149192764053685,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999675866681247,
      "loss": 6.3242,
      "mean_token_accuracy": 0.12403198331594467,
      "num_tokens": 2932821.0,
      "step": 1690
    },
    {
      "entropy": 6.244329929351807,
      "epoch": 0.1318809570122544,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999671145324286,
      "loss": 6.2049,
      "mean_token_accuracy": 0.12842011004686354,
      "num_tokens": 2940926.0,
      "step": 1695
    },
    {
      "entropy": 6.194862365722656,
      "epoch": 0.132269986383972,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999666389832144,
      "loss": 6.1331,
      "mean_token_accuracy": 0.13136012181639672,
      "num_tokens": 2950028.0,
      "step": 1700
    },
    {
      "entropy": 6.151176118850708,
      "epoch": 0.13265901575568956,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999661600204897,
      "loss": 6.1368,
      "mean_token_accuracy": 0.13054827898740767,
      "num_tokens": 2957636.0,
      "step": 1705
    },
    {
      "entropy": 6.0364806175231935,
      "epoch": 0.13304804512740712,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999656776442615,
      "loss": 5.9255,
      "mean_token_accuracy": 0.1334216207265854,
      "num_tokens": 2966544.0,
      "step": 1710
    },
    {
      "entropy": 6.1242743015289305,
      "epoch": 0.13343707449912467,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999651918545372,
      "loss": 6.1605,
      "mean_token_accuracy": 0.12257000207901,
      "num_tokens": 2976236.0,
      "step": 1715
    },
    {
      "entropy": 6.152473640441895,
      "epoch": 0.13382610387084226,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999647026513243,
      "loss": 6.1172,
      "mean_token_accuracy": 0.13368836417794228,
      "num_tokens": 2984261.0,
      "step": 1720
    },
    {
      "entropy": 6.08830189704895,
      "epoch": 0.13421513324255982,
      "grad_norm": 1.078125,
      "learning_rate": 0.00049996421003463,
      "loss": 6.0843,
      "mean_token_accuracy": 0.12732574567198754,
      "num_tokens": 2992970.0,
      "step": 1725
    },
    {
      "entropy": 6.202375268936157,
      "epoch": 0.13460416261427738,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999637140044619,
      "loss": 6.1223,
      "mean_token_accuracy": 0.13581531047821044,
      "num_tokens": 3001356.0,
      "step": 1730
    },
    {
      "entropy": 6.075127840042114,
      "epoch": 0.13499319198599494,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999632145608275,
      "loss": 6.1247,
      "mean_token_accuracy": 0.13326116278767586,
      "num_tokens": 3009509.0,
      "step": 1735
    },
    {
      "entropy": 6.151353645324707,
      "epoch": 0.1353822213577125,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999627117037343,
      "loss": 6.1486,
      "mean_token_accuracy": 0.1305352844297886,
      "num_tokens": 3017675.0,
      "step": 1740
    },
    {
      "entropy": 6.159458446502685,
      "epoch": 0.13577125072943008,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00049996220543319,
      "loss": 6.1111,
      "mean_token_accuracy": 0.13545044139027596,
      "num_tokens": 3025527.0,
      "step": 1745
    },
    {
      "entropy": 6.122771835327148,
      "epoch": 0.13616028010114764,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999616957492025,
      "loss": 6.0723,
      "mean_token_accuracy": 0.13396283537149428,
      "num_tokens": 3034251.0,
      "step": 1750
    },
    {
      "entropy": 6.054795837402343,
      "epoch": 0.1365493094728652,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999611826517793,
      "loss": 6.3029,
      "mean_token_accuracy": 0.12316446378827095,
      "num_tokens": 3043272.0,
      "step": 1755
    },
    {
      "entropy": 6.303162145614624,
      "epoch": 0.13693833884458276,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999606661409281,
      "loss": 6.1523,
      "mean_token_accuracy": 0.1299624763429165,
      "num_tokens": 3051092.0,
      "step": 1760
    },
    {
      "entropy": 6.103368616104126,
      "epoch": 0.13732736821630034,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999601462166569,
      "loss": 6.1178,
      "mean_token_accuracy": 0.12677409127354622,
      "num_tokens": 3060584.0,
      "step": 1765
    },
    {
      "entropy": 6.161142778396607,
      "epoch": 0.1377163975880179,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999596228789736,
      "loss": 6.1185,
      "mean_token_accuracy": 0.13394217044115067,
      "num_tokens": 3069990.0,
      "step": 1770
    },
    {
      "entropy": 6.1580225944519045,
      "epoch": 0.13810542695973546,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999590961278859,
      "loss": 6.0852,
      "mean_token_accuracy": 0.12508078217506408,
      "num_tokens": 3078837.0,
      "step": 1775
    },
    {
      "entropy": 6.0550312995910645,
      "epoch": 0.13849445633145302,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999585659634021,
      "loss": 6.0777,
      "mean_token_accuracy": 0.1352244459092617,
      "num_tokens": 3087557.0,
      "step": 1780
    },
    {
      "entropy": 6.0994940280914305,
      "epoch": 0.13888348570317058,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999580323855301,
      "loss": 6.1028,
      "mean_token_accuracy": 0.12971555292606354,
      "num_tokens": 3096524.0,
      "step": 1785
    },
    {
      "entropy": 6.242001056671143,
      "epoch": 0.13927251507488816,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999574953942781,
      "loss": 6.164,
      "mean_token_accuracy": 0.12785928025841714,
      "num_tokens": 3105271.0,
      "step": 1790
    },
    {
      "entropy": 6.110553503036499,
      "epoch": 0.13966154444660572,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999569549896541,
      "loss": 6.071,
      "mean_token_accuracy": 0.12340522482991219,
      "num_tokens": 3114479.0,
      "step": 1795
    },
    {
      "entropy": 6.123258876800537,
      "epoch": 0.14005057381832328,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999564111716665,
      "loss": 6.0558,
      "mean_token_accuracy": 0.14115509390830994,
      "num_tokens": 3123319.0,
      "step": 1800
    },
    {
      "entropy": 6.0733929634094235,
      "epoch": 0.14043960319004084,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999558639403232,
      "loss": 6.0825,
      "mean_token_accuracy": 0.13198988437652587,
      "num_tokens": 3132532.0,
      "step": 1805
    },
    {
      "entropy": 6.229814195632935,
      "epoch": 0.1408286325617584,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999553132956328,
      "loss": 6.1592,
      "mean_token_accuracy": 0.12872777804732322,
      "num_tokens": 3140798.0,
      "step": 1810
    },
    {
      "entropy": 6.103444957733155,
      "epoch": 0.14121766193347599,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999547592376035,
      "loss": 6.1475,
      "mean_token_accuracy": 0.13363951444625854,
      "num_tokens": 3149607.0,
      "step": 1815
    },
    {
      "entropy": 6.03386869430542,
      "epoch": 0.14160669130519354,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999542017662438,
      "loss": 5.9717,
      "mean_token_accuracy": 0.13687212467193605,
      "num_tokens": 3157537.0,
      "step": 1820
    },
    {
      "entropy": 6.089433431625366,
      "epoch": 0.1419957206769111,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999536408815623,
      "loss": 6.1376,
      "mean_token_accuracy": 0.12610816434025765,
      "num_tokens": 3166351.0,
      "step": 1825
    },
    {
      "entropy": 6.060422134399414,
      "epoch": 0.14238475004862866,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999530765835672,
      "loss": 5.9849,
      "mean_token_accuracy": 0.13618941605091095,
      "num_tokens": 3174690.0,
      "step": 1830
    },
    {
      "entropy": 6.075569534301758,
      "epoch": 0.14277377942034625,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999525088722673,
      "loss": 6.1061,
      "mean_token_accuracy": 0.1265280582010746,
      "num_tokens": 3183460.0,
      "step": 1835
    },
    {
      "entropy": 6.1444073677062985,
      "epoch": 0.1431628087920638,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999519377476711,
      "loss": 6.0722,
      "mean_token_accuracy": 0.1317371055483818,
      "num_tokens": 3192183.0,
      "step": 1840
    },
    {
      "entropy": 6.090657949447632,
      "epoch": 0.14355183816378136,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999513632097872,
      "loss": 6.1671,
      "mean_token_accuracy": 0.13313975185155869,
      "num_tokens": 3200875.0,
      "step": 1845
    },
    {
      "entropy": 6.174860572814941,
      "epoch": 0.14394086753549892,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999507852586245,
      "loss": 6.0212,
      "mean_token_accuracy": 0.13179150596261024,
      "num_tokens": 3209797.0,
      "step": 1850
    },
    {
      "entropy": 6.107384920120239,
      "epoch": 0.14432989690721648,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999502038941917,
      "loss": 6.1357,
      "mean_token_accuracy": 0.126478423923254,
      "num_tokens": 3218612.0,
      "step": 1855
    },
    {
      "entropy": 6.0968400001525875,
      "epoch": 0.14471892627893407,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999496191164975,
      "loss": 6.0554,
      "mean_token_accuracy": 0.1293291188776493,
      "num_tokens": 3227364.0,
      "step": 1860
    },
    {
      "entropy": 6.112162446975708,
      "epoch": 0.14510795565065163,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499949030925551,
      "loss": 6.0761,
      "mean_token_accuracy": 0.12927125319838523,
      "num_tokens": 3236448.0,
      "step": 1865
    },
    {
      "entropy": 6.098051738739014,
      "epoch": 0.14549698502236919,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999484393213609,
      "loss": 6.012,
      "mean_token_accuracy": 0.13607736453413963,
      "num_tokens": 3244957.0,
      "step": 1870
    },
    {
      "entropy": 6.022885847091675,
      "epoch": 0.14588601439408674,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999478443039363,
      "loss": 6.1038,
      "mean_token_accuracy": 0.12877627462148666,
      "num_tokens": 3253250.0,
      "step": 1875
    },
    {
      "entropy": 6.15300350189209,
      "epoch": 0.14627504376580433,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999472458732862,
      "loss": 5.9976,
      "mean_token_accuracy": 0.14316268488764763,
      "num_tokens": 3261455.0,
      "step": 1880
    },
    {
      "entropy": 6.0133215427398685,
      "epoch": 0.1466640731375219,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999466440294198,
      "loss": 6.0574,
      "mean_token_accuracy": 0.13383367732167245,
      "num_tokens": 3270108.0,
      "step": 1885
    },
    {
      "entropy": 6.168636178970337,
      "epoch": 0.14705310250923945,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499946038772346,
      "loss": 6.1715,
      "mean_token_accuracy": 0.12984701097011567,
      "num_tokens": 3278425.0,
      "step": 1890
    },
    {
      "entropy": 6.1114959716796875,
      "epoch": 0.147442131880957,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999454301020741,
      "loss": 6.1128,
      "mean_token_accuracy": 0.12607192546129226,
      "num_tokens": 3288358.0,
      "step": 1895
    },
    {
      "entropy": 6.106830310821533,
      "epoch": 0.14783116125267456,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999448180186133,
      "loss": 6.1045,
      "mean_token_accuracy": 0.12614032998681068,
      "num_tokens": 3297199.0,
      "step": 1900
    },
    {
      "entropy": 6.098498821258545,
      "epoch": 0.14822019062439215,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499944202521973,
      "loss": 6.1129,
      "mean_token_accuracy": 0.12640552148222922,
      "num_tokens": 3306350.0,
      "step": 1905
    },
    {
      "entropy": 6.05791392326355,
      "epoch": 0.1486092199961097,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999435836121624,
      "loss": 5.9412,
      "mean_token_accuracy": 0.13614155799150468,
      "num_tokens": 3314921.0,
      "step": 1910
    },
    {
      "entropy": 5.942939949035645,
      "epoch": 0.14899824936782727,
      "grad_norm": 1.1875,
      "learning_rate": 0.000499942961289191,
      "loss": 6.0897,
      "mean_token_accuracy": 0.13519461452960968,
      "num_tokens": 3323324.0,
      "step": 1915
    },
    {
      "entropy": 6.268859624862671,
      "epoch": 0.14938727873954483,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999423355530683,
      "loss": 6.2168,
      "mean_token_accuracy": 0.12899630516767502,
      "num_tokens": 3331614.0,
      "step": 1920
    },
    {
      "entropy": 6.068661069869995,
      "epoch": 0.1497763081112624,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004999417064038035,
      "loss": 6.048,
      "mean_token_accuracy": 0.13625814467668534,
      "num_tokens": 3339573.0,
      "step": 1925
    },
    {
      "entropy": 6.113537693023682,
      "epoch": 0.15016533748297997,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999410738414065,
      "loss": 6.019,
      "mean_token_accuracy": 0.13822999522089957,
      "num_tokens": 3347775.0,
      "step": 1930
    },
    {
      "entropy": 6.026789045333862,
      "epoch": 0.15055436685469753,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999404378658866,
      "loss": 5.9738,
      "mean_token_accuracy": 0.13702433034777642,
      "num_tokens": 3356947.0,
      "step": 1935
    },
    {
      "entropy": 6.101319646835327,
      "epoch": 0.1509433962264151,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999397984772536,
      "loss": 6.067,
      "mean_token_accuracy": 0.1303373880684376,
      "num_tokens": 3365420.0,
      "step": 1940
    },
    {
      "entropy": 6.0900736331939695,
      "epoch": 0.15133242559813265,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999391556755172,
      "loss": 6.1591,
      "mean_token_accuracy": 0.12977703511714936,
      "num_tokens": 3373483.0,
      "step": 1945
    },
    {
      "entropy": 6.1492267608642575,
      "epoch": 0.15172145496985023,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999385094606872,
      "loss": 6.1473,
      "mean_token_accuracy": 0.12746963351964952,
      "num_tokens": 3382557.0,
      "step": 1950
    },
    {
      "entropy": 6.168496370315552,
      "epoch": 0.1521104843415678,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999378598327733,
      "loss": 6.0838,
      "mean_token_accuracy": 0.12692655250430107,
      "num_tokens": 3392079.0,
      "step": 1955
    },
    {
      "entropy": 6.093343210220337,
      "epoch": 0.15249951371328535,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999372067917854,
      "loss": 6.1531,
      "mean_token_accuracy": 0.13067096173763276,
      "num_tokens": 3401070.0,
      "step": 1960
    },
    {
      "entropy": 6.078389596939087,
      "epoch": 0.1528885430850029,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999365503377335,
      "loss": 6.0271,
      "mean_token_accuracy": 0.1348519966006279,
      "num_tokens": 3409921.0,
      "step": 1965
    },
    {
      "entropy": 6.104602289199829,
      "epoch": 0.1532775724567205,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999358904706273,
      "loss": 6.0616,
      "mean_token_accuracy": 0.1285829097032547,
      "num_tokens": 3419149.0,
      "step": 1970
    },
    {
      "entropy": 6.030079126358032,
      "epoch": 0.15366660182843805,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999352271904771,
      "loss": 5.9462,
      "mean_token_accuracy": 0.13964028134942055,
      "num_tokens": 3427435.0,
      "step": 1975
    },
    {
      "entropy": 6.123502540588379,
      "epoch": 0.1540556312001556,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999345604972928,
      "loss": 6.1541,
      "mean_token_accuracy": 0.12671427205204963,
      "num_tokens": 3436377.0,
      "step": 1980
    },
    {
      "entropy": 6.069276142120361,
      "epoch": 0.15444466057187317,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999338903910846,
      "loss": 6.0279,
      "mean_token_accuracy": 0.13194865062832833,
      "num_tokens": 3444997.0,
      "step": 1985
    },
    {
      "entropy": 6.103115558624268,
      "epoch": 0.15483368994359073,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999332168718626,
      "loss": 6.104,
      "mean_token_accuracy": 0.1315036803483963,
      "num_tokens": 3454254.0,
      "step": 1990
    },
    {
      "entropy": 6.131479167938233,
      "epoch": 0.15522271931530832,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999325399396371,
      "loss": 6.0603,
      "mean_token_accuracy": 0.1318090170621872,
      "num_tokens": 3462853.0,
      "step": 1995
    },
    {
      "entropy": 5.984948539733887,
      "epoch": 0.15561174868702587,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999318595944183,
      "loss": 6.0059,
      "mean_token_accuracy": 0.13085325956344604,
      "num_tokens": 3471288.0,
      "step": 2000
    },
    {
      "entropy": 5.977605199813842,
      "epoch": 0.15600077805874343,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999311758362166,
      "loss": 5.9,
      "mean_token_accuracy": 0.1340898171067238,
      "num_tokens": 3479479.0,
      "step": 2005
    },
    {
      "entropy": 6.000810623168945,
      "epoch": 0.156389807430461,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999304886650422,
      "loss": 5.9227,
      "mean_token_accuracy": 0.13736169710755347,
      "num_tokens": 3488621.0,
      "step": 2010
    },
    {
      "entropy": 6.011513090133667,
      "epoch": 0.15677883680217858,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999297980809058,
      "loss": 5.9912,
      "mean_token_accuracy": 0.1385435238480568,
      "num_tokens": 3497792.0,
      "step": 2015
    },
    {
      "entropy": 5.96446213722229,
      "epoch": 0.15716786617389614,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999291040838176,
      "loss": 5.9932,
      "mean_token_accuracy": 0.13427436128258705,
      "num_tokens": 3506548.0,
      "step": 2020
    },
    {
      "entropy": 6.073589420318603,
      "epoch": 0.1575568955456137,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999284066737884,
      "loss": 5.9874,
      "mean_token_accuracy": 0.14162887558341025,
      "num_tokens": 3515086.0,
      "step": 2025
    },
    {
      "entropy": 6.079417896270752,
      "epoch": 0.15794592491733125,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999277058508285,
      "loss": 6.025,
      "mean_token_accuracy": 0.13054124414920806,
      "num_tokens": 3523985.0,
      "step": 2030
    },
    {
      "entropy": 6.108053159713745,
      "epoch": 0.1583349542890488,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999270016149486,
      "loss": 6.1046,
      "mean_token_accuracy": 0.1336774356663227,
      "num_tokens": 3532575.0,
      "step": 2035
    },
    {
      "entropy": 6.097129058837891,
      "epoch": 0.1587239836607664,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999262939661596,
      "loss": 6.0793,
      "mean_token_accuracy": 0.13377865329384803,
      "num_tokens": 3541558.0,
      "step": 2040
    },
    {
      "entropy": 6.012680435180664,
      "epoch": 0.15911301303248396,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999255829044721,
      "loss": 5.9951,
      "mean_token_accuracy": 0.1400863155722618,
      "num_tokens": 3549697.0,
      "step": 2045
    },
    {
      "entropy": 6.0912693500518795,
      "epoch": 0.15950204240420152,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999248684298968,
      "loss": 6.1595,
      "mean_token_accuracy": 0.12602965235710145,
      "num_tokens": 3558673.0,
      "step": 2050
    },
    {
      "entropy": 6.175625467300415,
      "epoch": 0.15989107177591907,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999241505424447,
      "loss": 6.022,
      "mean_token_accuracy": 0.13359369337558746,
      "num_tokens": 3567176.0,
      "step": 2055
    },
    {
      "entropy": 6.115584897994995,
      "epoch": 0.16028010114763663,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999234292421265,
      "loss": 6.1588,
      "mean_token_accuracy": 0.12792168632149697,
      "num_tokens": 3576020.0,
      "step": 2060
    },
    {
      "entropy": 6.0277955532073975,
      "epoch": 0.16066913051935422,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999227045289535,
      "loss": 5.983,
      "mean_token_accuracy": 0.13384223580360413,
      "num_tokens": 3584705.0,
      "step": 2065
    },
    {
      "entropy": 6.09326057434082,
      "epoch": 0.16105815989107178,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999219764029363,
      "loss": 6.1786,
      "mean_token_accuracy": 0.12971396073698999,
      "num_tokens": 3593339.0,
      "step": 2070
    },
    {
      "entropy": 6.074710321426392,
      "epoch": 0.16144718926278934,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999212448640861,
      "loss": 5.9169,
      "mean_token_accuracy": 0.14326073378324508,
      "num_tokens": 3601206.0,
      "step": 2075
    },
    {
      "entropy": 5.917049407958984,
      "epoch": 0.1618362186345069,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999205099124141,
      "loss": 6.0137,
      "mean_token_accuracy": 0.1296630918979645,
      "num_tokens": 3610371.0,
      "step": 2080
    },
    {
      "entropy": 6.166858434677124,
      "epoch": 0.16222524800622448,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999197715479313,
      "loss": 6.0274,
      "mean_token_accuracy": 0.13334443271160126,
      "num_tokens": 3618834.0,
      "step": 2085
    },
    {
      "entropy": 5.907374048233033,
      "epoch": 0.16261427737794204,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999190297706489,
      "loss": 6.0599,
      "mean_token_accuracy": 0.1330284595489502,
      "num_tokens": 3627426.0,
      "step": 2090
    },
    {
      "entropy": 6.1726648807525635,
      "epoch": 0.1630033067496596,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004999182845805783,
      "loss": 6.0783,
      "mean_token_accuracy": 0.1352132059633732,
      "num_tokens": 3636020.0,
      "step": 2095
    },
    {
      "entropy": 6.094464588165283,
      "epoch": 0.16339233612137716,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999175359777306,
      "loss": 6.0024,
      "mean_token_accuracy": 0.13878497257828712,
      "num_tokens": 3644419.0,
      "step": 2100
    },
    {
      "entropy": 5.983648729324341,
      "epoch": 0.16378136549309472,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999167839621174,
      "loss": 6.1132,
      "mean_token_accuracy": 0.12433986812829971,
      "num_tokens": 3653223.0,
      "step": 2105
    },
    {
      "entropy": 6.119035482406616,
      "epoch": 0.1641703948648123,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999160285337501,
      "loss": 6.1165,
      "mean_token_accuracy": 0.12628517150878907,
      "num_tokens": 3662409.0,
      "step": 2110
    },
    {
      "entropy": 6.122584199905395,
      "epoch": 0.16455942423652986,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004999152696926399,
      "loss": 6.0838,
      "mean_token_accuracy": 0.12654283344745637,
      "num_tokens": 3670635.0,
      "step": 2115
    },
    {
      "entropy": 5.966174030303955,
      "epoch": 0.16494845360824742,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999145074387985,
      "loss": 6.0089,
      "mean_token_accuracy": 0.14118071272969246,
      "num_tokens": 3679303.0,
      "step": 2120
    },
    {
      "entropy": 6.192105531692505,
      "epoch": 0.16533748297996498,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999137417722374,
      "loss": 6.1857,
      "mean_token_accuracy": 0.1251898981630802,
      "num_tokens": 3689299.0,
      "step": 2125
    },
    {
      "entropy": 6.109566402435303,
      "epoch": 0.16572651235168256,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999129726929684,
      "loss": 6.088,
      "mean_token_accuracy": 0.13834489807486533,
      "num_tokens": 3697843.0,
      "step": 2130
    },
    {
      "entropy": 6.1182887077331545,
      "epoch": 0.16611554172340012,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999122002010029,
      "loss": 6.0233,
      "mean_token_accuracy": 0.1364858292043209,
      "num_tokens": 3705876.0,
      "step": 2135
    },
    {
      "entropy": 6.109085321426392,
      "epoch": 0.16650457109511768,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999114242963527,
      "loss": 6.1332,
      "mean_token_accuracy": 0.12421078756451606,
      "num_tokens": 3714835.0,
      "step": 2140
    },
    {
      "entropy": 6.028184270858764,
      "epoch": 0.16689360046683524,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999106449790298,
      "loss": 5.9968,
      "mean_token_accuracy": 0.13307098522782326,
      "num_tokens": 3723421.0,
      "step": 2145
    },
    {
      "entropy": 6.027436256408691,
      "epoch": 0.1672826298385528,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999098622490458,
      "loss": 6.0106,
      "mean_token_accuracy": 0.13246056139469148,
      "num_tokens": 3731590.0,
      "step": 2150
    },
    {
      "entropy": 6.005514812469483,
      "epoch": 0.16767165921027039,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999090761064125,
      "loss": 5.9387,
      "mean_token_accuracy": 0.13851954340934752,
      "num_tokens": 3740700.0,
      "step": 2155
    },
    {
      "entropy": 6.0273407936096195,
      "epoch": 0.16806068858198794,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499908286551142,
      "loss": 6.0128,
      "mean_token_accuracy": 0.1371268466114998,
      "num_tokens": 3749516.0,
      "step": 2160
    },
    {
      "entropy": 6.007265949249268,
      "epoch": 0.1684497179537055,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999074935832463,
      "loss": 5.9506,
      "mean_token_accuracy": 0.14042013585567475,
      "num_tokens": 3757696.0,
      "step": 2165
    },
    {
      "entropy": 6.036279582977295,
      "epoch": 0.16883874732542306,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999066972027373,
      "loss": 6.0279,
      "mean_token_accuracy": 0.13567456975579262,
      "num_tokens": 3765786.0,
      "step": 2170
    },
    {
      "entropy": 5.970781230926514,
      "epoch": 0.16922777669714065,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999058974096271,
      "loss": 5.9077,
      "mean_token_accuracy": 0.1398158609867096,
      "num_tokens": 3773265.0,
      "step": 2175
    },
    {
      "entropy": 6.03927173614502,
      "epoch": 0.1696168060688582,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499905094203928,
      "loss": 6.0607,
      "mean_token_accuracy": 0.13150161504745483,
      "num_tokens": 3782402.0,
      "step": 2180
    },
    {
      "entropy": 6.0491118907928465,
      "epoch": 0.17000583544057576,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499904287585652,
      "loss": 6.1228,
      "mean_token_accuracy": 0.1305118791759014,
      "num_tokens": 3791901.0,
      "step": 2185
    },
    {
      "entropy": 6.073975229263306,
      "epoch": 0.17039486481229332,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999034775548113,
      "loss": 5.9415,
      "mean_token_accuracy": 0.14102055951952935,
      "num_tokens": 3799872.0,
      "step": 2190
    },
    {
      "entropy": 5.967797756195068,
      "epoch": 0.17078389418401088,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999026641114185,
      "loss": 5.8987,
      "mean_token_accuracy": 0.13859765678644181,
      "num_tokens": 3808949.0,
      "step": 2195
    },
    {
      "entropy": 6.072041463851929,
      "epoch": 0.17117292355572847,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999018472554857,
      "loss": 6.0097,
      "mean_token_accuracy": 0.13413021862506866,
      "num_tokens": 3817429.0,
      "step": 2200
    },
    {
      "entropy": 6.208354568481445,
      "epoch": 0.17156195292744603,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999010269870253,
      "loss": 6.2949,
      "mean_token_accuracy": 0.12197455763816833,
      "num_tokens": 3827101.0,
      "step": 2205
    },
    {
      "entropy": 6.107734298706054,
      "epoch": 0.17195098229916359,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999002033060498,
      "loss": 6.0075,
      "mean_token_accuracy": 0.1353926807641983,
      "num_tokens": 3836147.0,
      "step": 2210
    },
    {
      "entropy": 5.9734382152557375,
      "epoch": 0.17234001167088114,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004998993762125716,
      "loss": 6.0336,
      "mean_token_accuracy": 0.13586182668805122,
      "num_tokens": 3845300.0,
      "step": 2215
    },
    {
      "entropy": 6.102244281768799,
      "epoch": 0.17272904104259873,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004998985457066034,
      "loss": 6.0182,
      "mean_token_accuracy": 0.1369253620505333,
      "num_tokens": 3853620.0,
      "step": 2220
    },
    {
      "entropy": 6.035351324081421,
      "epoch": 0.1731180704143163,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004998977117881576,
      "loss": 6.0209,
      "mean_token_accuracy": 0.13868235647678376,
      "num_tokens": 3861734.0,
      "step": 2225
    },
    {
      "entropy": 6.056541538238525,
      "epoch": 0.17350709978603385,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998968744572472,
      "loss": 6.0255,
      "mean_token_accuracy": 0.13340835496783257,
      "num_tokens": 3870753.0,
      "step": 2230
    },
    {
      "entropy": 6.01226224899292,
      "epoch": 0.1738961291577514,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998960337138845,
      "loss": 5.901,
      "mean_token_accuracy": 0.1386076405644417,
      "num_tokens": 3879382.0,
      "step": 2235
    },
    {
      "entropy": 6.042281532287598,
      "epoch": 0.17428515852946896,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998951895580826,
      "loss": 6.0833,
      "mean_token_accuracy": 0.12960439622402192,
      "num_tokens": 3888417.0,
      "step": 2240
    },
    {
      "entropy": 6.019704818725586,
      "epoch": 0.17467418790118655,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998943419898542,
      "loss": 5.9544,
      "mean_token_accuracy": 0.13442030027508736,
      "num_tokens": 3897083.0,
      "step": 2245
    },
    {
      "entropy": 5.953819322586059,
      "epoch": 0.1750632172729041,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998934910092119,
      "loss": 5.9233,
      "mean_token_accuracy": 0.13848119750618934,
      "num_tokens": 3905564.0,
      "step": 2250
    },
    {
      "entropy": 6.029740190505981,
      "epoch": 0.17545224664462167,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998926366161689,
      "loss": 6.0581,
      "mean_token_accuracy": 0.1252748928964138,
      "num_tokens": 3914896.0,
      "step": 2255
    },
    {
      "entropy": 6.078394889831543,
      "epoch": 0.17584127601633923,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004998917788107383,
      "loss": 6.0724,
      "mean_token_accuracy": 0.13376214429736139,
      "num_tokens": 3924151.0,
      "step": 2260
    },
    {
      "entropy": 6.082054328918457,
      "epoch": 0.17623030538805678,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998909175929326,
      "loss": 6.0164,
      "mean_token_accuracy": 0.13435427844524384,
      "num_tokens": 3933163.0,
      "step": 2265
    },
    {
      "entropy": 6.003792142868042,
      "epoch": 0.17661933475977437,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004998900529627653,
      "loss": 5.9297,
      "mean_token_accuracy": 0.13886286765336991,
      "num_tokens": 3941120.0,
      "step": 2270
    },
    {
      "entropy": 5.892064476013184,
      "epoch": 0.17700836413149193,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998891849202495,
      "loss": 5.8359,
      "mean_token_accuracy": 0.1366935320198536,
      "num_tokens": 3949896.0,
      "step": 2275
    },
    {
      "entropy": 6.072416210174561,
      "epoch": 0.1773973935032095,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499888313465398,
      "loss": 5.974,
      "mean_token_accuracy": 0.14090991467237474,
      "num_tokens": 3958794.0,
      "step": 2280
    },
    {
      "entropy": 5.998848390579224,
      "epoch": 0.17778642287492705,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998874385982247,
      "loss": 6.0345,
      "mean_token_accuracy": 0.1342969536781311,
      "num_tokens": 3967267.0,
      "step": 2285
    },
    {
      "entropy": 5.955915451049805,
      "epoch": 0.17817545224664463,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004998865603187421,
      "loss": 5.9559,
      "mean_token_accuracy": 0.1364803656935692,
      "num_tokens": 3976536.0,
      "step": 2290
    },
    {
      "entropy": 6.106266164779663,
      "epoch": 0.1785644816183622,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499885678626964,
      "loss": 6.022,
      "mean_token_accuracy": 0.1321044869720936,
      "num_tokens": 3986006.0,
      "step": 2295
    },
    {
      "entropy": 5.940597105026245,
      "epoch": 0.17895351099007975,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004998847935229038,
      "loss": 5.9638,
      "mean_token_accuracy": 0.13107028752565383,
      "num_tokens": 3994478.0,
      "step": 2300
    },
    {
      "entropy": 6.0922130107879635,
      "epoch": 0.1793425403617973,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998839050065746,
      "loss": 5.9898,
      "mean_token_accuracy": 0.13526349514722824,
      "num_tokens": 4003190.0,
      "step": 2305
    },
    {
      "entropy": 6.025421810150147,
      "epoch": 0.17973156973351487,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998830130779902,
      "loss": 5.9347,
      "mean_token_accuracy": 0.13667116016149522,
      "num_tokens": 4011886.0,
      "step": 2310
    },
    {
      "entropy": 5.891482496261597,
      "epoch": 0.18012059910523245,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499882117737164,
      "loss": 5.8326,
      "mean_token_accuracy": 0.13485828712582587,
      "num_tokens": 4020792.0,
      "step": 2315
    },
    {
      "entropy": 5.931954765319825,
      "epoch": 0.18050962847695,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998812189841096,
      "loss": 5.8219,
      "mean_token_accuracy": 0.14047594219446183,
      "num_tokens": 4029235.0,
      "step": 2320
    },
    {
      "entropy": 5.931398057937622,
      "epoch": 0.18089865784866757,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004998803168188407,
      "loss": 6.0385,
      "mean_token_accuracy": 0.13752583786845207,
      "num_tokens": 4037341.0,
      "step": 2325
    },
    {
      "entropy": 5.9774956703186035,
      "epoch": 0.18128768722038513,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004998794112413708,
      "loss": 5.9435,
      "mean_token_accuracy": 0.13515810221433638,
      "num_tokens": 4046324.0,
      "step": 2330
    },
    {
      "entropy": 5.988140821456909,
      "epoch": 0.18167671659210272,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998785022517137,
      "loss": 5.9935,
      "mean_token_accuracy": 0.13900482952594756,
      "num_tokens": 4054579.0,
      "step": 2335
    },
    {
      "entropy": 6.065921449661255,
      "epoch": 0.18206574596382027,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004998775898498835,
      "loss": 5.9492,
      "mean_token_accuracy": 0.14194765612483023,
      "num_tokens": 4063087.0,
      "step": 2340
    },
    {
      "entropy": 6.072832870483398,
      "epoch": 0.18245477533553783,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004998766740358936,
      "loss": 6.0289,
      "mean_token_accuracy": 0.13890346884727478,
      "num_tokens": 4072309.0,
      "step": 2345
    },
    {
      "entropy": 6.000887250900268,
      "epoch": 0.1828438047072554,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998757548097582,
      "loss": 6.0795,
      "mean_token_accuracy": 0.13404593393206596,
      "num_tokens": 4081017.0,
      "step": 2350
    },
    {
      "entropy": 6.056308937072754,
      "epoch": 0.18323283407897295,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004998748321714911,
      "loss": 6.083,
      "mean_token_accuracy": 0.13330470472574235,
      "num_tokens": 4089799.0,
      "step": 2355
    },
    {
      "entropy": 6.006195831298828,
      "epoch": 0.18362186345069054,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004998739061211065,
      "loss": 6.0072,
      "mean_token_accuracy": 0.1390180394053459,
      "num_tokens": 4097948.0,
      "step": 2360
    },
    {
      "entropy": 6.019151639938355,
      "epoch": 0.1840108928224081,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998729766586181,
      "loss": 5.8641,
      "mean_token_accuracy": 0.13988614454865456,
      "num_tokens": 4107050.0,
      "step": 2365
    },
    {
      "entropy": 5.934751653671265,
      "epoch": 0.18439992219412565,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998720437840403,
      "loss": 5.904,
      "mean_token_accuracy": 0.13505125269293786,
      "num_tokens": 4115547.0,
      "step": 2370
    },
    {
      "entropy": 5.9610343933105465,
      "epoch": 0.1847889515658432,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998711074973871,
      "loss": 5.9015,
      "mean_token_accuracy": 0.13932024538517,
      "num_tokens": 4124047.0,
      "step": 2375
    },
    {
      "entropy": 5.893201923370361,
      "epoch": 0.1851779809375608,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998701677986728,
      "loss": 5.8559,
      "mean_token_accuracy": 0.1403021939098835,
      "num_tokens": 4132005.0,
      "step": 2380
    },
    {
      "entropy": 6.073483228683472,
      "epoch": 0.18556701030927836,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998692246879116,
      "loss": 6.0277,
      "mean_token_accuracy": 0.1299221158027649,
      "num_tokens": 4140199.0,
      "step": 2385
    },
    {
      "entropy": 5.97588005065918,
      "epoch": 0.18595603968099592,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998682781651178,
      "loss": 5.9748,
      "mean_token_accuracy": 0.1365482524037361,
      "num_tokens": 4149353.0,
      "step": 2390
    },
    {
      "entropy": 5.944373273849488,
      "epoch": 0.18634506905271347,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998673282303059,
      "loss": 5.9579,
      "mean_token_accuracy": 0.13471838161349298,
      "num_tokens": 4158531.0,
      "step": 2395
    },
    {
      "entropy": 5.880623435974121,
      "epoch": 0.18673409842443103,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998663748834901,
      "loss": 5.7368,
      "mean_token_accuracy": 0.14892419204115867,
      "num_tokens": 4166743.0,
      "step": 2400
    },
    {
      "entropy": 6.018037462234497,
      "epoch": 0.18712312779614862,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998654181246851,
      "loss": 6.1001,
      "mean_token_accuracy": 0.1294975347816944,
      "num_tokens": 4175416.0,
      "step": 2405
    },
    {
      "entropy": 6.046519470214844,
      "epoch": 0.18751215716786618,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998644579539052,
      "loss": 6.0364,
      "mean_token_accuracy": 0.1348044417798519,
      "num_tokens": 4185324.0,
      "step": 2410
    },
    {
      "entropy": 5.996607494354248,
      "epoch": 0.18790118653958374,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499863494371165,
      "loss": 5.9524,
      "mean_token_accuracy": 0.13784608691930772,
      "num_tokens": 4193361.0,
      "step": 2415
    },
    {
      "entropy": 6.035355854034424,
      "epoch": 0.1882902159113013,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998625273764793,
      "loss": 6.0001,
      "mean_token_accuracy": 0.1302669644355774,
      "num_tokens": 4202553.0,
      "step": 2420
    },
    {
      "entropy": 5.929054594039917,
      "epoch": 0.18867924528301888,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004998615569698626,
      "loss": 5.9479,
      "mean_token_accuracy": 0.1433779552578926,
      "num_tokens": 4211585.0,
      "step": 2425
    },
    {
      "entropy": 6.020910692214966,
      "epoch": 0.18906827465473644,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004998605831513296,
      "loss": 5.98,
      "mean_token_accuracy": 0.13211928457021713,
      "num_tokens": 4220829.0,
      "step": 2430
    },
    {
      "entropy": 5.951228189468384,
      "epoch": 0.189457304026454,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004998596059208953,
      "loss": 5.8953,
      "mean_token_accuracy": 0.13831101283431052,
      "num_tokens": 4228658.0,
      "step": 2435
    },
    {
      "entropy": 5.926197052001953,
      "epoch": 0.18984633339817156,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998586252785743,
      "loss": 5.9772,
      "mean_token_accuracy": 0.14194808825850486,
      "num_tokens": 4237650.0,
      "step": 2440
    },
    {
      "entropy": 6.0081957340240475,
      "epoch": 0.19023536276988912,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998576412243816,
      "loss": 5.9425,
      "mean_token_accuracy": 0.13898246958851815,
      "num_tokens": 4246701.0,
      "step": 2445
    },
    {
      "entropy": 5.968480920791626,
      "epoch": 0.1906243921416067,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004998566537583321,
      "loss": 5.9247,
      "mean_token_accuracy": 0.13750442117452621,
      "num_tokens": 4254541.0,
      "step": 2450
    },
    {
      "entropy": 5.9757208824157715,
      "epoch": 0.19101342151332426,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998556628804408,
      "loss": 5.9488,
      "mean_token_accuracy": 0.13661977127194405,
      "num_tokens": 4263111.0,
      "step": 2455
    },
    {
      "entropy": 5.903095483779907,
      "epoch": 0.19140245088504182,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004998546685907225,
      "loss": 5.98,
      "mean_token_accuracy": 0.13575077131390573,
      "num_tokens": 4272796.0,
      "step": 2460
    },
    {
      "entropy": 5.948573637008667,
      "epoch": 0.19179148025675938,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998536708891927,
      "loss": 5.8919,
      "mean_token_accuracy": 0.1417429715394974,
      "num_tokens": 4281361.0,
      "step": 2465
    },
    {
      "entropy": 6.026061391830444,
      "epoch": 0.19218050962847694,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004998526697758663,
      "loss": 5.9372,
      "mean_token_accuracy": 0.13869260102510453,
      "num_tokens": 4290470.0,
      "step": 2470
    },
    {
      "entropy": 5.973368835449219,
      "epoch": 0.19256953900019452,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998516652507585,
      "loss": 5.9896,
      "mean_token_accuracy": 0.1328210227191448,
      "num_tokens": 4299087.0,
      "step": 2475
    },
    {
      "entropy": 5.942784452438355,
      "epoch": 0.19295856837191208,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004998506573138846,
      "loss": 5.8646,
      "mean_token_accuracy": 0.1429189197719097,
      "num_tokens": 4307572.0,
      "step": 2480
    },
    {
      "entropy": 5.963555192947387,
      "epoch": 0.19334759774362964,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998496459652598,
      "loss": 5.9277,
      "mean_token_accuracy": 0.14059290289878845,
      "num_tokens": 4316337.0,
      "step": 2485
    },
    {
      "entropy": 6.075507640838623,
      "epoch": 0.1937366271153472,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998486312048995,
      "loss": 6.0435,
      "mean_token_accuracy": 0.13950250148773194,
      "num_tokens": 4326258.0,
      "step": 2490
    },
    {
      "entropy": 5.918562126159668,
      "epoch": 0.19412565648706478,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998476130328193,
      "loss": 5.8813,
      "mean_token_accuracy": 0.14137958139181137,
      "num_tokens": 4334711.0,
      "step": 2495
    },
    {
      "entropy": 5.970353269577027,
      "epoch": 0.19451468585878234,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998465914490343,
      "loss": 5.9743,
      "mean_token_accuracy": 0.13537674471735955,
      "num_tokens": 4343739.0,
      "step": 2500
    },
    {
      "entropy": 5.954079532623291,
      "epoch": 0.1949037152304999,
      "grad_norm": 1.046875,
      "learning_rate": 0.00049984556645356,
      "loss": 5.9093,
      "mean_token_accuracy": 0.1404120497405529,
      "num_tokens": 4352905.0,
      "step": 2505
    },
    {
      "entropy": 6.014039754867554,
      "epoch": 0.19529274460221746,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998445380464124,
      "loss": 5.9394,
      "mean_token_accuracy": 0.14536958336830139,
      "num_tokens": 4361333.0,
      "step": 2510
    },
    {
      "entropy": 5.888523912429809,
      "epoch": 0.19568177397393502,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998435062276066,
      "loss": 5.8983,
      "mean_token_accuracy": 0.13909753113985063,
      "num_tokens": 4369872.0,
      "step": 2515
    },
    {
      "entropy": 5.987714862823486,
      "epoch": 0.1960708033456526,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998424709971586,
      "loss": 5.965,
      "mean_token_accuracy": 0.1441866084933281,
      "num_tokens": 4378097.0,
      "step": 2520
    },
    {
      "entropy": 5.949221229553222,
      "epoch": 0.19645983271737016,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998414323550839,
      "loss": 5.8936,
      "mean_token_accuracy": 0.13887374848127365,
      "num_tokens": 4387414.0,
      "step": 2525
    },
    {
      "entropy": 6.000862646102905,
      "epoch": 0.19684886208908772,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004998403903013984,
      "loss": 5.9835,
      "mean_token_accuracy": 0.14411986842751504,
      "num_tokens": 4395922.0,
      "step": 2530
    },
    {
      "entropy": 5.882594108581543,
      "epoch": 0.19723789146080528,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998393448361179,
      "loss": 5.9476,
      "mean_token_accuracy": 0.13559090122580528,
      "num_tokens": 4404500.0,
      "step": 2535
    },
    {
      "entropy": 5.838614749908447,
      "epoch": 0.19762692083252287,
      "grad_norm": 1.1875,
      "learning_rate": 0.000499838295959258,
      "loss": 5.9285,
      "mean_token_accuracy": 0.1404655933380127,
      "num_tokens": 4413433.0,
      "step": 2540
    },
    {
      "entropy": 5.985950040817261,
      "epoch": 0.19801595020424043,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499837243670835,
      "loss": 5.9686,
      "mean_token_accuracy": 0.14125767052173616,
      "num_tokens": 4421789.0,
      "step": 2545
    },
    {
      "entropy": 5.908414649963379,
      "epoch": 0.19840497957595798,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998361879708646,
      "loss": 5.842,
      "mean_token_accuracy": 0.14500181004405022,
      "num_tokens": 4430256.0,
      "step": 2550
    },
    {
      "entropy": 5.934167671203613,
      "epoch": 0.19879400894767554,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499835128859363,
      "loss": 5.8883,
      "mean_token_accuracy": 0.13772125393152237,
      "num_tokens": 4438431.0,
      "step": 2555
    },
    {
      "entropy": 5.989070034027099,
      "epoch": 0.1991830383193931,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998340663363461,
      "loss": 5.9525,
      "mean_token_accuracy": 0.13467848151922227,
      "num_tokens": 4447150.0,
      "step": 2560
    },
    {
      "entropy": 5.908329772949219,
      "epoch": 0.1995720676911107,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004998330004018301,
      "loss": 5.8286,
      "mean_token_accuracy": 0.1427350550889969,
      "num_tokens": 4455181.0,
      "step": 2565
    },
    {
      "entropy": 5.937408065795898,
      "epoch": 0.19996109706282825,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998319310558312,
      "loss": 5.8475,
      "mean_token_accuracy": 0.14098502844572067,
      "num_tokens": 4463361.0,
      "step": 2570
    },
    {
      "entropy": 5.972924375534058,
      "epoch": 0.2003501264345458,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004998308582983657,
      "loss": 5.9067,
      "mean_token_accuracy": 0.14643627852201463,
      "num_tokens": 4471853.0,
      "step": 2575
    },
    {
      "entropy": 5.81297550201416,
      "epoch": 0.20073915580626336,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004998297821294497,
      "loss": 5.8622,
      "mean_token_accuracy": 0.14285662174224853,
      "num_tokens": 4480812.0,
      "step": 2580
    },
    {
      "entropy": 6.017388820648193,
      "epoch": 0.20112818517798095,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998287025490995,
      "loss": 5.8837,
      "mean_token_accuracy": 0.1472306877374649,
      "num_tokens": 4489180.0,
      "step": 2585
    },
    {
      "entropy": 5.9446248531341555,
      "epoch": 0.2015172145496985,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004998276195573317,
      "loss": 5.9769,
      "mean_token_accuracy": 0.13497108072042466,
      "num_tokens": 4498013.0,
      "step": 2590
    },
    {
      "entropy": 6.0001615524292,
      "epoch": 0.20190624392141607,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004998265331541627,
      "loss": 6.0107,
      "mean_token_accuracy": 0.12803680896759034,
      "num_tokens": 4506767.0,
      "step": 2595
    },
    {
      "entropy": 6.023769521713257,
      "epoch": 0.20229527329313363,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998254433396088,
      "loss": 6.0163,
      "mean_token_accuracy": 0.13594405427575112,
      "num_tokens": 4516657.0,
      "step": 2600
    },
    {
      "entropy": 5.886989116668701,
      "epoch": 0.20268430266485118,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004998243501136867,
      "loss": 5.8502,
      "mean_token_accuracy": 0.14113568887114525,
      "num_tokens": 4524980.0,
      "step": 2605
    },
    {
      "entropy": 5.9184175491333,
      "epoch": 0.20307333203656877,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998232534764129,
      "loss": 5.9077,
      "mean_token_accuracy": 0.14252403378486633,
      "num_tokens": 4533236.0,
      "step": 2610
    },
    {
      "entropy": 5.863450002670288,
      "epoch": 0.20346236140828633,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004998221534278041,
      "loss": 5.8162,
      "mean_token_accuracy": 0.14335810765624046,
      "num_tokens": 4541778.0,
      "step": 2615
    },
    {
      "entropy": 5.959815788269043,
      "epoch": 0.2038513907800039,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998210499678769,
      "loss": 5.923,
      "mean_token_accuracy": 0.13540767207741738,
      "num_tokens": 4551131.0,
      "step": 2620
    },
    {
      "entropy": 5.985937166213989,
      "epoch": 0.20424042015172145,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998199430966481,
      "loss": 5.9087,
      "mean_token_accuracy": 0.13660723865032195,
      "num_tokens": 4559530.0,
      "step": 2625
    },
    {
      "entropy": 5.76199860572815,
      "epoch": 0.20462944952343903,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004998188328141346,
      "loss": 5.7991,
      "mean_token_accuracy": 0.14168990179896354,
      "num_tokens": 4567533.0,
      "step": 2630
    },
    {
      "entropy": 6.024087429046631,
      "epoch": 0.2050184788951566,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998177191203531,
      "loss": 5.9456,
      "mean_token_accuracy": 0.13533081263303756,
      "num_tokens": 4576414.0,
      "step": 2635
    },
    {
      "entropy": 5.922002506256104,
      "epoch": 0.20540750826687415,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998166020153204,
      "loss": 5.8799,
      "mean_token_accuracy": 0.1403877317905426,
      "num_tokens": 4585331.0,
      "step": 2640
    },
    {
      "entropy": 5.909115886688232,
      "epoch": 0.2057965376385917,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004998154814990538,
      "loss": 5.9162,
      "mean_token_accuracy": 0.14347948133945465,
      "num_tokens": 4593526.0,
      "step": 2645
    },
    {
      "entropy": 5.876611804962158,
      "epoch": 0.20618556701030927,
      "grad_norm": 1.21875,
      "learning_rate": 0.00049981435757157,
      "loss": 5.9093,
      "mean_token_accuracy": 0.1371060587465763,
      "num_tokens": 4602400.0,
      "step": 2650
    },
    {
      "entropy": 5.864795446395874,
      "epoch": 0.20657459638202685,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004998132302328862,
      "loss": 5.8555,
      "mean_token_accuracy": 0.14304873943328858,
      "num_tokens": 4611037.0,
      "step": 2655
    },
    {
      "entropy": 5.906079483032227,
      "epoch": 0.2069636257537444,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004998120994830195,
      "loss": 5.8705,
      "mean_token_accuracy": 0.1427972808480263,
      "num_tokens": 4619030.0,
      "step": 2660
    },
    {
      "entropy": 5.921322631835937,
      "epoch": 0.20735265512546197,
      "grad_norm": 1.125,
      "learning_rate": 0.000499810965321987,
      "loss": 5.864,
      "mean_token_accuracy": 0.1356546849012375,
      "num_tokens": 4628649.0,
      "step": 2665
    },
    {
      "entropy": 5.863415861129761,
      "epoch": 0.20774168449717953,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004998098277498059,
      "loss": 5.7696,
      "mean_token_accuracy": 0.14580075666308404,
      "num_tokens": 4636935.0,
      "step": 2670
    },
    {
      "entropy": 5.887133026123047,
      "epoch": 0.2081307138688971,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004998086867664934,
      "loss": 5.9707,
      "mean_token_accuracy": 0.13916994631290436,
      "num_tokens": 4645532.0,
      "step": 2675
    },
    {
      "entropy": 6.022394037246704,
      "epoch": 0.20851974324061467,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499807542372067,
      "loss": 6.0362,
      "mean_token_accuracy": 0.13294412568211555,
      "num_tokens": 4654433.0,
      "step": 2680
    },
    {
      "entropy": 6.088439798355102,
      "epoch": 0.20890877261233223,
      "grad_norm": 1.25,
      "learning_rate": 0.0004998063945665439,
      "loss": 6.0405,
      "mean_token_accuracy": 0.1323567308485508,
      "num_tokens": 4662913.0,
      "step": 2685
    },
    {
      "entropy": 5.968340587615967,
      "epoch": 0.2092978019840498,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998052433499416,
      "loss": 6.0029,
      "mean_token_accuracy": 0.13033748045563698,
      "num_tokens": 4672270.0,
      "step": 2690
    },
    {
      "entropy": 6.009377145767212,
      "epoch": 0.20968683135576735,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004998040887222776,
      "loss": 6.045,
      "mean_token_accuracy": 0.1328607678413391,
      "num_tokens": 4681428.0,
      "step": 2695
    },
    {
      "entropy": 5.998253345489502,
      "epoch": 0.21007586072748494,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004998029306835693,
      "loss": 5.8696,
      "mean_token_accuracy": 0.14181804582476615,
      "num_tokens": 4690492.0,
      "step": 2700
    },
    {
      "entropy": 5.976545810699463,
      "epoch": 0.2104648900992025,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998017692338344,
      "loss": 5.9857,
      "mean_token_accuracy": 0.13384222462773324,
      "num_tokens": 4699667.0,
      "step": 2705
    },
    {
      "entropy": 5.75888295173645,
      "epoch": 0.21085391947092005,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004998006043730905,
      "loss": 5.6552,
      "mean_token_accuracy": 0.1563888520002365,
      "num_tokens": 4708529.0,
      "step": 2710
    },
    {
      "entropy": 5.798262882232666,
      "epoch": 0.2112429488426376,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004997994361013551,
      "loss": 5.9154,
      "mean_token_accuracy": 0.13552722483873367,
      "num_tokens": 4717684.0,
      "step": 2715
    },
    {
      "entropy": 5.907504463195801,
      "epoch": 0.21163197821435517,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997982644186461,
      "loss": 5.8459,
      "mean_token_accuracy": 0.14533798694610595,
      "num_tokens": 4727067.0,
      "step": 2720
    },
    {
      "entropy": 5.893569469451904,
      "epoch": 0.21202100758607276,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004997970893249813,
      "loss": 5.8291,
      "mean_token_accuracy": 0.14064672365784645,
      "num_tokens": 4735434.0,
      "step": 2725
    },
    {
      "entropy": 5.894484281539917,
      "epoch": 0.21241003695779032,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004997959108203785,
      "loss": 5.8669,
      "mean_token_accuracy": 0.14163699969649315,
      "num_tokens": 4743843.0,
      "step": 2730
    },
    {
      "entropy": 6.0105119228363035,
      "epoch": 0.21279906632950787,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997947289048554,
      "loss": 5.9142,
      "mean_token_accuracy": 0.1434435725212097,
      "num_tokens": 4753228.0,
      "step": 2735
    },
    {
      "entropy": 5.9337730884552,
      "epoch": 0.21318809570122543,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004997935435784302,
      "loss": 5.9959,
      "mean_token_accuracy": 0.13996100500226022,
      "num_tokens": 4763401.0,
      "step": 2740
    },
    {
      "entropy": 5.934045076370239,
      "epoch": 0.21357712507294302,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004997923548411208,
      "loss": 5.8468,
      "mean_token_accuracy": 0.14104977995157242,
      "num_tokens": 4771812.0,
      "step": 2745
    },
    {
      "entropy": 5.906756114959717,
      "epoch": 0.21396615444466058,
      "grad_norm": 1.15625,
      "learning_rate": 0.000499791162692945,
      "loss": 5.8769,
      "mean_token_accuracy": 0.14148839563131332,
      "num_tokens": 4780616.0,
      "step": 2750
    },
    {
      "entropy": 5.880796670913696,
      "epoch": 0.21435518381637814,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997899671339214,
      "loss": 5.8714,
      "mean_token_accuracy": 0.1407567411661148,
      "num_tokens": 4789886.0,
      "step": 2755
    },
    {
      "entropy": 5.902111768722534,
      "epoch": 0.2147442131880957,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997887681640675,
      "loss": 5.8659,
      "mean_token_accuracy": 0.1388687163591385,
      "num_tokens": 4799189.0,
      "step": 2760
    },
    {
      "entropy": 5.9799223899841305,
      "epoch": 0.21513324255981325,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004997875657834021,
      "loss": 5.9074,
      "mean_token_accuracy": 0.14284486845135688,
      "num_tokens": 4807850.0,
      "step": 2765
    },
    {
      "entropy": 5.920933198928833,
      "epoch": 0.21552227193153084,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004997863599919432,
      "loss": 5.8206,
      "mean_token_accuracy": 0.14280714765191077,
      "num_tokens": 4815742.0,
      "step": 2770
    },
    {
      "entropy": 5.91602783203125,
      "epoch": 0.2159113013032484,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997851507897089,
      "loss": 5.9167,
      "mean_token_accuracy": 0.14030473083257675,
      "num_tokens": 4824761.0,
      "step": 2775
    },
    {
      "entropy": 6.005980825424194,
      "epoch": 0.21630033067496596,
      "grad_norm": 1.25,
      "learning_rate": 0.0004997839381767178,
      "loss": 6.023,
      "mean_token_accuracy": 0.13371331468224526,
      "num_tokens": 4832805.0,
      "step": 2780
    },
    {
      "entropy": 5.939170408248901,
      "epoch": 0.21668936004668352,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997827221529882,
      "loss": 5.8843,
      "mean_token_accuracy": 0.14137587547302247,
      "num_tokens": 4841938.0,
      "step": 2785
    },
    {
      "entropy": 5.92487063407898,
      "epoch": 0.2170783894184011,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997815027185386,
      "loss": 5.9011,
      "mean_token_accuracy": 0.13681731075048448,
      "num_tokens": 4849932.0,
      "step": 2790
    },
    {
      "entropy": 5.821724700927734,
      "epoch": 0.21746741879011866,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004997802798733874,
      "loss": 5.779,
      "mean_token_accuracy": 0.14767844080924988,
      "num_tokens": 4859054.0,
      "step": 2795
    },
    {
      "entropy": 6.00215711593628,
      "epoch": 0.21785644816183622,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004997790536175534,
      "loss": 5.9348,
      "mean_token_accuracy": 0.14400338754057884,
      "num_tokens": 4867415.0,
      "step": 2800
    },
    {
      "entropy": 5.9811138153076175,
      "epoch": 0.21824547753355378,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997778239510548,
      "loss": 5.9799,
      "mean_token_accuracy": 0.1346263289451599,
      "num_tokens": 4875965.0,
      "step": 2805
    },
    {
      "entropy": 5.946578025817871,
      "epoch": 0.21863450690527134,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997765908739106,
      "loss": 5.868,
      "mean_token_accuracy": 0.1421209491789341,
      "num_tokens": 4884629.0,
      "step": 2810
    },
    {
      "entropy": 5.920269346237182,
      "epoch": 0.21902353627698892,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004997753543861395,
      "loss": 5.9236,
      "mean_token_accuracy": 0.14194096624851227,
      "num_tokens": 4893573.0,
      "step": 2815
    },
    {
      "entropy": 5.871851634979248,
      "epoch": 0.21941256564870648,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00049977411448776,
      "loss": 5.8508,
      "mean_token_accuracy": 0.14088700339198112,
      "num_tokens": 4902831.0,
      "step": 2820
    },
    {
      "entropy": 6.002893781661987,
      "epoch": 0.21980159502042404,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997728711787912,
      "loss": 5.9012,
      "mean_token_accuracy": 0.13369439616799356,
      "num_tokens": 4911360.0,
      "step": 2825
    },
    {
      "entropy": 5.933447265625,
      "epoch": 0.2201906243921416,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004997716244592518,
      "loss": 5.856,
      "mean_token_accuracy": 0.1380616083741188,
      "num_tokens": 4919758.0,
      "step": 2830
    },
    {
      "entropy": 5.814820194244385,
      "epoch": 0.22057965376385918,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997703743291607,
      "loss": 5.7524,
      "mean_token_accuracy": 0.15023381486535073,
      "num_tokens": 4928934.0,
      "step": 2835
    },
    {
      "entropy": 5.785764932632446,
      "epoch": 0.22096868313557674,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499769120788537,
      "loss": 5.8125,
      "mean_token_accuracy": 0.1431200847029686,
      "num_tokens": 4937780.0,
      "step": 2840
    },
    {
      "entropy": 6.037427234649658,
      "epoch": 0.2213577125072943,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004997678638373995,
      "loss": 5.9051,
      "mean_token_accuracy": 0.1383948467671871,
      "num_tokens": 4946862.0,
      "step": 2845
    },
    {
      "entropy": 5.868165636062622,
      "epoch": 0.22174674187901186,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997666034757676,
      "loss": 5.887,
      "mean_token_accuracy": 0.14369800984859465,
      "num_tokens": 4954883.0,
      "step": 2850
    },
    {
      "entropy": 5.797883749008179,
      "epoch": 0.22213577125072942,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004997653397036603,
      "loss": 5.6813,
      "mean_token_accuracy": 0.15522562339901924,
      "num_tokens": 4962840.0,
      "step": 2855
    },
    {
      "entropy": 5.835611629486084,
      "epoch": 0.222524800622447,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004997640725210965,
      "loss": 5.9223,
      "mean_token_accuracy": 0.14193681627511978,
      "num_tokens": 4972334.0,
      "step": 2860
    },
    {
      "entropy": 5.9330864429473875,
      "epoch": 0.22291382999416456,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004997628019280958,
      "loss": 5.8871,
      "mean_token_accuracy": 0.14560381472110748,
      "num_tokens": 4981032.0,
      "step": 2865
    },
    {
      "entropy": 5.886129903793335,
      "epoch": 0.22330285936588212,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004997615279246773,
      "loss": 5.914,
      "mean_token_accuracy": 0.14388557896018028,
      "num_tokens": 4990502.0,
      "step": 2870
    },
    {
      "entropy": 6.024803590774536,
      "epoch": 0.22369188873759968,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004997602505108603,
      "loss": 6.0752,
      "mean_token_accuracy": 0.13173774257302284,
      "num_tokens": 4999121.0,
      "step": 2875
    },
    {
      "entropy": 5.972172689437866,
      "epoch": 0.22408091810931724,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997589696866644,
      "loss": 5.8406,
      "mean_token_accuracy": 0.14171420633792878,
      "num_tokens": 5007271.0,
      "step": 2880
    },
    {
      "entropy": 5.839230108261108,
      "epoch": 0.22446994748103483,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004997576854521088,
      "loss": 5.8827,
      "mean_token_accuracy": 0.13617974072694777,
      "num_tokens": 5015940.0,
      "step": 2885
    },
    {
      "entropy": 5.812696170806885,
      "epoch": 0.22485897685275238,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499756397807213,
      "loss": 5.781,
      "mean_token_accuracy": 0.14219485297799112,
      "num_tokens": 5024294.0,
      "step": 2890
    },
    {
      "entropy": 5.876422786712647,
      "epoch": 0.22524800622446994,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004997551067519966,
      "loss": 5.7615,
      "mean_token_accuracy": 0.15088033750653268,
      "num_tokens": 5033401.0,
      "step": 2895
    },
    {
      "entropy": 5.839484453201294,
      "epoch": 0.2256370355961875,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004997538122864792,
      "loss": 5.8552,
      "mean_token_accuracy": 0.1417061373591423,
      "num_tokens": 5041955.0,
      "step": 2900
    },
    {
      "entropy": 5.92848014831543,
      "epoch": 0.2260260649679051,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997525144106804,
      "loss": 5.898,
      "mean_token_accuracy": 0.13416951149702072,
      "num_tokens": 5051132.0,
      "step": 2905
    },
    {
      "entropy": 5.7870166301727295,
      "epoch": 0.22641509433962265,
      "grad_norm": 1.09375,
      "learning_rate": 0.00049975121312462,
      "loss": 5.8046,
      "mean_token_accuracy": 0.14889086335897445,
      "num_tokens": 5059492.0,
      "step": 2910
    },
    {
      "entropy": 5.853955602645874,
      "epoch": 0.2268041237113402,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004997499084283177,
      "loss": 5.8692,
      "mean_token_accuracy": 0.143447807431221,
      "num_tokens": 5069015.0,
      "step": 2915
    },
    {
      "entropy": 5.916986465454102,
      "epoch": 0.22719315308305776,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004997486003217932,
      "loss": 5.8649,
      "mean_token_accuracy": 0.14138873815536498,
      "num_tokens": 5077614.0,
      "step": 2920
    },
    {
      "entropy": 5.828479671478272,
      "epoch": 0.22758218245477532,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997472888050664,
      "loss": 5.8007,
      "mean_token_accuracy": 0.1477051630616188,
      "num_tokens": 5086300.0,
      "step": 2925
    },
    {
      "entropy": 5.84206714630127,
      "epoch": 0.2279712118264929,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004997459738781573,
      "loss": 5.8453,
      "mean_token_accuracy": 0.1453349806368351,
      "num_tokens": 5095967.0,
      "step": 2930
    },
    {
      "entropy": 5.8847919464111325,
      "epoch": 0.22836024119821047,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004997446555410857,
      "loss": 5.8065,
      "mean_token_accuracy": 0.14551942944526672,
      "num_tokens": 5104611.0,
      "step": 2935
    },
    {
      "entropy": 5.818680095672607,
      "epoch": 0.22874927056992803,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997433337938716,
      "loss": 5.7603,
      "mean_token_accuracy": 0.14727237671613694,
      "num_tokens": 5113452.0,
      "step": 2940
    },
    {
      "entropy": 5.7687592029571535,
      "epoch": 0.22913829994164558,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997420086365351,
      "loss": 5.7552,
      "mean_token_accuracy": 0.1463024787604809,
      "num_tokens": 5121558.0,
      "step": 2945
    },
    {
      "entropy": 5.800099563598633,
      "epoch": 0.22952732931336317,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004997406800690966,
      "loss": 5.7291,
      "mean_token_accuracy": 0.158209228515625,
      "num_tokens": 5129942.0,
      "step": 2950
    },
    {
      "entropy": 5.918742609024048,
      "epoch": 0.22991635868508073,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997393480915758,
      "loss": 5.8627,
      "mean_token_accuracy": 0.13658782318234444,
      "num_tokens": 5139012.0,
      "step": 2955
    },
    {
      "entropy": 5.898570966720581,
      "epoch": 0.2303053880567983,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004997380127039931,
      "loss": 5.8623,
      "mean_token_accuracy": 0.1460738256573677,
      "num_tokens": 5148185.0,
      "step": 2960
    },
    {
      "entropy": 5.961733865737915,
      "epoch": 0.23069441742851585,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997366739063688,
      "loss": 5.8949,
      "mean_token_accuracy": 0.1355762153863907,
      "num_tokens": 5157086.0,
      "step": 2965
    },
    {
      "entropy": 5.97607455253601,
      "epoch": 0.2310834468002334,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004997353316987231,
      "loss": 5.9666,
      "mean_token_accuracy": 0.13954457640647888,
      "num_tokens": 5165842.0,
      "step": 2970
    },
    {
      "entropy": 5.937599039077758,
      "epoch": 0.231472476171951,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997339860810765,
      "loss": 5.9787,
      "mean_token_accuracy": 0.1414493590593338,
      "num_tokens": 5174161.0,
      "step": 2975
    },
    {
      "entropy": 5.968914127349853,
      "epoch": 0.23186150554366855,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997326370534495,
      "loss": 5.8987,
      "mean_token_accuracy": 0.1409076064825058,
      "num_tokens": 5183830.0,
      "step": 2980
    },
    {
      "entropy": 5.895692634582519,
      "epoch": 0.2322505349153861,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004997312846158622,
      "loss": 5.8161,
      "mean_token_accuracy": 0.14208230078220369,
      "num_tokens": 5192248.0,
      "step": 2985
    },
    {
      "entropy": 5.817179870605469,
      "epoch": 0.23263956428710367,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997299287683355,
      "loss": 5.7776,
      "mean_token_accuracy": 0.1490553967654705,
      "num_tokens": 5201212.0,
      "step": 2990
    },
    {
      "entropy": 5.853628253936767,
      "epoch": 0.23302859365882125,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997285695108898,
      "loss": 5.8421,
      "mean_token_accuracy": 0.14157011210918427,
      "num_tokens": 5209338.0,
      "step": 2995
    },
    {
      "entropy": 5.974374198913575,
      "epoch": 0.2334176230305388,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997272068435458,
      "loss": 5.8576,
      "mean_token_accuracy": 0.14298054277896882,
      "num_tokens": 5217590.0,
      "step": 3000
    },
    {
      "epoch": 0.2334176230305388,
      "eval_entropy": 5.72153007830417,
      "eval_loss": 5.904752254486084,
      "eval_mean_token_accuracy": 0.14778396588919596,
      "eval_num_tokens": 5217590.0,
      "eval_runtime": 28.7542,
      "eval_samples_per_second": 1445.286,
      "eval_steps_per_second": 180.669,
      "step": 3000
    },
    {
      "entropy": 5.8005329132080075,
      "epoch": 0.23380665240225637,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499725840766324,
      "loss": 5.7915,
      "mean_token_accuracy": 0.14410673528909684,
      "num_tokens": 5226383.0,
      "step": 3005
    },
    {
      "entropy": 5.920779180526734,
      "epoch": 0.23419568177397393,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997244712792453,
      "loss": 5.8243,
      "mean_token_accuracy": 0.1450793609023094,
      "num_tokens": 5235517.0,
      "step": 3010
    },
    {
      "entropy": 5.9037463665008545,
      "epoch": 0.2345847111456915,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997230983823305,
      "loss": 5.799,
      "mean_token_accuracy": 0.15086298882961274,
      "num_tokens": 5243812.0,
      "step": 3015
    },
    {
      "entropy": 5.84418740272522,
      "epoch": 0.23497374051740907,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997217220756003,
      "loss": 5.817,
      "mean_token_accuracy": 0.14579055458307266,
      "num_tokens": 5252399.0,
      "step": 3020
    },
    {
      "entropy": 5.86341438293457,
      "epoch": 0.23536276988912663,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004997203423590757,
      "loss": 5.8496,
      "mean_token_accuracy": 0.14410731792449952,
      "num_tokens": 5260847.0,
      "step": 3025
    },
    {
      "entropy": 5.9321190357208256,
      "epoch": 0.2357517992608442,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004997189592327775,
      "loss": 5.8993,
      "mean_token_accuracy": 0.13984145000576972,
      "num_tokens": 5268847.0,
      "step": 3030
    },
    {
      "entropy": 5.912533950805664,
      "epoch": 0.23614082863256175,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997175726967268,
      "loss": 5.8449,
      "mean_token_accuracy": 0.1432294562458992,
      "num_tokens": 5277989.0,
      "step": 3035
    },
    {
      "entropy": 5.807249355316162,
      "epoch": 0.23652985800427934,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004997161827509447,
      "loss": 5.8298,
      "mean_token_accuracy": 0.14517134800553322,
      "num_tokens": 5286272.0,
      "step": 3040
    },
    {
      "entropy": 5.843175888061523,
      "epoch": 0.2369188873759969,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004997147893954521,
      "loss": 5.8561,
      "mean_token_accuracy": 0.14687449485063553,
      "num_tokens": 5295281.0,
      "step": 3045
    },
    {
      "entropy": 5.927417421340943,
      "epoch": 0.23730791674771445,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004997133926302702,
      "loss": 5.7811,
      "mean_token_accuracy": 0.147725660353899,
      "num_tokens": 5303905.0,
      "step": 3050
    },
    {
      "entropy": 5.8274088382720945,
      "epoch": 0.237696946119432,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004997119924554204,
      "loss": 5.7698,
      "mean_token_accuracy": 0.14279649034142494,
      "num_tokens": 5313146.0,
      "step": 3055
    },
    {
      "entropy": 5.812492322921753,
      "epoch": 0.23808597549114957,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997105888709236,
      "loss": 5.8213,
      "mean_token_accuracy": 0.1478417322039604,
      "num_tokens": 5320838.0,
      "step": 3060
    },
    {
      "entropy": 5.884876251220703,
      "epoch": 0.23847500486286716,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004997091818768015,
      "loss": 5.8272,
      "mean_token_accuracy": 0.14175651222467422,
      "num_tokens": 5329312.0,
      "step": 3065
    },
    {
      "entropy": 5.842220640182495,
      "epoch": 0.23886403423458472,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000499707771473075,
      "loss": 5.8239,
      "mean_token_accuracy": 0.14224815219640732,
      "num_tokens": 5337747.0,
      "step": 3070
    },
    {
      "entropy": 5.890048694610596,
      "epoch": 0.23925306360630227,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997063576597659,
      "loss": 5.82,
      "mean_token_accuracy": 0.13902755007147788,
      "num_tokens": 5346928.0,
      "step": 3075
    },
    {
      "entropy": 5.860515022277832,
      "epoch": 0.23964209297801983,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004997049404368954,
      "loss": 5.8373,
      "mean_token_accuracy": 0.13793924003839492,
      "num_tokens": 5355827.0,
      "step": 3080
    },
    {
      "entropy": 5.991226482391357,
      "epoch": 0.2400311223497374,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499703519804485,
      "loss": 5.9741,
      "mean_token_accuracy": 0.13731370642781257,
      "num_tokens": 5364309.0,
      "step": 3085
    },
    {
      "entropy": 5.884651184082031,
      "epoch": 0.24042015172145498,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004997020957625564,
      "loss": 5.868,
      "mean_token_accuracy": 0.1458408921957016,
      "num_tokens": 5372979.0,
      "step": 3090
    },
    {
      "entropy": 5.857665157318115,
      "epoch": 0.24080918109317254,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997006683111312,
      "loss": 5.8287,
      "mean_token_accuracy": 0.14318791329860686,
      "num_tokens": 5381809.0,
      "step": 3095
    },
    {
      "entropy": 5.925519943237305,
      "epoch": 0.2411982104648901,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000499699237450231,
      "loss": 5.8886,
      "mean_token_accuracy": 0.14159021526575089,
      "num_tokens": 5390036.0,
      "step": 3100
    },
    {
      "entropy": 5.988127946853638,
      "epoch": 0.24158723983660765,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996978031798774,
      "loss": 5.9362,
      "mean_token_accuracy": 0.13715847954154015,
      "num_tokens": 5398739.0,
      "step": 3105
    },
    {
      "entropy": 5.866925144195557,
      "epoch": 0.24197626920832524,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004996963655000924,
      "loss": 5.8508,
      "mean_token_accuracy": 0.13922770768404008,
      "num_tokens": 5407441.0,
      "step": 3110
    },
    {
      "entropy": 5.887317657470703,
      "epoch": 0.2423652985800428,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004996949244108977,
      "loss": 5.8102,
      "mean_token_accuracy": 0.14160098806023597,
      "num_tokens": 5416534.0,
      "step": 3115
    },
    {
      "entropy": 5.842632818222046,
      "epoch": 0.24275432795176036,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499693479912315,
      "loss": 5.8628,
      "mean_token_accuracy": 0.14435580223798752,
      "num_tokens": 5425170.0,
      "step": 3120
    },
    {
      "entropy": 5.8731931209564205,
      "epoch": 0.24314335732347792,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004996920320043666,
      "loss": 5.826,
      "mean_token_accuracy": 0.13852888271212577,
      "num_tokens": 5433721.0,
      "step": 3125
    },
    {
      "entropy": 5.938917779922486,
      "epoch": 0.24353238669519547,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004996905806870742,
      "loss": 5.8817,
      "mean_token_accuracy": 0.1431050829589367,
      "num_tokens": 5441708.0,
      "step": 3130
    },
    {
      "entropy": 5.844237279891968,
      "epoch": 0.24392141606691306,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004996891259604598,
      "loss": 5.7866,
      "mean_token_accuracy": 0.14370601251721382,
      "num_tokens": 5449942.0,
      "step": 3135
    },
    {
      "entropy": 5.880588436126709,
      "epoch": 0.24431044543863062,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004996876678245455,
      "loss": 5.7923,
      "mean_token_accuracy": 0.1475344941020012,
      "num_tokens": 5458004.0,
      "step": 3140
    },
    {
      "entropy": 5.9699300765991214,
      "epoch": 0.24469947481034818,
      "grad_norm": 1.125,
      "learning_rate": 0.0004996862062793536,
      "loss": 5.8779,
      "mean_token_accuracy": 0.1441954031586647,
      "num_tokens": 5467071.0,
      "step": 3145
    },
    {
      "entropy": 5.882994031906128,
      "epoch": 0.24508850418206574,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996847413249061,
      "loss": 5.8592,
      "mean_token_accuracy": 0.14534298777580262,
      "num_tokens": 5476221.0,
      "step": 3150
    },
    {
      "entropy": 5.889163970947266,
      "epoch": 0.24547753355378332,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004996832729612252,
      "loss": 5.7781,
      "mean_token_accuracy": 0.1421756461262703,
      "num_tokens": 5484549.0,
      "step": 3155
    },
    {
      "entropy": 5.770286703109742,
      "epoch": 0.24586656292550088,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004996818011883333,
      "loss": 5.7796,
      "mean_token_accuracy": 0.1462826058268547,
      "num_tokens": 5492909.0,
      "step": 3160
    },
    {
      "entropy": 5.878460454940796,
      "epoch": 0.24625559229721844,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004996803260062527,
      "loss": 5.8447,
      "mean_token_accuracy": 0.1453042894601822,
      "num_tokens": 5501418.0,
      "step": 3165
    },
    {
      "entropy": 5.947468614578247,
      "epoch": 0.246644621668936,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004996788474150057,
      "loss": 5.8269,
      "mean_token_accuracy": 0.14394153654575348,
      "num_tokens": 5510609.0,
      "step": 3170
    },
    {
      "entropy": 5.731346225738525,
      "epoch": 0.24703365104065356,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004996773654146149,
      "loss": 5.6813,
      "mean_token_accuracy": 0.1574917584657669,
      "num_tokens": 5518349.0,
      "step": 3175
    },
    {
      "entropy": 5.728027200698852,
      "epoch": 0.24742268041237114,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004996758800051026,
      "loss": 5.625,
      "mean_token_accuracy": 0.15064212381839753,
      "num_tokens": 5526972.0,
      "step": 3180
    },
    {
      "entropy": 5.777186107635498,
      "epoch": 0.2478117097840887,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004996743911864916,
      "loss": 5.8014,
      "mean_token_accuracy": 0.14768730401992797,
      "num_tokens": 5535683.0,
      "step": 3185
    },
    {
      "entropy": 5.830631589889526,
      "epoch": 0.24820073915580626,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004996728989588041,
      "loss": 5.7698,
      "mean_token_accuracy": 0.15127997547388078,
      "num_tokens": 5544258.0,
      "step": 3190
    },
    {
      "entropy": 5.7605572700500485,
      "epoch": 0.24858976852752382,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004996714033220632,
      "loss": 5.7839,
      "mean_token_accuracy": 0.14324464723467828,
      "num_tokens": 5552608.0,
      "step": 3195
    },
    {
      "entropy": 5.882555913925171,
      "epoch": 0.2489787978992414,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996699042762911,
      "loss": 5.911,
      "mean_token_accuracy": 0.13898452371358871,
      "num_tokens": 5561357.0,
      "step": 3200
    },
    {
      "entropy": 5.895790910720825,
      "epoch": 0.24936782727095896,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004996684018215111,
      "loss": 5.7385,
      "mean_token_accuracy": 0.15563871711492538,
      "num_tokens": 5569862.0,
      "step": 3205
    },
    {
      "entropy": 5.797485113143921,
      "epoch": 0.24975685664267652,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004996668959577453,
      "loss": 5.803,
      "mean_token_accuracy": 0.148903951048851,
      "num_tokens": 5578485.0,
      "step": 3210
    },
    {
      "entropy": 5.792536973953247,
      "epoch": 0.2501458860143941,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996653866850173,
      "loss": 5.7815,
      "mean_token_accuracy": 0.13914278522133827,
      "num_tokens": 5587713.0,
      "step": 3215
    },
    {
      "entropy": 5.876435041427612,
      "epoch": 0.25053491538611167,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004996638740033495,
      "loss": 5.8683,
      "mean_token_accuracy": 0.14536678940057754,
      "num_tokens": 5596541.0,
      "step": 3220
    },
    {
      "entropy": 5.836820125579834,
      "epoch": 0.2509239447578292,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996623579127651,
      "loss": 5.7374,
      "mean_token_accuracy": 0.1486624523997307,
      "num_tokens": 5605573.0,
      "step": 3225
    },
    {
      "entropy": 5.815556573867798,
      "epoch": 0.2513129741295468,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004996608384132868,
      "loss": 5.7302,
      "mean_token_accuracy": 0.15229415744543076,
      "num_tokens": 5614016.0,
      "step": 3230
    },
    {
      "entropy": 5.929158639907837,
      "epoch": 0.25170200350126437,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499659315504938,
      "loss": 5.9714,
      "mean_token_accuracy": 0.1410625085234642,
      "num_tokens": 5622451.0,
      "step": 3235
    },
    {
      "entropy": 5.871538591384888,
      "epoch": 0.2520910328729819,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004996577891877417,
      "loss": 5.8263,
      "mean_token_accuracy": 0.1432577043771744,
      "num_tokens": 5631384.0,
      "step": 3240
    },
    {
      "entropy": 5.845392656326294,
      "epoch": 0.2524800622446995,
      "grad_norm": 1.125,
      "learning_rate": 0.0004996562594617209,
      "loss": 5.787,
      "mean_token_accuracy": 0.14552973359823226,
      "num_tokens": 5640726.0,
      "step": 3245
    },
    {
      "entropy": 5.880887746810913,
      "epoch": 0.252869091616417,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004996547263268991,
      "loss": 5.7875,
      "mean_token_accuracy": 0.14155513122677804,
      "num_tokens": 5648948.0,
      "step": 3250
    },
    {
      "entropy": 5.727876091003418,
      "epoch": 0.2532581209881346,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004996531897832993,
      "loss": 5.7484,
      "mean_token_accuracy": 0.14809300750494003,
      "num_tokens": 5657951.0,
      "step": 3255
    },
    {
      "entropy": 5.896551561355591,
      "epoch": 0.2536471503598522,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499651649830945,
      "loss": 5.8633,
      "mean_token_accuracy": 0.14910533055663108,
      "num_tokens": 5667476.0,
      "step": 3260
    },
    {
      "entropy": 5.904336595535279,
      "epoch": 0.2540361797315697,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996501064698593,
      "loss": 5.8708,
      "mean_token_accuracy": 0.14054447636008263,
      "num_tokens": 5676627.0,
      "step": 3265
    },
    {
      "entropy": 5.7955468654632565,
      "epoch": 0.2544252091032873,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000499648559700066,
      "loss": 5.7879,
      "mean_token_accuracy": 0.15155547559261323,
      "num_tokens": 5684376.0,
      "step": 3270
    },
    {
      "entropy": 5.905218029022217,
      "epoch": 0.25481423847500484,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004996470095215884,
      "loss": 5.9569,
      "mean_token_accuracy": 0.133226178586483,
      "num_tokens": 5693629.0,
      "step": 3275
    },
    {
      "entropy": 5.966583728790283,
      "epoch": 0.2552032678467224,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004996454559344498,
      "loss": 5.8397,
      "mean_token_accuracy": 0.14316044747829437,
      "num_tokens": 5702010.0,
      "step": 3280
    },
    {
      "entropy": 5.7592079639434814,
      "epoch": 0.25559229721844,
      "grad_norm": 1.25,
      "learning_rate": 0.0004996438989386741,
      "loss": 5.7497,
      "mean_token_accuracy": 0.14288737773895263,
      "num_tokens": 5710823.0,
      "step": 3285
    },
    {
      "entropy": 5.851301670074463,
      "epoch": 0.25598132659015754,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004996423385342847,
      "loss": 5.7971,
      "mean_token_accuracy": 0.15071944296360015,
      "num_tokens": 5719389.0,
      "step": 3290
    },
    {
      "entropy": 5.921173000335694,
      "epoch": 0.25637035596187513,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004996407747213055,
      "loss": 5.7838,
      "mean_token_accuracy": 0.14567249342799188,
      "num_tokens": 5727327.0,
      "step": 3295
    },
    {
      "entropy": 5.809784698486328,
      "epoch": 0.25675938533359266,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00049963920749976,
      "loss": 5.6969,
      "mean_token_accuracy": 0.15062555074691772,
      "num_tokens": 5735886.0,
      "step": 3300
    },
    {
      "entropy": 5.797888612747192,
      "epoch": 0.25714841470531025,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004996376368696721,
      "loss": 5.7431,
      "mean_token_accuracy": 0.1468089461326599,
      "num_tokens": 5744656.0,
      "step": 3305
    },
    {
      "entropy": 5.88851490020752,
      "epoch": 0.25753744407702783,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004996360628310656,
      "loss": 5.9045,
      "mean_token_accuracy": 0.13661069944500923,
      "num_tokens": 5753026.0,
      "step": 3310
    },
    {
      "entropy": 5.876181364059448,
      "epoch": 0.25792647344874536,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004996344853839644,
      "loss": 5.8452,
      "mean_token_accuracy": 0.14359349012374878,
      "num_tokens": 5762771.0,
      "step": 3315
    },
    {
      "entropy": 5.826467847824096,
      "epoch": 0.25831550282046295,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004996329045283924,
      "loss": 5.8225,
      "mean_token_accuracy": 0.15036021620035173,
      "num_tokens": 5771151.0,
      "step": 3320
    },
    {
      "entropy": 5.829469013214111,
      "epoch": 0.25870453219218054,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004996313202643737,
      "loss": 5.7301,
      "mean_token_accuracy": 0.14670131206512452,
      "num_tokens": 5780648.0,
      "step": 3325
    },
    {
      "entropy": 5.801216411590576,
      "epoch": 0.25909356156389807,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004996297325919321,
      "loss": 5.7089,
      "mean_token_accuracy": 0.14597562104463577,
      "num_tokens": 5789698.0,
      "step": 3330
    },
    {
      "entropy": 5.902236461639404,
      "epoch": 0.25948259093561565,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004996281415110919,
      "loss": 5.9083,
      "mean_token_accuracy": 0.13697285056114197,
      "num_tokens": 5799218.0,
      "step": 3335
    },
    {
      "entropy": 5.800975608825683,
      "epoch": 0.2598716203073332,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996265470218773,
      "loss": 5.7761,
      "mean_token_accuracy": 0.14101554378867148,
      "num_tokens": 5808455.0,
      "step": 3340
    },
    {
      "entropy": 5.86399564743042,
      "epoch": 0.26026064967905077,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004996249491243122,
      "loss": 5.8026,
      "mean_token_accuracy": 0.14507102519273757,
      "num_tokens": 5816464.0,
      "step": 3345
    },
    {
      "entropy": 5.831243515014648,
      "epoch": 0.26064967905076836,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004996233478184211,
      "loss": 5.7665,
      "mean_token_accuracy": 0.1419643133878708,
      "num_tokens": 5825188.0,
      "step": 3350
    },
    {
      "entropy": 5.816180896759033,
      "epoch": 0.2610387084224859,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004996217431042282,
      "loss": 5.8148,
      "mean_token_accuracy": 0.14716667979955672,
      "num_tokens": 5834150.0,
      "step": 3355
    },
    {
      "entropy": 5.8475151538848875,
      "epoch": 0.2614277377942035,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000499620134981758,
      "loss": 5.6758,
      "mean_token_accuracy": 0.15172670781612396,
      "num_tokens": 5842801.0,
      "step": 3360
    },
    {
      "entropy": 5.8005210876464846,
      "epoch": 0.261816767165921,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004996185234510346,
      "loss": 5.7071,
      "mean_token_accuracy": 0.15291839390993117,
      "num_tokens": 5851547.0,
      "step": 3365
    },
    {
      "entropy": 5.862095355987549,
      "epoch": 0.2622057965376386,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004996169085120828,
      "loss": 5.8569,
      "mean_token_accuracy": 0.1441195249557495,
      "num_tokens": 5860430.0,
      "step": 3370
    },
    {
      "entropy": 5.778133964538574,
      "epoch": 0.2625948259093562,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004996152901649268,
      "loss": 5.7645,
      "mean_token_accuracy": 0.1446751818060875,
      "num_tokens": 5869648.0,
      "step": 3375
    },
    {
      "entropy": 5.897901725769043,
      "epoch": 0.2629838552810737,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004996136684095913,
      "loss": 5.7754,
      "mean_token_accuracy": 0.14787262827157974,
      "num_tokens": 5878606.0,
      "step": 3380
    },
    {
      "entropy": 5.817872047424316,
      "epoch": 0.2633728846527913,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004996120432461009,
      "loss": 5.8403,
      "mean_token_accuracy": 0.1442799985408783,
      "num_tokens": 5886772.0,
      "step": 3385
    },
    {
      "entropy": 5.80500373840332,
      "epoch": 0.2637619140245088,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004996104146744804,
      "loss": 5.7437,
      "mean_token_accuracy": 0.1436271235346794,
      "num_tokens": 5895007.0,
      "step": 3390
    },
    {
      "entropy": 5.937970876693726,
      "epoch": 0.2641509433962264,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004996087826947541,
      "loss": 5.8675,
      "mean_token_accuracy": 0.13643872514367103,
      "num_tokens": 5904249.0,
      "step": 3395
    },
    {
      "entropy": 5.748004293441772,
      "epoch": 0.264539972767944,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004996071473069471,
      "loss": 5.8357,
      "mean_token_accuracy": 0.1466163992881775,
      "num_tokens": 5911416.0,
      "step": 3400
    },
    {
      "entropy": 5.873762226104736,
      "epoch": 0.26492900213966153,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004996055085110842,
      "loss": 5.875,
      "mean_token_accuracy": 0.14148805439472198,
      "num_tokens": 5920370.0,
      "step": 3405
    },
    {
      "entropy": 5.922806882858277,
      "epoch": 0.2653180315113791,
      "grad_norm": 1.25,
      "learning_rate": 0.0004996038663071902,
      "loss": 5.8398,
      "mean_token_accuracy": 0.14385833963751793,
      "num_tokens": 5928884.0,
      "step": 3410
    },
    {
      "entropy": 5.857811307907104,
      "epoch": 0.26570706088309665,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004996022206952901,
      "loss": 5.857,
      "mean_token_accuracy": 0.13694723397493364,
      "num_tokens": 5938456.0,
      "step": 3415
    },
    {
      "entropy": 5.915801191329956,
      "epoch": 0.26609609025481423,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004996005716754086,
      "loss": 5.8785,
      "mean_token_accuracy": 0.14502357617020606,
      "num_tokens": 5947157.0,
      "step": 3420
    },
    {
      "entropy": 5.856646108627319,
      "epoch": 0.2664851196265318,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000499598919247571,
      "loss": 5.7374,
      "mean_token_accuracy": 0.15310335159301758,
      "num_tokens": 5956084.0,
      "step": 3425
    },
    {
      "entropy": 5.784454250335694,
      "epoch": 0.26687414899824935,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004995972634118023,
      "loss": 5.7743,
      "mean_token_accuracy": 0.14791329205036163,
      "num_tokens": 5964782.0,
      "step": 3430
    },
    {
      "entropy": 5.885459470748901,
      "epoch": 0.26726317836996694,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004995956041681277,
      "loss": 5.9786,
      "mean_token_accuracy": 0.13400535508990288,
      "num_tokens": 5974013.0,
      "step": 3435
    },
    {
      "entropy": 5.869965076446533,
      "epoch": 0.2676522077416845,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004995939415165722,
      "loss": 5.85,
      "mean_token_accuracy": 0.14296992719173432,
      "num_tokens": 5983351.0,
      "step": 3440
    },
    {
      "entropy": 5.8040210723876955,
      "epoch": 0.26804123711340205,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004995922754571611,
      "loss": 5.6754,
      "mean_token_accuracy": 0.15375011265277863,
      "num_tokens": 5992084.0,
      "step": 3445
    },
    {
      "entropy": 5.711496829986572,
      "epoch": 0.26843026648511964,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995906059899197,
      "loss": 5.7014,
      "mean_token_accuracy": 0.1504288472235203,
      "num_tokens": 6001043.0,
      "step": 3450
    },
    {
      "entropy": 5.7762144088745115,
      "epoch": 0.26881929585683717,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004995889331148733,
      "loss": 5.7247,
      "mean_token_accuracy": 0.14711771607398988,
      "num_tokens": 6009620.0,
      "step": 3455
    },
    {
      "entropy": 5.789427089691162,
      "epoch": 0.26920832522855476,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004995872568320474,
      "loss": 5.7209,
      "mean_token_accuracy": 0.1434781163930893,
      "num_tokens": 6018278.0,
      "step": 3460
    },
    {
      "entropy": 5.775634527206421,
      "epoch": 0.26959735460027234,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004995855771414673,
      "loss": 5.7694,
      "mean_token_accuracy": 0.14877430200576783,
      "num_tokens": 6026741.0,
      "step": 3465
    },
    {
      "entropy": 5.8912004947662355,
      "epoch": 0.2699863839719899,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004995838940431585,
      "loss": 5.7922,
      "mean_token_accuracy": 0.14324328750371934,
      "num_tokens": 6035370.0,
      "step": 3470
    },
    {
      "entropy": 5.831309652328491,
      "epoch": 0.27037541334370746,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004995822075371466,
      "loss": 5.7808,
      "mean_token_accuracy": 0.1441688671708107,
      "num_tokens": 6044372.0,
      "step": 3475
    },
    {
      "entropy": 5.8337959289550785,
      "epoch": 0.270764442715425,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004995805176234571,
      "loss": 5.7907,
      "mean_token_accuracy": 0.1507210314273834,
      "num_tokens": 6052940.0,
      "step": 3480
    },
    {
      "entropy": 5.863040447235107,
      "epoch": 0.2711534720871426,
      "grad_norm": 1.25,
      "learning_rate": 0.0004995788243021158,
      "loss": 5.7137,
      "mean_token_accuracy": 0.1456463798880577,
      "num_tokens": 6062048.0,
      "step": 3485
    },
    {
      "entropy": 5.742090702056885,
      "epoch": 0.27154250145886016,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004995771275731483,
      "loss": 5.7422,
      "mean_token_accuracy": 0.14996225386857986,
      "num_tokens": 6071478.0,
      "step": 3490
    },
    {
      "entropy": 5.718463277816772,
      "epoch": 0.2719315308305777,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004995754274365802,
      "loss": 5.6752,
      "mean_token_accuracy": 0.15684566050767898,
      "num_tokens": 6080495.0,
      "step": 3495
    },
    {
      "entropy": 5.880794525146484,
      "epoch": 0.2723205602022953,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004995737238924376,
      "loss": 5.8302,
      "mean_token_accuracy": 0.13611356019973755,
      "num_tokens": 6090141.0,
      "step": 3500
    },
    {
      "entropy": 5.855888843536377,
      "epoch": 0.2727095895740128,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004995720169407461,
      "loss": 5.7909,
      "mean_token_accuracy": 0.14242940098047258,
      "num_tokens": 6099215.0,
      "step": 3505
    },
    {
      "entropy": 5.864683771133423,
      "epoch": 0.2730986189457304,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004995703065815317,
      "loss": 5.8744,
      "mean_token_accuracy": 0.13853381276130677,
      "num_tokens": 6108802.0,
      "step": 3510
    },
    {
      "entropy": 5.982160949707032,
      "epoch": 0.273487648317448,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004995685928148203,
      "loss": 5.8708,
      "mean_token_accuracy": 0.14367551431059838,
      "num_tokens": 6117486.0,
      "step": 3515
    },
    {
      "entropy": 5.751517820358276,
      "epoch": 0.2738766776891655,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000499566875640638,
      "loss": 5.691,
      "mean_token_accuracy": 0.15760284960269927,
      "num_tokens": 6125248.0,
      "step": 3520
    },
    {
      "entropy": 5.816810512542725,
      "epoch": 0.2742657070608831,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004995651550590108,
      "loss": 5.7458,
      "mean_token_accuracy": 0.14610794484615325,
      "num_tokens": 6134831.0,
      "step": 3525
    },
    {
      "entropy": 5.793434810638428,
      "epoch": 0.2746547364326007,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004995634310699647,
      "loss": 5.8221,
      "mean_token_accuracy": 0.1443501353263855,
      "num_tokens": 6143248.0,
      "step": 3530
    },
    {
      "entropy": 5.8795264720916744,
      "epoch": 0.2750437658043182,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004995617036735261,
      "loss": 5.829,
      "mean_token_accuracy": 0.14278091713786126,
      "num_tokens": 6151768.0,
      "step": 3535
    },
    {
      "entropy": 5.806972646713257,
      "epoch": 0.2754327951760358,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004995599728697211,
      "loss": 5.716,
      "mean_token_accuracy": 0.15529972016811372,
      "num_tokens": 6160091.0,
      "step": 3540
    },
    {
      "entropy": 5.80297966003418,
      "epoch": 0.27582182454775334,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004995582386585759,
      "loss": 5.8261,
      "mean_token_accuracy": 0.1379936493933201,
      "num_tokens": 6169440.0,
      "step": 3545
    },
    {
      "entropy": 5.845169639587402,
      "epoch": 0.2762108539194709,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004995565010401167,
      "loss": 5.7921,
      "mean_token_accuracy": 0.14506490528583527,
      "num_tokens": 6177604.0,
      "step": 3550
    },
    {
      "entropy": 5.884006500244141,
      "epoch": 0.2765998832911885,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004995547600143702,
      "loss": 5.8005,
      "mean_token_accuracy": 0.13861700221896173,
      "num_tokens": 6186252.0,
      "step": 3555
    },
    {
      "entropy": 5.80251636505127,
      "epoch": 0.27698891266290604,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004995530155813625,
      "loss": 5.7424,
      "mean_token_accuracy": 0.1510542720556259,
      "num_tokens": 6194641.0,
      "step": 3560
    },
    {
      "entropy": 5.8658629894256595,
      "epoch": 0.2773779420346236,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004995512677411203,
      "loss": 5.8756,
      "mean_token_accuracy": 0.14317311719059944,
      "num_tokens": 6203925.0,
      "step": 3565
    },
    {
      "entropy": 5.895973205566406,
      "epoch": 0.27776697140634116,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004995495164936698,
      "loss": 5.7566,
      "mean_token_accuracy": 0.14684337750077248,
      "num_tokens": 6213001.0,
      "step": 3570
    },
    {
      "entropy": 5.795819807052612,
      "epoch": 0.27815600077805874,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995477618390381,
      "loss": 5.7686,
      "mean_token_accuracy": 0.1391209714114666,
      "num_tokens": 6221754.0,
      "step": 3575
    },
    {
      "entropy": 5.763400554656982,
      "epoch": 0.27854503014977633,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004995460037772513,
      "loss": 5.6719,
      "mean_token_accuracy": 0.14914391040802003,
      "num_tokens": 6229792.0,
      "step": 3580
    },
    {
      "entropy": 5.773159742355347,
      "epoch": 0.27893405952149386,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004995442423083365,
      "loss": 5.7772,
      "mean_token_accuracy": 0.14420934915542602,
      "num_tokens": 6238839.0,
      "step": 3585
    },
    {
      "entropy": 5.84812970161438,
      "epoch": 0.27932308889321145,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004995424774323201,
      "loss": 5.6807,
      "mean_token_accuracy": 0.14983972907066345,
      "num_tokens": 6246939.0,
      "step": 3590
    },
    {
      "entropy": 5.830242109298706,
      "epoch": 0.279712118264929,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499540709149229,
      "loss": 5.8123,
      "mean_token_accuracy": 0.1413383439183235,
      "num_tokens": 6255858.0,
      "step": 3595
    },
    {
      "entropy": 5.772054052352905,
      "epoch": 0.28010114763664656,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004995389374590901,
      "loss": 5.7289,
      "mean_token_accuracy": 0.14664228558540343,
      "num_tokens": 6265137.0,
      "step": 3600
    },
    {
      "entropy": 5.795121526718139,
      "epoch": 0.28049017700836415,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004995371623619301,
      "loss": 5.727,
      "mean_token_accuracy": 0.14856354743242264,
      "num_tokens": 6273286.0,
      "step": 3605
    },
    {
      "entropy": 5.699319839477539,
      "epoch": 0.2808792063800817,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000499535383857776,
      "loss": 5.7434,
      "mean_token_accuracy": 0.1431848406791687,
      "num_tokens": 6282477.0,
      "step": 3610
    },
    {
      "entropy": 5.889150619506836,
      "epoch": 0.28126823575179927,
      "grad_norm": 1.1875,
      "learning_rate": 0.000499533601946655,
      "loss": 5.7949,
      "mean_token_accuracy": 0.14193343818187715,
      "num_tokens": 6291756.0,
      "step": 3615
    },
    {
      "entropy": 5.8616869926452635,
      "epoch": 0.2816572651235168,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004995318166285938,
      "loss": 5.8663,
      "mean_token_accuracy": 0.13642492145299911,
      "num_tokens": 6301093.0,
      "step": 3620
    },
    {
      "entropy": 5.8186732769012455,
      "epoch": 0.2820462944952344,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004995300279036199,
      "loss": 5.6829,
      "mean_token_accuracy": 0.15127351135015488,
      "num_tokens": 6310379.0,
      "step": 3625
    },
    {
      "entropy": 5.775457906723022,
      "epoch": 0.28243532386695197,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00049952823577176,
      "loss": 5.7066,
      "mean_token_accuracy": 0.1457876168191433,
      "num_tokens": 6318983.0,
      "step": 3630
    },
    {
      "entropy": 5.753813028335571,
      "epoch": 0.2828243532386695,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004995264402330416,
      "loss": 5.7176,
      "mean_token_accuracy": 0.15276553630828857,
      "num_tokens": 6327816.0,
      "step": 3635
    },
    {
      "entropy": 5.794705772399903,
      "epoch": 0.2832133826103871,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004995246412874919,
      "loss": 5.8038,
      "mean_token_accuracy": 0.14400455206632615,
      "num_tokens": 6336693.0,
      "step": 3640
    },
    {
      "entropy": 5.853319835662842,
      "epoch": 0.2836024119821047,
      "grad_norm": 1.203125,
      "learning_rate": 0.000499522838935138,
      "loss": 5.7451,
      "mean_token_accuracy": 0.14893249571323394,
      "num_tokens": 6345456.0,
      "step": 3645
    },
    {
      "entropy": 5.794118928909302,
      "epoch": 0.2839914413538222,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004995210331760074,
      "loss": 5.7217,
      "mean_token_accuracy": 0.15264002978801727,
      "num_tokens": 6353584.0,
      "step": 3650
    },
    {
      "entropy": 5.727006578445435,
      "epoch": 0.2843804707255398,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004995192240101275,
      "loss": 5.7469,
      "mean_token_accuracy": 0.15091292411088944,
      "num_tokens": 6362258.0,
      "step": 3655
    },
    {
      "entropy": 5.738786602020264,
      "epoch": 0.2847695000972573,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004995174114375258,
      "loss": 5.6761,
      "mean_token_accuracy": 0.15246595144271852,
      "num_tokens": 6371373.0,
      "step": 3660
    },
    {
      "entropy": 5.864908266067505,
      "epoch": 0.2851585294689749,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004995155954582297,
      "loss": 5.7375,
      "mean_token_accuracy": 0.15102535858750343,
      "num_tokens": 6380431.0,
      "step": 3665
    },
    {
      "entropy": 5.861855220794678,
      "epoch": 0.2855475588406925,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004995137760722668,
      "loss": 5.9038,
      "mean_token_accuracy": 0.1457961119711399,
      "num_tokens": 6389305.0,
      "step": 3670
    },
    {
      "entropy": 5.851635265350342,
      "epoch": 0.28593658821241,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004995119532796646,
      "loss": 5.7213,
      "mean_token_accuracy": 0.15150809288024902,
      "num_tokens": 6397065.0,
      "step": 3675
    },
    {
      "entropy": 5.806237506866455,
      "epoch": 0.2863256175841276,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000499510127080451,
      "loss": 5.8472,
      "mean_token_accuracy": 0.13979469537734984,
      "num_tokens": 6407053.0,
      "step": 3680
    },
    {
      "entropy": 5.867288637161255,
      "epoch": 0.28671464695584514,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004995082974746535,
      "loss": 5.8145,
      "mean_token_accuracy": 0.1415085345506668,
      "num_tokens": 6415165.0,
      "step": 3685
    },
    {
      "entropy": 5.869475412368774,
      "epoch": 0.28710367632756273,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004995064644622999,
      "loss": 5.795,
      "mean_token_accuracy": 0.1459844619035721,
      "num_tokens": 6423646.0,
      "step": 3690
    },
    {
      "entropy": 5.868303823471069,
      "epoch": 0.2874927056992803,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004995046280434181,
      "loss": 5.8131,
      "mean_token_accuracy": 0.1465369090437889,
      "num_tokens": 6432134.0,
      "step": 3695
    },
    {
      "entropy": 5.818410873413086,
      "epoch": 0.28788173507099785,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004995027882180358,
      "loss": 5.7027,
      "mean_token_accuracy": 0.1474486768245697,
      "num_tokens": 6440456.0,
      "step": 3700
    },
    {
      "entropy": 5.809641408920288,
      "epoch": 0.28827076444271543,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004995009449861812,
      "loss": 5.7866,
      "mean_token_accuracy": 0.13990482091903686,
      "num_tokens": 6448952.0,
      "step": 3705
    },
    {
      "entropy": 5.804608154296875,
      "epoch": 0.28865979381443296,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000499499098347882,
      "loss": 5.6512,
      "mean_token_accuracy": 0.16037264764308928,
      "num_tokens": 6457225.0,
      "step": 3710
    },
    {
      "entropy": 5.810186100006104,
      "epoch": 0.28904882318615055,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004994972483031662,
      "loss": 5.7833,
      "mean_token_accuracy": 0.1467502385377884,
      "num_tokens": 6465450.0,
      "step": 3715
    },
    {
      "entropy": 5.810656881332397,
      "epoch": 0.28943785255786814,
      "grad_norm": 1.21875,
      "learning_rate": 0.000499495394852062,
      "loss": 5.8163,
      "mean_token_accuracy": 0.1399491086602211,
      "num_tokens": 6473736.0,
      "step": 3720
    },
    {
      "entropy": 5.83799033164978,
      "epoch": 0.28982688192958567,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004994935379945977,
      "loss": 5.7577,
      "mean_token_accuracy": 0.1468055784702301,
      "num_tokens": 6482092.0,
      "step": 3725
    },
    {
      "entropy": 5.8806088924407955,
      "epoch": 0.29021591130130325,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004994916777308012,
      "loss": 5.7831,
      "mean_token_accuracy": 0.14831546545028687,
      "num_tokens": 6490704.0,
      "step": 3730
    },
    {
      "entropy": 5.850636386871338,
      "epoch": 0.29060494067302084,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004994898140607007,
      "loss": 5.7201,
      "mean_token_accuracy": 0.15038958936929703,
      "num_tokens": 6499634.0,
      "step": 3735
    },
    {
      "entropy": 5.743886566162109,
      "epoch": 0.29099397004473837,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004994879469843247,
      "loss": 5.8767,
      "mean_token_accuracy": 0.13921897560358049,
      "num_tokens": 6509261.0,
      "step": 3740
    },
    {
      "entropy": 5.8543627738952635,
      "epoch": 0.29138299941645596,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004994860765017013,
      "loss": 5.7098,
      "mean_token_accuracy": 0.14895353764295577,
      "num_tokens": 6517542.0,
      "step": 3745
    },
    {
      "entropy": 5.767716026306152,
      "epoch": 0.2917720287881735,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499484202612859,
      "loss": 5.7096,
      "mean_token_accuracy": 0.14601053446531295,
      "num_tokens": 6525945.0,
      "step": 3750
    },
    {
      "entropy": 5.740671157836914,
      "epoch": 0.2921610581598911,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004994823253178264,
      "loss": 5.7276,
      "mean_token_accuracy": 0.1483728989958763,
      "num_tokens": 6534744.0,
      "step": 3755
    },
    {
      "entropy": 5.7915150165557865,
      "epoch": 0.29255008753160866,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004994804446166317,
      "loss": 5.6894,
      "mean_token_accuracy": 0.15682424753904342,
      "num_tokens": 6542640.0,
      "step": 3760
    },
    {
      "entropy": 5.801311922073364,
      "epoch": 0.2929391169033262,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004994785605093035,
      "loss": 5.747,
      "mean_token_accuracy": 0.15235013365745545,
      "num_tokens": 6551292.0,
      "step": 3765
    },
    {
      "entropy": 5.7961516857147215,
      "epoch": 0.2933281462750438,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004994766729958705,
      "loss": 5.8392,
      "mean_token_accuracy": 0.1455648086965084,
      "num_tokens": 6560760.0,
      "step": 3770
    },
    {
      "entropy": 5.925059127807617,
      "epoch": 0.2937171756467613,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004994747820763612,
      "loss": 5.8572,
      "mean_token_accuracy": 0.13900205940008165,
      "num_tokens": 6570618.0,
      "step": 3775
    },
    {
      "entropy": 5.86215295791626,
      "epoch": 0.2941062050184789,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004994728877508046,
      "loss": 5.7023,
      "mean_token_accuracy": 0.14909525960683823,
      "num_tokens": 6579484.0,
      "step": 3780
    },
    {
      "entropy": 5.818112993240357,
      "epoch": 0.2944952343901965,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004994709900192289,
      "loss": 5.7803,
      "mean_token_accuracy": 0.1460746668279171,
      "num_tokens": 6588531.0,
      "step": 3785
    },
    {
      "entropy": 5.769961595535278,
      "epoch": 0.294884263761914,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004994690888816635,
      "loss": 5.8196,
      "mean_token_accuracy": 0.14704116955399513,
      "num_tokens": 6596862.0,
      "step": 3790
    },
    {
      "entropy": 5.825479936599732,
      "epoch": 0.2952732931336316,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004994671843381367,
      "loss": 5.723,
      "mean_token_accuracy": 0.14900503158569336,
      "num_tokens": 6605594.0,
      "step": 3795
    },
    {
      "entropy": 5.8890174388885494,
      "epoch": 0.29566232250534913,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004994652763886778,
      "loss": 5.7712,
      "mean_token_accuracy": 0.14753982722759246,
      "num_tokens": 6613971.0,
      "step": 3800
    },
    {
      "entropy": 5.812981510162354,
      "epoch": 0.2960513518770667,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004994633650333155,
      "loss": 5.8431,
      "mean_token_accuracy": 0.1393670342862606,
      "num_tokens": 6623116.0,
      "step": 3805
    },
    {
      "entropy": 5.812322998046875,
      "epoch": 0.2964403812487843,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004994614502720792,
      "loss": 5.7277,
      "mean_token_accuracy": 0.1557611882686615,
      "num_tokens": 6631910.0,
      "step": 3810
    },
    {
      "entropy": 5.8097423076629635,
      "epoch": 0.29682941062050183,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004994595321049974,
      "loss": 5.7528,
      "mean_token_accuracy": 0.15263307020068168,
      "num_tokens": 6640882.0,
      "step": 3815
    },
    {
      "entropy": 5.902229881286621,
      "epoch": 0.2972184399922194,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004994576105320994,
      "loss": 5.8396,
      "mean_token_accuracy": 0.1413460522890091,
      "num_tokens": 6649665.0,
      "step": 3820
    },
    {
      "entropy": 5.703270339965821,
      "epoch": 0.29760746936393695,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004994556855534146,
      "loss": 5.6241,
      "mean_token_accuracy": 0.154371440410614,
      "num_tokens": 6658725.0,
      "step": 3825
    },
    {
      "entropy": 5.67880334854126,
      "epoch": 0.29799649873565454,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004994537571689719,
      "loss": 5.6708,
      "mean_token_accuracy": 0.1537424236536026,
      "num_tokens": 6667403.0,
      "step": 3830
    },
    {
      "entropy": 5.888912105560303,
      "epoch": 0.2983855281073721,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004994518253788007,
      "loss": 5.756,
      "mean_token_accuracy": 0.14269450083374977,
      "num_tokens": 6676673.0,
      "step": 3835
    },
    {
      "entropy": 5.810962390899658,
      "epoch": 0.29877455747908965,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004994498901829304,
      "loss": 5.7563,
      "mean_token_accuracy": 0.14370476379990577,
      "num_tokens": 6686212.0,
      "step": 3840
    },
    {
      "entropy": 5.744761562347412,
      "epoch": 0.29916358685080724,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004994479515813903,
      "loss": 5.7617,
      "mean_token_accuracy": 0.14795507341623307,
      "num_tokens": 6694812.0,
      "step": 3845
    },
    {
      "entropy": 5.79722900390625,
      "epoch": 0.2995526162225248,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004994460095742096,
      "loss": 5.7061,
      "mean_token_accuracy": 0.14745592027902604,
      "num_tokens": 6703439.0,
      "step": 3850
    },
    {
      "entropy": 5.837129878997803,
      "epoch": 0.29994164559424236,
      "grad_norm": 1.265625,
      "learning_rate": 0.000499444064161418,
      "loss": 5.7082,
      "mean_token_accuracy": 0.14777314588427543,
      "num_tokens": 6711465.0,
      "step": 3855
    },
    {
      "entropy": 5.729570055007935,
      "epoch": 0.30033067496595994,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499442115343045,
      "loss": 5.7249,
      "mean_token_accuracy": 0.14670694693922998,
      "num_tokens": 6720526.0,
      "step": 3860
    },
    {
      "entropy": 5.893850612640381,
      "epoch": 0.3007197043376775,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004994401631191202,
      "loss": 5.9148,
      "mean_token_accuracy": 0.1395176962018013,
      "num_tokens": 6728696.0,
      "step": 3865
    },
    {
      "entropy": 5.797284650802612,
      "epoch": 0.30110873370939506,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004994382074896731,
      "loss": 5.7731,
      "mean_token_accuracy": 0.14963408410549164,
      "num_tokens": 6737955.0,
      "step": 3870
    },
    {
      "entropy": 5.84608416557312,
      "epoch": 0.30149776308111265,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004994362484547335,
      "loss": 5.8347,
      "mean_token_accuracy": 0.14240911602973938,
      "num_tokens": 6746413.0,
      "step": 3875
    },
    {
      "entropy": 5.765618562698364,
      "epoch": 0.3018867924528302,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004994342860143309,
      "loss": 5.6949,
      "mean_token_accuracy": 0.14801103472709656,
      "num_tokens": 6754341.0,
      "step": 3880
    },
    {
      "entropy": 5.667687511444091,
      "epoch": 0.30227582182454776,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004994323201684953,
      "loss": 5.5675,
      "mean_token_accuracy": 0.15506181120872498,
      "num_tokens": 6762623.0,
      "step": 3885
    },
    {
      "entropy": 5.747409772872925,
      "epoch": 0.3026648511962653,
      "grad_norm": 1.125,
      "learning_rate": 0.0004994303509172566,
      "loss": 5.7186,
      "mean_token_accuracy": 0.14648325890302658,
      "num_tokens": 6771270.0,
      "step": 3890
    },
    {
      "entropy": 5.833155012130737,
      "epoch": 0.3030538805679829,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004994283782606444,
      "loss": 5.7834,
      "mean_token_accuracy": 0.14363724514842033,
      "num_tokens": 6780292.0,
      "step": 3895
    },
    {
      "entropy": 5.880225801467896,
      "epoch": 0.30344290993970047,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004994264021986888,
      "loss": 5.6681,
      "mean_token_accuracy": 0.15166061818599702,
      "num_tokens": 6788580.0,
      "step": 3900
    },
    {
      "entropy": 5.639140987396241,
      "epoch": 0.303831939311418,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004994244227314197,
      "loss": 5.6524,
      "mean_token_accuracy": 0.14872112795710563,
      "num_tokens": 6796801.0,
      "step": 3905
    },
    {
      "entropy": 5.7354284763336185,
      "epoch": 0.3042209686831356,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004994224398588672,
      "loss": 5.6678,
      "mean_token_accuracy": 0.14618211537599562,
      "num_tokens": 6805630.0,
      "step": 3910
    },
    {
      "entropy": 5.683595514297485,
      "epoch": 0.3046099980548531,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004994204535810615,
      "loss": 5.6425,
      "mean_token_accuracy": 0.15028658956289292,
      "num_tokens": 6814316.0,
      "step": 3915
    },
    {
      "entropy": 5.896274042129517,
      "epoch": 0.3049990274265707,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004994184638980326,
      "loss": 5.8131,
      "mean_token_accuracy": 0.14128096029162407,
      "num_tokens": 6823224.0,
      "step": 3920
    },
    {
      "entropy": 5.82032413482666,
      "epoch": 0.3053880567982883,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004994164708098107,
      "loss": 5.8399,
      "mean_token_accuracy": 0.1436244174838066,
      "num_tokens": 6832104.0,
      "step": 3925
    },
    {
      "entropy": 5.832739591598511,
      "epoch": 0.3057770861700058,
      "grad_norm": 1.15625,
      "learning_rate": 0.000499414474316426,
      "loss": 5.7432,
      "mean_token_accuracy": 0.1472333326935768,
      "num_tokens": 6841208.0,
      "step": 3930
    },
    {
      "entropy": 5.652276039123535,
      "epoch": 0.3061661155417234,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004994124744179088,
      "loss": 5.546,
      "mean_token_accuracy": 0.15998484641313554,
      "num_tokens": 6849622.0,
      "step": 3935
    },
    {
      "entropy": 5.7208733558654785,
      "epoch": 0.306555144913441,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004994104711142896,
      "loss": 5.7513,
      "mean_token_accuracy": 0.1525330901145935,
      "num_tokens": 6857355.0,
      "step": 3940
    },
    {
      "entropy": 5.833088970184326,
      "epoch": 0.3069441742851585,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004994084644055986,
      "loss": 5.764,
      "mean_token_accuracy": 0.15032006576657295,
      "num_tokens": 6866173.0,
      "step": 3945
    },
    {
      "entropy": 5.745790719985962,
      "epoch": 0.3073332036568761,
      "grad_norm": 1.25,
      "learning_rate": 0.0004994064542918664,
      "loss": 5.7618,
      "mean_token_accuracy": 0.14327194094657897,
      "num_tokens": 6874438.0,
      "step": 3950
    },
    {
      "entropy": 5.838888788223267,
      "epoch": 0.30772223302859364,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004994044407731235,
      "loss": 5.7559,
      "mean_token_accuracy": 0.14924216270446777,
      "num_tokens": 6882790.0,
      "step": 3955
    },
    {
      "entropy": 5.77858567237854,
      "epoch": 0.3081112624003112,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004994024238494002,
      "loss": 5.8004,
      "mean_token_accuracy": 0.13810172826051711,
      "num_tokens": 6891375.0,
      "step": 3960
    },
    {
      "entropy": 5.672658538818359,
      "epoch": 0.3085002917720288,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004994004035207274,
      "loss": 5.7009,
      "mean_token_accuracy": 0.15040683150291442,
      "num_tokens": 6899510.0,
      "step": 3965
    },
    {
      "entropy": 5.849552297592163,
      "epoch": 0.30888932114374634,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004993983797871356,
      "loss": 5.7546,
      "mean_token_accuracy": 0.15272879973053932,
      "num_tokens": 6908311.0,
      "step": 3970
    },
    {
      "entropy": 5.77811074256897,
      "epoch": 0.30927835051546393,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004993963526486555,
      "loss": 5.7107,
      "mean_token_accuracy": 0.14666830450296403,
      "num_tokens": 6917064.0,
      "step": 3975
    },
    {
      "entropy": 5.765193367004395,
      "epoch": 0.30966737988718146,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000499394322105318,
      "loss": 5.7055,
      "mean_token_accuracy": 0.14877342879772187,
      "num_tokens": 6925705.0,
      "step": 3980
    },
    {
      "entropy": 5.929165410995483,
      "epoch": 0.31005640925889905,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004993922881571538,
      "loss": 5.8505,
      "mean_token_accuracy": 0.14009160473942756,
      "num_tokens": 6934232.0,
      "step": 3985
    },
    {
      "entropy": 5.842931127548217,
      "epoch": 0.31044543863061663,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004993902508041939,
      "loss": 5.7825,
      "mean_token_accuracy": 0.1494826927781105,
      "num_tokens": 6943143.0,
      "step": 3990
    },
    {
      "entropy": 5.784590435028076,
      "epoch": 0.31083446800233416,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000499388210046469,
      "loss": 5.6572,
      "mean_token_accuracy": 0.15119214802980424,
      "num_tokens": 6951613.0,
      "step": 3995
    },
    {
      "entropy": 5.781867980957031,
      "epoch": 0.31122349737405175,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004993861658840102,
      "loss": 5.8635,
      "mean_token_accuracy": 0.144657726585865,
      "num_tokens": 6960531.0,
      "step": 4000
    },
    {
      "entropy": 5.813362073898316,
      "epoch": 0.3116125267457693,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004993841183168484,
      "loss": 5.7836,
      "mean_token_accuracy": 0.14522304087877275,
      "num_tokens": 6969211.0,
      "step": 4005
    },
    {
      "entropy": 5.923810148239136,
      "epoch": 0.31200155611748687,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004993820673450148,
      "loss": 5.8097,
      "mean_token_accuracy": 0.14293079674243928,
      "num_tokens": 6978414.0,
      "step": 4010
    },
    {
      "entropy": 5.676694631576538,
      "epoch": 0.31239058548920445,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004993800129685404,
      "loss": 5.5568,
      "mean_token_accuracy": 0.15368023365736008,
      "num_tokens": 6987350.0,
      "step": 4015
    },
    {
      "entropy": 5.725512933731079,
      "epoch": 0.312779614860922,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004993779551874565,
      "loss": 5.7006,
      "mean_token_accuracy": 0.15659889280796052,
      "num_tokens": 6995817.0,
      "step": 4020
    },
    {
      "entropy": 5.77770471572876,
      "epoch": 0.31316864423263957,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004993758940017943,
      "loss": 5.7712,
      "mean_token_accuracy": 0.14786949157714843,
      "num_tokens": 7004454.0,
      "step": 4025
    },
    {
      "entropy": 5.818613433837891,
      "epoch": 0.31355767360435716,
      "grad_norm": 1.296875,
      "learning_rate": 0.000499373829411585,
      "loss": 5.712,
      "mean_token_accuracy": 0.1497095689177513,
      "num_tokens": 7012126.0,
      "step": 4030
    },
    {
      "entropy": 5.80422043800354,
      "epoch": 0.3139467029760747,
      "grad_norm": 1.15625,
      "learning_rate": 0.00049937176141686,
      "loss": 5.7819,
      "mean_token_accuracy": 0.15036623552441597,
      "num_tokens": 7021232.0,
      "step": 4035
    },
    {
      "entropy": 5.852689599990844,
      "epoch": 0.3143357323477923,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004993696900176506,
      "loss": 5.7645,
      "mean_token_accuracy": 0.14852062985301018,
      "num_tokens": 7029488.0,
      "step": 4040
    },
    {
      "entropy": 5.820941209793091,
      "epoch": 0.3147247617195098,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004993676152139883,
      "loss": 5.7985,
      "mean_token_accuracy": 0.1442652441561222,
      "num_tokens": 7037323.0,
      "step": 4045
    },
    {
      "entropy": 5.7845946788787845,
      "epoch": 0.3151137910912274,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004993655370059046,
      "loss": 5.7531,
      "mean_token_accuracy": 0.14139595925807952,
      "num_tokens": 7045754.0,
      "step": 4050
    },
    {
      "entropy": 5.7826457023620605,
      "epoch": 0.315502820462945,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004993634553934309,
      "loss": 5.6884,
      "mean_token_accuracy": 0.14923796579241752,
      "num_tokens": 7054026.0,
      "step": 4055
    },
    {
      "entropy": 5.715740585327149,
      "epoch": 0.3158918498346625,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004993613703765989,
      "loss": 5.7329,
      "mean_token_accuracy": 0.1490222007036209,
      "num_tokens": 7063383.0,
      "step": 4060
    },
    {
      "entropy": 5.74744987487793,
      "epoch": 0.3162808792063801,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004993592819554402,
      "loss": 5.7028,
      "mean_token_accuracy": 0.15151856541633607,
      "num_tokens": 7071952.0,
      "step": 4065
    },
    {
      "entropy": 5.862038898468017,
      "epoch": 0.3166699085780976,
      "grad_norm": 1.375,
      "learning_rate": 0.0004993571901299866,
      "loss": 5.7424,
      "mean_token_accuracy": 0.1466784045100212,
      "num_tokens": 7080283.0,
      "step": 4070
    },
    {
      "entropy": 5.658473825454712,
      "epoch": 0.3170589379498152,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004993550949002697,
      "loss": 5.6853,
      "mean_token_accuracy": 0.14666975438594818,
      "num_tokens": 7088669.0,
      "step": 4075
    },
    {
      "entropy": 5.697621583938599,
      "epoch": 0.3174479673215328,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004993529962663213,
      "loss": 5.5697,
      "mean_token_accuracy": 0.16409860402345658,
      "num_tokens": 7097201.0,
      "step": 4080
    },
    {
      "entropy": 5.748375940322876,
      "epoch": 0.31783699669325033,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004993508942281732,
      "loss": 5.7527,
      "mean_token_accuracy": 0.14704224616289138,
      "num_tokens": 7106509.0,
      "step": 4085
    },
    {
      "entropy": 5.772694444656372,
      "epoch": 0.3182260260649679,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004993487887858575,
      "loss": 5.662,
      "mean_token_accuracy": 0.15324160903692247,
      "num_tokens": 7114550.0,
      "step": 4090
    },
    {
      "entropy": 5.823660564422608,
      "epoch": 0.31861505543668545,
      "grad_norm": 1.125,
      "learning_rate": 0.000499346679939406,
      "loss": 5.7734,
      "mean_token_accuracy": 0.15043377876281738,
      "num_tokens": 7122875.0,
      "step": 4095
    },
    {
      "entropy": 5.785100364685059,
      "epoch": 0.31900408480840303,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004993445676888507,
      "loss": 5.5858,
      "mean_token_accuracy": 0.14903404712677001,
      "num_tokens": 7132263.0,
      "step": 4100
    },
    {
      "entropy": 5.708623313903809,
      "epoch": 0.3193931141801206,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004993424520342236,
      "loss": 5.8122,
      "mean_token_accuracy": 0.14726694226264953,
      "num_tokens": 7141448.0,
      "step": 4105
    },
    {
      "entropy": 5.757666683197021,
      "epoch": 0.31978214355183815,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004993403329755569,
      "loss": 5.8158,
      "mean_token_accuracy": 0.14173928201198577,
      "num_tokens": 7151021.0,
      "step": 4110
    },
    {
      "entropy": 5.908272171020508,
      "epoch": 0.32017117292355574,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004993382105128828,
      "loss": 5.7724,
      "mean_token_accuracy": 0.14403056129813194,
      "num_tokens": 7160103.0,
      "step": 4115
    },
    {
      "entropy": 5.794647169113159,
      "epoch": 0.32056020229527327,
      "grad_norm": 1.25,
      "learning_rate": 0.0004993360846462333,
      "loss": 5.7575,
      "mean_token_accuracy": 0.14108945354819297,
      "num_tokens": 7169003.0,
      "step": 4120
    },
    {
      "entropy": 5.856556415557861,
      "epoch": 0.32094923166699085,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004993339553756408,
      "loss": 5.8024,
      "mean_token_accuracy": 0.15243273302912713,
      "num_tokens": 7177086.0,
      "step": 4125
    },
    {
      "entropy": 5.769464302062988,
      "epoch": 0.32133826103870844,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004993318227011378,
      "loss": 5.674,
      "mean_token_accuracy": 0.15314000099897385,
      "num_tokens": 7185772.0,
      "step": 4130
    },
    {
      "entropy": 5.854861259460449,
      "epoch": 0.32172729041042597,
      "grad_norm": 1.25,
      "learning_rate": 0.0004993296866227562,
      "loss": 5.8114,
      "mean_token_accuracy": 0.14875666797161102,
      "num_tokens": 7194394.0,
      "step": 4135
    },
    {
      "entropy": 5.803802633285523,
      "epoch": 0.32211631978214356,
      "grad_norm": 1.25,
      "learning_rate": 0.0004993275471405288,
      "loss": 5.6327,
      "mean_token_accuracy": 0.15715198665857316,
      "num_tokens": 7202143.0,
      "step": 4140
    },
    {
      "entropy": 5.6745908737182615,
      "epoch": 0.32250534915386114,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004993254042544879,
      "loss": 5.737,
      "mean_token_accuracy": 0.14759556651115419,
      "num_tokens": 7210645.0,
      "step": 4145
    },
    {
      "entropy": 5.773699808120727,
      "epoch": 0.3228943785255787,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000499323257964666,
      "loss": 5.6711,
      "mean_token_accuracy": 0.15104791671037673,
      "num_tokens": 7218872.0,
      "step": 4150
    },
    {
      "entropy": 5.864423131942749,
      "epoch": 0.32328340789729626,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004993211082710957,
      "loss": 5.8581,
      "mean_token_accuracy": 0.14464571326971054,
      "num_tokens": 7226977.0,
      "step": 4155
    },
    {
      "entropy": 5.803394317626953,
      "epoch": 0.3236724372690138,
      "grad_norm": 1.25,
      "learning_rate": 0.0004993189551738097,
      "loss": 5.6977,
      "mean_token_accuracy": 0.14795618653297424,
      "num_tokens": 7235635.0,
      "step": 4160
    },
    {
      "entropy": 5.760642957687378,
      "epoch": 0.3240614666407314,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004993167986728405,
      "loss": 5.6863,
      "mean_token_accuracy": 0.15549185127019882,
      "num_tokens": 7244506.0,
      "step": 4165
    },
    {
      "entropy": 5.751373052597046,
      "epoch": 0.32445049601244896,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000499314638768221,
      "loss": 5.7222,
      "mean_token_accuracy": 0.149913389980793,
      "num_tokens": 7252883.0,
      "step": 4170
    },
    {
      "entropy": 5.821081495285034,
      "epoch": 0.3248395253841665,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004993124754599839,
      "loss": 5.6798,
      "mean_token_accuracy": 0.14590843692421912,
      "num_tokens": 7261580.0,
      "step": 4175
    },
    {
      "entropy": 5.793644857406616,
      "epoch": 0.3252285547558841,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004993103087481619,
      "loss": 5.7723,
      "mean_token_accuracy": 0.1446990892291069,
      "num_tokens": 7270346.0,
      "step": 4180
    },
    {
      "entropy": 5.804017353057861,
      "epoch": 0.3256175841276016,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004993081386327881,
      "loss": 5.8424,
      "mean_token_accuracy": 0.14238107204437256,
      "num_tokens": 7279393.0,
      "step": 4185
    },
    {
      "entropy": 5.733974027633667,
      "epoch": 0.3260066134993192,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004993059651138953,
      "loss": 5.6846,
      "mean_token_accuracy": 0.14898681640625,
      "num_tokens": 7287929.0,
      "step": 4190
    },
    {
      "entropy": 5.858212947845459,
      "epoch": 0.3263956428710368,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004993037881915165,
      "loss": 5.7203,
      "mean_token_accuracy": 0.14945944994688035,
      "num_tokens": 7296336.0,
      "step": 4195
    },
    {
      "entropy": 5.687768888473511,
      "epoch": 0.3267846722427543,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004993016078656847,
      "loss": 5.6126,
      "mean_token_accuracy": 0.15262411832809447,
      "num_tokens": 7304624.0,
      "step": 4200
    },
    {
      "entropy": 5.690052080154419,
      "epoch": 0.3271737016144719,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499299424136433,
      "loss": 5.7504,
      "mean_token_accuracy": 0.14595601409673692,
      "num_tokens": 7312945.0,
      "step": 4205
    },
    {
      "entropy": 5.824770212173462,
      "epoch": 0.32756273098618943,
      "grad_norm": 1.25,
      "learning_rate": 0.0004992972370037946,
      "loss": 5.669,
      "mean_token_accuracy": 0.15038443952798844,
      "num_tokens": 7321273.0,
      "step": 4210
    },
    {
      "entropy": 5.8330482006072994,
      "epoch": 0.327951760357907,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004992950464678026,
      "loss": 5.7675,
      "mean_token_accuracy": 0.15167799592018127,
      "num_tokens": 7330446.0,
      "step": 4215
    },
    {
      "entropy": 5.669593095779419,
      "epoch": 0.3283407897296246,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004992928525284903,
      "loss": 5.704,
      "mean_token_accuracy": 0.1521438404917717,
      "num_tokens": 7339282.0,
      "step": 4220
    },
    {
      "entropy": 5.752499961853028,
      "epoch": 0.32872981910134214,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000499290655185891,
      "loss": 5.6707,
      "mean_token_accuracy": 0.14109978452324867,
      "num_tokens": 7347848.0,
      "step": 4225
    },
    {
      "entropy": 5.679991912841797,
      "epoch": 0.3291188484730597,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000499288454440038,
      "loss": 5.6748,
      "mean_token_accuracy": 0.15071465373039244,
      "num_tokens": 7356678.0,
      "step": 4230
    },
    {
      "entropy": 5.791227054595947,
      "epoch": 0.3295078778447773,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004992862502909645,
      "loss": 5.7363,
      "mean_token_accuracy": 0.15067468732595443,
      "num_tokens": 7365411.0,
      "step": 4235
    },
    {
      "entropy": 5.725597524642945,
      "epoch": 0.32989690721649484,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004992840427387043,
      "loss": 5.6558,
      "mean_token_accuracy": 0.15863664150238038,
      "num_tokens": 7373635.0,
      "step": 4240
    },
    {
      "entropy": 5.772441816329956,
      "epoch": 0.3302859365882124,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004992818317832908,
      "loss": 5.7525,
      "mean_token_accuracy": 0.15104876458644867,
      "num_tokens": 7382469.0,
      "step": 4245
    },
    {
      "entropy": 5.791848611831665,
      "epoch": 0.33067496595992996,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004992796174247574,
      "loss": 5.7456,
      "mean_token_accuracy": 0.14663487672805786,
      "num_tokens": 7390712.0,
      "step": 4250
    },
    {
      "entropy": 5.83238000869751,
      "epoch": 0.33106399533164754,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004992773996631378,
      "loss": 5.6763,
      "mean_token_accuracy": 0.14819879531860353,
      "num_tokens": 7398905.0,
      "step": 4255
    },
    {
      "entropy": 5.7469171524047855,
      "epoch": 0.33145302470336513,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004992751784984656,
      "loss": 5.7521,
      "mean_token_accuracy": 0.14595615193247796,
      "num_tokens": 7407242.0,
      "step": 4260
    },
    {
      "entropy": 5.742021656036377,
      "epoch": 0.33184205407508266,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004992729539307745,
      "loss": 5.7694,
      "mean_token_accuracy": 0.1436052680015564,
      "num_tokens": 7416489.0,
      "step": 4265
    },
    {
      "entropy": 5.906120681762696,
      "epoch": 0.33223108344680025,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004992707259600984,
      "loss": 5.7559,
      "mean_token_accuracy": 0.1527278333902359,
      "num_tokens": 7424507.0,
      "step": 4270
    },
    {
      "entropy": 5.736468029022217,
      "epoch": 0.3326201128185178,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004992684945864709,
      "loss": 5.746,
      "mean_token_accuracy": 0.1425243631005287,
      "num_tokens": 7433343.0,
      "step": 4275
    },
    {
      "entropy": 5.7437169551849365,
      "epoch": 0.33300914219023536,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004992662598099261,
      "loss": 5.6856,
      "mean_token_accuracy": 0.15159007757902146,
      "num_tokens": 7441452.0,
      "step": 4280
    },
    {
      "entropy": 5.82387146949768,
      "epoch": 0.33339817156195295,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004992640216304975,
      "loss": 5.7383,
      "mean_token_accuracy": 0.15166881456971168,
      "num_tokens": 7449541.0,
      "step": 4285
    },
    {
      "entropy": 5.772676610946656,
      "epoch": 0.3337872009336705,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004992617800482196,
      "loss": 5.7508,
      "mean_token_accuracy": 0.1446318544447422,
      "num_tokens": 7458812.0,
      "step": 4290
    },
    {
      "entropy": 5.80000319480896,
      "epoch": 0.33417623030538807,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000499259535063126,
      "loss": 5.6754,
      "mean_token_accuracy": 0.14619821310043335,
      "num_tokens": 7467607.0,
      "step": 4295
    },
    {
      "entropy": 5.712003326416015,
      "epoch": 0.3345652596771056,
      "grad_norm": 1.34375,
      "learning_rate": 0.000499257286675251,
      "loss": 5.6227,
      "mean_token_accuracy": 0.1527673527598381,
      "num_tokens": 7476047.0,
      "step": 4300
    },
    {
      "entropy": 5.78293080329895,
      "epoch": 0.3349542890488232,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004992550348846285,
      "loss": 5.6868,
      "mean_token_accuracy": 0.14961570352315903,
      "num_tokens": 7484496.0,
      "step": 4305
    },
    {
      "entropy": 5.65687518119812,
      "epoch": 0.33534331842054077,
      "grad_norm": 1.25,
      "learning_rate": 0.0004992527796912928,
      "loss": 5.612,
      "mean_token_accuracy": 0.15613239780068397,
      "num_tokens": 7493161.0,
      "step": 4310
    },
    {
      "entropy": 5.699185752868653,
      "epoch": 0.3357323477922583,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004992505210952782,
      "loss": 5.7685,
      "mean_token_accuracy": 0.14365074336528777,
      "num_tokens": 7502122.0,
      "step": 4315
    },
    {
      "entropy": 5.795812940597534,
      "epoch": 0.3361213771639759,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004992482590966188,
      "loss": 5.7234,
      "mean_token_accuracy": 0.14752976670861245,
      "num_tokens": 7510369.0,
      "step": 4320
    },
    {
      "entropy": 5.68804521560669,
      "epoch": 0.3365104065356934,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004992459936953489,
      "loss": 5.6099,
      "mean_token_accuracy": 0.158554507791996,
      "num_tokens": 7519321.0,
      "step": 4325
    },
    {
      "entropy": 5.735086917877197,
      "epoch": 0.336899435907411,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004992437248915032,
      "loss": 5.655,
      "mean_token_accuracy": 0.1553067222237587,
      "num_tokens": 7528196.0,
      "step": 4330
    },
    {
      "entropy": 5.724933862686157,
      "epoch": 0.3372884652791286,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004992414526851157,
      "loss": 5.7136,
      "mean_token_accuracy": 0.1463082104921341,
      "num_tokens": 7536124.0,
      "step": 4335
    },
    {
      "entropy": 5.84370985031128,
      "epoch": 0.3376774946508461,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004992391770762212,
      "loss": 5.8058,
      "mean_token_accuracy": 0.1447911098599434,
      "num_tokens": 7544409.0,
      "step": 4340
    },
    {
      "entropy": 5.809447193145752,
      "epoch": 0.3380665240225637,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499236898064854,
      "loss": 5.7001,
      "mean_token_accuracy": 0.1530858114361763,
      "num_tokens": 7552803.0,
      "step": 4345
    },
    {
      "entropy": 5.6515076637268065,
      "epoch": 0.3384555533942813,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004992346156510488,
      "loss": 5.6644,
      "mean_token_accuracy": 0.1459401488304138,
      "num_tokens": 7561909.0,
      "step": 4350
    },
    {
      "entropy": 5.825453233718872,
      "epoch": 0.3388445827659988,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004992323298348403,
      "loss": 5.7289,
      "mean_token_accuracy": 0.14583828970789908,
      "num_tokens": 7571759.0,
      "step": 4355
    },
    {
      "entropy": 5.8584259986877445,
      "epoch": 0.3392336121377164,
      "grad_norm": 1.1875,
      "learning_rate": 0.000499230040616263,
      "loss": 5.8566,
      "mean_token_accuracy": 0.15016695931553842,
      "num_tokens": 7580010.0,
      "step": 4360
    },
    {
      "entropy": 5.782916736602783,
      "epoch": 0.33962264150943394,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004992277479953518,
      "loss": 5.6514,
      "mean_token_accuracy": 0.14859042316675186,
      "num_tokens": 7587841.0,
      "step": 4365
    },
    {
      "entropy": 5.720286655426025,
      "epoch": 0.34001167088115153,
      "grad_norm": 1.25,
      "learning_rate": 0.0004992254519721414,
      "loss": 5.6889,
      "mean_token_accuracy": 0.14878612980246544,
      "num_tokens": 7596187.0,
      "step": 4370
    },
    {
      "entropy": 5.681316328048706,
      "epoch": 0.3404007002528691,
      "grad_norm": 1.25,
      "learning_rate": 0.0004992231525466666,
      "loss": 5.6074,
      "mean_token_accuracy": 0.15483227521181106,
      "num_tokens": 7604361.0,
      "step": 4375
    },
    {
      "entropy": 5.770654630661011,
      "epoch": 0.34078972962458665,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004992208497189624,
      "loss": 5.6927,
      "mean_token_accuracy": 0.1531132161617279,
      "num_tokens": 7613173.0,
      "step": 4380
    },
    {
      "entropy": 5.7282201766967775,
      "epoch": 0.34117875899630423,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004992185434890637,
      "loss": 5.7741,
      "mean_token_accuracy": 0.14267738461494445,
      "num_tokens": 7621730.0,
      "step": 4385
    },
    {
      "entropy": 5.8589198112487795,
      "epoch": 0.34156778836802176,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004992162338570055,
      "loss": 5.7414,
      "mean_token_accuracy": 0.15011457800865174,
      "num_tokens": 7630362.0,
      "step": 4390
    },
    {
      "entropy": 5.7542157649993895,
      "epoch": 0.34195681773973935,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004992139208228227,
      "loss": 5.7186,
      "mean_token_accuracy": 0.14609325155615807,
      "num_tokens": 7639368.0,
      "step": 4395
    },
    {
      "entropy": 5.7741481304168705,
      "epoch": 0.34234584711145694,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004992116043865506,
      "loss": 5.7489,
      "mean_token_accuracy": 0.1471161961555481,
      "num_tokens": 7648399.0,
      "step": 4400
    },
    {
      "entropy": 5.7407738208770756,
      "epoch": 0.34273487648317447,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004992092845482244,
      "loss": 5.5847,
      "mean_token_accuracy": 0.15751371681690216,
      "num_tokens": 7657298.0,
      "step": 4405
    },
    {
      "entropy": 5.788975954055786,
      "epoch": 0.34312390585489205,
      "grad_norm": 1.25,
      "learning_rate": 0.0004992069613078791,
      "loss": 5.7176,
      "mean_token_accuracy": 0.15156893134117128,
      "num_tokens": 7665328.0,
      "step": 4410
    },
    {
      "entropy": 5.724450349807739,
      "epoch": 0.3435129352266096,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004992046346655499,
      "loss": 5.6566,
      "mean_token_accuracy": 0.1507851794362068,
      "num_tokens": 7674610.0,
      "step": 4415
    },
    {
      "entropy": 5.772369146347046,
      "epoch": 0.34390196459832717,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004992023046212723,
      "loss": 5.6962,
      "mean_token_accuracy": 0.15413508117198943,
      "num_tokens": 7682727.0,
      "step": 4420
    },
    {
      "entropy": 5.71473445892334,
      "epoch": 0.34429099397004476,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004991999711750816,
      "loss": 5.6621,
      "mean_token_accuracy": 0.15658219903707504,
      "num_tokens": 7691547.0,
      "step": 4425
    },
    {
      "entropy": 5.73526964187622,
      "epoch": 0.3446800233417623,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004991976343270133,
      "loss": 5.6102,
      "mean_token_accuracy": 0.15017578452825547,
      "num_tokens": 7699987.0,
      "step": 4430
    },
    {
      "entropy": 5.739548635482788,
      "epoch": 0.3450690527134799,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004991952940771026,
      "loss": 5.6789,
      "mean_token_accuracy": 0.15383388996124267,
      "num_tokens": 7708397.0,
      "step": 4435
    },
    {
      "entropy": 5.745605754852295,
      "epoch": 0.34545808208519746,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004991929504253852,
      "loss": 5.6689,
      "mean_token_accuracy": 0.14823864698410033,
      "num_tokens": 7716795.0,
      "step": 4440
    },
    {
      "entropy": 5.750987958908081,
      "epoch": 0.345847111456915,
      "grad_norm": 1.25,
      "learning_rate": 0.0004991906033718966,
      "loss": 5.6828,
      "mean_token_accuracy": 0.14957790747284888,
      "num_tokens": 7725663.0,
      "step": 4445
    },
    {
      "entropy": 5.786761999130249,
      "epoch": 0.3462361408286326,
      "grad_norm": 1.25,
      "learning_rate": 0.0004991882529166724,
      "loss": 5.8069,
      "mean_token_accuracy": 0.14756733924150467,
      "num_tokens": 7735175.0,
      "step": 4450
    },
    {
      "entropy": 5.77248911857605,
      "epoch": 0.3466251702003501,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004991858990597483,
      "loss": 5.6742,
      "mean_token_accuracy": 0.15260669961571693,
      "num_tokens": 7744041.0,
      "step": 4455
    },
    {
      "entropy": 5.813203620910644,
      "epoch": 0.3470141995720677,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004991835418011601,
      "loss": 5.7328,
      "mean_token_accuracy": 0.14599995985627173,
      "num_tokens": 7752168.0,
      "step": 4460
    },
    {
      "entropy": 5.774614095687866,
      "epoch": 0.3474032289437853,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004991811811409434,
      "loss": 5.6653,
      "mean_token_accuracy": 0.1525656133890152,
      "num_tokens": 7760746.0,
      "step": 4465
    },
    {
      "entropy": 5.690586090087891,
      "epoch": 0.3477922583155028,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004991788170791341,
      "loss": 5.6782,
      "mean_token_accuracy": 0.15173280835151673,
      "num_tokens": 7769960.0,
      "step": 4470
    },
    {
      "entropy": 5.805200624465942,
      "epoch": 0.3481812876872204,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004991764496157682,
      "loss": 5.7113,
      "mean_token_accuracy": 0.14903471767902374,
      "num_tokens": 7778475.0,
      "step": 4475
    },
    {
      "entropy": 5.694829988479614,
      "epoch": 0.34857031705893793,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004991740787508814,
      "loss": 5.571,
      "mean_token_accuracy": 0.15922194421291352,
      "num_tokens": 7787590.0,
      "step": 4480
    },
    {
      "entropy": 5.716617298126221,
      "epoch": 0.3489593464306555,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004991717044845097,
      "loss": 5.7509,
      "mean_token_accuracy": 0.14409140273928642,
      "num_tokens": 7796841.0,
      "step": 4485
    },
    {
      "entropy": 5.844481468200684,
      "epoch": 0.3493483758023731,
      "grad_norm": 1.375,
      "learning_rate": 0.0004991693268166892,
      "loss": 5.752,
      "mean_token_accuracy": 0.1473579525947571,
      "num_tokens": 7804889.0,
      "step": 4490
    },
    {
      "entropy": 5.751714563369751,
      "epoch": 0.34973740517409063,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004991669457474559,
      "loss": 5.7307,
      "mean_token_accuracy": 0.1499755159020424,
      "num_tokens": 7813185.0,
      "step": 4495
    },
    {
      "entropy": 5.805147790908814,
      "epoch": 0.3501264345458082,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004991645612768461,
      "loss": 5.7197,
      "mean_token_accuracy": 0.15086604356765748,
      "num_tokens": 7821630.0,
      "step": 4500
    },
    {
      "entropy": 5.836012315750122,
      "epoch": 0.35051546391752575,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004991621734048958,
      "loss": 5.7522,
      "mean_token_accuracy": 0.15427149161696435,
      "num_tokens": 7829941.0,
      "step": 4505
    },
    {
      "entropy": 5.707473087310791,
      "epoch": 0.35090449328924334,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004991597821316415,
      "loss": 5.6936,
      "mean_token_accuracy": 0.15636295527219773,
      "num_tokens": 7838515.0,
      "step": 4510
    },
    {
      "entropy": 5.8055788516998295,
      "epoch": 0.3512935226609609,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004991573874571192,
      "loss": 5.6896,
      "mean_token_accuracy": 0.15325772613286973,
      "num_tokens": 7846575.0,
      "step": 4515
    },
    {
      "entropy": 5.651686334609986,
      "epoch": 0.35168255203267845,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004991549893813653,
      "loss": 5.5803,
      "mean_token_accuracy": 0.1577739804983139,
      "num_tokens": 7854970.0,
      "step": 4520
    },
    {
      "entropy": 5.672727489471436,
      "epoch": 0.35207158140439604,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004991525879044162,
      "loss": 5.6722,
      "mean_token_accuracy": 0.1503447487950325,
      "num_tokens": 7863293.0,
      "step": 4525
    },
    {
      "entropy": 5.777629470825195,
      "epoch": 0.35246061077611357,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004991501830263084,
      "loss": 5.6652,
      "mean_token_accuracy": 0.15367957353591918,
      "num_tokens": 7871029.0,
      "step": 4530
    },
    {
      "entropy": 5.671704244613648,
      "epoch": 0.35284964014783116,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004991477747470783,
      "loss": 5.5294,
      "mean_token_accuracy": 0.1555831491947174,
      "num_tokens": 7879797.0,
      "step": 4535
    },
    {
      "entropy": 5.820374870300293,
      "epoch": 0.35323866951954874,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004991453630667625,
      "loss": 5.8385,
      "mean_token_accuracy": 0.14210431426763534,
      "num_tokens": 7888600.0,
      "step": 4540
    },
    {
      "entropy": 5.752278423309326,
      "epoch": 0.3536276988912663,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004991429479853976,
      "loss": 5.6771,
      "mean_token_accuracy": 0.15561284869909286,
      "num_tokens": 7897794.0,
      "step": 4545
    },
    {
      "entropy": 5.773390102386474,
      "epoch": 0.35401672826298386,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004991405295030202,
      "loss": 5.658,
      "mean_token_accuracy": 0.15940722376108168,
      "num_tokens": 7905822.0,
      "step": 4550
    },
    {
      "entropy": 5.700776815414429,
      "epoch": 0.35440575763470145,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004991381076196671,
      "loss": 5.7223,
      "mean_token_accuracy": 0.1521821677684784,
      "num_tokens": 7914560.0,
      "step": 4555
    },
    {
      "entropy": 5.661709499359131,
      "epoch": 0.354794787006419,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004991356823353748,
      "loss": 5.577,
      "mean_token_accuracy": 0.15658768862485886,
      "num_tokens": 7924058.0,
      "step": 4560
    },
    {
      "entropy": 5.762267398834228,
      "epoch": 0.35518381637813656,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004991332536501804,
      "loss": 5.7225,
      "mean_token_accuracy": 0.1503455936908722,
      "num_tokens": 7932979.0,
      "step": 4565
    },
    {
      "entropy": 5.655368661880493,
      "epoch": 0.3555728457498541,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004991308215641205,
      "loss": 5.4999,
      "mean_token_accuracy": 0.16385250836610793,
      "num_tokens": 7941357.0,
      "step": 4570
    },
    {
      "entropy": 5.6736588954925535,
      "epoch": 0.3559618751215717,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004991283860772323,
      "loss": 5.6739,
      "mean_token_accuracy": 0.15402223020792008,
      "num_tokens": 7949997.0,
      "step": 4575
    },
    {
      "entropy": 5.716558218002319,
      "epoch": 0.35635090449328927,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004991259471895525,
      "loss": 5.6288,
      "mean_token_accuracy": 0.14936613291502,
      "num_tokens": 7958246.0,
      "step": 4580
    },
    {
      "entropy": 5.683485794067383,
      "epoch": 0.3567399338650068,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004991235049011182,
      "loss": 5.7203,
      "mean_token_accuracy": 0.14906762838363646,
      "num_tokens": 7967176.0,
      "step": 4585
    },
    {
      "entropy": 5.747391319274902,
      "epoch": 0.3571289632367244,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004991210592119663,
      "loss": 5.5664,
      "mean_token_accuracy": 0.16120724231004716,
      "num_tokens": 7974761.0,
      "step": 4590
    },
    {
      "entropy": 5.708501148223877,
      "epoch": 0.3575179926084419,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004991186101221342,
      "loss": 5.643,
      "mean_token_accuracy": 0.1528185173869133,
      "num_tokens": 7983155.0,
      "step": 4595
    },
    {
      "entropy": 5.7329905986785885,
      "epoch": 0.3579070219801595,
      "grad_norm": 1.21875,
      "learning_rate": 0.000499116157631659,
      "loss": 5.6438,
      "mean_token_accuracy": 0.14957136809825897,
      "num_tokens": 7991276.0,
      "step": 4600
    },
    {
      "entropy": 5.790548658370971,
      "epoch": 0.3582960513518771,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004991137017405777,
      "loss": 5.7855,
      "mean_token_accuracy": 0.1485683262348175,
      "num_tokens": 7999488.0,
      "step": 4605
    },
    {
      "entropy": 5.799504184722901,
      "epoch": 0.3586850807235946,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004991112424489277,
      "loss": 5.6554,
      "mean_token_accuracy": 0.1543542042374611,
      "num_tokens": 8008553.0,
      "step": 4610
    },
    {
      "entropy": 5.698079681396484,
      "epoch": 0.3590741100953122,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004991087797567462,
      "loss": 5.6266,
      "mean_token_accuracy": 0.15792726278305053,
      "num_tokens": 8017890.0,
      "step": 4615
    },
    {
      "entropy": 5.748518419265747,
      "epoch": 0.35946313946702974,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004991063136640709,
      "loss": 5.7805,
      "mean_token_accuracy": 0.14351232275366782,
      "num_tokens": 8025782.0,
      "step": 4620
    },
    {
      "entropy": 5.807697534561157,
      "epoch": 0.3598521688387473,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004991038441709388,
      "loss": 5.6364,
      "mean_token_accuracy": 0.15547887086868287,
      "num_tokens": 8034623.0,
      "step": 4625
    },
    {
      "entropy": 5.715529870986939,
      "epoch": 0.3602411982104649,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004991013712773877,
      "loss": 5.6676,
      "mean_token_accuracy": 0.148578180372715,
      "num_tokens": 8043789.0,
      "step": 4630
    },
    {
      "entropy": 5.687366724014282,
      "epoch": 0.36063022758218244,
      "grad_norm": 1.25,
      "learning_rate": 0.0004990988949834549,
      "loss": 5.6296,
      "mean_token_accuracy": 0.1519090548157692,
      "num_tokens": 8053456.0,
      "step": 4635
    },
    {
      "entropy": 5.633113861083984,
      "epoch": 0.3610192569539,
      "grad_norm": 1.40625,
      "learning_rate": 0.000499096415289178,
      "loss": 5.6033,
      "mean_token_accuracy": 0.15289738029241562,
      "num_tokens": 8061692.0,
      "step": 4640
    },
    {
      "entropy": 5.604199457168579,
      "epoch": 0.3614082863256176,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004990939321945948,
      "loss": 5.5536,
      "mean_token_accuracy": 0.15800046771764756,
      "num_tokens": 8069564.0,
      "step": 4645
    },
    {
      "entropy": 5.718245315551758,
      "epoch": 0.36179731569733514,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004990914456997427,
      "loss": 5.6435,
      "mean_token_accuracy": 0.14634049832820892,
      "num_tokens": 8078490.0,
      "step": 4650
    },
    {
      "entropy": 5.722595119476319,
      "epoch": 0.36218634506905273,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004990889558046598,
      "loss": 5.7247,
      "mean_token_accuracy": 0.14826465845108033,
      "num_tokens": 8086858.0,
      "step": 4655
    },
    {
      "entropy": 5.757637453079224,
      "epoch": 0.36257537444077026,
      "grad_norm": 1.375,
      "learning_rate": 0.0004990864625093836,
      "loss": 5.6636,
      "mean_token_accuracy": 0.1571757525205612,
      "num_tokens": 8094542.0,
      "step": 4660
    },
    {
      "entropy": 5.779032802581787,
      "epoch": 0.36296440381248785,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004990839658139519,
      "loss": 5.7169,
      "mean_token_accuracy": 0.15186462104320525,
      "num_tokens": 8102756.0,
      "step": 4665
    },
    {
      "entropy": 5.825753402709961,
      "epoch": 0.36335343318420543,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004990814657184028,
      "loss": 5.7476,
      "mean_token_accuracy": 0.15559100806713105,
      "num_tokens": 8111258.0,
      "step": 4670
    },
    {
      "entropy": 5.857345867156982,
      "epoch": 0.36374246255592296,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004990789622227741,
      "loss": 5.8447,
      "mean_token_accuracy": 0.144379186630249,
      "num_tokens": 8119624.0,
      "step": 4675
    },
    {
      "entropy": 5.7407505989074705,
      "epoch": 0.36413149192764055,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004990764553271038,
      "loss": 5.6134,
      "mean_token_accuracy": 0.15366872400045395,
      "num_tokens": 8128156.0,
      "step": 4680
    },
    {
      "entropy": 5.72602686882019,
      "epoch": 0.3645205212993581,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004990739450314299,
      "loss": 5.6786,
      "mean_token_accuracy": 0.14902170971035958,
      "num_tokens": 8136080.0,
      "step": 4685
    },
    {
      "entropy": 5.751847553253174,
      "epoch": 0.36490955067107567,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004990714313357906,
      "loss": 5.6585,
      "mean_token_accuracy": 0.15590455532073974,
      "num_tokens": 8144742.0,
      "step": 4690
    },
    {
      "entropy": 5.737034893035888,
      "epoch": 0.36529858004279325,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000499068914240224,
      "loss": 5.7293,
      "mean_token_accuracy": 0.15128549188375473,
      "num_tokens": 8153377.0,
      "step": 4695
    },
    {
      "entropy": 5.796698474884034,
      "epoch": 0.3656876094145108,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004990663937447682,
      "loss": 5.6537,
      "mean_token_accuracy": 0.15566126704216005,
      "num_tokens": 8162169.0,
      "step": 4700
    },
    {
      "entropy": 5.7169088363647464,
      "epoch": 0.36607663878622837,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004990638698494615,
      "loss": 5.6176,
      "mean_token_accuracy": 0.15707356631755828,
      "num_tokens": 8170727.0,
      "step": 4705
    },
    {
      "entropy": 5.605865859985352,
      "epoch": 0.3664656681579459,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004990613425543423,
      "loss": 5.5769,
      "mean_token_accuracy": 0.15790215730667115,
      "num_tokens": 8179013.0,
      "step": 4710
    },
    {
      "entropy": 5.608945417404175,
      "epoch": 0.3668546975296635,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004990588118594487,
      "loss": 5.5506,
      "mean_token_accuracy": 0.16253751665353774,
      "num_tokens": 8187030.0,
      "step": 4715
    },
    {
      "entropy": 5.626094484329224,
      "epoch": 0.3672437269013811,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004990562777648194,
      "loss": 5.5305,
      "mean_token_accuracy": 0.16110363602638245,
      "num_tokens": 8195073.0,
      "step": 4720
    },
    {
      "entropy": 5.708602380752564,
      "epoch": 0.3676327562730986,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004990537402704928,
      "loss": 5.6886,
      "mean_token_accuracy": 0.15165950506925582,
      "num_tokens": 8203626.0,
      "step": 4725
    },
    {
      "entropy": 5.740704441070557,
      "epoch": 0.3680217856448162,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004990511993765071,
      "loss": 5.6818,
      "mean_token_accuracy": 0.15222800001502038,
      "num_tokens": 8212750.0,
      "step": 4730
    },
    {
      "entropy": 5.796080303192139,
      "epoch": 0.3684108150165337,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004990486550829011,
      "loss": 5.6619,
      "mean_token_accuracy": 0.14802802354097366,
      "num_tokens": 8221542.0,
      "step": 4735
    },
    {
      "entropy": 5.622773456573486,
      "epoch": 0.3687998443882513,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004990461073897134,
      "loss": 5.6049,
      "mean_token_accuracy": 0.14905112236738205,
      "num_tokens": 8230574.0,
      "step": 4740
    },
    {
      "entropy": 5.730619859695435,
      "epoch": 0.3691888737599689,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004990435562969827,
      "loss": 5.6972,
      "mean_token_accuracy": 0.14874366521835328,
      "num_tokens": 8239400.0,
      "step": 4745
    },
    {
      "entropy": 5.72901029586792,
      "epoch": 0.3695779031316864,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004990410018047475,
      "loss": 5.5251,
      "mean_token_accuracy": 0.1579698383808136,
      "num_tokens": 8247537.0,
      "step": 4750
    },
    {
      "entropy": 5.6703370094299315,
      "epoch": 0.369966932503404,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004990384439130467,
      "loss": 5.6396,
      "mean_token_accuracy": 0.15181135013699532,
      "num_tokens": 8256551.0,
      "step": 4755
    },
    {
      "entropy": 5.7400750637054445,
      "epoch": 0.3703559618751216,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004990358826219192,
      "loss": 5.6844,
      "mean_token_accuracy": 0.1494295671582222,
      "num_tokens": 8265492.0,
      "step": 4760
    },
    {
      "entropy": 5.737407207489014,
      "epoch": 0.37074499124683913,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004990333179314038,
      "loss": 5.6124,
      "mean_token_accuracy": 0.15380854159593582,
      "num_tokens": 8273798.0,
      "step": 4765
    },
    {
      "entropy": 5.640963745117188,
      "epoch": 0.3711340206185567,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004990307498415393,
      "loss": 5.5995,
      "mean_token_accuracy": 0.1561552718281746,
      "num_tokens": 8281998.0,
      "step": 4770
    },
    {
      "entropy": 5.725772285461426,
      "epoch": 0.37152304999027425,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004990281783523648,
      "loss": 5.7066,
      "mean_token_accuracy": 0.14791832715272904,
      "num_tokens": 8289833.0,
      "step": 4775
    },
    {
      "entropy": 5.670173215866089,
      "epoch": 0.37191207936199183,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004990256034639192,
      "loss": 5.6611,
      "mean_token_accuracy": 0.1593535929918289,
      "num_tokens": 8297562.0,
      "step": 4780
    },
    {
      "entropy": 5.801353979110718,
      "epoch": 0.3723011087337094,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004990230251762418,
      "loss": 5.6676,
      "mean_token_accuracy": 0.15265004262328147,
      "num_tokens": 8305262.0,
      "step": 4785
    },
    {
      "entropy": 5.720134830474853,
      "epoch": 0.37269013810542695,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004990204434893713,
      "loss": 5.5911,
      "mean_token_accuracy": 0.1556418776512146,
      "num_tokens": 8313949.0,
      "step": 4790
    },
    {
      "entropy": 5.650853681564331,
      "epoch": 0.37307916747714454,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004990178584033474,
      "loss": 5.63,
      "mean_token_accuracy": 0.153776678442955,
      "num_tokens": 8322801.0,
      "step": 4795
    },
    {
      "entropy": 5.779564762115479,
      "epoch": 0.37346819684886207,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004990152699182089,
      "loss": 5.62,
      "mean_token_accuracy": 0.1598824217915535,
      "num_tokens": 8331118.0,
      "step": 4800
    },
    {
      "entropy": 5.7409477710723875,
      "epoch": 0.37385722622057965,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004990126780339953,
      "loss": 5.6884,
      "mean_token_accuracy": 0.15528960078954696,
      "num_tokens": 8340498.0,
      "step": 4805
    },
    {
      "entropy": 5.785793113708496,
      "epoch": 0.37424625559229724,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004990100827507459,
      "loss": 5.8207,
      "mean_token_accuracy": 0.14700310155749322,
      "num_tokens": 8349235.0,
      "step": 4810
    },
    {
      "entropy": 5.755542755126953,
      "epoch": 0.37463528496401477,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004990074840684999,
      "loss": 5.6936,
      "mean_token_accuracy": 0.1547679141163826,
      "num_tokens": 8358484.0,
      "step": 4815
    },
    {
      "entropy": 5.776040840148926,
      "epoch": 0.37502431433573236,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004990048819872969,
      "loss": 5.6245,
      "mean_token_accuracy": 0.15924780070781708,
      "num_tokens": 8366894.0,
      "step": 4820
    },
    {
      "entropy": 5.662250757217407,
      "epoch": 0.3754133437074499,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004990022765071765,
      "loss": 5.631,
      "mean_token_accuracy": 0.1452302061021328,
      "num_tokens": 8375271.0,
      "step": 4825
    },
    {
      "entropy": 5.68854022026062,
      "epoch": 0.3758023730791675,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000498999667628178,
      "loss": 5.5764,
      "mean_token_accuracy": 0.15381914377212524,
      "num_tokens": 8383714.0,
      "step": 4830
    },
    {
      "entropy": 5.651234245300293,
      "epoch": 0.37619140245088506,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004989970553503411,
      "loss": 5.6853,
      "mean_token_accuracy": 0.15144770443439484,
      "num_tokens": 8392022.0,
      "step": 4835
    },
    {
      "entropy": 5.759373378753662,
      "epoch": 0.3765804318226026,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004989944396737054,
      "loss": 5.65,
      "mean_token_accuracy": 0.14829441905021667,
      "num_tokens": 8401282.0,
      "step": 4840
    },
    {
      "entropy": 5.729718780517578,
      "epoch": 0.3769694611943202,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004989918205983106,
      "loss": 5.7199,
      "mean_token_accuracy": 0.14889771938323976,
      "num_tokens": 8411147.0,
      "step": 4845
    },
    {
      "entropy": 5.8805125713348385,
      "epoch": 0.37735849056603776,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004989891981241964,
      "loss": 5.7159,
      "mean_token_accuracy": 0.1495855137705803,
      "num_tokens": 8420434.0,
      "step": 4850
    },
    {
      "entropy": 5.745460796356201,
      "epoch": 0.3777475199377553,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004989865722514027,
      "loss": 5.6811,
      "mean_token_accuracy": 0.1509254492819309,
      "num_tokens": 8430022.0,
      "step": 4855
    },
    {
      "entropy": 5.540482044219971,
      "epoch": 0.3781365493094729,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004989839429799692,
      "loss": 5.4636,
      "mean_token_accuracy": 0.16462698578834534,
      "num_tokens": 8438898.0,
      "step": 4860
    },
    {
      "entropy": 5.618959999084472,
      "epoch": 0.3785255786811904,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000498981310309936,
      "loss": 5.5649,
      "mean_token_accuracy": 0.15923128873109818,
      "num_tokens": 8446931.0,
      "step": 4865
    },
    {
      "entropy": 5.75497031211853,
      "epoch": 0.378914608052908,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004989786742413428,
      "loss": 5.638,
      "mean_token_accuracy": 0.15647049695253373,
      "num_tokens": 8455515.0,
      "step": 4870
    },
    {
      "entropy": 5.714741182327271,
      "epoch": 0.3793036374246256,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004989760347742298,
      "loss": 5.639,
      "mean_token_accuracy": 0.15014245137572288,
      "num_tokens": 8464135.0,
      "step": 4875
    },
    {
      "entropy": 5.76575117111206,
      "epoch": 0.3796926667963431,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004989733919086369,
      "loss": 5.7329,
      "mean_token_accuracy": 0.1491445854306221,
      "num_tokens": 8473222.0,
      "step": 4880
    },
    {
      "entropy": 5.757653379440308,
      "epoch": 0.3800816961680607,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004989707456446043,
      "loss": 5.7107,
      "mean_token_accuracy": 0.15802261978387833,
      "num_tokens": 8481966.0,
      "step": 4885
    },
    {
      "entropy": 5.763038921356201,
      "epoch": 0.38047072553977823,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004989680959821721,
      "loss": 5.7354,
      "mean_token_accuracy": 0.1485402375459671,
      "num_tokens": 8491276.0,
      "step": 4890
    },
    {
      "entropy": 5.71974368095398,
      "epoch": 0.3808597549114958,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004989654429213805,
      "loss": 5.6814,
      "mean_token_accuracy": 0.15291936323046684,
      "num_tokens": 8499926.0,
      "step": 4895
    },
    {
      "entropy": 5.741155004501342,
      "epoch": 0.3812487842832134,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004989627864622699,
      "loss": 5.5594,
      "mean_token_accuracy": 0.15729069113731384,
      "num_tokens": 8508544.0,
      "step": 4900
    },
    {
      "entropy": 5.758600425720215,
      "epoch": 0.38163781365493094,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004989601266048804,
      "loss": 5.682,
      "mean_token_accuracy": 0.1453734003007412,
      "num_tokens": 8517326.0,
      "step": 4905
    },
    {
      "entropy": 5.64643874168396,
      "epoch": 0.3820268430266485,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004989574633492525,
      "loss": 5.6628,
      "mean_token_accuracy": 0.15118559151887895,
      "num_tokens": 8526218.0,
      "step": 4910
    },
    {
      "entropy": 5.787919235229492,
      "epoch": 0.38241587239836605,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004989547966954266,
      "loss": 5.6956,
      "mean_token_accuracy": 0.1465151771903038,
      "num_tokens": 8534708.0,
      "step": 4915
    },
    {
      "entropy": 5.723714780807495,
      "epoch": 0.38280490177008364,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000498952126643443,
      "loss": 5.6392,
      "mean_token_accuracy": 0.15696539580821992,
      "num_tokens": 8543302.0,
      "step": 4920
    },
    {
      "entropy": 5.613685512542725,
      "epoch": 0.3831939311418012,
      "grad_norm": 1.25,
      "learning_rate": 0.0004989494531933424,
      "loss": 5.5588,
      "mean_token_accuracy": 0.15237713605165482,
      "num_tokens": 8552586.0,
      "step": 4925
    },
    {
      "entropy": 5.64679446220398,
      "epoch": 0.38358296051351876,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004989467763451652,
      "loss": 5.5374,
      "mean_token_accuracy": 0.15823079496622086,
      "num_tokens": 8561720.0,
      "step": 4930
    },
    {
      "entropy": 5.698852682113648,
      "epoch": 0.38397198988523634,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004989440960989523,
      "loss": 5.529,
      "mean_token_accuracy": 0.15683713406324387,
      "num_tokens": 8570510.0,
      "step": 4935
    },
    {
      "entropy": 5.667657041549683,
      "epoch": 0.3843610192569539,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000498941412454744,
      "loss": 5.6631,
      "mean_token_accuracy": 0.15189100801944733,
      "num_tokens": 8579986.0,
      "step": 4940
    },
    {
      "entropy": 5.660448598861694,
      "epoch": 0.38475004862867146,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004989387254125813,
      "loss": 5.6278,
      "mean_token_accuracy": 0.14973406940698625,
      "num_tokens": 8588507.0,
      "step": 4945
    },
    {
      "entropy": 5.676297044754028,
      "epoch": 0.38513907800038905,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004989360349725049,
      "loss": 5.5545,
      "mean_token_accuracy": 0.1551334172487259,
      "num_tokens": 8596651.0,
      "step": 4950
    },
    {
      "entropy": 5.7129144191741945,
      "epoch": 0.3855281073721066,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004989333411345555,
      "loss": 5.5836,
      "mean_token_accuracy": 0.15729959458112716,
      "num_tokens": 8605080.0,
      "step": 4955
    },
    {
      "entropy": 5.641625881195068,
      "epoch": 0.38591713674382416,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004989306438987742,
      "loss": 5.548,
      "mean_token_accuracy": 0.1537535160779953,
      "num_tokens": 8613017.0,
      "step": 4960
    },
    {
      "entropy": 5.702543449401856,
      "epoch": 0.38630616611554175,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004989279432652018,
      "loss": 5.7241,
      "mean_token_accuracy": 0.1546338826417923,
      "num_tokens": 8621954.0,
      "step": 4965
    },
    {
      "entropy": 5.809708023071289,
      "epoch": 0.3866951954872593,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004989252392338791,
      "loss": 5.6677,
      "mean_token_accuracy": 0.1566025510430336,
      "num_tokens": 8630213.0,
      "step": 4970
    },
    {
      "entropy": 5.784931945800781,
      "epoch": 0.38708422485897687,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004989225318048475,
      "loss": 5.6861,
      "mean_token_accuracy": 0.15659048557281494,
      "num_tokens": 8638360.0,
      "step": 4975
    },
    {
      "entropy": 5.757261753082275,
      "epoch": 0.3874732542306944,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004989198209781478,
      "loss": 5.723,
      "mean_token_accuracy": 0.15046840906143188,
      "num_tokens": 8647778.0,
      "step": 4980
    },
    {
      "entropy": 5.73045597076416,
      "epoch": 0.387862283602412,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004989171067538211,
      "loss": 5.6507,
      "mean_token_accuracy": 0.15228100568056108,
      "num_tokens": 8655584.0,
      "step": 4985
    },
    {
      "entropy": 5.673905515670777,
      "epoch": 0.38825131297412957,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000498914389131909,
      "loss": 5.6874,
      "mean_token_accuracy": 0.1532427988946438,
      "num_tokens": 8664159.0,
      "step": 4990
    },
    {
      "entropy": 5.718062448501587,
      "epoch": 0.3886403423458471,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004989116681124523,
      "loss": 5.5877,
      "mean_token_accuracy": 0.1540216326713562,
      "num_tokens": 8672488.0,
      "step": 4995
    },
    {
      "entropy": 5.754921197891235,
      "epoch": 0.3890293717175647,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004989089436954924,
      "loss": 5.7203,
      "mean_token_accuracy": 0.1480067603290081,
      "num_tokens": 8681492.0,
      "step": 5000
    },
    {
      "entropy": 5.677997732162476,
      "epoch": 0.3894184010892822,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004989062158810706,
      "loss": 5.606,
      "mean_token_accuracy": 0.15589867532253265,
      "num_tokens": 8690453.0,
      "step": 5005
    },
    {
      "entropy": 5.791937589645386,
      "epoch": 0.3898074304609998,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004989034846692284,
      "loss": 5.6959,
      "mean_token_accuracy": 0.15180487632751466,
      "num_tokens": 8699349.0,
      "step": 5010
    },
    {
      "entropy": 5.6987401962280275,
      "epoch": 0.3901964598327174,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004989007500600073,
      "loss": 5.6942,
      "mean_token_accuracy": 0.1500448077917099,
      "num_tokens": 8707512.0,
      "step": 5015
    },
    {
      "entropy": 5.793087816238403,
      "epoch": 0.3905854892044349,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004988980120534486,
      "loss": 5.717,
      "mean_token_accuracy": 0.14697987586259842,
      "num_tokens": 8715847.0,
      "step": 5020
    },
    {
      "entropy": 5.72705192565918,
      "epoch": 0.3909745185761525,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000498895270649594,
      "loss": 5.6111,
      "mean_token_accuracy": 0.1572824612259865,
      "num_tokens": 8725548.0,
      "step": 5025
    },
    {
      "entropy": 5.638960075378418,
      "epoch": 0.39136354794787004,
      "grad_norm": 1.9921875,
      "learning_rate": 0.000498892525848485,
      "loss": 5.564,
      "mean_token_accuracy": 0.16159805804491043,
      "num_tokens": 8733935.0,
      "step": 5030
    },
    {
      "entropy": 5.7204138278961185,
      "epoch": 0.3917525773195876,
      "grad_norm": 1.25,
      "learning_rate": 0.0004988897776501633,
      "loss": 5.6876,
      "mean_token_accuracy": 0.15322512090206147,
      "num_tokens": 8743110.0,
      "step": 5035
    },
    {
      "entropy": 5.641509246826172,
      "epoch": 0.3921416066913052,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004988870260546705,
      "loss": 5.5112,
      "mean_token_accuracy": 0.1590324655175209,
      "num_tokens": 8751166.0,
      "step": 5040
    },
    {
      "entropy": 5.681423711776733,
      "epoch": 0.39253063606302274,
      "grad_norm": 1.25,
      "learning_rate": 0.0004988842710620486,
      "loss": 5.7162,
      "mean_token_accuracy": 0.15126417875289916,
      "num_tokens": 8759929.0,
      "step": 5045
    },
    {
      "entropy": 5.837319183349609,
      "epoch": 0.39291966543474033,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004988815126723391,
      "loss": 5.7138,
      "mean_token_accuracy": 0.15048869475722312,
      "num_tokens": 8768541.0,
      "step": 5050
    },
    {
      "entropy": 5.765131187438965,
      "epoch": 0.3933086948064579,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004988787508855841,
      "loss": 5.6237,
      "mean_token_accuracy": 0.153934982419014,
      "num_tokens": 8776803.0,
      "step": 5055
    },
    {
      "entropy": 5.62071008682251,
      "epoch": 0.39369772417817545,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004988759857018253,
      "loss": 5.5955,
      "mean_token_accuracy": 0.1592630684375763,
      "num_tokens": 8784773.0,
      "step": 5060
    },
    {
      "entropy": 5.751138544082641,
      "epoch": 0.39408675354989303,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004988732171211048,
      "loss": 5.655,
      "mean_token_accuracy": 0.15452175438404084,
      "num_tokens": 8793797.0,
      "step": 5065
    },
    {
      "entropy": 5.625418424606323,
      "epoch": 0.39447578292161056,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004988704451434644,
      "loss": 5.5384,
      "mean_token_accuracy": 0.15230093747377396,
      "num_tokens": 8802366.0,
      "step": 5070
    },
    {
      "entropy": 5.739065170288086,
      "epoch": 0.39486481229332815,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004988676697689465,
      "loss": 5.6729,
      "mean_token_accuracy": 0.15218451768159866,
      "num_tokens": 8810975.0,
      "step": 5075
    },
    {
      "entropy": 5.769051790237427,
      "epoch": 0.39525384166504574,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000498864890997593,
      "loss": 5.6945,
      "mean_token_accuracy": 0.1473867617547512,
      "num_tokens": 8819137.0,
      "step": 5080
    },
    {
      "entropy": 5.685499238967895,
      "epoch": 0.39564287103676327,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004988621088294461,
      "loss": 5.6939,
      "mean_token_accuracy": 0.1510564461350441,
      "num_tokens": 8827676.0,
      "step": 5085
    },
    {
      "entropy": 5.709568500518799,
      "epoch": 0.39603190040848085,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000498859323264548,
      "loss": 5.6426,
      "mean_token_accuracy": 0.15703052133321763,
      "num_tokens": 8836005.0,
      "step": 5090
    },
    {
      "entropy": 5.636990165710449,
      "epoch": 0.3964209297801984,
      "grad_norm": 1.421875,
      "learning_rate": 0.000498856534302941,
      "loss": 5.5295,
      "mean_token_accuracy": 0.1587708003818989,
      "num_tokens": 8844153.0,
      "step": 5095
    },
    {
      "entropy": 5.71082124710083,
      "epoch": 0.39680995915191597,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004988537419446673,
      "loss": 5.644,
      "mean_token_accuracy": 0.153173828125,
      "num_tokens": 8852520.0,
      "step": 5100
    },
    {
      "entropy": 5.714675331115723,
      "epoch": 0.39719898852363356,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004988509461897694,
      "loss": 5.4872,
      "mean_token_accuracy": 0.1576193779706955,
      "num_tokens": 8860609.0,
      "step": 5105
    },
    {
      "entropy": 5.648047876358032,
      "epoch": 0.3975880178953511,
      "grad_norm": 1.375,
      "learning_rate": 0.0004988481470382897,
      "loss": 5.6664,
      "mean_token_accuracy": 0.1524683952331543,
      "num_tokens": 8869185.0,
      "step": 5110
    },
    {
      "entropy": 5.684601879119873,
      "epoch": 0.3979770472670687,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004988453444902708,
      "loss": 5.537,
      "mean_token_accuracy": 0.152810937166214,
      "num_tokens": 8877758.0,
      "step": 5115
    },
    {
      "entropy": 5.723855924606323,
      "epoch": 0.3983660766387862,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004988425385457549,
      "loss": 5.7168,
      "mean_token_accuracy": 0.15099436789751053,
      "num_tokens": 8887236.0,
      "step": 5120
    },
    {
      "entropy": 5.716874504089356,
      "epoch": 0.3987551060105038,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004988397292047848,
      "loss": 5.6896,
      "mean_token_accuracy": 0.15267200022935867,
      "num_tokens": 8895808.0,
      "step": 5125
    },
    {
      "entropy": 5.7161201477050785,
      "epoch": 0.3991441353822214,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004988369164674032,
      "loss": 5.6626,
      "mean_token_accuracy": 0.1510595954954624,
      "num_tokens": 8904091.0,
      "step": 5130
    },
    {
      "entropy": 5.5925610065460205,
      "epoch": 0.3995331647539389,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004988341003336526,
      "loss": 5.4659,
      "mean_token_accuracy": 0.15727559775114058,
      "num_tokens": 8912046.0,
      "step": 5135
    },
    {
      "entropy": 5.770779085159302,
      "epoch": 0.3999221941256565,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004988312808035757,
      "loss": 5.681,
      "mean_token_accuracy": 0.14915614053606988,
      "num_tokens": 8921060.0,
      "step": 5140
    },
    {
      "entropy": 5.727402973175049,
      "epoch": 0.400311223497374,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004988284578772155,
      "loss": 5.629,
      "mean_token_accuracy": 0.15255586802959442,
      "num_tokens": 8930757.0,
      "step": 5145
    },
    {
      "entropy": 5.825820398330689,
      "epoch": 0.4007002528690916,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004988256315546146,
      "loss": 5.7951,
      "mean_token_accuracy": 0.15042151659727096,
      "num_tokens": 8939544.0,
      "step": 5150
    },
    {
      "entropy": 5.696900844573975,
      "epoch": 0.4010892822408092,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004988228018358161,
      "loss": 5.6018,
      "mean_token_accuracy": 0.15607771426439285,
      "num_tokens": 8949347.0,
      "step": 5155
    },
    {
      "entropy": 5.725059270858765,
      "epoch": 0.40147831161252673,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004988199687208628,
      "loss": 5.6584,
      "mean_token_accuracy": 0.15238844454288483,
      "num_tokens": 8957697.0,
      "step": 5160
    },
    {
      "entropy": 5.740241527557373,
      "epoch": 0.4018673409842443,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004988171322097977,
      "loss": 5.6755,
      "mean_token_accuracy": 0.14611184522509574,
      "num_tokens": 8966107.0,
      "step": 5165
    },
    {
      "entropy": 5.675651693344117,
      "epoch": 0.4022563703559619,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004988142923026638,
      "loss": 5.5825,
      "mean_token_accuracy": 0.15178962647914887,
      "num_tokens": 8974128.0,
      "step": 5170
    },
    {
      "entropy": 5.6408562660217285,
      "epoch": 0.40264539972767943,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004988114489995044,
      "loss": 5.5668,
      "mean_token_accuracy": 0.15450093746185303,
      "num_tokens": 8982564.0,
      "step": 5175
    },
    {
      "entropy": 5.664896154403687,
      "epoch": 0.403034429099397,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004988086023003624,
      "loss": 5.6069,
      "mean_token_accuracy": 0.1550554156303406,
      "num_tokens": 8990924.0,
      "step": 5180
    },
    {
      "entropy": 5.820183563232422,
      "epoch": 0.40342345847111455,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004988057522052811,
      "loss": 5.698,
      "mean_token_accuracy": 0.14887791126966476,
      "num_tokens": 8999827.0,
      "step": 5185
    },
    {
      "entropy": 5.7121435642242435,
      "epoch": 0.40381248784283214,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004988028987143037,
      "loss": 5.5321,
      "mean_token_accuracy": 0.1553960233926773,
      "num_tokens": 9007885.0,
      "step": 5190
    },
    {
      "entropy": 5.6945513725280765,
      "epoch": 0.4042015172145497,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004988000418274736,
      "loss": 5.6351,
      "mean_token_accuracy": 0.15085338950157165,
      "num_tokens": 9016752.0,
      "step": 5195
    },
    {
      "entropy": 5.713669633865356,
      "epoch": 0.40459054658626725,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004987971815448341,
      "loss": 5.6434,
      "mean_token_accuracy": 0.15495601445436477,
      "num_tokens": 9025547.0,
      "step": 5200
    },
    {
      "entropy": 5.741844606399536,
      "epoch": 0.40497957595798484,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004987943178664285,
      "loss": 5.6778,
      "mean_token_accuracy": 0.15183082818984986,
      "num_tokens": 9034711.0,
      "step": 5205
    },
    {
      "entropy": 5.718159914016724,
      "epoch": 0.40536860532970237,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004987914507923004,
      "loss": 5.59,
      "mean_token_accuracy": 0.15771124362945557,
      "num_tokens": 9042565.0,
      "step": 5210
    },
    {
      "entropy": 5.592754888534546,
      "epoch": 0.40575763470141996,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004987885803224931,
      "loss": 5.6583,
      "mean_token_accuracy": 0.14471440464258195,
      "num_tokens": 9051590.0,
      "step": 5215
    },
    {
      "entropy": 5.819624614715576,
      "epoch": 0.40614666407313754,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004987857064570505,
      "loss": 5.7309,
      "mean_token_accuracy": 0.15148103088140488,
      "num_tokens": 9060415.0,
      "step": 5220
    },
    {
      "entropy": 5.640044164657593,
      "epoch": 0.4065356934448551,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004987828291960158,
      "loss": 5.516,
      "mean_token_accuracy": 0.15662309378385544,
      "num_tokens": 9069492.0,
      "step": 5225
    },
    {
      "entropy": 5.608748483657837,
      "epoch": 0.40692472281657266,
      "grad_norm": 1.1875,
      "learning_rate": 0.000498779948539433,
      "loss": 5.5735,
      "mean_token_accuracy": 0.16103582680225373,
      "num_tokens": 9078183.0,
      "step": 5230
    },
    {
      "entropy": 5.808728265762329,
      "epoch": 0.4073137521882902,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004987770644873455,
      "loss": 5.7835,
      "mean_token_accuracy": 0.14806534647941588,
      "num_tokens": 9086661.0,
      "step": 5235
    },
    {
      "entropy": 5.690426826477051,
      "epoch": 0.4077027815600078,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004987741770397974,
      "loss": 5.5175,
      "mean_token_accuracy": 0.1621639087796211,
      "num_tokens": 9095449.0,
      "step": 5240
    },
    {
      "entropy": 5.665590906143189,
      "epoch": 0.40809181093172536,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004987712861968323,
      "loss": 5.6577,
      "mean_token_accuracy": 0.15061125010252,
      "num_tokens": 9103347.0,
      "step": 5245
    },
    {
      "entropy": 5.572380304336548,
      "epoch": 0.4084808403034429,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000498768391958494,
      "loss": 5.5632,
      "mean_token_accuracy": 0.1605206623673439,
      "num_tokens": 9112382.0,
      "step": 5250
    },
    {
      "entropy": 5.73834261894226,
      "epoch": 0.4088698696751605,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004987654943248266,
      "loss": 5.5031,
      "mean_token_accuracy": 0.15901872366666794,
      "num_tokens": 9120628.0,
      "step": 5255
    },
    {
      "entropy": 5.75648136138916,
      "epoch": 0.40925889904687807,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004987625932958739,
      "loss": 5.6933,
      "mean_token_accuracy": 0.15073462724685668,
      "num_tokens": 9130280.0,
      "step": 5260
    },
    {
      "entropy": 5.856960916519165,
      "epoch": 0.4096479284185956,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004987596888716801,
      "loss": 5.7914,
      "mean_token_accuracy": 0.1458967387676239,
      "num_tokens": 9139720.0,
      "step": 5265
    },
    {
      "entropy": 5.688859844207764,
      "epoch": 0.4100369577903132,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004987567810522892,
      "loss": 5.7051,
      "mean_token_accuracy": 0.149783219397068,
      "num_tokens": 9148537.0,
      "step": 5270
    },
    {
      "entropy": 5.752209806442261,
      "epoch": 0.4104259871620307,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004987538698377451,
      "loss": 5.7694,
      "mean_token_accuracy": 0.15261628925800325,
      "num_tokens": 9156965.0,
      "step": 5275
    },
    {
      "entropy": 5.672241067886352,
      "epoch": 0.4108150165337483,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004987509552280924,
      "loss": 5.4953,
      "mean_token_accuracy": 0.1591440185904503,
      "num_tokens": 9164982.0,
      "step": 5280
    },
    {
      "entropy": 5.738524103164673,
      "epoch": 0.4112040459054659,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000498748037223375,
      "loss": 5.7207,
      "mean_token_accuracy": 0.14531775191426277,
      "num_tokens": 9173383.0,
      "step": 5285
    },
    {
      "entropy": 5.788292264938354,
      "epoch": 0.4115930752771834,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004987451158236372,
      "loss": 5.7358,
      "mean_token_accuracy": 0.15552923828363419,
      "num_tokens": 9182743.0,
      "step": 5290
    },
    {
      "entropy": 5.747622489929199,
      "epoch": 0.411982104648901,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004987421910289234,
      "loss": 5.6239,
      "mean_token_accuracy": 0.15142476558685303,
      "num_tokens": 9191770.0,
      "step": 5295
    },
    {
      "entropy": 5.5697931289672855,
      "epoch": 0.41237113402061853,
      "grad_norm": 1.25,
      "learning_rate": 0.0004987392628392781,
      "loss": 5.5785,
      "mean_token_accuracy": 0.15786964744329451,
      "num_tokens": 9200885.0,
      "step": 5300
    },
    {
      "entropy": 5.832540702819824,
      "epoch": 0.4127601633923361,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004987363312547456,
      "loss": 5.6686,
      "mean_token_accuracy": 0.1493069648742676,
      "num_tokens": 9210114.0,
      "step": 5305
    },
    {
      "entropy": 5.745130014419556,
      "epoch": 0.4131491927640537,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004987333962753703,
      "loss": 5.6199,
      "mean_token_accuracy": 0.15895049124956132,
      "num_tokens": 9218608.0,
      "step": 5310
    },
    {
      "entropy": 5.698068571090698,
      "epoch": 0.41353822213577124,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004987304579011967,
      "loss": 5.6659,
      "mean_token_accuracy": 0.15064293891191483,
      "num_tokens": 9227737.0,
      "step": 5315
    },
    {
      "entropy": 5.731441307067871,
      "epoch": 0.4139272515074888,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004987275161322697,
      "loss": 5.5737,
      "mean_token_accuracy": 0.15715447142720224,
      "num_tokens": 9236589.0,
      "step": 5320
    },
    {
      "entropy": 5.67019567489624,
      "epoch": 0.41431628087920636,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004987245709686337,
      "loss": 5.6204,
      "mean_token_accuracy": 0.16002295315265655,
      "num_tokens": 9245313.0,
      "step": 5325
    },
    {
      "entropy": 5.662066316604614,
      "epoch": 0.41470531025092394,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004987216224103334,
      "loss": 5.655,
      "mean_token_accuracy": 0.15002036690711976,
      "num_tokens": 9254001.0,
      "step": 5330
    },
    {
      "entropy": 5.80666069984436,
      "epoch": 0.41509433962264153,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004987186704574136,
      "loss": 5.7102,
      "mean_token_accuracy": 0.14658118188381195,
      "num_tokens": 9262485.0,
      "step": 5335
    },
    {
      "entropy": 5.689135217666626,
      "epoch": 0.41548336899435906,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004987157151099189,
      "loss": 5.5907,
      "mean_token_accuracy": 0.1536131516098976,
      "num_tokens": 9271490.0,
      "step": 5340
    },
    {
      "entropy": 5.625450229644775,
      "epoch": 0.41587239836607665,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004987127563678945,
      "loss": 5.6282,
      "mean_token_accuracy": 0.15292252898216246,
      "num_tokens": 9280595.0,
      "step": 5345
    },
    {
      "entropy": 5.719094228744507,
      "epoch": 0.4162614277377942,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004987097942313852,
      "loss": 5.5414,
      "mean_token_accuracy": 0.1592733532190323,
      "num_tokens": 9289260.0,
      "step": 5350
    },
    {
      "entropy": 5.755507230758667,
      "epoch": 0.41665045710951176,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004987068287004355,
      "loss": 5.7281,
      "mean_token_accuracy": 0.15436695963144303,
      "num_tokens": 9297842.0,
      "step": 5355
    },
    {
      "entropy": 5.752491664886475,
      "epoch": 0.41703948648122935,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004987038597750909,
      "loss": 5.7035,
      "mean_token_accuracy": 0.1510893762111664,
      "num_tokens": 9306533.0,
      "step": 5360
    },
    {
      "entropy": 5.7559339046478275,
      "epoch": 0.4174285158529469,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004987008874553964,
      "loss": 5.5968,
      "mean_token_accuracy": 0.1522625654935837,
      "num_tokens": 9314534.0,
      "step": 5365
    },
    {
      "entropy": 5.749003839492798,
      "epoch": 0.41781754522466447,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004986979117413969,
      "loss": 5.758,
      "mean_token_accuracy": 0.15595654249191285,
      "num_tokens": 9323179.0,
      "step": 5370
    },
    {
      "entropy": 5.687869930267334,
      "epoch": 0.41820657459638205,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004986949326331376,
      "loss": 5.5471,
      "mean_token_accuracy": 0.16193525791168212,
      "num_tokens": 9331265.0,
      "step": 5375
    },
    {
      "entropy": 5.665004444122315,
      "epoch": 0.4185956039680996,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004986919501306638,
      "loss": 5.6089,
      "mean_token_accuracy": 0.15000245124101638,
      "num_tokens": 9340856.0,
      "step": 5380
    },
    {
      "entropy": 5.683504724502564,
      "epoch": 0.41898463333981717,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004986889642340207,
      "loss": 5.6271,
      "mean_token_accuracy": 0.15085209980607034,
      "num_tokens": 9350728.0,
      "step": 5385
    },
    {
      "entropy": 5.813549613952636,
      "epoch": 0.4193736627115347,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004986859749432536,
      "loss": 5.7624,
      "mean_token_accuracy": 0.14495969265699388,
      "num_tokens": 9360109.0,
      "step": 5390
    },
    {
      "entropy": 5.783225584030151,
      "epoch": 0.4197626920832523,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004986829822584078,
      "loss": 5.7253,
      "mean_token_accuracy": 0.1517239674925804,
      "num_tokens": 9369101.0,
      "step": 5395
    },
    {
      "entropy": 5.762203168869019,
      "epoch": 0.4201517214549699,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004986799861795289,
      "loss": 5.6917,
      "mean_token_accuracy": 0.14891715198755265,
      "num_tokens": 9377686.0,
      "step": 5400
    },
    {
      "entropy": 5.774033546447754,
      "epoch": 0.4205407508266874,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004986769867066622,
      "loss": 5.683,
      "mean_token_accuracy": 0.1555148705840111,
      "num_tokens": 9386439.0,
      "step": 5405
    },
    {
      "entropy": 5.68793888092041,
      "epoch": 0.420929780198405,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004986739838398532,
      "loss": 5.717,
      "mean_token_accuracy": 0.15431362837553025,
      "num_tokens": 9395086.0,
      "step": 5410
    },
    {
      "entropy": 5.714366340637207,
      "epoch": 0.4213188095701225,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004986709775791475,
      "loss": 5.5778,
      "mean_token_accuracy": 0.15415338277816773,
      "num_tokens": 9403762.0,
      "step": 5415
    },
    {
      "entropy": 5.662868070602417,
      "epoch": 0.4217078389418401,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004986679679245907,
      "loss": 5.6173,
      "mean_token_accuracy": 0.1550531730055809,
      "num_tokens": 9412386.0,
      "step": 5420
    },
    {
      "entropy": 5.671959304809571,
      "epoch": 0.4220968683135577,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004986649548762286,
      "loss": 5.5757,
      "mean_token_accuracy": 0.16350484043359756,
      "num_tokens": 9420538.0,
      "step": 5425
    },
    {
      "entropy": 5.661460494995117,
      "epoch": 0.4224858976852752,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004986619384341066,
      "loss": 5.5702,
      "mean_token_accuracy": 0.15350655913352967,
      "num_tokens": 9429083.0,
      "step": 5430
    },
    {
      "entropy": 5.733646297454834,
      "epoch": 0.4228749270569928,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004986589185982708,
      "loss": 5.6627,
      "mean_token_accuracy": 0.15776549130678177,
      "num_tokens": 9436892.0,
      "step": 5435
    },
    {
      "entropy": 5.726545667648315,
      "epoch": 0.42326395642871034,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004986558953687671,
      "loss": 5.6756,
      "mean_token_accuracy": 0.1516024351119995,
      "num_tokens": 9445267.0,
      "step": 5440
    },
    {
      "entropy": 5.721020984649658,
      "epoch": 0.42365298580042793,
      "grad_norm": 1.625,
      "learning_rate": 0.0004986528687456409,
      "loss": 5.685,
      "mean_token_accuracy": 0.15134571343660355,
      "num_tokens": 9454292.0,
      "step": 5445
    },
    {
      "entropy": 5.707912731170654,
      "epoch": 0.4240420151721455,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004986498387289384,
      "loss": 5.5852,
      "mean_token_accuracy": 0.16004859358072282,
      "num_tokens": 9463126.0,
      "step": 5450
    },
    {
      "entropy": 5.679943418502807,
      "epoch": 0.42443104454386305,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004986468053187058,
      "loss": 5.6632,
      "mean_token_accuracy": 0.15489592105150224,
      "num_tokens": 9471703.0,
      "step": 5455
    },
    {
      "entropy": 5.664609098434449,
      "epoch": 0.42482007391558063,
      "grad_norm": 1.5,
      "learning_rate": 0.0004986437685149887,
      "loss": 5.5365,
      "mean_token_accuracy": 0.1670185387134552,
      "num_tokens": 9480029.0,
      "step": 5460
    },
    {
      "entropy": 5.743315315246582,
      "epoch": 0.4252091032872982,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004986407283178334,
      "loss": 5.6452,
      "mean_token_accuracy": 0.15648874193429946,
      "num_tokens": 9488277.0,
      "step": 5465
    },
    {
      "entropy": 5.725664234161377,
      "epoch": 0.42559813265901575,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004986376847272861,
      "loss": 5.6974,
      "mean_token_accuracy": 0.1544747918844223,
      "num_tokens": 9496096.0,
      "step": 5470
    },
    {
      "entropy": 5.682655239105225,
      "epoch": 0.42598716203073334,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004986346377433929,
      "loss": 5.6116,
      "mean_token_accuracy": 0.16111418157815932,
      "num_tokens": 9504653.0,
      "step": 5475
    },
    {
      "entropy": 5.7783355712890625,
      "epoch": 0.42637619140245087,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004986315873661999,
      "loss": 5.5822,
      "mean_token_accuracy": 0.15282244384288787,
      "num_tokens": 9513088.0,
      "step": 5480
    },
    {
      "entropy": 5.619540166854859,
      "epoch": 0.42676522077416845,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004986285335957536,
      "loss": 5.5013,
      "mean_token_accuracy": 0.15677933245897294,
      "num_tokens": 9521714.0,
      "step": 5485
    },
    {
      "entropy": 5.613918161392212,
      "epoch": 0.42715425014588604,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004986254764321002,
      "loss": 5.6033,
      "mean_token_accuracy": 0.1575162246823311,
      "num_tokens": 9530629.0,
      "step": 5490
    },
    {
      "entropy": 5.701473379135132,
      "epoch": 0.42754327951760357,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004986224158752861,
      "loss": 5.6084,
      "mean_token_accuracy": 0.162319153547287,
      "num_tokens": 9539293.0,
      "step": 5495
    },
    {
      "entropy": 5.623651647567749,
      "epoch": 0.42793230888932116,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004986193519253578,
      "loss": 5.5598,
      "mean_token_accuracy": 0.15547969192266464,
      "num_tokens": 9547628.0,
      "step": 5500
    },
    {
      "entropy": 5.656365871429443,
      "epoch": 0.4283213382610387,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004986162845823618,
      "loss": 5.5529,
      "mean_token_accuracy": 0.16461351811885833,
      "num_tokens": 9557260.0,
      "step": 5505
    },
    {
      "entropy": 5.761644983291626,
      "epoch": 0.4287103676327563,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004986132138463446,
      "loss": 5.5998,
      "mean_token_accuracy": 0.1509658232331276,
      "num_tokens": 9565923.0,
      "step": 5510
    },
    {
      "entropy": 5.653045129776001,
      "epoch": 0.42909939700447386,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004986101397173528,
      "loss": 5.6036,
      "mean_token_accuracy": 0.15777654498815535,
      "num_tokens": 9574373.0,
      "step": 5515
    },
    {
      "entropy": 5.683512020111084,
      "epoch": 0.4294884263761914,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000498607062195433,
      "loss": 5.6989,
      "mean_token_accuracy": 0.15172723829746246,
      "num_tokens": 9583652.0,
      "step": 5520
    },
    {
      "entropy": 5.723854303359985,
      "epoch": 0.429877455747909,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000498603981280632,
      "loss": 5.6535,
      "mean_token_accuracy": 0.1534549355506897,
      "num_tokens": 9592625.0,
      "step": 5525
    },
    {
      "entropy": 5.6727447509765625,
      "epoch": 0.4302664851196265,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004986008969729964,
      "loss": 5.5531,
      "mean_token_accuracy": 0.15823169946670532,
      "num_tokens": 9600934.0,
      "step": 5530
    },
    {
      "entropy": 5.625131464004516,
      "epoch": 0.4306555144913441,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004985978092725731,
      "loss": 5.5533,
      "mean_token_accuracy": 0.16186170428991317,
      "num_tokens": 9609481.0,
      "step": 5535
    },
    {
      "entropy": 5.626396179199219,
      "epoch": 0.4310445438630617,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000498594718179409,
      "loss": 5.525,
      "mean_token_accuracy": 0.16443886905908583,
      "num_tokens": 9618063.0,
      "step": 5540
    },
    {
      "entropy": 5.672539138793946,
      "epoch": 0.4314335732347792,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004985916236935508,
      "loss": 5.6056,
      "mean_token_accuracy": 0.15570127665996553,
      "num_tokens": 9626144.0,
      "step": 5545
    },
    {
      "entropy": 5.569118499755859,
      "epoch": 0.4318226026064968,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004985885258150458,
      "loss": 5.5639,
      "mean_token_accuracy": 0.1569093570113182,
      "num_tokens": 9635328.0,
      "step": 5550
    },
    {
      "entropy": 5.7548158168792725,
      "epoch": 0.43221163197821433,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004985854245439408,
      "loss": 5.7049,
      "mean_token_accuracy": 0.14901985377073287,
      "num_tokens": 9644529.0,
      "step": 5555
    },
    {
      "entropy": 5.743087673187256,
      "epoch": 0.4326006613499319,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004985823198802827,
      "loss": 5.6839,
      "mean_token_accuracy": 0.15375212132930755,
      "num_tokens": 9653763.0,
      "step": 5560
    },
    {
      "entropy": 5.674944353103638,
      "epoch": 0.4329896907216495,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004985792118241188,
      "loss": 5.5095,
      "mean_token_accuracy": 0.1607091173529625,
      "num_tokens": 9661433.0,
      "step": 5565
    },
    {
      "entropy": 5.601434946060181,
      "epoch": 0.43337872009336703,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004985761003754961,
      "loss": 5.5038,
      "mean_token_accuracy": 0.15965722799301146,
      "num_tokens": 9669688.0,
      "step": 5570
    },
    {
      "entropy": 5.7145667552948,
      "epoch": 0.4337677494650846,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004985729855344622,
      "loss": 5.6217,
      "mean_token_accuracy": 0.155311219394207,
      "num_tokens": 9678403.0,
      "step": 5575
    },
    {
      "entropy": 5.7246904373168945,
      "epoch": 0.4341567788368022,
      "grad_norm": 1.28125,
      "learning_rate": 0.000498569867301064,
      "loss": 5.7205,
      "mean_token_accuracy": 0.14789665564894677,
      "num_tokens": 9687504.0,
      "step": 5580
    },
    {
      "entropy": 5.7681337833404545,
      "epoch": 0.43454580820851973,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004985667456753489,
      "loss": 5.5273,
      "mean_token_accuracy": 0.1617080509662628,
      "num_tokens": 9695772.0,
      "step": 5585
    },
    {
      "entropy": 5.617122650146484,
      "epoch": 0.4349348375802373,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004985636206573642,
      "loss": 5.5952,
      "mean_token_accuracy": 0.1571861684322357,
      "num_tokens": 9704150.0,
      "step": 5590
    },
    {
      "entropy": 5.557275772094727,
      "epoch": 0.43532386695195485,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004985604922471575,
      "loss": 5.5016,
      "mean_token_accuracy": 0.16332297772169113,
      "num_tokens": 9712245.0,
      "step": 5595
    },
    {
      "entropy": 5.662200260162353,
      "epoch": 0.43571289632367244,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004985573604447761,
      "loss": 5.6432,
      "mean_token_accuracy": 0.1524854212999344,
      "num_tokens": 9721027.0,
      "step": 5600
    },
    {
      "entropy": 5.731251001358032,
      "epoch": 0.43610192569539,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004985542252502676,
      "loss": 5.5657,
      "mean_token_accuracy": 0.1564818188548088,
      "num_tokens": 9728939.0,
      "step": 5605
    },
    {
      "entropy": 5.763981342315674,
      "epoch": 0.43649095506710756,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004985510866636796,
      "loss": 5.7119,
      "mean_token_accuracy": 0.15163636654615403,
      "num_tokens": 9737863.0,
      "step": 5610
    },
    {
      "entropy": 5.688189840316772,
      "epoch": 0.43687998443882514,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004985479446850596,
      "loss": 5.7083,
      "mean_token_accuracy": 0.1486771047115326,
      "num_tokens": 9747666.0,
      "step": 5615
    },
    {
      "entropy": 5.658762121200562,
      "epoch": 0.4372690138105427,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004985447993144554,
      "loss": 5.5646,
      "mean_token_accuracy": 0.1599259227514267,
      "num_tokens": 9756595.0,
      "step": 5620
    },
    {
      "entropy": 5.730542516708374,
      "epoch": 0.43765804318226026,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004985416505519147,
      "loss": 5.5981,
      "mean_token_accuracy": 0.15622531548142432,
      "num_tokens": 9764747.0,
      "step": 5625
    },
    {
      "entropy": 5.640635251998901,
      "epoch": 0.43804707255397785,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004985384983974853,
      "loss": 5.5542,
      "mean_token_accuracy": 0.15800215303897858,
      "num_tokens": 9772992.0,
      "step": 5630
    },
    {
      "entropy": 5.633075332641601,
      "epoch": 0.4384361019256954,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004985353428512148,
      "loss": 5.5335,
      "mean_token_accuracy": 0.1635892003774643,
      "num_tokens": 9781107.0,
      "step": 5635
    },
    {
      "entropy": 5.624766874313354,
      "epoch": 0.43882513129741296,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004985321839131514,
      "loss": 5.6325,
      "mean_token_accuracy": 0.15101755112409593,
      "num_tokens": 9790618.0,
      "step": 5640
    },
    {
      "entropy": 5.734963464736938,
      "epoch": 0.4392141606691305,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004985290215833428,
      "loss": 5.6187,
      "mean_token_accuracy": 0.150974078476429,
      "num_tokens": 9800147.0,
      "step": 5645
    },
    {
      "entropy": 5.785725402832031,
      "epoch": 0.4396031900408481,
      "grad_norm": 1.25,
      "learning_rate": 0.0004985258558618372,
      "loss": 5.6791,
      "mean_token_accuracy": 0.15372121408581735,
      "num_tokens": 9809056.0,
      "step": 5650
    },
    {
      "entropy": 5.627626085281372,
      "epoch": 0.43999221941256567,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004985226867486825,
      "loss": 5.5326,
      "mean_token_accuracy": 0.15544520765542985,
      "num_tokens": 9817188.0,
      "step": 5655
    },
    {
      "entropy": 5.6710656642913815,
      "epoch": 0.4403812487842832,
      "grad_norm": 1.25,
      "learning_rate": 0.0004985195142439267,
      "loss": 5.7274,
      "mean_token_accuracy": 0.1486324056982994,
      "num_tokens": 9826904.0,
      "step": 5660
    },
    {
      "entropy": 5.741341066360474,
      "epoch": 0.4407702781560008,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004985163383476181,
      "loss": 5.6834,
      "mean_token_accuracy": 0.1525462031364441,
      "num_tokens": 9835937.0,
      "step": 5665
    },
    {
      "entropy": 5.69437313079834,
      "epoch": 0.44115930752771837,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004985131590598048,
      "loss": 5.5055,
      "mean_token_accuracy": 0.1657673478126526,
      "num_tokens": 9843695.0,
      "step": 5670
    },
    {
      "entropy": 5.693076705932617,
      "epoch": 0.4415483368994359,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004985099763805352,
      "loss": 5.6019,
      "mean_token_accuracy": 0.15521264225244522,
      "num_tokens": 9852693.0,
      "step": 5675
    },
    {
      "entropy": 5.6236875534057615,
      "epoch": 0.4419373662711535,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004985067903098574,
      "loss": 5.6014,
      "mean_token_accuracy": 0.15343275368213655,
      "num_tokens": 9860251.0,
      "step": 5680
    },
    {
      "entropy": 5.5922425270080565,
      "epoch": 0.442326395642871,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004985036008478197,
      "loss": 5.5379,
      "mean_token_accuracy": 0.15147336274385453,
      "num_tokens": 9868319.0,
      "step": 5685
    },
    {
      "entropy": 5.629176235198974,
      "epoch": 0.4427154250145886,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004985004079944707,
      "loss": 5.5678,
      "mean_token_accuracy": 0.15566625073552132,
      "num_tokens": 9877119.0,
      "step": 5690
    },
    {
      "entropy": 5.791639518737793,
      "epoch": 0.4431044543863062,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004984972117498587,
      "loss": 5.6647,
      "mean_token_accuracy": 0.15812490731477738,
      "num_tokens": 9886015.0,
      "step": 5695
    },
    {
      "entropy": 5.674712800979615,
      "epoch": 0.4434934837580237,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004984940121140323,
      "loss": 5.602,
      "mean_token_accuracy": 0.1511713407933712,
      "num_tokens": 9894829.0,
      "step": 5700
    },
    {
      "entropy": 5.8527120590209964,
      "epoch": 0.4438825131297413,
      "grad_norm": 1.359375,
      "learning_rate": 0.00049849080908704,
      "loss": 5.8629,
      "mean_token_accuracy": 0.1408378891646862,
      "num_tokens": 9904840.0,
      "step": 5705
    },
    {
      "entropy": 5.819258117675782,
      "epoch": 0.44427154250145884,
      "grad_norm": 1.25,
      "learning_rate": 0.0004984876026689302,
      "loss": 5.6971,
      "mean_token_accuracy": 0.1521285057067871,
      "num_tokens": 9913535.0,
      "step": 5710
    },
    {
      "entropy": 5.7410163402557375,
      "epoch": 0.4446605718731764,
      "grad_norm": 1.265625,
      "learning_rate": 0.000498484392859752,
      "loss": 5.5853,
      "mean_token_accuracy": 0.1592594563961029,
      "num_tokens": 9921676.0,
      "step": 5715
    },
    {
      "entropy": 5.634752941131592,
      "epoch": 0.445049601244894,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004984811796595538,
      "loss": 5.6579,
      "mean_token_accuracy": 0.1542586490511894,
      "num_tokens": 9930321.0,
      "step": 5720
    },
    {
      "entropy": 5.693371438980103,
      "epoch": 0.44543863061661154,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004984779630683843,
      "loss": 5.5919,
      "mean_token_accuracy": 0.15253776609897612,
      "num_tokens": 9939324.0,
      "step": 5725
    },
    {
      "entropy": 5.716857051849365,
      "epoch": 0.44582765998832913,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004984747430862924,
      "loss": 5.6196,
      "mean_token_accuracy": 0.16022590100765227,
      "num_tokens": 9948603.0,
      "step": 5730
    },
    {
      "entropy": 5.600257349014282,
      "epoch": 0.44621668936004666,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004984715197133271,
      "loss": 5.5224,
      "mean_token_accuracy": 0.1616570070385933,
      "num_tokens": 9957165.0,
      "step": 5735
    },
    {
      "entropy": 5.739328241348266,
      "epoch": 0.44660571873176425,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004984682929495371,
      "loss": 5.6943,
      "mean_token_accuracy": 0.14594707265496254,
      "num_tokens": 9966064.0,
      "step": 5740
    },
    {
      "entropy": 5.640423107147217,
      "epoch": 0.44699474810348183,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004984650627949715,
      "loss": 5.5367,
      "mean_token_accuracy": 0.15209029316902162,
      "num_tokens": 9975146.0,
      "step": 5745
    },
    {
      "entropy": 5.680380773544312,
      "epoch": 0.44738377747519936,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004984618292496792,
      "loss": 5.6202,
      "mean_token_accuracy": 0.15602054297924042,
      "num_tokens": 9983725.0,
      "step": 5750
    },
    {
      "entropy": 5.733370542526245,
      "epoch": 0.44777280684691695,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004984585923137093,
      "loss": 5.6747,
      "mean_token_accuracy": 0.15484231412410737,
      "num_tokens": 9992627.0,
      "step": 5755
    },
    {
      "entropy": 5.630478382110596,
      "epoch": 0.4481618362186345,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000498455351987111,
      "loss": 5.4979,
      "mean_token_accuracy": 0.15697283297777176,
      "num_tokens": 10000801.0,
      "step": 5760
    },
    {
      "entropy": 5.698464012145996,
      "epoch": 0.44855086559035207,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004984521082699333,
      "loss": 5.6096,
      "mean_token_accuracy": 0.160343961417675,
      "num_tokens": 10009421.0,
      "step": 5765
    },
    {
      "entropy": 5.667017126083374,
      "epoch": 0.44893989496206965,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004984488611622256,
      "loss": 5.6474,
      "mean_token_accuracy": 0.15203245729207993,
      "num_tokens": 10018527.0,
      "step": 5770
    },
    {
      "entropy": 5.714696216583252,
      "epoch": 0.4493289243337872,
      "grad_norm": 1.265625,
      "learning_rate": 0.000498445610664037,
      "loss": 5.5737,
      "mean_token_accuracy": 0.16014605462551118,
      "num_tokens": 10026906.0,
      "step": 5775
    },
    {
      "entropy": 5.624043941497803,
      "epoch": 0.44971795370550477,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000498442356775417,
      "loss": 5.5475,
      "mean_token_accuracy": 0.1580877646803856,
      "num_tokens": 10035972.0,
      "step": 5780
    },
    {
      "entropy": 5.652503919601441,
      "epoch": 0.45010698307722236,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004984390994964148,
      "loss": 5.6497,
      "mean_token_accuracy": 0.14800439924001693,
      "num_tokens": 10044544.0,
      "step": 5785
    },
    {
      "entropy": 5.768403387069702,
      "epoch": 0.4504960124489399,
      "grad_norm": 1.375,
      "learning_rate": 0.00049843583882708,
      "loss": 5.6335,
      "mean_token_accuracy": 0.15916564911603928,
      "num_tokens": 10052880.0,
      "step": 5790
    },
    {
      "entropy": 5.7311272621154785,
      "epoch": 0.4508850418206575,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000498432574767462,
      "loss": 5.6963,
      "mean_token_accuracy": 0.14732673913240432,
      "num_tokens": 10062373.0,
      "step": 5795
    },
    {
      "entropy": 5.6710062503814695,
      "epoch": 0.451274071192375,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004984293073176103,
      "loss": 5.5125,
      "mean_token_accuracy": 0.16145335882902145,
      "num_tokens": 10070473.0,
      "step": 5800
    },
    {
      "entropy": 5.725533723831177,
      "epoch": 0.4516631005640926,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004984260364775744,
      "loss": 5.6547,
      "mean_token_accuracy": 0.15550251305103302,
      "num_tokens": 10079857.0,
      "step": 5805
    },
    {
      "entropy": 5.70276026725769,
      "epoch": 0.4520521299358102,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000498422762247404,
      "loss": 5.6039,
      "mean_token_accuracy": 0.1518784373998642,
      "num_tokens": 10088326.0,
      "step": 5810
    },
    {
      "entropy": 5.675834608078003,
      "epoch": 0.4524411593075277,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004984194846271489,
      "loss": 5.5466,
      "mean_token_accuracy": 0.15411396324634552,
      "num_tokens": 10097208.0,
      "step": 5815
    },
    {
      "entropy": 5.686789417266846,
      "epoch": 0.4528301886792453,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004984162036168586,
      "loss": 5.6393,
      "mean_token_accuracy": 0.14983187168836593,
      "num_tokens": 10106057.0,
      "step": 5820
    },
    {
      "entropy": 5.659746503829956,
      "epoch": 0.4532192180509628,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004984129192165831,
      "loss": 5.5515,
      "mean_token_accuracy": 0.1533331200480461,
      "num_tokens": 10115384.0,
      "step": 5825
    },
    {
      "entropy": 5.728514289855957,
      "epoch": 0.4536082474226804,
      "grad_norm": 1.25,
      "learning_rate": 0.0004984096314263722,
      "loss": 5.6365,
      "mean_token_accuracy": 0.14838182777166367,
      "num_tokens": 10123958.0,
      "step": 5830
    },
    {
      "entropy": 5.685654067993164,
      "epoch": 0.453997276794398,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004984063402462757,
      "loss": 5.5811,
      "mean_token_accuracy": 0.16384691447019578,
      "num_tokens": 10133199.0,
      "step": 5835
    },
    {
      "entropy": 5.608744812011719,
      "epoch": 0.45438630616611553,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004984030456763435,
      "loss": 5.5209,
      "mean_token_accuracy": 0.16693407595157622,
      "num_tokens": 10141796.0,
      "step": 5840
    },
    {
      "entropy": 5.632417869567871,
      "epoch": 0.4547753355378331,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004983997477166257,
      "loss": 5.5498,
      "mean_token_accuracy": 0.16132254749536515,
      "num_tokens": 10149941.0,
      "step": 5845
    },
    {
      "entropy": 5.706463146209717,
      "epoch": 0.45516436490955064,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004983964463671723,
      "loss": 5.6488,
      "mean_token_accuracy": 0.15120118111371994,
      "num_tokens": 10158796.0,
      "step": 5850
    },
    {
      "entropy": 5.7029928207397464,
      "epoch": 0.45555339428126823,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004983931416280334,
      "loss": 5.5775,
      "mean_token_accuracy": 0.1588720500469208,
      "num_tokens": 10167328.0,
      "step": 5855
    },
    {
      "entropy": 5.687714672088623,
      "epoch": 0.4559424236529858,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004983898334992591,
      "loss": 5.6381,
      "mean_token_accuracy": 0.1526034250855446,
      "num_tokens": 10175937.0,
      "step": 5860
    },
    {
      "entropy": 5.681948041915893,
      "epoch": 0.45633145302470335,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004983865219808998,
      "loss": 5.5921,
      "mean_token_accuracy": 0.14964849948883058,
      "num_tokens": 10184828.0,
      "step": 5865
    },
    {
      "entropy": 5.637333917617798,
      "epoch": 0.45672048239642093,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004983832070730055,
      "loss": 5.5963,
      "mean_token_accuracy": 0.1577572539448738,
      "num_tokens": 10193871.0,
      "step": 5870
    },
    {
      "entropy": 5.634118127822876,
      "epoch": 0.4571095117681385,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004983798887756265,
      "loss": 5.6008,
      "mean_token_accuracy": 0.15409007370471955,
      "num_tokens": 10202740.0,
      "step": 5875
    },
    {
      "entropy": 5.706058931350708,
      "epoch": 0.45749854113985605,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004983765670888133,
      "loss": 5.5434,
      "mean_token_accuracy": 0.1615920588374138,
      "num_tokens": 10210903.0,
      "step": 5880
    },
    {
      "entropy": 5.608094549179077,
      "epoch": 0.45788757051157364,
      "grad_norm": 1.25,
      "learning_rate": 0.0004983732420126163,
      "loss": 5.4997,
      "mean_token_accuracy": 0.16220609247684478,
      "num_tokens": 10219305.0,
      "step": 5885
    },
    {
      "entropy": 5.657922029495239,
      "epoch": 0.45827659988329117,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004983699135470858,
      "loss": 5.6679,
      "mean_token_accuracy": 0.14259497374296187,
      "num_tokens": 10228575.0,
      "step": 5890
    },
    {
      "entropy": 5.6840509414672855,
      "epoch": 0.45866562925500876,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004983665816922723,
      "loss": 5.6437,
      "mean_token_accuracy": 0.15515442043542862,
      "num_tokens": 10237813.0,
      "step": 5895
    },
    {
      "entropy": 5.765459394454956,
      "epoch": 0.45905465862672634,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004983632464482267,
      "loss": 5.639,
      "mean_token_accuracy": 0.151662015914917,
      "num_tokens": 10246606.0,
      "step": 5900
    },
    {
      "entropy": 5.665762281417846,
      "epoch": 0.4594436879984439,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004983599078149991,
      "loss": 5.5937,
      "mean_token_accuracy": 0.1628975346684456,
      "num_tokens": 10254996.0,
      "step": 5905
    },
    {
      "entropy": 5.648976230621338,
      "epoch": 0.45983271737016146,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004983565657926405,
      "loss": 5.6435,
      "mean_token_accuracy": 0.15243904292583466,
      "num_tokens": 10263570.0,
      "step": 5910
    },
    {
      "entropy": 5.629678344726562,
      "epoch": 0.460221746741879,
      "grad_norm": 1.125,
      "learning_rate": 0.0004983532203812017,
      "loss": 5.4759,
      "mean_token_accuracy": 0.16085339486598968,
      "num_tokens": 10272237.0,
      "step": 5915
    },
    {
      "entropy": 5.653153705596924,
      "epoch": 0.4606107761135966,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004983498715807331,
      "loss": 5.6017,
      "mean_token_accuracy": 0.15792696475982665,
      "num_tokens": 10281215.0,
      "step": 5920
    },
    {
      "entropy": 5.727438068389892,
      "epoch": 0.46099980548531416,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004983465193912857,
      "loss": 5.6669,
      "mean_token_accuracy": 0.15387242883443833,
      "num_tokens": 10291054.0,
      "step": 5925
    },
    {
      "entropy": 5.810876703262329,
      "epoch": 0.4613888348570317,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004983431638129104,
      "loss": 5.6472,
      "mean_token_accuracy": 0.1563032776117325,
      "num_tokens": 10299847.0,
      "step": 5930
    },
    {
      "entropy": 5.660638475418091,
      "epoch": 0.4617778642287493,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004983398048456581,
      "loss": 5.5824,
      "mean_token_accuracy": 0.15701979771256447,
      "num_tokens": 10309195.0,
      "step": 5935
    },
    {
      "entropy": 5.640789127349853,
      "epoch": 0.4621668936004668,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004983364424895796,
      "loss": 5.6019,
      "mean_token_accuracy": 0.15272444784641265,
      "num_tokens": 10318386.0,
      "step": 5940
    },
    {
      "entropy": 5.626794528961182,
      "epoch": 0.4625559229721844,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004983330767447262,
      "loss": 5.5244,
      "mean_token_accuracy": 0.1628873825073242,
      "num_tokens": 10327230.0,
      "step": 5945
    },
    {
      "entropy": 5.629368782043457,
      "epoch": 0.462944952343902,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004983297076111489,
      "loss": 5.5165,
      "mean_token_accuracy": 0.15757380425930023,
      "num_tokens": 10335617.0,
      "step": 5950
    },
    {
      "entropy": 5.661604261398315,
      "epoch": 0.4633339817156195,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004983263350888987,
      "loss": 5.5725,
      "mean_token_accuracy": 0.15994318276643754,
      "num_tokens": 10344302.0,
      "step": 5955
    },
    {
      "entropy": 5.681477737426758,
      "epoch": 0.4637230110873371,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004983229591780268,
      "loss": 5.6433,
      "mean_token_accuracy": 0.1517888329923153,
      "num_tokens": 10352490.0,
      "step": 5960
    },
    {
      "entropy": 5.655088281631469,
      "epoch": 0.46411204045905463,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004983195798785844,
      "loss": 5.5766,
      "mean_token_accuracy": 0.16044610887765884,
      "num_tokens": 10361249.0,
      "step": 5965
    },
    {
      "entropy": 5.74783935546875,
      "epoch": 0.4645010698307722,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004983161971906228,
      "loss": 5.6829,
      "mean_token_accuracy": 0.14617585390806198,
      "num_tokens": 10370367.0,
      "step": 5970
    },
    {
      "entropy": 5.724832582473755,
      "epoch": 0.4648900992024898,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004983128111141935,
      "loss": 5.6284,
      "mean_token_accuracy": 0.14997454285621642,
      "num_tokens": 10378429.0,
      "step": 5975
    },
    {
      "entropy": 5.623139333724976,
      "epoch": 0.46527912857420733,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004983094216493475,
      "loss": 5.5512,
      "mean_token_accuracy": 0.15990946292877198,
      "num_tokens": 10386762.0,
      "step": 5980
    },
    {
      "entropy": 5.657111024856567,
      "epoch": 0.4656681579459249,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004983060287961367,
      "loss": 5.5573,
      "mean_token_accuracy": 0.15611529648303984,
      "num_tokens": 10395150.0,
      "step": 5985
    },
    {
      "entropy": 5.686012268066406,
      "epoch": 0.4660571873176425,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004983026325546123,
      "loss": 5.6061,
      "mean_token_accuracy": 0.1571057692170143,
      "num_tokens": 10403941.0,
      "step": 5990
    },
    {
      "entropy": 5.790388917922973,
      "epoch": 0.46644621668936004,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004982992329248257,
      "loss": 5.6699,
      "mean_token_accuracy": 0.14361992180347444,
      "num_tokens": 10413955.0,
      "step": 5995
    },
    {
      "entropy": 5.740258312225341,
      "epoch": 0.4668352460610776,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004982958299068289,
      "loss": 5.576,
      "mean_token_accuracy": 0.15433995723724364,
      "num_tokens": 10421676.0,
      "step": 6000
    },
    {
      "epoch": 0.4668352460610776,
      "eval_entropy": 5.496154374681606,
      "eval_loss": 5.649525165557861,
      "eval_mean_token_accuracy": 0.16116220440296053,
      "eval_num_tokens": 10421676.0,
      "eval_runtime": 28.6748,
      "eval_samples_per_second": 1449.287,
      "eval_steps_per_second": 181.17,
      "step": 6000
    },
    {
      "entropy": 5.599911737442016,
      "epoch": 0.46722427543279516,
      "grad_norm": 1.46875,
      "learning_rate": 0.000498292423500673,
      "loss": 5.5414,
      "mean_token_accuracy": 0.15246708691120148,
      "num_tokens": 10430714.0,
      "step": 6005
    },
    {
      "entropy": 5.7318932056427006,
      "epoch": 0.46761330480451274,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004982890137064102,
      "loss": 5.6676,
      "mean_token_accuracy": 0.15135798901319503,
      "num_tokens": 10439089.0,
      "step": 6010
    },
    {
      "entropy": 5.671796178817749,
      "epoch": 0.46800233417623033,
      "grad_norm": 1.328125,
      "learning_rate": 0.000498285600524092,
      "loss": 5.5886,
      "mean_token_accuracy": 0.1586393378674984,
      "num_tokens": 10447455.0,
      "step": 6015
    },
    {
      "entropy": 5.6312915802001955,
      "epoch": 0.46839136354794786,
      "grad_norm": 1.375,
      "learning_rate": 0.0004982821839537701,
      "loss": 5.5752,
      "mean_token_accuracy": 0.154066501557827,
      "num_tokens": 10455936.0,
      "step": 6020
    },
    {
      "entropy": 5.781944608688354,
      "epoch": 0.46878039291966545,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004982787639954966,
      "loss": 5.7007,
      "mean_token_accuracy": 0.14853788390755654,
      "num_tokens": 10465389.0,
      "step": 6025
    },
    {
      "entropy": 5.8027105808258055,
      "epoch": 0.469169422291383,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004982753406493232,
      "loss": 5.5786,
      "mean_token_accuracy": 0.15479041785001754,
      "num_tokens": 10474709.0,
      "step": 6030
    },
    {
      "entropy": 5.6630011081695555,
      "epoch": 0.46955845166310056,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004982719139153018,
      "loss": 5.5814,
      "mean_token_accuracy": 0.152097749710083,
      "num_tokens": 10482972.0,
      "step": 6035
    },
    {
      "entropy": 5.663127374649048,
      "epoch": 0.46994748103481815,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004982684837934845,
      "loss": 5.5797,
      "mean_token_accuracy": 0.15856870114803315,
      "num_tokens": 10490897.0,
      "step": 6040
    },
    {
      "entropy": 5.644709920883178,
      "epoch": 0.4703365104065357,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004982650502839234,
      "loss": 5.5886,
      "mean_token_accuracy": 0.15786385238170625,
      "num_tokens": 10499679.0,
      "step": 6045
    },
    {
      "entropy": 5.725128126144409,
      "epoch": 0.47072553977825327,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004982616133866705,
      "loss": 5.6421,
      "mean_token_accuracy": 0.15326734334230424,
      "num_tokens": 10508775.0,
      "step": 6050
    },
    {
      "entropy": 5.618102359771728,
      "epoch": 0.4711145691499708,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004982581731017779,
      "loss": 5.5447,
      "mean_token_accuracy": 0.16346119344234467,
      "num_tokens": 10517600.0,
      "step": 6055
    },
    {
      "entropy": 5.5527385711669925,
      "epoch": 0.4715035985216884,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004982547294292979,
      "loss": 5.5035,
      "mean_token_accuracy": 0.16293236017227172,
      "num_tokens": 10525674.0,
      "step": 6060
    },
    {
      "entropy": 5.699562311172485,
      "epoch": 0.47189262789340597,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004982512823692828,
      "loss": 5.7015,
      "mean_token_accuracy": 0.14990407973527908,
      "num_tokens": 10533819.0,
      "step": 6065
    },
    {
      "entropy": 5.661665582656861,
      "epoch": 0.4722816572651235,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004982478319217848,
      "loss": 5.5065,
      "mean_token_accuracy": 0.1595573216676712,
      "num_tokens": 10542901.0,
      "step": 6070
    },
    {
      "entropy": 5.691421365737915,
      "epoch": 0.4726706866368411,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004982443780868562,
      "loss": 5.5825,
      "mean_token_accuracy": 0.16080685406923295,
      "num_tokens": 10552388.0,
      "step": 6075
    },
    {
      "entropy": 5.728979778289795,
      "epoch": 0.4730597160085587,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004982409208645496,
      "loss": 5.6839,
      "mean_token_accuracy": 0.15371156632900237,
      "num_tokens": 10561676.0,
      "step": 6080
    },
    {
      "entropy": 5.769681835174561,
      "epoch": 0.4734487453802762,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004982374602549173,
      "loss": 5.5888,
      "mean_token_accuracy": 0.15942396819591523,
      "num_tokens": 10570037.0,
      "step": 6085
    },
    {
      "entropy": 5.650616025924682,
      "epoch": 0.4738377747519938,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004982339962580119,
      "loss": 5.5862,
      "mean_token_accuracy": 0.15037061125040055,
      "num_tokens": 10579140.0,
      "step": 6090
    },
    {
      "entropy": 5.655625486373902,
      "epoch": 0.4742268041237113,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004982305288738859,
      "loss": 5.6288,
      "mean_token_accuracy": 0.15578012466430663,
      "num_tokens": 10587689.0,
      "step": 6095
    },
    {
      "entropy": 5.667337083816529,
      "epoch": 0.4746158334954289,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004982270581025919,
      "loss": 5.5078,
      "mean_token_accuracy": 0.15836172550916672,
      "num_tokens": 10596002.0,
      "step": 6100
    },
    {
      "entropy": 5.702313661575317,
      "epoch": 0.4750048628671465,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004982235839441826,
      "loss": 5.7311,
      "mean_token_accuracy": 0.14916150867938996,
      "num_tokens": 10605895.0,
      "step": 6105
    },
    {
      "entropy": 5.763257837295532,
      "epoch": 0.475393892238864,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004982201063987108,
      "loss": 5.6638,
      "mean_token_accuracy": 0.14756975322961807,
      "num_tokens": 10614641.0,
      "step": 6110
    },
    {
      "entropy": 5.660020637512207,
      "epoch": 0.4757829216105816,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004982166254662292,
      "loss": 5.6378,
      "mean_token_accuracy": 0.1533852770924568,
      "num_tokens": 10624381.0,
      "step": 6115
    },
    {
      "entropy": 5.706807374954224,
      "epoch": 0.47617195098229914,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004982131411467904,
      "loss": 5.5952,
      "mean_token_accuracy": 0.15574658811092376,
      "num_tokens": 10632493.0,
      "step": 6120
    },
    {
      "entropy": 5.704343557357788,
      "epoch": 0.4765609803540167,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004982096534404476,
      "loss": 5.6232,
      "mean_token_accuracy": 0.14914219677448273,
      "num_tokens": 10641556.0,
      "step": 6125
    },
    {
      "entropy": 5.67762017250061,
      "epoch": 0.4769500097257343,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004982061623472535,
      "loss": 5.6904,
      "mean_token_accuracy": 0.15633485242724418,
      "num_tokens": 10649736.0,
      "step": 6130
    },
    {
      "entropy": 5.684682178497314,
      "epoch": 0.47733903909745184,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004982026678672612,
      "loss": 5.5009,
      "mean_token_accuracy": 0.16547404527664183,
      "num_tokens": 10657991.0,
      "step": 6135
    },
    {
      "entropy": 5.596237707138061,
      "epoch": 0.47772806846916943,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004981991700005238,
      "loss": 5.5706,
      "mean_token_accuracy": 0.15123811960220337,
      "num_tokens": 10667310.0,
      "step": 6140
    },
    {
      "entropy": 5.741936254501343,
      "epoch": 0.47811709784088696,
      "grad_norm": 1.328125,
      "learning_rate": 0.000498195668747094,
      "loss": 5.6213,
      "mean_token_accuracy": 0.15695492923259735,
      "num_tokens": 10675658.0,
      "step": 6145
    },
    {
      "entropy": 5.638043642044067,
      "epoch": 0.47850612721260455,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004981921641070254,
      "loss": 5.5102,
      "mean_token_accuracy": 0.16681503355503083,
      "num_tokens": 10684016.0,
      "step": 6150
    },
    {
      "entropy": 5.666017532348633,
      "epoch": 0.47889515658432213,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004981886560803708,
      "loss": 5.6281,
      "mean_token_accuracy": 0.15898596942424775,
      "num_tokens": 10692814.0,
      "step": 6155
    },
    {
      "entropy": 5.71913104057312,
      "epoch": 0.47928418595603967,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004981851446671838,
      "loss": 5.6469,
      "mean_token_accuracy": 0.15398247838020324,
      "num_tokens": 10701571.0,
      "step": 6160
    },
    {
      "entropy": 5.743273782730102,
      "epoch": 0.47967321532775725,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004981816298675173,
      "loss": 5.6321,
      "mean_token_accuracy": 0.15258511602878572,
      "num_tokens": 10710569.0,
      "step": 6165
    },
    {
      "entropy": 5.699658441543579,
      "epoch": 0.4800622446994748,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004981781116814248,
      "loss": 5.6556,
      "mean_token_accuracy": 0.15451659709215165,
      "num_tokens": 10719277.0,
      "step": 6170
    },
    {
      "entropy": 5.6413978099822994,
      "epoch": 0.48045127407119237,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004981745901089596,
      "loss": 5.523,
      "mean_token_accuracy": 0.1553031951189041,
      "num_tokens": 10727174.0,
      "step": 6175
    },
    {
      "entropy": 5.652820158004761,
      "epoch": 0.48084030344290996,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004981710651501753,
      "loss": 5.6308,
      "mean_token_accuracy": 0.1499924898147583,
      "num_tokens": 10736072.0,
      "step": 6180
    },
    {
      "entropy": 5.77542290687561,
      "epoch": 0.4812293328146275,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004981675368051254,
      "loss": 5.6366,
      "mean_token_accuracy": 0.1591070920228958,
      "num_tokens": 10745375.0,
      "step": 6185
    },
    {
      "entropy": 5.6705575466156,
      "epoch": 0.4816183621863451,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004981640050738633,
      "loss": 5.6372,
      "mean_token_accuracy": 0.15502333045005798,
      "num_tokens": 10754049.0,
      "step": 6190
    },
    {
      "entropy": 5.6272234439849855,
      "epoch": 0.48200739155806266,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004981604699564426,
      "loss": 5.5709,
      "mean_token_accuracy": 0.15725173354148864,
      "num_tokens": 10763475.0,
      "step": 6195
    },
    {
      "entropy": 5.723629283905029,
      "epoch": 0.4823964209297802,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004981569314529169,
      "loss": 5.6537,
      "mean_token_accuracy": 0.15193783938884736,
      "num_tokens": 10772100.0,
      "step": 6200
    },
    {
      "entropy": 5.624773168563843,
      "epoch": 0.4827854503014978,
      "grad_norm": 1.421875,
      "learning_rate": 0.00049815338956334,
      "loss": 5.586,
      "mean_token_accuracy": 0.1571911796927452,
      "num_tokens": 10780605.0,
      "step": 6205
    },
    {
      "entropy": 5.688734722137451,
      "epoch": 0.4831744796732153,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004981498442877656,
      "loss": 5.57,
      "mean_token_accuracy": 0.15786497220396994,
      "num_tokens": 10789235.0,
      "step": 6210
    },
    {
      "entropy": 5.692664527893067,
      "epoch": 0.4835635090449329,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004981462956262474,
      "loss": 5.5806,
      "mean_token_accuracy": 0.15071646571159364,
      "num_tokens": 10798078.0,
      "step": 6215
    },
    {
      "entropy": 5.687393140792847,
      "epoch": 0.4839525384166505,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004981427435788396,
      "loss": 5.6054,
      "mean_token_accuracy": 0.14804309904575347,
      "num_tokens": 10807667.0,
      "step": 6220
    },
    {
      "entropy": 5.652494049072265,
      "epoch": 0.484341567788368,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004981391881455957,
      "loss": 5.5685,
      "mean_token_accuracy": 0.14992497563362123,
      "num_tokens": 10816574.0,
      "step": 6225
    },
    {
      "entropy": 5.658774518966675,
      "epoch": 0.4847305971600856,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004981356293265696,
      "loss": 5.5465,
      "mean_token_accuracy": 0.15477658212184905,
      "num_tokens": 10824675.0,
      "step": 6230
    },
    {
      "entropy": 5.645687818527222,
      "epoch": 0.4851196265318031,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004981320671218157,
      "loss": 5.5049,
      "mean_token_accuracy": 0.16456515043973924,
      "num_tokens": 10832855.0,
      "step": 6235
    },
    {
      "entropy": 5.572377777099609,
      "epoch": 0.4855086559035207,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004981285015313877,
      "loss": 5.4122,
      "mean_token_accuracy": 0.1659885361790657,
      "num_tokens": 10841603.0,
      "step": 6240
    },
    {
      "entropy": 5.74435248374939,
      "epoch": 0.4858976852752383,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004981249325553399,
      "loss": 5.7553,
      "mean_token_accuracy": 0.14605401381850242,
      "num_tokens": 10850348.0,
      "step": 6245
    },
    {
      "entropy": 5.7315455913543705,
      "epoch": 0.48628671464695583,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004981213601937264,
      "loss": 5.4984,
      "mean_token_accuracy": 0.15666187256574632,
      "num_tokens": 10858955.0,
      "step": 6250
    },
    {
      "entropy": 5.634166193008423,
      "epoch": 0.4866757440186734,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004981177844466013,
      "loss": 5.5928,
      "mean_token_accuracy": 0.1482228606939316,
      "num_tokens": 10867336.0,
      "step": 6255
    },
    {
      "entropy": 5.68542799949646,
      "epoch": 0.48706477339039095,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004981142053140192,
      "loss": 5.6738,
      "mean_token_accuracy": 0.15138802379369737,
      "num_tokens": 10877043.0,
      "step": 6260
    },
    {
      "entropy": 5.799023103713989,
      "epoch": 0.48745380276210853,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004981106227960339,
      "loss": 5.6314,
      "mean_token_accuracy": 0.15212034434080124,
      "num_tokens": 10885580.0,
      "step": 6265
    },
    {
      "entropy": 5.727660846710205,
      "epoch": 0.4878428321338261,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004981070368927001,
      "loss": 5.7355,
      "mean_token_accuracy": 0.1497201591730118,
      "num_tokens": 10894289.0,
      "step": 6270
    },
    {
      "entropy": 5.723119068145752,
      "epoch": 0.48823186150554365,
      "grad_norm": 1.375,
      "learning_rate": 0.000498103447604072,
      "loss": 5.6595,
      "mean_token_accuracy": 0.14703867062926293,
      "num_tokens": 10902384.0,
      "step": 6275
    },
    {
      "entropy": 5.73089189529419,
      "epoch": 0.48862089087726124,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004980998549302044,
      "loss": 5.5618,
      "mean_token_accuracy": 0.1498264640569687,
      "num_tokens": 10911160.0,
      "step": 6280
    },
    {
      "entropy": 5.621191596984863,
      "epoch": 0.4890099202489788,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004980962588711516,
      "loss": 5.4562,
      "mean_token_accuracy": 0.16945781260728837,
      "num_tokens": 10920066.0,
      "step": 6285
    },
    {
      "entropy": 5.694791412353515,
      "epoch": 0.48939894962069636,
      "grad_norm": 1.375,
      "learning_rate": 0.000498092659426968,
      "loss": 5.635,
      "mean_token_accuracy": 0.15655620992183686,
      "num_tokens": 10928969.0,
      "step": 6290
    },
    {
      "entropy": 5.752533292770385,
      "epoch": 0.48978797899241394,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004980890565977085,
      "loss": 5.627,
      "mean_token_accuracy": 0.1562311053276062,
      "num_tokens": 10936939.0,
      "step": 6295
    },
    {
      "entropy": 5.633117532730102,
      "epoch": 0.4901770083641315,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004980854503834276,
      "loss": 5.5868,
      "mean_token_accuracy": 0.15852609723806382,
      "num_tokens": 10945929.0,
      "step": 6300
    },
    {
      "entropy": 5.663782453536987,
      "epoch": 0.49056603773584906,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004980818407841802,
      "loss": 5.5863,
      "mean_token_accuracy": 0.1573207527399063,
      "num_tokens": 10954849.0,
      "step": 6305
    },
    {
      "entropy": 5.678795623779297,
      "epoch": 0.49095506710756665,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004980782278000207,
      "loss": 5.6407,
      "mean_token_accuracy": 0.15423305928707123,
      "num_tokens": 10963800.0,
      "step": 6310
    },
    {
      "entropy": 5.839993190765381,
      "epoch": 0.4913440964792842,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004980746114310043,
      "loss": 5.6953,
      "mean_token_accuracy": 0.15321774333715438,
      "num_tokens": 10972477.0,
      "step": 6315
    },
    {
      "entropy": 5.829525756835937,
      "epoch": 0.49173312585100176,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004980709916771858,
      "loss": 5.6855,
      "mean_token_accuracy": 0.1555396370589733,
      "num_tokens": 10981437.0,
      "step": 6320
    },
    {
      "entropy": 5.64800500869751,
      "epoch": 0.4921221552227193,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004980673685386198,
      "loss": 5.5357,
      "mean_token_accuracy": 0.15727183669805528,
      "num_tokens": 10989973.0,
      "step": 6325
    },
    {
      "entropy": 5.586803579330445,
      "epoch": 0.4925111845944369,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004980637420153618,
      "loss": 5.5574,
      "mean_token_accuracy": 0.15266730338335038,
      "num_tokens": 10998160.0,
      "step": 6330
    },
    {
      "entropy": 5.676742506027222,
      "epoch": 0.49290021396615447,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004980601121074664,
      "loss": 5.7116,
      "mean_token_accuracy": 0.1521316148340702,
      "num_tokens": 11007127.0,
      "step": 6335
    },
    {
      "entropy": 5.7719337940216064,
      "epoch": 0.493289243337872,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004980564788149889,
      "loss": 5.5615,
      "mean_token_accuracy": 0.15340966284275054,
      "num_tokens": 11015518.0,
      "step": 6340
    },
    {
      "entropy": 5.611857223510742,
      "epoch": 0.4936782727095896,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004980528421379844,
      "loss": 5.4632,
      "mean_token_accuracy": 0.16847451776266098,
      "num_tokens": 11023739.0,
      "step": 6345
    },
    {
      "entropy": 5.586323976516724,
      "epoch": 0.4940673020813071,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000498049202076508,
      "loss": 5.4961,
      "mean_token_accuracy": 0.16648571491241454,
      "num_tokens": 11032034.0,
      "step": 6350
    },
    {
      "entropy": 5.642065715789795,
      "epoch": 0.4944563314530247,
      "grad_norm": 1.375,
      "learning_rate": 0.0004980455586306149,
      "loss": 5.4984,
      "mean_token_accuracy": 0.1611510455608368,
      "num_tokens": 11040319.0,
      "step": 6355
    },
    {
      "entropy": 5.604467725753784,
      "epoch": 0.4948453608247423,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004980419118003605,
      "loss": 5.5654,
      "mean_token_accuracy": 0.15740724951028823,
      "num_tokens": 11049459.0,
      "step": 6360
    },
    {
      "entropy": 5.663004446029663,
      "epoch": 0.4952343901964598,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004980382615858001,
      "loss": 5.5489,
      "mean_token_accuracy": 0.15955363661050798,
      "num_tokens": 11058402.0,
      "step": 6365
    },
    {
      "entropy": 5.593869209289551,
      "epoch": 0.4956234195681774,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004980346079869892,
      "loss": 5.4955,
      "mean_token_accuracy": 0.1616832584142685,
      "num_tokens": 11067430.0,
      "step": 6370
    },
    {
      "entropy": 5.668562316894532,
      "epoch": 0.49601244893989493,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000498030951003983,
      "loss": 5.492,
      "mean_token_accuracy": 0.16149785220623017,
      "num_tokens": 11075594.0,
      "step": 6375
    },
    {
      "entropy": 5.595048952102661,
      "epoch": 0.4964014783116125,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004980272906368371,
      "loss": 5.4584,
      "mean_token_accuracy": 0.1725854143500328,
      "num_tokens": 11083509.0,
      "step": 6380
    },
    {
      "entropy": 5.637046384811401,
      "epoch": 0.4967905076833301,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004980236268856071,
      "loss": 5.6269,
      "mean_token_accuracy": 0.15187471285462378,
      "num_tokens": 11091938.0,
      "step": 6385
    },
    {
      "entropy": 5.681407499313354,
      "epoch": 0.49717953705504764,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004980199597503487,
      "loss": 5.6592,
      "mean_token_accuracy": 0.15273707509040832,
      "num_tokens": 11099974.0,
      "step": 6390
    },
    {
      "entropy": 5.583227300643921,
      "epoch": 0.4975685664267652,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004980162892311171,
      "loss": 5.3097,
      "mean_token_accuracy": 0.1723850205540657,
      "num_tokens": 11107944.0,
      "step": 6395
    },
    {
      "entropy": 5.559449100494385,
      "epoch": 0.4979575957984828,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004980126153279684,
      "loss": 5.603,
      "mean_token_accuracy": 0.1561395511031151,
      "num_tokens": 11116401.0,
      "step": 6400
    },
    {
      "entropy": 5.660170412063598,
      "epoch": 0.49834662517020034,
      "grad_norm": 1.375,
      "learning_rate": 0.0004980089380409583,
      "loss": 5.5709,
      "mean_token_accuracy": 0.15186020135879516,
      "num_tokens": 11125430.0,
      "step": 6405
    },
    {
      "entropy": 5.715793418884277,
      "epoch": 0.4987356545419179,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004980052573701424,
      "loss": 5.6372,
      "mean_token_accuracy": 0.148736110329628,
      "num_tokens": 11133949.0,
      "step": 6410
    },
    {
      "entropy": 5.731146192550659,
      "epoch": 0.49912468391363546,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004980015733155767,
      "loss": 5.5675,
      "mean_token_accuracy": 0.16028442829847336,
      "num_tokens": 11141749.0,
      "step": 6415
    },
    {
      "entropy": 5.649786901473999,
      "epoch": 0.49951371328535304,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004979978858773171,
      "loss": 5.5657,
      "mean_token_accuracy": 0.15979141891002654,
      "num_tokens": 11150618.0,
      "step": 6420
    },
    {
      "entropy": 5.609336900711059,
      "epoch": 0.49990274265707063,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004979941950554195,
      "loss": 5.5968,
      "mean_token_accuracy": 0.15541461259126663,
      "num_tokens": 11158549.0,
      "step": 6425
    },
    {
      "entropy": 5.672288656234741,
      "epoch": 0.5002917720287882,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004979905008499399,
      "loss": 5.5365,
      "mean_token_accuracy": 0.16468718647956848,
      "num_tokens": 11167131.0,
      "step": 6430
    },
    {
      "entropy": 5.669491386413574,
      "epoch": 0.5006808014005057,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004979868032609344,
      "loss": 5.6143,
      "mean_token_accuracy": 0.15387326255440711,
      "num_tokens": 11175766.0,
      "step": 6435
    },
    {
      "entropy": 5.688392448425293,
      "epoch": 0.5010698307722233,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004979831022884591,
      "loss": 5.5899,
      "mean_token_accuracy": 0.15998782515525817,
      "num_tokens": 11184799.0,
      "step": 6440
    },
    {
      "entropy": 5.594391107559204,
      "epoch": 0.5014588601439409,
      "grad_norm": 1.375,
      "learning_rate": 0.0004979793979325701,
      "loss": 5.5329,
      "mean_token_accuracy": 0.1567195326089859,
      "num_tokens": 11193083.0,
      "step": 6445
    },
    {
      "entropy": 5.582624864578247,
      "epoch": 0.5018478895156584,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004979756901933236,
      "loss": 5.5276,
      "mean_token_accuracy": 0.16227489709854126,
      "num_tokens": 11201753.0,
      "step": 6450
    },
    {
      "entropy": 5.760551834106446,
      "epoch": 0.502236918887376,
      "grad_norm": 1.296875,
      "learning_rate": 0.000497971979070776,
      "loss": 5.619,
      "mean_token_accuracy": 0.15672486275434494,
      "num_tokens": 11209965.0,
      "step": 6455
    },
    {
      "entropy": 5.6459235668182375,
      "epoch": 0.5026259482590936,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004979682645649834,
      "loss": 5.517,
      "mean_token_accuracy": 0.1660740926861763,
      "num_tokens": 11217961.0,
      "step": 6460
    },
    {
      "entropy": 5.661819362640381,
      "epoch": 0.5030149776308112,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004979645466760025,
      "loss": 5.6222,
      "mean_token_accuracy": 0.15498576909303666,
      "num_tokens": 11225803.0,
      "step": 6465
    },
    {
      "entropy": 5.619951629638672,
      "epoch": 0.5034040070025287,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004979608254038892,
      "loss": 5.4606,
      "mean_token_accuracy": 0.16229200810194017,
      "num_tokens": 11234584.0,
      "step": 6470
    },
    {
      "entropy": 5.680429172515869,
      "epoch": 0.5037930363742462,
      "grad_norm": 1.375,
      "learning_rate": 0.0004979571007487003,
      "loss": 5.6651,
      "mean_token_accuracy": 0.15144428014755248,
      "num_tokens": 11243340.0,
      "step": 6475
    },
    {
      "entropy": 5.62803840637207,
      "epoch": 0.5041820657459638,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004979533727104924,
      "loss": 5.531,
      "mean_token_accuracy": 0.1532328888773918,
      "num_tokens": 11252383.0,
      "step": 6480
    },
    {
      "entropy": 5.630037784576416,
      "epoch": 0.5045710951176814,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000497949641289322,
      "loss": 5.6087,
      "mean_token_accuracy": 0.15235372483730317,
      "num_tokens": 11261419.0,
      "step": 6485
    },
    {
      "entropy": 5.653547477722168,
      "epoch": 0.504960124489399,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004979459064852456,
      "loss": 5.574,
      "mean_token_accuracy": 0.15386256873607634,
      "num_tokens": 11270743.0,
      "step": 6490
    },
    {
      "entropy": 5.706313419342041,
      "epoch": 0.5053491538611166,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004979421682983197,
      "loss": 5.4868,
      "mean_token_accuracy": 0.15976502299308776,
      "num_tokens": 11279666.0,
      "step": 6495
    },
    {
      "entropy": 5.7113391876220705,
      "epoch": 0.505738183232834,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004979384267286015,
      "loss": 5.6601,
      "mean_token_accuracy": 0.14999309778213502,
      "num_tokens": 11288710.0,
      "step": 6500
    },
    {
      "entropy": 5.6965361595153805,
      "epoch": 0.5061272126045516,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004979346817761475,
      "loss": 5.5297,
      "mean_token_accuracy": 0.15320013761520385,
      "num_tokens": 11297665.0,
      "step": 6505
    },
    {
      "entropy": 5.7402442455291744,
      "epoch": 0.5065162419762692,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004979309334410144,
      "loss": 5.6928,
      "mean_token_accuracy": 0.15129803717136384,
      "num_tokens": 11306751.0,
      "step": 6510
    },
    {
      "entropy": 5.690135049819946,
      "epoch": 0.5069052713479868,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004979271817232594,
      "loss": 5.6844,
      "mean_token_accuracy": 0.15013830214738846,
      "num_tokens": 11315659.0,
      "step": 6515
    },
    {
      "entropy": 5.701597547531128,
      "epoch": 0.5072943007197044,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004979234266229391,
      "loss": 5.5613,
      "mean_token_accuracy": 0.1573285400867462,
      "num_tokens": 11323362.0,
      "step": 6520
    },
    {
      "entropy": 5.6727128505706785,
      "epoch": 0.5076833300914219,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004979196681401106,
      "loss": 5.5589,
      "mean_token_accuracy": 0.15374930649995805,
      "num_tokens": 11332343.0,
      "step": 6525
    },
    {
      "entropy": 5.6307044506073,
      "epoch": 0.5080723594631394,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000497915906274831,
      "loss": 5.5097,
      "mean_token_accuracy": 0.153728187084198,
      "num_tokens": 11340816.0,
      "step": 6530
    },
    {
      "entropy": 5.596715879440308,
      "epoch": 0.508461388834857,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004979121410271574,
      "loss": 5.5076,
      "mean_token_accuracy": 0.1596982181072235,
      "num_tokens": 11349728.0,
      "step": 6535
    },
    {
      "entropy": 5.6834207534790036,
      "epoch": 0.5088504182065746,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004979083723971468,
      "loss": 5.6381,
      "mean_token_accuracy": 0.1554409921169281,
      "num_tokens": 11358477.0,
      "step": 6540
    },
    {
      "entropy": 5.589273118972779,
      "epoch": 0.5092394475782922,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004979046003848564,
      "loss": 5.4559,
      "mean_token_accuracy": 0.166815909743309,
      "num_tokens": 11367183.0,
      "step": 6545
    },
    {
      "entropy": 5.6982262134552,
      "epoch": 0.5096284769500097,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004979008249903435,
      "loss": 5.6663,
      "mean_token_accuracy": 0.1446331888437271,
      "num_tokens": 11376703.0,
      "step": 6550
    },
    {
      "entropy": 5.741359567642212,
      "epoch": 0.5100175063217273,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004978970462136655,
      "loss": 5.6684,
      "mean_token_accuracy": 0.1546018734574318,
      "num_tokens": 11385502.0,
      "step": 6555
    },
    {
      "entropy": 5.686350870132446,
      "epoch": 0.5104065356934449,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004978932640548794,
      "loss": 5.5298,
      "mean_token_accuracy": 0.1591428518295288,
      "num_tokens": 11393947.0,
      "step": 6560
    },
    {
      "entropy": 5.59600796699524,
      "epoch": 0.5107955650651624,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004978894785140429,
      "loss": 5.5188,
      "mean_token_accuracy": 0.164163675904274,
      "num_tokens": 11403077.0,
      "step": 6565
    },
    {
      "entropy": 5.57859468460083,
      "epoch": 0.51118459443688,
      "grad_norm": 1.375,
      "learning_rate": 0.0004978856895912132,
      "loss": 5.5,
      "mean_token_accuracy": 0.15813149958848954,
      "num_tokens": 11411863.0,
      "step": 6570
    },
    {
      "entropy": 5.681532716751098,
      "epoch": 0.5115736238085975,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004978818972864481,
      "loss": 5.4611,
      "mean_token_accuracy": 0.16052573472261428,
      "num_tokens": 11420030.0,
      "step": 6575
    },
    {
      "entropy": 5.702321243286133,
      "epoch": 0.5119626531803151,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004978781015998048,
      "loss": 5.5644,
      "mean_token_accuracy": 0.15766163170337677,
      "num_tokens": 11428534.0,
      "step": 6580
    },
    {
      "entropy": 5.664851665496826,
      "epoch": 0.5123516825520327,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004978743025313413,
      "loss": 5.5533,
      "mean_token_accuracy": 0.15721214860677718,
      "num_tokens": 11436641.0,
      "step": 6585
    },
    {
      "entropy": 5.631090784072876,
      "epoch": 0.5127407119237503,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004978705000811148,
      "loss": 5.5312,
      "mean_token_accuracy": 0.1598398506641388,
      "num_tokens": 11444595.0,
      "step": 6590
    },
    {
      "entropy": 5.577209997177124,
      "epoch": 0.5131297412954678,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004978666942491832,
      "loss": 5.3865,
      "mean_token_accuracy": 0.16264140456914902,
      "num_tokens": 11453257.0,
      "step": 6595
    },
    {
      "entropy": 5.589799070358277,
      "epoch": 0.5135187706671853,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004978628850356043,
      "loss": 5.5615,
      "mean_token_accuracy": 0.15875291973352432,
      "num_tokens": 11461762.0,
      "step": 6600
    },
    {
      "entropy": 5.649018383026123,
      "epoch": 0.5139078000389029,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004978590724404358,
      "loss": 5.4686,
      "mean_token_accuracy": 0.15878367722034453,
      "num_tokens": 11470602.0,
      "step": 6605
    },
    {
      "entropy": 5.635024499893189,
      "epoch": 0.5142968294106205,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004978552564637356,
      "loss": 5.5583,
      "mean_token_accuracy": 0.15673930644989015,
      "num_tokens": 11479826.0,
      "step": 6610
    },
    {
      "entropy": 5.680516052246094,
      "epoch": 0.5146858587823381,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004978514371055616,
      "loss": 5.6371,
      "mean_token_accuracy": 0.15062430500984192,
      "num_tokens": 11488698.0,
      "step": 6615
    },
    {
      "entropy": 5.728860330581665,
      "epoch": 0.5150748881540557,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004978476143659718,
      "loss": 5.6055,
      "mean_token_accuracy": 0.15783123075962066,
      "num_tokens": 11497434.0,
      "step": 6620
    },
    {
      "entropy": 5.626099395751953,
      "epoch": 0.5154639175257731,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004978437882450241,
      "loss": 5.6226,
      "mean_token_accuracy": 0.15543157532811164,
      "num_tokens": 11506173.0,
      "step": 6625
    },
    {
      "entropy": 5.63395414352417,
      "epoch": 0.5158529468974907,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004978399587427766,
      "loss": 5.5153,
      "mean_token_accuracy": 0.1609303444623947,
      "num_tokens": 11514824.0,
      "step": 6630
    },
    {
      "entropy": 5.634879398345947,
      "epoch": 0.5162419762692083,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004978361258592874,
      "loss": 5.5784,
      "mean_token_accuracy": 0.15130043774843216,
      "num_tokens": 11523634.0,
      "step": 6635
    },
    {
      "entropy": 5.803075313568115,
      "epoch": 0.5166310056409259,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004978322895946147,
      "loss": 5.7006,
      "mean_token_accuracy": 0.15314190834760666,
      "num_tokens": 11531916.0,
      "step": 6640
    },
    {
      "entropy": 5.699688673019409,
      "epoch": 0.5170200350126435,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004978284499488167,
      "loss": 5.5159,
      "mean_token_accuracy": 0.16227565407752992,
      "num_tokens": 11540000.0,
      "step": 6645
    },
    {
      "entropy": 5.699098587036133,
      "epoch": 0.5174090643843611,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004978246069219516,
      "loss": 5.6113,
      "mean_token_accuracy": 0.15647616535425185,
      "num_tokens": 11548760.0,
      "step": 6650
    },
    {
      "entropy": 5.587365961074829,
      "epoch": 0.5177980937560785,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004978207605140777,
      "loss": 5.4321,
      "mean_token_accuracy": 0.16562076658010483,
      "num_tokens": 11557991.0,
      "step": 6655
    },
    {
      "entropy": 5.608901739120483,
      "epoch": 0.5181871231277961,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004978169107252534,
      "loss": 5.4985,
      "mean_token_accuracy": 0.16201310306787492,
      "num_tokens": 11566638.0,
      "step": 6660
    },
    {
      "entropy": 5.53792028427124,
      "epoch": 0.5185761524995137,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004978130575555373,
      "loss": 5.4198,
      "mean_token_accuracy": 0.16330699175596236,
      "num_tokens": 11575106.0,
      "step": 6665
    },
    {
      "entropy": 5.5862060546875,
      "epoch": 0.5189651818712313,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004978092010049877,
      "loss": 5.5869,
      "mean_token_accuracy": 0.1444960653781891,
      "num_tokens": 11583971.0,
      "step": 6670
    },
    {
      "entropy": 5.6814610958099365,
      "epoch": 0.5193542112429489,
      "grad_norm": 1.328125,
      "learning_rate": 0.000497805341073663,
      "loss": 5.5944,
      "mean_token_accuracy": 0.1549292489886284,
      "num_tokens": 11592936.0,
      "step": 6675
    },
    {
      "entropy": 5.649386644363403,
      "epoch": 0.5197432406146664,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004978014777616219,
      "loss": 5.4676,
      "mean_token_accuracy": 0.159424264729023,
      "num_tokens": 11602055.0,
      "step": 6680
    },
    {
      "entropy": 5.592701244354248,
      "epoch": 0.520132269986384,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004977976110689229,
      "loss": 5.4963,
      "mean_token_accuracy": 0.15947575122117996,
      "num_tokens": 11610211.0,
      "step": 6685
    },
    {
      "entropy": 5.6688920021057125,
      "epoch": 0.5205212993581015,
      "grad_norm": 1.515625,
      "learning_rate": 0.000497793740995625,
      "loss": 5.5579,
      "mean_token_accuracy": 0.15732331275939943,
      "num_tokens": 11619096.0,
      "step": 6690
    },
    {
      "entropy": 5.658513736724854,
      "epoch": 0.5209103287298191,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004977898675417866,
      "loss": 5.6363,
      "mean_token_accuracy": 0.15888825207948684,
      "num_tokens": 11628142.0,
      "step": 6695
    },
    {
      "entropy": 5.700950670242309,
      "epoch": 0.5212993581015367,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004977859907074664,
      "loss": 5.485,
      "mean_token_accuracy": 0.16149185299873353,
      "num_tokens": 11635961.0,
      "step": 6700
    },
    {
      "entropy": 5.6735364437103275,
      "epoch": 0.5216883874732542,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004977821104927236,
      "loss": 5.6212,
      "mean_token_accuracy": 0.1562149181962013,
      "num_tokens": 11644778.0,
      "step": 6705
    },
    {
      "entropy": 5.657710838317871,
      "epoch": 0.5220774168449718,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004977782268976167,
      "loss": 5.5925,
      "mean_token_accuracy": 0.15472943633794783,
      "num_tokens": 11654043.0,
      "step": 6710
    },
    {
      "entropy": 5.711113595962525,
      "epoch": 0.5224664462166894,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004977743399222048,
      "loss": 5.5755,
      "mean_token_accuracy": 0.15707167237997055,
      "num_tokens": 11662794.0,
      "step": 6715
    },
    {
      "entropy": 5.631554222106933,
      "epoch": 0.522855475588407,
      "grad_norm": 1.625,
      "learning_rate": 0.0004977704495665471,
      "loss": 5.5782,
      "mean_token_accuracy": 0.1574481889605522,
      "num_tokens": 11670624.0,
      "step": 6720
    },
    {
      "entropy": 5.58207483291626,
      "epoch": 0.5232445049601245,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004977665558307023,
      "loss": 5.482,
      "mean_token_accuracy": 0.15776702016592026,
      "num_tokens": 11679366.0,
      "step": 6725
    },
    {
      "entropy": 5.526380920410157,
      "epoch": 0.523633534331842,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004977626587147295,
      "loss": 5.3984,
      "mean_token_accuracy": 0.16687122136354446,
      "num_tokens": 11688137.0,
      "step": 6730
    },
    {
      "entropy": 5.559792566299438,
      "epoch": 0.5240225637035596,
      "grad_norm": 1.609375,
      "learning_rate": 0.000497758758218688,
      "loss": 5.5255,
      "mean_token_accuracy": 0.16067289113998412,
      "num_tokens": 11697236.0,
      "step": 6735
    },
    {
      "entropy": 5.669983386993408,
      "epoch": 0.5244115930752772,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004977548543426368,
      "loss": 5.5201,
      "mean_token_accuracy": 0.159443636238575,
      "num_tokens": 11706513.0,
      "step": 6740
    },
    {
      "entropy": 5.647953271865845,
      "epoch": 0.5248006224469948,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004977509470866353,
      "loss": 5.4815,
      "mean_token_accuracy": 0.16161043643951417,
      "num_tokens": 11715239.0,
      "step": 6745
    },
    {
      "entropy": 5.624539852142334,
      "epoch": 0.5251896518187124,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004977470364507427,
      "loss": 5.5067,
      "mean_token_accuracy": 0.1691492259502411,
      "num_tokens": 11724086.0,
      "step": 6750
    },
    {
      "entropy": 5.574176502227783,
      "epoch": 0.5255786811904298,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004977431224350184,
      "loss": 5.5399,
      "mean_token_accuracy": 0.16304653882980347,
      "num_tokens": 11731975.0,
      "step": 6755
    },
    {
      "entropy": 5.606770372390747,
      "epoch": 0.5259677105621474,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004977392050395217,
      "loss": 5.5292,
      "mean_token_accuracy": 0.15769765973091127,
      "num_tokens": 11741172.0,
      "step": 6760
    },
    {
      "entropy": 5.635431337356567,
      "epoch": 0.526356739933865,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004977352842643121,
      "loss": 5.497,
      "mean_token_accuracy": 0.16328567042946815,
      "num_tokens": 11750609.0,
      "step": 6765
    },
    {
      "entropy": 5.648671865463257,
      "epoch": 0.5267457693055826,
      "grad_norm": 1.5546875,
      "learning_rate": 0.000497731360109449,
      "loss": 5.4808,
      "mean_token_accuracy": 0.164144603908062,
      "num_tokens": 11758901.0,
      "step": 6770
    },
    {
      "entropy": 5.56562385559082,
      "epoch": 0.5271347986773002,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004977274325749922,
      "loss": 5.4329,
      "mean_token_accuracy": 0.16389899849891662,
      "num_tokens": 11767202.0,
      "step": 6775
    },
    {
      "entropy": 5.573253870010376,
      "epoch": 0.5275238280490177,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004977235016610011,
      "loss": 5.5385,
      "mean_token_accuracy": 0.15641603320837022,
      "num_tokens": 11776164.0,
      "step": 6780
    },
    {
      "entropy": 5.63574366569519,
      "epoch": 0.5279128574207352,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004977195673675353,
      "loss": 5.4803,
      "mean_token_accuracy": 0.16279528737068177,
      "num_tokens": 11784548.0,
      "step": 6785
    },
    {
      "entropy": 5.597996187210083,
      "epoch": 0.5283018867924528,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004977156296946545,
      "loss": 5.5534,
      "mean_token_accuracy": 0.15933460593223572,
      "num_tokens": 11793249.0,
      "step": 6790
    },
    {
      "entropy": 5.689667177200318,
      "epoch": 0.5286909161641704,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004977116886424187,
      "loss": 5.6144,
      "mean_token_accuracy": 0.14791540801525116,
      "num_tokens": 11801512.0,
      "step": 6795
    },
    {
      "entropy": 5.639977645874024,
      "epoch": 0.529079945535888,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004977077442108873,
      "loss": 5.4758,
      "mean_token_accuracy": 0.16063254177570344,
      "num_tokens": 11810248.0,
      "step": 6800
    },
    {
      "entropy": 5.584291219711304,
      "epoch": 0.5294689749076055,
      "grad_norm": 1.375,
      "learning_rate": 0.0004977037964001204,
      "loss": 5.4666,
      "mean_token_accuracy": 0.16245136857032777,
      "num_tokens": 11819190.0,
      "step": 6805
    },
    {
      "entropy": 5.538795900344849,
      "epoch": 0.5298580042793231,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000497699845210178,
      "loss": 5.5398,
      "mean_token_accuracy": 0.15167467221617698,
      "num_tokens": 11827585.0,
      "step": 6810
    },
    {
      "entropy": 5.7072672843933105,
      "epoch": 0.5302470336510406,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004976958906411197,
      "loss": 5.5459,
      "mean_token_accuracy": 0.1585001215338707,
      "num_tokens": 11836256.0,
      "step": 6815
    },
    {
      "entropy": 5.716243410110474,
      "epoch": 0.5306360630227582,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004976919326930059,
      "loss": 5.6621,
      "mean_token_accuracy": 0.1554935097694397,
      "num_tokens": 11845022.0,
      "step": 6820
    },
    {
      "entropy": 5.5915556907653805,
      "epoch": 0.5310250923944758,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004976879713658964,
      "loss": 5.5293,
      "mean_token_accuracy": 0.15867503434419633,
      "num_tokens": 11853289.0,
      "step": 6825
    },
    {
      "entropy": 5.57533483505249,
      "epoch": 0.5314141217661933,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004976840066598514,
      "loss": 5.4742,
      "mean_token_accuracy": 0.16293567717075347,
      "num_tokens": 11861589.0,
      "step": 6830
    },
    {
      "entropy": 5.715223932266236,
      "epoch": 0.5318031511379109,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004976800385749309,
      "loss": 5.5798,
      "mean_token_accuracy": 0.15770153254270552,
      "num_tokens": 11870313.0,
      "step": 6835
    },
    {
      "entropy": 5.734091663360596,
      "epoch": 0.5321921805096285,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004976760671111954,
      "loss": 5.5738,
      "mean_token_accuracy": 0.15089233294129373,
      "num_tokens": 11879558.0,
      "step": 6840
    },
    {
      "entropy": 5.651639175415039,
      "epoch": 0.532581209881346,
      "grad_norm": 1.421875,
      "learning_rate": 0.000497672092268705,
      "loss": 5.5776,
      "mean_token_accuracy": 0.15512731820344924,
      "num_tokens": 11887951.0,
      "step": 6845
    },
    {
      "entropy": 5.729362916946411,
      "epoch": 0.5329702392530636,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00049766811404752,
      "loss": 5.7686,
      "mean_token_accuracy": 0.15023297369480132,
      "num_tokens": 11896900.0,
      "step": 6850
    },
    {
      "entropy": 5.703069496154785,
      "epoch": 0.5333592686247812,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004976641324477008,
      "loss": 5.5873,
      "mean_token_accuracy": 0.1571554109454155,
      "num_tokens": 11905681.0,
      "step": 6855
    },
    {
      "entropy": 5.608556461334229,
      "epoch": 0.5337482979964987,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004976601474693077,
      "loss": 5.5402,
      "mean_token_accuracy": 0.15932610929012297,
      "num_tokens": 11914662.0,
      "step": 6860
    },
    {
      "entropy": 5.5953371047973635,
      "epoch": 0.5341373273682163,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004976561591124014,
      "loss": 5.5401,
      "mean_token_accuracy": 0.16420037299394608,
      "num_tokens": 11923298.0,
      "step": 6865
    },
    {
      "entropy": 5.679533910751343,
      "epoch": 0.5345263567399339,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004976521673770421,
      "loss": 5.4428,
      "mean_token_accuracy": 0.16081129461526872,
      "num_tokens": 11932206.0,
      "step": 6870
    },
    {
      "entropy": 5.593166303634644,
      "epoch": 0.5349153861116515,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004976481722632907,
      "loss": 5.5182,
      "mean_token_accuracy": 0.16239935159683228,
      "num_tokens": 11940786.0,
      "step": 6875
    },
    {
      "entropy": 5.602063179016113,
      "epoch": 0.535304415483369,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004976441737712076,
      "loss": 5.5302,
      "mean_token_accuracy": 0.15598469376564025,
      "num_tokens": 11949619.0,
      "step": 6880
    },
    {
      "entropy": 5.618106126785278,
      "epoch": 0.5356934448550865,
      "grad_norm": 1.875,
      "learning_rate": 0.0004976401719008536,
      "loss": 5.4673,
      "mean_token_accuracy": 0.16513773053884506,
      "num_tokens": 11957696.0,
      "step": 6885
    },
    {
      "entropy": 5.6020430564880375,
      "epoch": 0.5360824742268041,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004976361666522893,
      "loss": 5.4777,
      "mean_token_accuracy": 0.16478777974843978,
      "num_tokens": 11965622.0,
      "step": 6890
    },
    {
      "entropy": 5.564819049835205,
      "epoch": 0.5364715035985217,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004976321580255755,
      "loss": 5.5604,
      "mean_token_accuracy": 0.15686802566051483,
      "num_tokens": 11973969.0,
      "step": 6895
    },
    {
      "entropy": 5.735695314407349,
      "epoch": 0.5368605329702393,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004976281460207731,
      "loss": 5.5424,
      "mean_token_accuracy": 0.1604687198996544,
      "num_tokens": 11982469.0,
      "step": 6900
    },
    {
      "entropy": 5.748460721969605,
      "epoch": 0.5372495623419569,
      "grad_norm": 1.515625,
      "learning_rate": 0.000497624130637943,
      "loss": 5.6423,
      "mean_token_accuracy": 0.15869580954313278,
      "num_tokens": 11990656.0,
      "step": 6905
    },
    {
      "entropy": 5.631741619110107,
      "epoch": 0.5376385917136743,
      "grad_norm": 1.6015625,
      "learning_rate": 0.000497620111877146,
      "loss": 5.6412,
      "mean_token_accuracy": 0.15766993165016174,
      "num_tokens": 11999394.0,
      "step": 6910
    },
    {
      "entropy": 5.693097543716431,
      "epoch": 0.5380276210853919,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004976160897384431,
      "loss": 5.5945,
      "mean_token_accuracy": 0.1619994305074215,
      "num_tokens": 12007692.0,
      "step": 6915
    },
    {
      "entropy": 5.696098041534424,
      "epoch": 0.5384166504571095,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004976120642218955,
      "loss": 5.581,
      "mean_token_accuracy": 0.15603196769952773,
      "num_tokens": 12016038.0,
      "step": 6920
    },
    {
      "entropy": 5.663388681411743,
      "epoch": 0.5388056798288271,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000497608035327564,
      "loss": 5.6578,
      "mean_token_accuracy": 0.15578502416610718,
      "num_tokens": 12024504.0,
      "step": 6925
    },
    {
      "entropy": 5.620071315765381,
      "epoch": 0.5391947092005447,
      "grad_norm": 1.578125,
      "learning_rate": 0.00049760400305551,
      "loss": 5.5747,
      "mean_token_accuracy": 0.1651473879814148,
      "num_tokens": 12032935.0,
      "step": 6930
    },
    {
      "entropy": 5.692807579040528,
      "epoch": 0.5395837385722622,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004975999674057944,
      "loss": 5.6058,
      "mean_token_accuracy": 0.1548255831003189,
      "num_tokens": 12041907.0,
      "step": 6935
    },
    {
      "entropy": 5.632751417160034,
      "epoch": 0.5399727679439797,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004975959283784787,
      "loss": 5.4339,
      "mean_token_accuracy": 0.1649518758058548,
      "num_tokens": 12050528.0,
      "step": 6940
    },
    {
      "entropy": 5.575006771087646,
      "epoch": 0.5403617973156973,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004975918859736241,
      "loss": 5.4231,
      "mean_token_accuracy": 0.16248443573713303,
      "num_tokens": 12059221.0,
      "step": 6945
    },
    {
      "entropy": 5.634440135955811,
      "epoch": 0.5407508266874149,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004975878401912919,
      "loss": 5.6479,
      "mean_token_accuracy": 0.15099793821573257,
      "num_tokens": 12068131.0,
      "step": 6950
    },
    {
      "entropy": 5.661639404296875,
      "epoch": 0.5411398560591325,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004975837910315436,
      "loss": 5.5298,
      "mean_token_accuracy": 0.1661975249648094,
      "num_tokens": 12076521.0,
      "step": 6955
    },
    {
      "entropy": 5.66230411529541,
      "epoch": 0.54152888543085,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004975797384944405,
      "loss": 5.7098,
      "mean_token_accuracy": 0.15018136128783227,
      "num_tokens": 12085306.0,
      "step": 6960
    },
    {
      "entropy": 5.655632781982422,
      "epoch": 0.5419179148025676,
      "grad_norm": 1.5703125,
      "learning_rate": 0.000497575682580044,
      "loss": 5.4907,
      "mean_token_accuracy": 0.15574343353509904,
      "num_tokens": 12094409.0,
      "step": 6965
    },
    {
      "entropy": 5.595790147781372,
      "epoch": 0.5423069441742852,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000497571623288416,
      "loss": 5.5227,
      "mean_token_accuracy": 0.15902038514614106,
      "num_tokens": 12103385.0,
      "step": 6970
    },
    {
      "entropy": 5.695128440856934,
      "epoch": 0.5426959735460027,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004975675606196177,
      "loss": 5.6476,
      "mean_token_accuracy": 0.15420727133750917,
      "num_tokens": 12112452.0,
      "step": 6975
    },
    {
      "entropy": 5.769511127471924,
      "epoch": 0.5430850029177203,
      "grad_norm": 1.328125,
      "learning_rate": 0.000497563494573711,
      "loss": 5.6461,
      "mean_token_accuracy": 0.15349312052130698,
      "num_tokens": 12120918.0,
      "step": 6980
    },
    {
      "entropy": 5.665116882324218,
      "epoch": 0.5434740322894378,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004975594251507575,
      "loss": 5.592,
      "mean_token_accuracy": 0.1524340584874153,
      "num_tokens": 12129797.0,
      "step": 6985
    },
    {
      "entropy": 5.638438987731933,
      "epoch": 0.5438630616611554,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000497555352350819,
      "loss": 5.5875,
      "mean_token_accuracy": 0.1590471237897873,
      "num_tokens": 12139245.0,
      "step": 6990
    },
    {
      "entropy": 5.724023914337158,
      "epoch": 0.544252091032873,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004975512761739572,
      "loss": 5.753,
      "mean_token_accuracy": 0.15055884122848512,
      "num_tokens": 12147409.0,
      "step": 6995
    },
    {
      "entropy": 5.770619010925293,
      "epoch": 0.5446411204045906,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000497547196620234,
      "loss": 5.5615,
      "mean_token_accuracy": 0.16063691973686217,
      "num_tokens": 12156121.0,
      "step": 7000
    },
    {
      "entropy": 5.601393699645996,
      "epoch": 0.5450301497763081,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004975431136897114,
      "loss": 5.4546,
      "mean_token_accuracy": 0.15758006274700165,
      "num_tokens": 12165307.0,
      "step": 7005
    },
    {
      "entropy": 5.608690309524536,
      "epoch": 0.5454191791480256,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004975390273824512,
      "loss": 5.5458,
      "mean_token_accuracy": 0.1556534193456173,
      "num_tokens": 12173899.0,
      "step": 7010
    },
    {
      "entropy": 5.594245529174804,
      "epoch": 0.5458082085197432,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004975349376985155,
      "loss": 5.532,
      "mean_token_accuracy": 0.157239693403244,
      "num_tokens": 12182898.0,
      "step": 7015
    },
    {
      "entropy": 5.639141654968261,
      "epoch": 0.5461972378914608,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004975308446379663,
      "loss": 5.4257,
      "mean_token_accuracy": 0.1626703605055809,
      "num_tokens": 12191500.0,
      "step": 7020
    },
    {
      "entropy": 5.6254807472229,
      "epoch": 0.5465862672631784,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004975267482008657,
      "loss": 5.5746,
      "mean_token_accuracy": 0.1632894828915596,
      "num_tokens": 12199420.0,
      "step": 7025
    },
    {
      "entropy": 5.611004114151001,
      "epoch": 0.546975296634896,
      "grad_norm": 1.4375,
      "learning_rate": 0.000497522648387276,
      "loss": 5.4928,
      "mean_token_accuracy": 0.16589215099811555,
      "num_tokens": 12207676.0,
      "step": 7030
    },
    {
      "entropy": 5.597789764404297,
      "epoch": 0.5473643260066134,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004975185451972592,
      "loss": 5.5569,
      "mean_token_accuracy": 0.15775526612997054,
      "num_tokens": 12216248.0,
      "step": 7035
    },
    {
      "entropy": 5.665650224685669,
      "epoch": 0.547753355378331,
      "grad_norm": 1.375,
      "learning_rate": 0.0004975144386308776,
      "loss": 5.588,
      "mean_token_accuracy": 0.15302129238843917,
      "num_tokens": 12224588.0,
      "step": 7040
    },
    {
      "entropy": 5.739396286010742,
      "epoch": 0.5481423847500486,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004975103286881936,
      "loss": 5.5993,
      "mean_token_accuracy": 0.15411234349012376,
      "num_tokens": 12234093.0,
      "step": 7045
    },
    {
      "entropy": 5.782606601715088,
      "epoch": 0.5485314141217662,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004975062153692696,
      "loss": 5.5289,
      "mean_token_accuracy": 0.16654838025569915,
      "num_tokens": 12242164.0,
      "step": 7050
    },
    {
      "entropy": 5.646966361999512,
      "epoch": 0.5489204434934838,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004975020986741678,
      "loss": 5.6246,
      "mean_token_accuracy": 0.15668172687292098,
      "num_tokens": 12251204.0,
      "step": 7055
    },
    {
      "entropy": 5.668913650512695,
      "epoch": 0.5493094728652014,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004974979786029509,
      "loss": 5.5663,
      "mean_token_accuracy": 0.1603537067770958,
      "num_tokens": 12259809.0,
      "step": 7060
    },
    {
      "entropy": 5.653614330291748,
      "epoch": 0.5496985022369189,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004974938551556814,
      "loss": 5.5587,
      "mean_token_accuracy": 0.16657759249210358,
      "num_tokens": 12268668.0,
      "step": 7065
    },
    {
      "entropy": 5.587209415435791,
      "epoch": 0.5500875316086364,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004974897283324217,
      "loss": 5.5027,
      "mean_token_accuracy": 0.16190332025289536,
      "num_tokens": 12276823.0,
      "step": 7070
    },
    {
      "entropy": 5.495704174041748,
      "epoch": 0.550476560980354,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004974855981332343,
      "loss": 5.4501,
      "mean_token_accuracy": 0.16237061023712157,
      "num_tokens": 12285950.0,
      "step": 7075
    },
    {
      "entropy": 5.627369546890259,
      "epoch": 0.5508655903520716,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004974814645581823,
      "loss": 5.5635,
      "mean_token_accuracy": 0.15469660311937333,
      "num_tokens": 12295265.0,
      "step": 7080
    },
    {
      "entropy": 5.76141324043274,
      "epoch": 0.5512546197237892,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004974773276073282,
      "loss": 5.5738,
      "mean_token_accuracy": 0.15849540680646895,
      "num_tokens": 12303649.0,
      "step": 7085
    },
    {
      "entropy": 5.638611364364624,
      "epoch": 0.5516436490955067,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004974731872807347,
      "loss": 5.6248,
      "mean_token_accuracy": 0.15033020302653313,
      "num_tokens": 12313303.0,
      "step": 7090
    },
    {
      "entropy": 5.6014965057373045,
      "epoch": 0.5520326784672243,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004974690435784647,
      "loss": 5.5833,
      "mean_token_accuracy": 0.16375116258859634,
      "num_tokens": 12321692.0,
      "step": 7095
    },
    {
      "entropy": 5.572794961929321,
      "epoch": 0.5524217078389418,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004974648965005811,
      "loss": 5.3883,
      "mean_token_accuracy": 0.1655963644385338,
      "num_tokens": 12329908.0,
      "step": 7100
    },
    {
      "entropy": 5.638770723342896,
      "epoch": 0.5528107372106594,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004974607460471466,
      "loss": 5.5623,
      "mean_token_accuracy": 0.15805137604475022,
      "num_tokens": 12339270.0,
      "step": 7105
    },
    {
      "entropy": 5.664997816085815,
      "epoch": 0.553199766582377,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004974565922182246,
      "loss": 5.4481,
      "mean_token_accuracy": 0.15830651968717574,
      "num_tokens": 12347586.0,
      "step": 7110
    },
    {
      "entropy": 5.583152627944946,
      "epoch": 0.5535887959540945,
      "grad_norm": 1.5,
      "learning_rate": 0.0004974524350138777,
      "loss": 5.4329,
      "mean_token_accuracy": 0.15977163314819337,
      "num_tokens": 12356291.0,
      "step": 7115
    },
    {
      "entropy": 5.618222665786743,
      "epoch": 0.5539778253258121,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004974482744341693,
      "loss": 5.6092,
      "mean_token_accuracy": 0.1476268246769905,
      "num_tokens": 12364948.0,
      "step": 7120
    },
    {
      "entropy": 5.7117523670196535,
      "epoch": 0.5543668546975297,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004974441104791624,
      "loss": 5.5624,
      "mean_token_accuracy": 0.15631393045186998,
      "num_tokens": 12373497.0,
      "step": 7125
    },
    {
      "entropy": 5.684345865249634,
      "epoch": 0.5547558840692473,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004974399431489201,
      "loss": 5.5101,
      "mean_token_accuracy": 0.15973858088254927,
      "num_tokens": 12382198.0,
      "step": 7130
    },
    {
      "entropy": 5.554830026626587,
      "epoch": 0.5551449134409648,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004974357724435057,
      "loss": 5.4866,
      "mean_token_accuracy": 0.1644471228122711,
      "num_tokens": 12391122.0,
      "step": 7135
    },
    {
      "entropy": 5.525238418579102,
      "epoch": 0.5555339428126823,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004974315983629825,
      "loss": 5.4274,
      "mean_token_accuracy": 0.16473395228385926,
      "num_tokens": 12399338.0,
      "step": 7140
    },
    {
      "entropy": 5.615933656692505,
      "epoch": 0.5559229721843999,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004974274209074139,
      "loss": 5.5348,
      "mean_token_accuracy": 0.14955220967531205,
      "num_tokens": 12408561.0,
      "step": 7145
    },
    {
      "entropy": 5.596295118331909,
      "epoch": 0.5563120015561175,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004974232400768632,
      "loss": 5.4188,
      "mean_token_accuracy": 0.16242959946393967,
      "num_tokens": 12417727.0,
      "step": 7150
    },
    {
      "entropy": 5.614875650405883,
      "epoch": 0.5567010309278351,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004974190558713939,
      "loss": 5.4376,
      "mean_token_accuracy": 0.16201478540897368,
      "num_tokens": 12426446.0,
      "step": 7155
    },
    {
      "entropy": 5.5842243194580075,
      "epoch": 0.5570900602995527,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004974148682910694,
      "loss": 5.495,
      "mean_token_accuracy": 0.16075697988271714,
      "num_tokens": 12435124.0,
      "step": 7160
    },
    {
      "entropy": 5.619173288345337,
      "epoch": 0.5574790896712701,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004974106773359533,
      "loss": 5.6183,
      "mean_token_accuracy": 0.15286070257425308,
      "num_tokens": 12444073.0,
      "step": 7165
    },
    {
      "entropy": 5.716760969161987,
      "epoch": 0.5578681190429877,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004974064830061091,
      "loss": 5.5898,
      "mean_token_accuracy": 0.1601371094584465,
      "num_tokens": 12452419.0,
      "step": 7170
    },
    {
      "entropy": 5.706222009658814,
      "epoch": 0.5582571484147053,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004974022853016006,
      "loss": 5.5499,
      "mean_token_accuracy": 0.15545956194400787,
      "num_tokens": 12461488.0,
      "step": 7175
    },
    {
      "entropy": 5.609360027313232,
      "epoch": 0.5586461777864229,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004973980842224913,
      "loss": 5.4911,
      "mean_token_accuracy": 0.15602228343486785,
      "num_tokens": 12470344.0,
      "step": 7180
    },
    {
      "entropy": 5.649265336990356,
      "epoch": 0.5590352071581405,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004973938797688452,
      "loss": 5.4932,
      "mean_token_accuracy": 0.15788175910711288,
      "num_tokens": 12479189.0,
      "step": 7185
    },
    {
      "entropy": 5.640626525878906,
      "epoch": 0.559424236529858,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004973896719407259,
      "loss": 5.5243,
      "mean_token_accuracy": 0.16059497147798538,
      "num_tokens": 12488199.0,
      "step": 7190
    },
    {
      "entropy": 5.557397651672363,
      "epoch": 0.5598132659015755,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004973854607381972,
      "loss": 5.4083,
      "mean_token_accuracy": 0.16683684289455414,
      "num_tokens": 12496704.0,
      "step": 7195
    },
    {
      "entropy": 5.537837314605713,
      "epoch": 0.5602022952732931,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004973812461613232,
      "loss": 5.4111,
      "mean_token_accuracy": 0.16901323050260544,
      "num_tokens": 12504724.0,
      "step": 7200
    },
    {
      "entropy": 5.606007385253906,
      "epoch": 0.5605913246450107,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004973770282101677,
      "loss": 5.595,
      "mean_token_accuracy": 0.15539490282535554,
      "num_tokens": 12514614.0,
      "step": 7205
    },
    {
      "entropy": 5.725541639328003,
      "epoch": 0.5609803540167283,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004973728068847947,
      "loss": 5.6217,
      "mean_token_accuracy": 0.1517338916659355,
      "num_tokens": 12522744.0,
      "step": 7210
    },
    {
      "entropy": 5.629028463363648,
      "epoch": 0.5613693833884458,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004973685821852684,
      "loss": 5.5292,
      "mean_token_accuracy": 0.16160846799612044,
      "num_tokens": 12531783.0,
      "step": 7215
    },
    {
      "entropy": 5.619050025939941,
      "epoch": 0.5617584127601634,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004973643541116528,
      "loss": 5.5904,
      "mean_token_accuracy": 0.1496799409389496,
      "num_tokens": 12541768.0,
      "step": 7220
    },
    {
      "entropy": 5.7859978675842285,
      "epoch": 0.562147442131881,
      "grad_norm": 1.453125,
      "learning_rate": 0.000497360122664012,
      "loss": 5.6416,
      "mean_token_accuracy": 0.15416546911001205,
      "num_tokens": 12550684.0,
      "step": 7225
    },
    {
      "entropy": 5.717968082427978,
      "epoch": 0.5625364715035985,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004973558878424104,
      "loss": 5.5466,
      "mean_token_accuracy": 0.15934956967830657,
      "num_tokens": 12559948.0,
      "step": 7230
    },
    {
      "entropy": 5.561044692993164,
      "epoch": 0.5629255008753161,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004973516496469119,
      "loss": 5.4869,
      "mean_token_accuracy": 0.16337195932865142,
      "num_tokens": 12568681.0,
      "step": 7235
    },
    {
      "entropy": 5.691223812103272,
      "epoch": 0.5633145302470336,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004973474080775811,
      "loss": 5.6114,
      "mean_token_accuracy": 0.152906334400177,
      "num_tokens": 12577724.0,
      "step": 7240
    },
    {
      "entropy": 5.76364803314209,
      "epoch": 0.5637035596187512,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004973431631344824,
      "loss": 5.5306,
      "mean_token_accuracy": 0.15633030086755753,
      "num_tokens": 12585994.0,
      "step": 7245
    },
    {
      "entropy": 5.6505897521972654,
      "epoch": 0.5640925889904688,
      "grad_norm": 1.234375,
      "learning_rate": 0.00049733891481768,
      "loss": 5.6015,
      "mean_token_accuracy": 0.15584657341241837,
      "num_tokens": 12595525.0,
      "step": 7250
    },
    {
      "entropy": 5.647922468185425,
      "epoch": 0.5644816183621864,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004973346631272384,
      "loss": 5.4333,
      "mean_token_accuracy": 0.16841700673103333,
      "num_tokens": 12603645.0,
      "step": 7255
    },
    {
      "entropy": 5.587335395812988,
      "epoch": 0.5648706477339039,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004973304080632222,
      "loss": 5.5055,
      "mean_token_accuracy": 0.15742655992507934,
      "num_tokens": 12612165.0,
      "step": 7260
    },
    {
      "entropy": 5.591152763366699,
      "epoch": 0.5652596771056215,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000497326149625696,
      "loss": 5.4214,
      "mean_token_accuracy": 0.16940227746963502,
      "num_tokens": 12620678.0,
      "step": 7265
    },
    {
      "entropy": 5.681361198425293,
      "epoch": 0.565648706477339,
      "grad_norm": 1.375,
      "learning_rate": 0.0004973218878147244,
      "loss": 5.6155,
      "mean_token_accuracy": 0.15588987991213799,
      "num_tokens": 12628745.0,
      "step": 7270
    },
    {
      "entropy": 5.70417628288269,
      "epoch": 0.5660377358490566,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004973176226303719,
      "loss": 5.6584,
      "mean_token_accuracy": 0.1473071962594986,
      "num_tokens": 12637448.0,
      "step": 7275
    },
    {
      "entropy": 5.679449367523193,
      "epoch": 0.5664267652207742,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004973133540727033,
      "loss": 5.5324,
      "mean_token_accuracy": 0.1581710919737816,
      "num_tokens": 12645804.0,
      "step": 7280
    },
    {
      "entropy": 5.655721378326416,
      "epoch": 0.5668157945924918,
      "grad_norm": 1.375,
      "learning_rate": 0.0004973090821417835,
      "loss": 5.4942,
      "mean_token_accuracy": 0.15874693989753724,
      "num_tokens": 12654156.0,
      "step": 7285
    },
    {
      "entropy": 5.6041888236999515,
      "epoch": 0.5672048239642093,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004973048068376772,
      "loss": 5.4653,
      "mean_token_accuracy": 0.16248227059841155,
      "num_tokens": 12662271.0,
      "step": 7290
    },
    {
      "entropy": 5.724769163131714,
      "epoch": 0.5675938533359268,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004973005281604492,
      "loss": 5.6371,
      "mean_token_accuracy": 0.15151395946741103,
      "num_tokens": 12671016.0,
      "step": 7295
    },
    {
      "entropy": 5.633071088790894,
      "epoch": 0.5679828827076444,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004972962461101646,
      "loss": 5.5932,
      "mean_token_accuracy": 0.15309520661830903,
      "num_tokens": 12679593.0,
      "step": 7300
    },
    {
      "entropy": 5.625471401214599,
      "epoch": 0.568371912079362,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004972919606868883,
      "loss": 5.5179,
      "mean_token_accuracy": 0.16120276600122452,
      "num_tokens": 12688715.0,
      "step": 7305
    },
    {
      "entropy": 5.643703126907349,
      "epoch": 0.5687609414510796,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004972876718906852,
      "loss": 5.5263,
      "mean_token_accuracy": 0.15836018621921538,
      "num_tokens": 12698078.0,
      "step": 7310
    },
    {
      "entropy": 5.5868239402771,
      "epoch": 0.5691499708227972,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004972833797216206,
      "loss": 5.5126,
      "mean_token_accuracy": 0.16403855681419371,
      "num_tokens": 12706375.0,
      "step": 7315
    },
    {
      "entropy": 5.724744892120361,
      "epoch": 0.5695390001945146,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004972790841797595,
      "loss": 5.7066,
      "mean_token_accuracy": 0.15244438797235488,
      "num_tokens": 12715727.0,
      "step": 7320
    },
    {
      "entropy": 5.67259030342102,
      "epoch": 0.5699280295662322,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000497274785265167,
      "loss": 5.6122,
      "mean_token_accuracy": 0.14762853384017943,
      "num_tokens": 12725565.0,
      "step": 7325
    },
    {
      "entropy": 5.743588399887085,
      "epoch": 0.5703170589379498,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004972704829779086,
      "loss": 5.631,
      "mean_token_accuracy": 0.15871698409318924,
      "num_tokens": 12733498.0,
      "step": 7330
    },
    {
      "entropy": 5.620137357711792,
      "epoch": 0.5707060883096674,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004972661773180492,
      "loss": 5.4476,
      "mean_token_accuracy": 0.1599652126431465,
      "num_tokens": 12741635.0,
      "step": 7335
    },
    {
      "entropy": 5.520822048187256,
      "epoch": 0.571095117681385,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004972618682856545,
      "loss": 5.4631,
      "mean_token_accuracy": 0.16328149437904357,
      "num_tokens": 12750973.0,
      "step": 7340
    },
    {
      "entropy": 5.571967029571534,
      "epoch": 0.5714841470531025,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004972575558807896,
      "loss": 5.5372,
      "mean_token_accuracy": 0.16574114114046096,
      "num_tokens": 12759075.0,
      "step": 7345
    },
    {
      "entropy": 5.700579309463501,
      "epoch": 0.57187317642482,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004972532401035201,
      "loss": 5.5225,
      "mean_token_accuracy": 0.15687755197286607,
      "num_tokens": 12767485.0,
      "step": 7350
    },
    {
      "entropy": 5.705129623413086,
      "epoch": 0.5722622057965376,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004972489209539114,
      "loss": 5.4884,
      "mean_token_accuracy": 0.16281367540359498,
      "num_tokens": 12775128.0,
      "step": 7355
    },
    {
      "entropy": 5.561953735351563,
      "epoch": 0.5726512351682552,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004972445984320292,
      "loss": 5.4076,
      "mean_token_accuracy": 0.1676819458603859,
      "num_tokens": 12783243.0,
      "step": 7360
    },
    {
      "entropy": 5.649951601028443,
      "epoch": 0.5730402645399728,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004972402725379388,
      "loss": 5.6279,
      "mean_token_accuracy": 0.16052431762218475,
      "num_tokens": 12791613.0,
      "step": 7365
    },
    {
      "entropy": 5.6840691566467285,
      "epoch": 0.5734292939116903,
      "grad_norm": 1.421875,
      "learning_rate": 0.000497235943271706,
      "loss": 5.559,
      "mean_token_accuracy": 0.15703178942203522,
      "num_tokens": 12799817.0,
      "step": 7370
    },
    {
      "entropy": 5.691116380691528,
      "epoch": 0.5738183232834079,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004972316106333966,
      "loss": 5.5644,
      "mean_token_accuracy": 0.15355917811393738,
      "num_tokens": 12808994.0,
      "step": 7375
    },
    {
      "entropy": 5.637736511230469,
      "epoch": 0.5742073526551255,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004972272746230761,
      "loss": 5.5256,
      "mean_token_accuracy": 0.1580606669187546,
      "num_tokens": 12817191.0,
      "step": 7380
    },
    {
      "entropy": 5.6404008865356445,
      "epoch": 0.574596382026843,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004972229352408104,
      "loss": 5.4956,
      "mean_token_accuracy": 0.15748804658651352,
      "num_tokens": 12825773.0,
      "step": 7385
    },
    {
      "entropy": 5.73073410987854,
      "epoch": 0.5749854113985606,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004972185924866655,
      "loss": 5.5644,
      "mean_token_accuracy": 0.16047893315553666,
      "num_tokens": 12834489.0,
      "step": 7390
    },
    {
      "entropy": 5.677036237716675,
      "epoch": 0.5753744407702781,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004972142463607071,
      "loss": 5.5498,
      "mean_token_accuracy": 0.1610588014125824,
      "num_tokens": 12843179.0,
      "step": 7395
    },
    {
      "entropy": 5.7206519603729244,
      "epoch": 0.5757634701419957,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004972098968630012,
      "loss": 5.5736,
      "mean_token_accuracy": 0.15535116493701934,
      "num_tokens": 12851243.0,
      "step": 7400
    },
    {
      "entropy": 5.598516273498535,
      "epoch": 0.5761524995137133,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004972055439936137,
      "loss": 5.6291,
      "mean_token_accuracy": 0.15630768090486527,
      "num_tokens": 12860267.0,
      "step": 7405
    },
    {
      "entropy": 5.757432508468628,
      "epoch": 0.5765415288854309,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004972011877526109,
      "loss": 5.6497,
      "mean_token_accuracy": 0.15755585208535194,
      "num_tokens": 12870022.0,
      "step": 7410
    },
    {
      "entropy": 5.694835472106933,
      "epoch": 0.5769305582571485,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004971968281400587,
      "loss": 5.5952,
      "mean_token_accuracy": 0.1492684304714203,
      "num_tokens": 12879227.0,
      "step": 7415
    },
    {
      "entropy": 5.661354351043701,
      "epoch": 0.5773195876288659,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004971924651560233,
      "loss": 5.5483,
      "mean_token_accuracy": 0.15837249904870987,
      "num_tokens": 12888180.0,
      "step": 7420
    },
    {
      "entropy": 5.675881195068359,
      "epoch": 0.5777086170005835,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004971880988005709,
      "loss": 5.5394,
      "mean_token_accuracy": 0.16022545620799064,
      "num_tokens": 12896767.0,
      "step": 7425
    },
    {
      "entropy": 5.618383169174194,
      "epoch": 0.5780976463723011,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004971837290737677,
      "loss": 5.4341,
      "mean_token_accuracy": 0.16697876006364823,
      "num_tokens": 12904766.0,
      "step": 7430
    },
    {
      "entropy": 5.529607534408569,
      "epoch": 0.5784866757440187,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004971793559756801,
      "loss": 5.5357,
      "mean_token_accuracy": 0.1580686628818512,
      "num_tokens": 12913808.0,
      "step": 7435
    },
    {
      "entropy": 5.590646886825562,
      "epoch": 0.5788757051157363,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004971749795063744,
      "loss": 5.5143,
      "mean_token_accuracy": 0.15939191728830338,
      "num_tokens": 12923103.0,
      "step": 7440
    },
    {
      "entropy": 5.6191099166870115,
      "epoch": 0.5792647344874537,
      "grad_norm": 1.546875,
      "learning_rate": 0.000497170599665917,
      "loss": 5.473,
      "mean_token_accuracy": 0.16246490627527238,
      "num_tokens": 12931168.0,
      "step": 7445
    },
    {
      "entropy": 5.548571968078614,
      "epoch": 0.5796537638591713,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004971662164543743,
      "loss": 5.5678,
      "mean_token_accuracy": 0.16391166001558305,
      "num_tokens": 12939875.0,
      "step": 7450
    },
    {
      "entropy": 5.61009259223938,
      "epoch": 0.5800427932308889,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004971618298718131,
      "loss": 5.4683,
      "mean_token_accuracy": 0.16916644275188447,
      "num_tokens": 12947883.0,
      "step": 7455
    },
    {
      "entropy": 5.789393091201783,
      "epoch": 0.5804318226026065,
      "grad_norm": 1.625,
      "learning_rate": 0.0004971574399182995,
      "loss": 5.7034,
      "mean_token_accuracy": 0.14763562753796577,
      "num_tokens": 12956940.0,
      "step": 7460
    },
    {
      "entropy": 5.626276636123658,
      "epoch": 0.5808208519743241,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004971530465939005,
      "loss": 5.4359,
      "mean_token_accuracy": 0.1634249910712242,
      "num_tokens": 12965003.0,
      "step": 7465
    },
    {
      "entropy": 5.603636360168457,
      "epoch": 0.5812098813460417,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004971486498986825,
      "loss": 5.4668,
      "mean_token_accuracy": 0.16253004223108292,
      "num_tokens": 12973323.0,
      "step": 7470
    },
    {
      "entropy": 5.600870990753174,
      "epoch": 0.5815989107177592,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004971442498327124,
      "loss": 5.5004,
      "mean_token_accuracy": 0.16327155977487565,
      "num_tokens": 12982654.0,
      "step": 7475
    },
    {
      "entropy": 5.62065544128418,
      "epoch": 0.5819879400894767,
      "grad_norm": 1.5,
      "learning_rate": 0.0004971398463960568,
      "loss": 5.5431,
      "mean_token_accuracy": 0.16062046885490416,
      "num_tokens": 12991781.0,
      "step": 7480
    },
    {
      "entropy": 5.694522523880005,
      "epoch": 0.5823769694611943,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004971354395887827,
      "loss": 5.6288,
      "mean_token_accuracy": 0.15725939124822616,
      "num_tokens": 13000539.0,
      "step": 7485
    },
    {
      "entropy": 5.702580308914184,
      "epoch": 0.5827659988329119,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004971310294109569,
      "loss": 5.4965,
      "mean_token_accuracy": 0.1644050046801567,
      "num_tokens": 13008638.0,
      "step": 7490
    },
    {
      "entropy": 5.61459527015686,
      "epoch": 0.5831550282046295,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004971266158626461,
      "loss": 5.4854,
      "mean_token_accuracy": 0.16098864525556564,
      "num_tokens": 13017217.0,
      "step": 7495
    },
    {
      "entropy": 5.55291600227356,
      "epoch": 0.583544057576347,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004971221989439177,
      "loss": 5.5311,
      "mean_token_accuracy": 0.15793122053146363,
      "num_tokens": 13025278.0,
      "step": 7500
    },
    {
      "entropy": 5.603717041015625,
      "epoch": 0.5839330869480646,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004971177786548384,
      "loss": 5.5575,
      "mean_token_accuracy": 0.15922897905111313,
      "num_tokens": 13034484.0,
      "step": 7505
    },
    {
      "entropy": 5.727790403366089,
      "epoch": 0.5843221163197821,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004971133549954753,
      "loss": 5.629,
      "mean_token_accuracy": 0.1499597117304802,
      "num_tokens": 13042831.0,
      "step": 7510
    },
    {
      "entropy": 5.727259254455566,
      "epoch": 0.5847111456914997,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004971089279658954,
      "loss": 5.5922,
      "mean_token_accuracy": 0.15369003713130952,
      "num_tokens": 13051614.0,
      "step": 7515
    },
    {
      "entropy": 5.634461355209351,
      "epoch": 0.5851001750632173,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004971044975661663,
      "loss": 5.5506,
      "mean_token_accuracy": 0.15672828108072281,
      "num_tokens": 13060179.0,
      "step": 7520
    },
    {
      "entropy": 5.575223398208618,
      "epoch": 0.5854892044349348,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004971000637963549,
      "loss": 5.4628,
      "mean_token_accuracy": 0.1732001096010208,
      "num_tokens": 13068328.0,
      "step": 7525
    },
    {
      "entropy": 5.537419319152832,
      "epoch": 0.5858782338066524,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004970956266565285,
      "loss": 5.5145,
      "mean_token_accuracy": 0.1610039070248604,
      "num_tokens": 13076879.0,
      "step": 7530
    },
    {
      "entropy": 5.68983154296875,
      "epoch": 0.58626726317837,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004970911861467544,
      "loss": 5.5583,
      "mean_token_accuracy": 0.1535510741174221,
      "num_tokens": 13085123.0,
      "step": 7535
    },
    {
      "entropy": 5.558996152877808,
      "epoch": 0.5866562925500876,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004970867422671002,
      "loss": 5.5188,
      "mean_token_accuracy": 0.16626303195953368,
      "num_tokens": 13093884.0,
      "step": 7540
    },
    {
      "entropy": 5.70104022026062,
      "epoch": 0.5870453219218051,
      "grad_norm": 1.359375,
      "learning_rate": 0.000497082295017633,
      "loss": 5.5751,
      "mean_token_accuracy": 0.16042875796556472,
      "num_tokens": 13102581.0,
      "step": 7545
    },
    {
      "entropy": 5.683917903900147,
      "epoch": 0.5874343512935226,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004970778443984206,
      "loss": 5.5609,
      "mean_token_accuracy": 0.15632733553647996,
      "num_tokens": 13111307.0,
      "step": 7550
    },
    {
      "entropy": 5.681131172180176,
      "epoch": 0.5878233806652402,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004970733904095303,
      "loss": 5.5639,
      "mean_token_accuracy": 0.15917730927467347,
      "num_tokens": 13119279.0,
      "step": 7555
    },
    {
      "entropy": 5.6647991180419925,
      "epoch": 0.5882124100369578,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004970689330510298,
      "loss": 5.5906,
      "mean_token_accuracy": 0.1582401856780052,
      "num_tokens": 13127753.0,
      "step": 7560
    },
    {
      "entropy": 5.680768918991089,
      "epoch": 0.5886014394086754,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004970644723229868,
      "loss": 5.636,
      "mean_token_accuracy": 0.15444589108228685,
      "num_tokens": 13136881.0,
      "step": 7565
    },
    {
      "entropy": 5.652431774139404,
      "epoch": 0.588990468780393,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004970600082254687,
      "loss": 5.4991,
      "mean_token_accuracy": 0.15578779727220535,
      "num_tokens": 13145291.0,
      "step": 7570
    },
    {
      "entropy": 5.579540681838989,
      "epoch": 0.5893794981521104,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004970555407585436,
      "loss": 5.4579,
      "mean_token_accuracy": 0.16331358551979064,
      "num_tokens": 13153560.0,
      "step": 7575
    },
    {
      "entropy": 5.626413154602051,
      "epoch": 0.589768527523828,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000497051069922279,
      "loss": 5.5139,
      "mean_token_accuracy": 0.15920064300298692,
      "num_tokens": 13162304.0,
      "step": 7580
    },
    {
      "entropy": 5.6824174404144285,
      "epoch": 0.5901575568955456,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004970465957167429,
      "loss": 5.5345,
      "mean_token_accuracy": 0.16362176686525345,
      "num_tokens": 13171404.0,
      "step": 7585
    },
    {
      "entropy": 5.739548587799073,
      "epoch": 0.5905465862672632,
      "grad_norm": 1.546875,
      "learning_rate": 0.000497042118142003,
      "loss": 5.6142,
      "mean_token_accuracy": 0.15650009736418724,
      "num_tokens": 13180001.0,
      "step": 7590
    },
    {
      "entropy": 5.686169528961182,
      "epoch": 0.5909356156389808,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004970376371981275,
      "loss": 5.552,
      "mean_token_accuracy": 0.15728999227285384,
      "num_tokens": 13189749.0,
      "step": 7595
    },
    {
      "entropy": 5.586195945739746,
      "epoch": 0.5913246450106983,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004970331528851842,
      "loss": 5.5261,
      "mean_token_accuracy": 0.16812347173690795,
      "num_tokens": 13199169.0,
      "step": 7600
    },
    {
      "entropy": 5.562873315811157,
      "epoch": 0.5917136743824158,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004970286652032412,
      "loss": 5.4885,
      "mean_token_accuracy": 0.16665935814380645,
      "num_tokens": 13207014.0,
      "step": 7605
    },
    {
      "entropy": 5.640641927719116,
      "epoch": 0.5921027037541334,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004970241741523667,
      "loss": 5.4406,
      "mean_token_accuracy": 0.1629709154367447,
      "num_tokens": 13215333.0,
      "step": 7610
    },
    {
      "entropy": 5.642063808441162,
      "epoch": 0.592491733125851,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004970196797326286,
      "loss": 5.5707,
      "mean_token_accuracy": 0.15491992235183716,
      "num_tokens": 13223740.0,
      "step": 7615
    },
    {
      "entropy": 5.596137189865113,
      "epoch": 0.5928807624975686,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004970151819440955,
      "loss": 5.5128,
      "mean_token_accuracy": 0.16230032742023467,
      "num_tokens": 13232065.0,
      "step": 7620
    },
    {
      "entropy": 5.629691648483276,
      "epoch": 0.5932697918692861,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004970106807868351,
      "loss": 5.5009,
      "mean_token_accuracy": 0.1631051391363144,
      "num_tokens": 13241171.0,
      "step": 7625
    },
    {
      "entropy": 5.5375292778015135,
      "epoch": 0.5936588212410037,
      "grad_norm": 1.375,
      "learning_rate": 0.000497006176260916,
      "loss": 5.4181,
      "mean_token_accuracy": 0.16693178415298462,
      "num_tokens": 13249553.0,
      "step": 7630
    },
    {
      "entropy": 5.618206977844238,
      "epoch": 0.5940478506127213,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004970016683664068,
      "loss": 5.4889,
      "mean_token_accuracy": 0.1628669649362564,
      "num_tokens": 13258585.0,
      "step": 7635
    },
    {
      "entropy": 5.637617492675782,
      "epoch": 0.5944368799844388,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004969971571033754,
      "loss": 5.5552,
      "mean_token_accuracy": 0.16011955440044404,
      "num_tokens": 13267317.0,
      "step": 7640
    },
    {
      "entropy": 5.575131750106811,
      "epoch": 0.5948259093561564,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004969926424718906,
      "loss": 5.37,
      "mean_token_accuracy": 0.16295740604400635,
      "num_tokens": 13275705.0,
      "step": 7645
    },
    {
      "entropy": 5.6024312496185305,
      "epoch": 0.5952149387278739,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004969881244720207,
      "loss": 5.5182,
      "mean_token_accuracy": 0.1615263268351555,
      "num_tokens": 13284046.0,
      "step": 7650
    },
    {
      "entropy": 5.58196268081665,
      "epoch": 0.5956039680995915,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004969836031038343,
      "loss": 5.5,
      "mean_token_accuracy": 0.16613336950540541,
      "num_tokens": 13292451.0,
      "step": 7655
    },
    {
      "entropy": 5.714156723022461,
      "epoch": 0.5959929974713091,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004969790783674002,
      "loss": 5.6349,
      "mean_token_accuracy": 0.1574876733124256,
      "num_tokens": 13300788.0,
      "step": 7660
    },
    {
      "entropy": 5.666077136993408,
      "epoch": 0.5963820268430267,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004969745502627868,
      "loss": 5.5477,
      "mean_token_accuracy": 0.16628242731094361,
      "num_tokens": 13309235.0,
      "step": 7665
    },
    {
      "entropy": 5.58871488571167,
      "epoch": 0.5967710562147442,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004969700187900629,
      "loss": 5.4521,
      "mean_token_accuracy": 0.1600484788417816,
      "num_tokens": 13317331.0,
      "step": 7670
    },
    {
      "entropy": 5.575796985626221,
      "epoch": 0.5971600855864618,
      "grad_norm": 1.75,
      "learning_rate": 0.0004969654839492973,
      "loss": 5.4885,
      "mean_token_accuracy": 0.1504474826157093,
      "num_tokens": 13327296.0,
      "step": 7675
    },
    {
      "entropy": 5.673458480834961,
      "epoch": 0.5975491149581793,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004969609457405588,
      "loss": 5.5549,
      "mean_token_accuracy": 0.15951465219259262,
      "num_tokens": 13336650.0,
      "step": 7680
    },
    {
      "entropy": 5.641073751449585,
      "epoch": 0.5979381443298969,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004969564041639163,
      "loss": 5.5005,
      "mean_token_accuracy": 0.15823953747749328,
      "num_tokens": 13345353.0,
      "step": 7685
    },
    {
      "entropy": 5.656012964248657,
      "epoch": 0.5983271737016145,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004969518592194386,
      "loss": 5.5111,
      "mean_token_accuracy": 0.16030848026275635,
      "num_tokens": 13354302.0,
      "step": 7690
    },
    {
      "entropy": 5.549968671798706,
      "epoch": 0.5987162030733321,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004969473109071946,
      "loss": 5.4504,
      "mean_token_accuracy": 0.163647423684597,
      "num_tokens": 13362950.0,
      "step": 7695
    },
    {
      "entropy": 5.585221242904663,
      "epoch": 0.5991052324450497,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004969427592272535,
      "loss": 5.5745,
      "mean_token_accuracy": 0.1560671731829643,
      "num_tokens": 13371383.0,
      "step": 7700
    },
    {
      "entropy": 5.669369792938232,
      "epoch": 0.5994942618167671,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004969382041796843,
      "loss": 5.5383,
      "mean_token_accuracy": 0.1628669172525406,
      "num_tokens": 13379767.0,
      "step": 7705
    },
    {
      "entropy": 5.655380296707153,
      "epoch": 0.5998832911884847,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000496933645764556,
      "loss": 5.5267,
      "mean_token_accuracy": 0.15433267802000045,
      "num_tokens": 13387970.0,
      "step": 7710
    },
    {
      "entropy": 5.545374393463135,
      "epoch": 0.6002723205602023,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004969290839819381,
      "loss": 5.4821,
      "mean_token_accuracy": 0.16639994978904724,
      "num_tokens": 13396605.0,
      "step": 7715
    },
    {
      "entropy": 5.582101249694825,
      "epoch": 0.6006613499319199,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004969245188318994,
      "loss": 5.4222,
      "mean_token_accuracy": 0.16884092390537261,
      "num_tokens": 13406028.0,
      "step": 7720
    },
    {
      "entropy": 5.643075370788575,
      "epoch": 0.6010503793036375,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004969199503145093,
      "loss": 5.4808,
      "mean_token_accuracy": 0.16709552854299545,
      "num_tokens": 13414927.0,
      "step": 7725
    },
    {
      "entropy": 5.538677787780761,
      "epoch": 0.601439408675355,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004969153784298374,
      "loss": 5.4197,
      "mean_token_accuracy": 0.16179912090301513,
      "num_tokens": 13423020.0,
      "step": 7730
    },
    {
      "entropy": 5.501712656021118,
      "epoch": 0.6018284380470725,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004969108031779527,
      "loss": 5.4259,
      "mean_token_accuracy": 0.16240470558404924,
      "num_tokens": 13431514.0,
      "step": 7735
    },
    {
      "entropy": 5.627806663513184,
      "epoch": 0.6022174674187901,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004969062245589247,
      "loss": 5.6042,
      "mean_token_accuracy": 0.15487990602850915,
      "num_tokens": 13440333.0,
      "step": 7740
    },
    {
      "entropy": 5.58061785697937,
      "epoch": 0.6026064967905077,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004969016425728231,
      "loss": 5.5314,
      "mean_token_accuracy": 0.1586664631962776,
      "num_tokens": 13448841.0,
      "step": 7745
    },
    {
      "entropy": 5.669718933105469,
      "epoch": 0.6029955261622253,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004968970572197172,
      "loss": 5.6055,
      "mean_token_accuracy": 0.15758649706840516,
      "num_tokens": 13457846.0,
      "step": 7750
    },
    {
      "entropy": 5.730754661560058,
      "epoch": 0.6033845555339428,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004968924684996765,
      "loss": 5.6091,
      "mean_token_accuracy": 0.15334318280220033,
      "num_tokens": 13466242.0,
      "step": 7755
    },
    {
      "entropy": 5.67420654296875,
      "epoch": 0.6037735849056604,
      "grad_norm": 1.46875,
      "learning_rate": 0.000496887876412771,
      "loss": 5.5405,
      "mean_token_accuracy": 0.15616600215435028,
      "num_tokens": 13474639.0,
      "step": 7760
    },
    {
      "entropy": 5.563692665100097,
      "epoch": 0.6041626142773779,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004968832809590701,
      "loss": 5.436,
      "mean_token_accuracy": 0.16429483294487,
      "num_tokens": 13483673.0,
      "step": 7765
    },
    {
      "entropy": 5.667019462585449,
      "epoch": 0.6045516436490955,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004968786821386435,
      "loss": 5.5529,
      "mean_token_accuracy": 0.15738777071237564,
      "num_tokens": 13493001.0,
      "step": 7770
    },
    {
      "entropy": 5.751283121109009,
      "epoch": 0.6049406730208131,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004968740799515611,
      "loss": 5.5688,
      "mean_token_accuracy": 0.15729465782642366,
      "num_tokens": 13500878.0,
      "step": 7775
    },
    {
      "entropy": 5.6461762428283695,
      "epoch": 0.6053297023925306,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004968694743978926,
      "loss": 5.6267,
      "mean_token_accuracy": 0.1560249738395214,
      "num_tokens": 13509945.0,
      "step": 7780
    },
    {
      "entropy": 5.580125570297241,
      "epoch": 0.6057187317642482,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004968648654777081,
      "loss": 5.5488,
      "mean_token_accuracy": 0.16145080476999282,
      "num_tokens": 13520024.0,
      "step": 7785
    },
    {
      "entropy": 5.669335126876831,
      "epoch": 0.6061077611359658,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004968602531910773,
      "loss": 5.644,
      "mean_token_accuracy": 0.15712960958480834,
      "num_tokens": 13529136.0,
      "step": 7790
    },
    {
      "entropy": 5.628845357894898,
      "epoch": 0.6064967905076833,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004968556375380704,
      "loss": 5.452,
      "mean_token_accuracy": 0.16029975712299346,
      "num_tokens": 13537001.0,
      "step": 7795
    },
    {
      "entropy": 5.619492673873902,
      "epoch": 0.6068858198794009,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004968510185187571,
      "loss": 5.4475,
      "mean_token_accuracy": 0.16057668179273604,
      "num_tokens": 13545520.0,
      "step": 7800
    },
    {
      "entropy": 5.581534576416016,
      "epoch": 0.6072748492511184,
      "grad_norm": 1.40625,
      "learning_rate": 0.000496846396133208,
      "loss": 5.5005,
      "mean_token_accuracy": 0.1582297757267952,
      "num_tokens": 13554770.0,
      "step": 7805
    },
    {
      "entropy": 5.701692295074463,
      "epoch": 0.607663878622836,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004968417703814928,
      "loss": 5.5476,
      "mean_token_accuracy": 0.15719548612833023,
      "num_tokens": 13562797.0,
      "step": 7810
    },
    {
      "entropy": 5.613367748260498,
      "epoch": 0.6080529079945536,
      "grad_norm": 1.5,
      "learning_rate": 0.0004968371412636818,
      "loss": 5.5004,
      "mean_token_accuracy": 0.16422986686229707,
      "num_tokens": 13572017.0,
      "step": 7815
    },
    {
      "entropy": 5.6862372875213625,
      "epoch": 0.6084419373662712,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004968325087798453,
      "loss": 5.5231,
      "mean_token_accuracy": 0.1657199665904045,
      "num_tokens": 13580245.0,
      "step": 7820
    },
    {
      "entropy": 5.597463941574096,
      "epoch": 0.6088309667379888,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004968278729300536,
      "loss": 5.4988,
      "mean_token_accuracy": 0.15661692023277282,
      "num_tokens": 13588719.0,
      "step": 7825
    },
    {
      "entropy": 5.66688551902771,
      "epoch": 0.6092199961097062,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004968232337143769,
      "loss": 5.5555,
      "mean_token_accuracy": 0.16355859488248825,
      "num_tokens": 13597345.0,
      "step": 7830
    },
    {
      "entropy": 5.707019233703614,
      "epoch": 0.6096090254814238,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004968185911328858,
      "loss": 5.5719,
      "mean_token_accuracy": 0.15881615430116652,
      "num_tokens": 13605468.0,
      "step": 7835
    },
    {
      "entropy": 5.659979629516601,
      "epoch": 0.6099980548531414,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004968139451856506,
      "loss": 5.5829,
      "mean_token_accuracy": 0.15562816858291625,
      "num_tokens": 13614085.0,
      "step": 7840
    },
    {
      "entropy": 5.596334028244018,
      "epoch": 0.610387084224859,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004968092958727419,
      "loss": 5.4023,
      "mean_token_accuracy": 0.16623687595129014,
      "num_tokens": 13622370.0,
      "step": 7845
    },
    {
      "entropy": 5.6452563285827635,
      "epoch": 0.6107761135965766,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00049680464319423,
      "loss": 5.5358,
      "mean_token_accuracy": 0.162747386097908,
      "num_tokens": 13630615.0,
      "step": 7850
    },
    {
      "entropy": 5.621069622039795,
      "epoch": 0.6111651429682942,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004967999871501858,
      "loss": 5.4597,
      "mean_token_accuracy": 0.16595776081085206,
      "num_tokens": 13638461.0,
      "step": 7855
    },
    {
      "entropy": 5.525365924835205,
      "epoch": 0.6115541723400116,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004967953277406798,
      "loss": 5.5588,
      "mean_token_accuracy": 0.15436985194683076,
      "num_tokens": 13647266.0,
      "step": 7860
    },
    {
      "entropy": 5.577764654159546,
      "epoch": 0.6119432017117292,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004967906649657828,
      "loss": 5.5385,
      "mean_token_accuracy": 0.15356022566556932,
      "num_tokens": 13655930.0,
      "step": 7865
    },
    {
      "entropy": 5.682845115661621,
      "epoch": 0.6123322310834468,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004967859988255655,
      "loss": 5.5266,
      "mean_token_accuracy": 0.1603233262896538,
      "num_tokens": 13664427.0,
      "step": 7870
    },
    {
      "entropy": 5.619559383392334,
      "epoch": 0.6127212604551644,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004967813293200985,
      "loss": 5.5442,
      "mean_token_accuracy": 0.1572070375084877,
      "num_tokens": 13673120.0,
      "step": 7875
    },
    {
      "entropy": 5.572880744934082,
      "epoch": 0.613110289826882,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004967766564494529,
      "loss": 5.3749,
      "mean_token_accuracy": 0.16913995444774627,
      "num_tokens": 13681918.0,
      "step": 7880
    },
    {
      "entropy": 5.554079675674439,
      "epoch": 0.6134993191985995,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004967719802136996,
      "loss": 5.4778,
      "mean_token_accuracy": 0.16967321187257767,
      "num_tokens": 13689904.0,
      "step": 7885
    },
    {
      "entropy": 5.60751166343689,
      "epoch": 0.613888348570317,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004967673006129094,
      "loss": 5.5141,
      "mean_token_accuracy": 0.16323916763067245,
      "num_tokens": 13697861.0,
      "step": 7890
    },
    {
      "entropy": 5.705719375610352,
      "epoch": 0.6142773779420346,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004967626176471536,
      "loss": 5.6527,
      "mean_token_accuracy": 0.15828764587640762,
      "num_tokens": 13706058.0,
      "step": 7895
    },
    {
      "entropy": 5.716040849685669,
      "epoch": 0.6146664073137522,
      "grad_norm": 1.5,
      "learning_rate": 0.0004967579313165028,
      "loss": 5.5623,
      "mean_token_accuracy": 0.1580530047416687,
      "num_tokens": 13715081.0,
      "step": 7900
    },
    {
      "entropy": 5.621213054656982,
      "epoch": 0.6150554366854698,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004967532416210284,
      "loss": 5.4588,
      "mean_token_accuracy": 0.16525533944368362,
      "num_tokens": 13723311.0,
      "step": 7905
    },
    {
      "entropy": 5.625402641296387,
      "epoch": 0.6154444660571873,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004967485485608016,
      "loss": 5.4866,
      "mean_token_accuracy": 0.15760971158742904,
      "num_tokens": 13731737.0,
      "step": 7910
    },
    {
      "entropy": 5.621528196334839,
      "epoch": 0.6158334954289049,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004967438521358934,
      "loss": 5.5461,
      "mean_token_accuracy": 0.16389579325914383,
      "num_tokens": 13740255.0,
      "step": 7915
    },
    {
      "entropy": 5.610876655578613,
      "epoch": 0.6162225248006225,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004967391523463754,
      "loss": 5.489,
      "mean_token_accuracy": 0.16789331734180452,
      "num_tokens": 13748656.0,
      "step": 7920
    },
    {
      "entropy": 5.564035367965698,
      "epoch": 0.61661155417234,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004967344491923185,
      "loss": 5.5076,
      "mean_token_accuracy": 0.15980468392372132,
      "num_tokens": 13757539.0,
      "step": 7925
    },
    {
      "entropy": 5.659132051467895,
      "epoch": 0.6170005835440576,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004967297426737943,
      "loss": 5.5129,
      "mean_token_accuracy": 0.15738938599824906,
      "num_tokens": 13766569.0,
      "step": 7930
    },
    {
      "entropy": 5.653924655914307,
      "epoch": 0.6173896129157751,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004967250327908742,
      "loss": 5.5366,
      "mean_token_accuracy": 0.16349831819534302,
      "num_tokens": 13775071.0,
      "step": 7935
    },
    {
      "entropy": 5.568353176116943,
      "epoch": 0.6177786422874927,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004967203195436297,
      "loss": 5.5299,
      "mean_token_accuracy": 0.15771082788705826,
      "num_tokens": 13783319.0,
      "step": 7940
    },
    {
      "entropy": 5.554364490509033,
      "epoch": 0.6181676716592103,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004967156029321322,
      "loss": 5.4737,
      "mean_token_accuracy": 0.1636737361550331,
      "num_tokens": 13792746.0,
      "step": 7945
    },
    {
      "entropy": 5.6123247146606445,
      "epoch": 0.6185567010309279,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004967108829564532,
      "loss": 5.4158,
      "mean_token_accuracy": 0.16776807606220245,
      "num_tokens": 13801632.0,
      "step": 7950
    },
    {
      "entropy": 5.608026552200317,
      "epoch": 0.6189457304026454,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004967061596166646,
      "loss": 5.554,
      "mean_token_accuracy": 0.164193557202816,
      "num_tokens": 13810527.0,
      "step": 7955
    },
    {
      "entropy": 5.641494846343994,
      "epoch": 0.6193347597743629,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004967014329128378,
      "loss": 5.4927,
      "mean_token_accuracy": 0.1584918275475502,
      "num_tokens": 13819885.0,
      "step": 7960
    },
    {
      "entropy": 5.602808618545533,
      "epoch": 0.6197237891460805,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004966967028450448,
      "loss": 5.4986,
      "mean_token_accuracy": 0.1619834065437317,
      "num_tokens": 13828509.0,
      "step": 7965
    },
    {
      "entropy": 5.639061784744262,
      "epoch": 0.6201128185177981,
      "grad_norm": 1.40625,
      "learning_rate": 0.000496691969413357,
      "loss": 5.4984,
      "mean_token_accuracy": 0.15714405626058578,
      "num_tokens": 13837196.0,
      "step": 7970
    },
    {
      "entropy": 5.626188468933106,
      "epoch": 0.6205018478895157,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004966872326178466,
      "loss": 5.5401,
      "mean_token_accuracy": 0.15958023220300674,
      "num_tokens": 13845849.0,
      "step": 7975
    },
    {
      "entropy": 5.598227500915527,
      "epoch": 0.6208908772612333,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004966824924585851,
      "loss": 5.4759,
      "mean_token_accuracy": 0.1665155291557312,
      "num_tokens": 13853430.0,
      "step": 7980
    },
    {
      "entropy": 5.629713916778565,
      "epoch": 0.6212799066329507,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004966777489356448,
      "loss": 5.6462,
      "mean_token_accuracy": 0.14797311276197433,
      "num_tokens": 13862520.0,
      "step": 7985
    },
    {
      "entropy": 5.67573127746582,
      "epoch": 0.6216689360046683,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004966730020490975,
      "loss": 5.5734,
      "mean_token_accuracy": 0.15468597412109375,
      "num_tokens": 13871064.0,
      "step": 7990
    },
    {
      "entropy": 5.598003816604614,
      "epoch": 0.6220579653763859,
      "grad_norm": 1.359375,
      "learning_rate": 0.000496668251799015,
      "loss": 5.4698,
      "mean_token_accuracy": 0.15911296308040618,
      "num_tokens": 13879952.0,
      "step": 7995
    },
    {
      "entropy": 5.624768495559692,
      "epoch": 0.6224469947481035,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004966634981854699,
      "loss": 5.6215,
      "mean_token_accuracy": 0.15347439646720887,
      "num_tokens": 13888054.0,
      "step": 8000
    },
    {
      "entropy": 5.720452976226807,
      "epoch": 0.6228360241198211,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004966587412085339,
      "loss": 5.5917,
      "mean_token_accuracy": 0.15596983581781387,
      "num_tokens": 13897051.0,
      "step": 8005
    },
    {
      "entropy": 5.644585418701172,
      "epoch": 0.6232250534915386,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004966539808682793,
      "loss": 5.5541,
      "mean_token_accuracy": 0.15398771762847902,
      "num_tokens": 13906175.0,
      "step": 8010
    },
    {
      "entropy": 5.594349575042725,
      "epoch": 0.6236140828632561,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004966492171647783,
      "loss": 5.5113,
      "mean_token_accuracy": 0.16241261661052703,
      "num_tokens": 13915283.0,
      "step": 8015
    },
    {
      "entropy": 5.613103246688842,
      "epoch": 0.6240031122349737,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004966444500981032,
      "loss": 5.5599,
      "mean_token_accuracy": 0.15472135916352273,
      "num_tokens": 13924007.0,
      "step": 8020
    },
    {
      "entropy": 5.635571241378784,
      "epoch": 0.6243921416066913,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004966396796683265,
      "loss": 5.4524,
      "mean_token_accuracy": 0.1615746721625328,
      "num_tokens": 13932694.0,
      "step": 8025
    },
    {
      "entropy": 5.6920552253723145,
      "epoch": 0.6247811709784089,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004966349058755204,
      "loss": 5.5668,
      "mean_token_accuracy": 0.1511019714176655,
      "num_tokens": 13941817.0,
      "step": 8030
    },
    {
      "entropy": 5.627501916885376,
      "epoch": 0.6251702003501264,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004966301287197573,
      "loss": 5.4196,
      "mean_token_accuracy": 0.16701409667730333,
      "num_tokens": 13950413.0,
      "step": 8035
    },
    {
      "entropy": 5.68371171951294,
      "epoch": 0.625559229721844,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004966253482011098,
      "loss": 5.5711,
      "mean_token_accuracy": 0.15596165508031845,
      "num_tokens": 13959415.0,
      "step": 8040
    },
    {
      "entropy": 5.586246681213379,
      "epoch": 0.6259482590935616,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004966205643196503,
      "loss": 5.4396,
      "mean_token_accuracy": 0.16483698040246964,
      "num_tokens": 13967889.0,
      "step": 8045
    },
    {
      "entropy": 5.451131200790405,
      "epoch": 0.6263372884652791,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004966157770754516,
      "loss": 5.4298,
      "mean_token_accuracy": 0.16041132509708406,
      "num_tokens": 13976190.0,
      "step": 8050
    },
    {
      "entropy": 5.743996858596802,
      "epoch": 0.6267263178369967,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000496610986468586,
      "loss": 5.5624,
      "mean_token_accuracy": 0.15470899939537047,
      "num_tokens": 13984940.0,
      "step": 8055
    },
    {
      "entropy": 5.711560821533203,
      "epoch": 0.6271153472087143,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004966061924991267,
      "loss": 5.5669,
      "mean_token_accuracy": 0.15803698003292083,
      "num_tokens": 13993740.0,
      "step": 8060
    },
    {
      "entropy": 5.649155855178833,
      "epoch": 0.6275043765804318,
      "grad_norm": 1.9609375,
      "learning_rate": 0.000496601395167146,
      "loss": 5.6063,
      "mean_token_accuracy": 0.15621764957904816,
      "num_tokens": 14002401.0,
      "step": 8065
    },
    {
      "entropy": 5.6560780048370365,
      "epoch": 0.6278934059521494,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004965965944727168,
      "loss": 5.589,
      "mean_token_accuracy": 0.16249808073043823,
      "num_tokens": 14011197.0,
      "step": 8070
    },
    {
      "entropy": 5.668906259536743,
      "epoch": 0.628282435323867,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004965917904159121,
      "loss": 5.4433,
      "mean_token_accuracy": 0.16768872141838073,
      "num_tokens": 14020344.0,
      "step": 8075
    },
    {
      "entropy": 5.573599004745484,
      "epoch": 0.6286714646955845,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004965869829968046,
      "loss": 5.4688,
      "mean_token_accuracy": 0.1629825420677662,
      "num_tokens": 14029117.0,
      "step": 8080
    },
    {
      "entropy": 5.550131130218506,
      "epoch": 0.6290604940673021,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004965821722154674,
      "loss": 5.3851,
      "mean_token_accuracy": 0.17250871807336807,
      "num_tokens": 14038395.0,
      "step": 8085
    },
    {
      "entropy": 5.654913759231567,
      "epoch": 0.6294495234390196,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004965773580719734,
      "loss": 5.4818,
      "mean_token_accuracy": 0.1610204964876175,
      "num_tokens": 14046983.0,
      "step": 8090
    },
    {
      "entropy": 5.532347106933594,
      "epoch": 0.6298385528107372,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004965725405663957,
      "loss": 5.3661,
      "mean_token_accuracy": 0.17185525000095367,
      "num_tokens": 14056261.0,
      "step": 8095
    },
    {
      "entropy": 5.560761213302612,
      "epoch": 0.6302275821824548,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004965677196988073,
      "loss": 5.5553,
      "mean_token_accuracy": 0.15718994289636612,
      "num_tokens": 14065371.0,
      "step": 8100
    },
    {
      "entropy": 5.667489385604858,
      "epoch": 0.6306166115541724,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004965628954692815,
      "loss": 5.6046,
      "mean_token_accuracy": 0.15320607796311378,
      "num_tokens": 14074007.0,
      "step": 8105
    },
    {
      "entropy": 5.6024304866790775,
      "epoch": 0.63100564092589,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004965580678778913,
      "loss": 5.4635,
      "mean_token_accuracy": 0.16192514896392823,
      "num_tokens": 14083259.0,
      "step": 8110
    },
    {
      "entropy": 5.5890124320983885,
      "epoch": 0.6313946702976074,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004965532369247102,
      "loss": 5.4472,
      "mean_token_accuracy": 0.16927091628313065,
      "num_tokens": 14090993.0,
      "step": 8115
    },
    {
      "entropy": 5.708404493331909,
      "epoch": 0.631783699669325,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004965484026098111,
      "loss": 5.5798,
      "mean_token_accuracy": 0.16099388897418976,
      "num_tokens": 14099722.0,
      "step": 8120
    },
    {
      "entropy": 5.694318532943726,
      "epoch": 0.6321727290410426,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004965435649332679,
      "loss": 5.5686,
      "mean_token_accuracy": 0.1553332358598709,
      "num_tokens": 14108483.0,
      "step": 8125
    },
    {
      "entropy": 5.700818920135498,
      "epoch": 0.6325617584127602,
      "grad_norm": 1.5,
      "learning_rate": 0.0004965387238951536,
      "loss": 5.5813,
      "mean_token_accuracy": 0.15432338714599608,
      "num_tokens": 14117187.0,
      "step": 8130
    },
    {
      "entropy": 5.629018354415893,
      "epoch": 0.6329507877844778,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004965338794955418,
      "loss": 5.4165,
      "mean_token_accuracy": 0.16764814108610154,
      "num_tokens": 14125424.0,
      "step": 8135
    },
    {
      "entropy": 5.578190708160401,
      "epoch": 0.6333398171561953,
      "grad_norm": 2.125,
      "learning_rate": 0.0004965290317345061,
      "loss": 5.4749,
      "mean_token_accuracy": 0.16633348613977433,
      "num_tokens": 14133038.0,
      "step": 8140
    },
    {
      "entropy": 5.669448947906494,
      "epoch": 0.6337288465279128,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004965241806121198,
      "loss": 5.635,
      "mean_token_accuracy": 0.15585788935422898,
      "num_tokens": 14141114.0,
      "step": 8145
    },
    {
      "entropy": 5.701724481582642,
      "epoch": 0.6341178758996304,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004965193261284567,
      "loss": 5.4913,
      "mean_token_accuracy": 0.16475782543420792,
      "num_tokens": 14149681.0,
      "step": 8150
    },
    {
      "entropy": 5.664780807495117,
      "epoch": 0.634506905271348,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004965144682835904,
      "loss": 5.521,
      "mean_token_accuracy": 0.15518609285354615,
      "num_tokens": 14158510.0,
      "step": 8155
    },
    {
      "entropy": 5.62149748802185,
      "epoch": 0.6348959346430656,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004965096070775946,
      "loss": 5.5722,
      "mean_token_accuracy": 0.16341715157032013,
      "num_tokens": 14166897.0,
      "step": 8160
    },
    {
      "entropy": 5.646671915054322,
      "epoch": 0.6352849640147831,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004965047425105431,
      "loss": 5.5112,
      "mean_token_accuracy": 0.16355205327272415,
      "num_tokens": 14174692.0,
      "step": 8165
    },
    {
      "entropy": 5.608004140853882,
      "epoch": 0.6356739933865007,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004964998745825097,
      "loss": 5.5145,
      "mean_token_accuracy": 0.1610356867313385,
      "num_tokens": 14183779.0,
      "step": 8170
    },
    {
      "entropy": 5.618157815933228,
      "epoch": 0.6360630227582182,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004964950032935682,
      "loss": 5.5653,
      "mean_token_accuracy": 0.1621095359325409,
      "num_tokens": 14191809.0,
      "step": 8175
    },
    {
      "entropy": 5.619154691696167,
      "epoch": 0.6364520521299358,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004964901286437927,
      "loss": 5.5212,
      "mean_token_accuracy": 0.1605943337082863,
      "num_tokens": 14200652.0,
      "step": 8180
    },
    {
      "entropy": 5.64874005317688,
      "epoch": 0.6368410815016534,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004964852506332568,
      "loss": 5.4994,
      "mean_token_accuracy": 0.1635741412639618,
      "num_tokens": 14209694.0,
      "step": 8185
    },
    {
      "entropy": 5.67854871749878,
      "epoch": 0.6372301108733709,
      "grad_norm": 1.71875,
      "learning_rate": 0.000496480369262035,
      "loss": 5.5653,
      "mean_token_accuracy": 0.15659616738557816,
      "num_tokens": 14218040.0,
      "step": 8190
    },
    {
      "entropy": 5.608455085754395,
      "epoch": 0.6376191402450885,
      "grad_norm": 1.453125,
      "learning_rate": 0.000496475484530201,
      "loss": 5.4904,
      "mean_token_accuracy": 0.16435641497373582,
      "num_tokens": 14225908.0,
      "step": 8195
    },
    {
      "entropy": 5.625738477706909,
      "epoch": 0.6380081696168061,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004964705964378292,
      "loss": 5.514,
      "mean_token_accuracy": 0.15561088025569916,
      "num_tokens": 14234140.0,
      "step": 8200
    },
    {
      "entropy": 5.676596355438233,
      "epoch": 0.6383971989885237,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004964657049849934,
      "loss": 5.5588,
      "mean_token_accuracy": 0.161154206097126,
      "num_tokens": 14243590.0,
      "step": 8205
    },
    {
      "entropy": 5.700593566894531,
      "epoch": 0.6387862283602412,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004964608101717681,
      "loss": 5.5236,
      "mean_token_accuracy": 0.1618498295545578,
      "num_tokens": 14252249.0,
      "step": 8210
    },
    {
      "entropy": 5.620331192016602,
      "epoch": 0.6391752577319587,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004964559119982275,
      "loss": 5.4516,
      "mean_token_accuracy": 0.1617949664592743,
      "num_tokens": 14261147.0,
      "step": 8215
    },
    {
      "entropy": 5.61994776725769,
      "epoch": 0.6395642871036763,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000496451010464446,
      "loss": 5.5242,
      "mean_token_accuracy": 0.16001735627651215,
      "num_tokens": 14270309.0,
      "step": 8220
    },
    {
      "entropy": 5.57561731338501,
      "epoch": 0.6399533164753939,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004964461055704978,
      "loss": 5.511,
      "mean_token_accuracy": 0.15979014486074447,
      "num_tokens": 14279379.0,
      "step": 8225
    },
    {
      "entropy": 5.560876750946045,
      "epoch": 0.6403423458471115,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004964411973164574,
      "loss": 5.4614,
      "mean_token_accuracy": 0.1537440910935402,
      "num_tokens": 14288136.0,
      "step": 8230
    },
    {
      "entropy": 5.591270303726196,
      "epoch": 0.6407313752188291,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004964362857023993,
      "loss": 5.3891,
      "mean_token_accuracy": 0.16101400256156922,
      "num_tokens": 14296437.0,
      "step": 8235
    },
    {
      "entropy": 5.613467645645142,
      "epoch": 0.6411204045905465,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000496431370728398,
      "loss": 5.4342,
      "mean_token_accuracy": 0.17272723615169525,
      "num_tokens": 14304861.0,
      "step": 8240
    },
    {
      "entropy": 5.53576021194458,
      "epoch": 0.6415094339622641,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004964264523945281,
      "loss": 5.4904,
      "mean_token_accuracy": 0.16508440673351288,
      "num_tokens": 14312861.0,
      "step": 8245
    },
    {
      "entropy": 5.503257942199707,
      "epoch": 0.6418984633339817,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004964215307008643,
      "loss": 5.3573,
      "mean_token_accuracy": 0.16898525953292848,
      "num_tokens": 14321651.0,
      "step": 8250
    },
    {
      "entropy": 5.606548023223877,
      "epoch": 0.6422874927056993,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004964166056474811,
      "loss": 5.5186,
      "mean_token_accuracy": 0.16981685757637024,
      "num_tokens": 14329234.0,
      "step": 8255
    },
    {
      "entropy": 5.63331151008606,
      "epoch": 0.6426765220774169,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004964116772344534,
      "loss": 5.641,
      "mean_token_accuracy": 0.15602685809135436,
      "num_tokens": 14337603.0,
      "step": 8260
    },
    {
      "entropy": 5.645445251464844,
      "epoch": 0.6430655514491345,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004964067454618559,
      "loss": 5.5221,
      "mean_token_accuracy": 0.16045982241630555,
      "num_tokens": 14346634.0,
      "step": 8265
    },
    {
      "entropy": 5.691012954711914,
      "epoch": 0.6434545808208519,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004964018103297634,
      "loss": 5.5444,
      "mean_token_accuracy": 0.15774981528520585,
      "num_tokens": 14355336.0,
      "step": 8270
    },
    {
      "entropy": 5.637445688247681,
      "epoch": 0.6438436101925695,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004963968718382509,
      "loss": 5.4931,
      "mean_token_accuracy": 0.1598949909210205,
      "num_tokens": 14364434.0,
      "step": 8275
    },
    {
      "entropy": 5.580499696731567,
      "epoch": 0.6442326395642871,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004963919299873931,
      "loss": 5.432,
      "mean_token_accuracy": 0.1723459243774414,
      "num_tokens": 14372454.0,
      "step": 8280
    },
    {
      "entropy": 5.626510953903198,
      "epoch": 0.6446216689360047,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004963869847772653,
      "loss": 5.4967,
      "mean_token_accuracy": 0.1624937891960144,
      "num_tokens": 14381381.0,
      "step": 8285
    },
    {
      "entropy": 5.6586121082305905,
      "epoch": 0.6450106983077223,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004963820362079423,
      "loss": 5.5855,
      "mean_token_accuracy": 0.15776764452457429,
      "num_tokens": 14390853.0,
      "step": 8290
    },
    {
      "entropy": 5.5826750755310055,
      "epoch": 0.6453997276794398,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004963770842794994,
      "loss": 5.4458,
      "mean_token_accuracy": 0.16542210429906845,
      "num_tokens": 14399897.0,
      "step": 8295
    },
    {
      "entropy": 5.546051073074341,
      "epoch": 0.6457887570511573,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004963721289920115,
      "loss": 5.5021,
      "mean_token_accuracy": 0.16482663303613662,
      "num_tokens": 14408192.0,
      "step": 8300
    },
    {
      "entropy": 5.548407506942749,
      "epoch": 0.6461777864228749,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004963671703455538,
      "loss": 5.3852,
      "mean_token_accuracy": 0.17058072835206986,
      "num_tokens": 14416004.0,
      "step": 8305
    },
    {
      "entropy": 5.585981321334839,
      "epoch": 0.6465668157945925,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004963622083402018,
      "loss": 5.4231,
      "mean_token_accuracy": 0.16638686507940292,
      "num_tokens": 14424573.0,
      "step": 8310
    },
    {
      "entropy": 5.536883068084717,
      "epoch": 0.6469558451663101,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004963572429760305,
      "loss": 5.4153,
      "mean_token_accuracy": 0.16832368671894074,
      "num_tokens": 14432756.0,
      "step": 8315
    },
    {
      "entropy": 5.62876329421997,
      "epoch": 0.6473448745380276,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004963522742531155,
      "loss": 5.5143,
      "mean_token_accuracy": 0.15935860425233841,
      "num_tokens": 14441695.0,
      "step": 8320
    },
    {
      "entropy": 5.589247512817383,
      "epoch": 0.6477339039097452,
      "grad_norm": 1.7109375,
      "learning_rate": 0.000496347302171532,
      "loss": 5.6141,
      "mean_token_accuracy": 0.15695839151740074,
      "num_tokens": 14451233.0,
      "step": 8325
    },
    {
      "entropy": 5.634691572189331,
      "epoch": 0.6481229332814628,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004963423267313554,
      "loss": 5.4841,
      "mean_token_accuracy": 0.16518651098012924,
      "num_tokens": 14459757.0,
      "step": 8330
    },
    {
      "entropy": 5.642829942703247,
      "epoch": 0.6485119626531803,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004963373479326614,
      "loss": 5.5698,
      "mean_token_accuracy": 0.1531238727271557,
      "num_tokens": 14468308.0,
      "step": 8335
    },
    {
      "entropy": 5.619338035583496,
      "epoch": 0.6489009920248979,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004963323657755255,
      "loss": 5.5011,
      "mean_token_accuracy": 0.1637391686439514,
      "num_tokens": 14476654.0,
      "step": 8340
    },
    {
      "entropy": 5.64993805885315,
      "epoch": 0.6492900213966154,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004963273802600231,
      "loss": 5.5943,
      "mean_token_accuracy": 0.16172325909137725,
      "num_tokens": 14484960.0,
      "step": 8345
    },
    {
      "entropy": 5.72280445098877,
      "epoch": 0.649679050768333,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00049632239138623,
      "loss": 5.5754,
      "mean_token_accuracy": 0.1508148118853569,
      "num_tokens": 14493708.0,
      "step": 8350
    },
    {
      "entropy": 5.721679782867431,
      "epoch": 0.6500680801400506,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004963173991542219,
      "loss": 5.531,
      "mean_token_accuracy": 0.16181753277778627,
      "num_tokens": 14502123.0,
      "step": 8355
    },
    {
      "entropy": 5.621743965148926,
      "epoch": 0.6504571095117682,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004963124035640746,
      "loss": 5.5046,
      "mean_token_accuracy": 0.16208697408437728,
      "num_tokens": 14510782.0,
      "step": 8360
    },
    {
      "entropy": 5.53667573928833,
      "epoch": 0.6508461388834857,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004963074046158637,
      "loss": 5.4504,
      "mean_token_accuracy": 0.1645912155508995,
      "num_tokens": 14518560.0,
      "step": 8365
    },
    {
      "entropy": 5.63091254234314,
      "epoch": 0.6512351682552032,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004963024023096652,
      "loss": 5.4836,
      "mean_token_accuracy": 0.1586213417351246,
      "num_tokens": 14527541.0,
      "step": 8370
    },
    {
      "entropy": 5.679382848739624,
      "epoch": 0.6516241976269208,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004962973966455551,
      "loss": 5.5212,
      "mean_token_accuracy": 0.16429576873779297,
      "num_tokens": 14535590.0,
      "step": 8375
    },
    {
      "entropy": 5.476749753952026,
      "epoch": 0.6520132269986384,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004962923876236092,
      "loss": 5.4237,
      "mean_token_accuracy": 0.15991545766592025,
      "num_tokens": 14544929.0,
      "step": 8380
    },
    {
      "entropy": 5.631164979934693,
      "epoch": 0.652402256370356,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004962873752439035,
      "loss": 5.5858,
      "mean_token_accuracy": 0.1527818873524666,
      "num_tokens": 14553883.0,
      "step": 8385
    },
    {
      "entropy": 5.696902561187744,
      "epoch": 0.6527912857420736,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004962823595065141,
      "loss": 5.6022,
      "mean_token_accuracy": 0.15535233467817305,
      "num_tokens": 14563085.0,
      "step": 8390
    },
    {
      "entropy": 5.681191062927246,
      "epoch": 0.653180315113791,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004962773404115172,
      "loss": 5.4552,
      "mean_token_accuracy": 0.16444580256938934,
      "num_tokens": 14571444.0,
      "step": 8395
    },
    {
      "entropy": 5.63764533996582,
      "epoch": 0.6535693444855086,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004962723179589886,
      "loss": 5.5539,
      "mean_token_accuracy": 0.16111520528793336,
      "num_tokens": 14580039.0,
      "step": 8400
    },
    {
      "entropy": 5.579935073852539,
      "epoch": 0.6539583738572262,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000496267292149005,
      "loss": 5.4336,
      "mean_token_accuracy": 0.16800265163183212,
      "num_tokens": 14588322.0,
      "step": 8405
    },
    {
      "entropy": 5.57730073928833,
      "epoch": 0.6543474032289438,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004962622629816423,
      "loss": 5.3722,
      "mean_token_accuracy": 0.1650988057255745,
      "num_tokens": 14597503.0,
      "step": 8410
    },
    {
      "entropy": 5.609830713272094,
      "epoch": 0.6547364326006614,
      "grad_norm": 1.703125,
      "learning_rate": 0.000496257230456977,
      "loss": 5.5037,
      "mean_token_accuracy": 0.15908412784337997,
      "num_tokens": 14605804.0,
      "step": 8415
    },
    {
      "entropy": 5.462103652954101,
      "epoch": 0.6551254619723789,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004962521945750855,
      "loss": 5.4273,
      "mean_token_accuracy": 0.17144200950860977,
      "num_tokens": 14614064.0,
      "step": 8420
    },
    {
      "entropy": 5.612936496734619,
      "epoch": 0.6555144913440965,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000496247155336044,
      "loss": 5.4338,
      "mean_token_accuracy": 0.16576751172542573,
      "num_tokens": 14622231.0,
      "step": 8425
    },
    {
      "entropy": 5.623867321014404,
      "epoch": 0.655903520715814,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004962421127399291,
      "loss": 5.5545,
      "mean_token_accuracy": 0.1569196030497551,
      "num_tokens": 14631244.0,
      "step": 8430
    },
    {
      "entropy": 5.587360048294068,
      "epoch": 0.6562925500875316,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004962370667868172,
      "loss": 5.4267,
      "mean_token_accuracy": 0.16126427203416824,
      "num_tokens": 14639575.0,
      "step": 8435
    },
    {
      "entropy": 5.58545503616333,
      "epoch": 0.6566815794592492,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004962320174767851,
      "loss": 5.4756,
      "mean_token_accuracy": 0.15965766459703445,
      "num_tokens": 14648810.0,
      "step": 8440
    },
    {
      "entropy": 5.555451202392578,
      "epoch": 0.6570706088309667,
      "grad_norm": 1.4375,
      "learning_rate": 0.000496226964809909,
      "loss": 5.3258,
      "mean_token_accuracy": 0.17258747071027755,
      "num_tokens": 14657783.0,
      "step": 8445
    },
    {
      "entropy": 5.624908018112182,
      "epoch": 0.6574596382026843,
      "grad_norm": 1.546875,
      "learning_rate": 0.000496221908786266,
      "loss": 5.5276,
      "mean_token_accuracy": 0.15910668671131134,
      "num_tokens": 14666872.0,
      "step": 8450
    },
    {
      "entropy": 5.645132446289063,
      "epoch": 0.6578486675744019,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004962168494059327,
      "loss": 5.6502,
      "mean_token_accuracy": 0.1516483262181282,
      "num_tokens": 14675762.0,
      "step": 8455
    },
    {
      "entropy": 5.561020088195801,
      "epoch": 0.6582376969461194,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004962117866689857,
      "loss": 5.3148,
      "mean_token_accuracy": 0.169151408970356,
      "num_tokens": 14683928.0,
      "step": 8460
    },
    {
      "entropy": 5.6118079662323,
      "epoch": 0.658626726317837,
      "grad_norm": 1.90625,
      "learning_rate": 0.000496206720575502,
      "loss": 5.5508,
      "mean_token_accuracy": 0.15828213691711426,
      "num_tokens": 14692411.0,
      "step": 8465
    },
    {
      "entropy": 5.686123466491699,
      "epoch": 0.6590157556895546,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004962016511255584,
      "loss": 5.5501,
      "mean_token_accuracy": 0.15016385316848754,
      "num_tokens": 14701002.0,
      "step": 8470
    },
    {
      "entropy": 5.7460590362548825,
      "epoch": 0.6594047850612721,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004961965783192318,
      "loss": 5.605,
      "mean_token_accuracy": 0.15275918543338776,
      "num_tokens": 14710172.0,
      "step": 8475
    },
    {
      "entropy": 5.595785903930664,
      "epoch": 0.6597938144329897,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004961915021565992,
      "loss": 5.5364,
      "mean_token_accuracy": 0.15851985067129135,
      "num_tokens": 14719446.0,
      "step": 8480
    },
    {
      "entropy": 5.684946155548095,
      "epoch": 0.6601828438047073,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004961864226377377,
      "loss": 5.5389,
      "mean_token_accuracy": 0.1646391198039055,
      "num_tokens": 14728193.0,
      "step": 8485
    },
    {
      "entropy": 5.620755958557129,
      "epoch": 0.6605718731764249,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004961813397627241,
      "loss": 5.434,
      "mean_token_accuracy": 0.15734331607818602,
      "num_tokens": 14736930.0,
      "step": 8490
    },
    {
      "entropy": 5.5294873237609865,
      "epoch": 0.6609609025481424,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004961762535316358,
      "loss": 5.4386,
      "mean_token_accuracy": 0.1667746901512146,
      "num_tokens": 14745995.0,
      "step": 8495
    },
    {
      "entropy": 5.519912528991699,
      "epoch": 0.6613499319198599,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004961711639445499,
      "loss": 5.488,
      "mean_token_accuracy": 0.1612710624933243,
      "num_tokens": 14754827.0,
      "step": 8500
    },
    {
      "entropy": 5.717486047744751,
      "epoch": 0.6617389612915775,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004961660710015435,
      "loss": 5.5755,
      "mean_token_accuracy": 0.15806931331753732,
      "num_tokens": 14763594.0,
      "step": 8505
    },
    {
      "entropy": 5.6062768459320065,
      "epoch": 0.6621279906632951,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004961609747026943,
      "loss": 5.3934,
      "mean_token_accuracy": 0.16947395503520965,
      "num_tokens": 14772366.0,
      "step": 8510
    },
    {
      "entropy": 5.5509490966796875,
      "epoch": 0.6625170200350127,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004961558750480791,
      "loss": 5.4908,
      "mean_token_accuracy": 0.1578017368912697,
      "num_tokens": 14780738.0,
      "step": 8515
    },
    {
      "entropy": 5.560414981842041,
      "epoch": 0.6629060494067303,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004961507720377756,
      "loss": 5.4666,
      "mean_token_accuracy": 0.15795683562755586,
      "num_tokens": 14788982.0,
      "step": 8520
    },
    {
      "entropy": 5.602201271057129,
      "epoch": 0.6632950787784477,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004961456656718611,
      "loss": 5.4556,
      "mean_token_accuracy": 0.1655229315161705,
      "num_tokens": 14797832.0,
      "step": 8525
    },
    {
      "entropy": 5.638581800460815,
      "epoch": 0.6636841081501653,
      "grad_norm": 1.6015625,
      "learning_rate": 0.000496140555950413,
      "loss": 5.4978,
      "mean_token_accuracy": 0.16762352734804153,
      "num_tokens": 14806391.0,
      "step": 8530
    },
    {
      "entropy": 5.6750490188598635,
      "epoch": 0.6640731375218829,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004961354428735091,
      "loss": 5.5066,
      "mean_token_accuracy": 0.1556483417749405,
      "num_tokens": 14814920.0,
      "step": 8535
    },
    {
      "entropy": 5.574772119522095,
      "epoch": 0.6644621668936005,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004961303264412267,
      "loss": 5.4551,
      "mean_token_accuracy": 0.16089894473552704,
      "num_tokens": 14823541.0,
      "step": 8540
    },
    {
      "entropy": 5.581590223312378,
      "epoch": 0.6648511962653181,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004961252066536437,
      "loss": 5.4229,
      "mean_token_accuracy": 0.1616658866405487,
      "num_tokens": 14832128.0,
      "step": 8545
    },
    {
      "entropy": 5.636154460906982,
      "epoch": 0.6652402256370356,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004961200835108375,
      "loss": 5.5883,
      "mean_token_accuracy": 0.15480114370584488,
      "num_tokens": 14841434.0,
      "step": 8550
    },
    {
      "entropy": 5.655720949172974,
      "epoch": 0.6656292550087531,
      "grad_norm": 1.734375,
      "learning_rate": 0.000496114957012886,
      "loss": 5.4972,
      "mean_token_accuracy": 0.16742078512907027,
      "num_tokens": 14850255.0,
      "step": 8555
    },
    {
      "entropy": 5.625656604766846,
      "epoch": 0.6660182843804707,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004961098271598671,
      "loss": 5.5812,
      "mean_token_accuracy": 0.1552358813583851,
      "num_tokens": 14858924.0,
      "step": 8560
    },
    {
      "entropy": 5.646219539642334,
      "epoch": 0.6664073137521883,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004961046939518585,
      "loss": 5.5106,
      "mean_token_accuracy": 0.16416953057050704,
      "num_tokens": 14867426.0,
      "step": 8565
    },
    {
      "entropy": 5.699926424026489,
      "epoch": 0.6667963431239059,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004960995573889379,
      "loss": 5.4714,
      "mean_token_accuracy": 0.16506651192903518,
      "num_tokens": 14876218.0,
      "step": 8570
    },
    {
      "entropy": 5.671121644973755,
      "epoch": 0.6671853724956234,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004960944174711836,
      "loss": 5.5289,
      "mean_token_accuracy": 0.15950822383165358,
      "num_tokens": 14883995.0,
      "step": 8575
    },
    {
      "entropy": 5.540333652496338,
      "epoch": 0.667574401867341,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004960892741986734,
      "loss": 5.5014,
      "mean_token_accuracy": 0.1624891757965088,
      "num_tokens": 14892659.0,
      "step": 8580
    },
    {
      "entropy": 5.664654541015625,
      "epoch": 0.6679634312390585,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004960841275714853,
      "loss": 5.5264,
      "mean_token_accuracy": 0.15727367997169495,
      "num_tokens": 14901518.0,
      "step": 8585
    },
    {
      "entropy": 5.607156229019165,
      "epoch": 0.6683524606107761,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004960789775896975,
      "loss": 5.509,
      "mean_token_accuracy": 0.16244361102581023,
      "num_tokens": 14910171.0,
      "step": 8590
    },
    {
      "entropy": 5.604865503311157,
      "epoch": 0.6687414899824937,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004960738242533881,
      "loss": 5.4873,
      "mean_token_accuracy": 0.16071989834308625,
      "num_tokens": 14918651.0,
      "step": 8595
    },
    {
      "entropy": 5.5974328994750975,
      "epoch": 0.6691305193542112,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004960686675626353,
      "loss": 5.4889,
      "mean_token_accuracy": 0.15908095240592957,
      "num_tokens": 14926951.0,
      "step": 8600
    },
    {
      "entropy": 5.726441287994385,
      "epoch": 0.6695195487259288,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004960635075175173,
      "loss": 5.5811,
      "mean_token_accuracy": 0.15933732837438583,
      "num_tokens": 14935528.0,
      "step": 8605
    },
    {
      "entropy": 5.610484170913696,
      "epoch": 0.6699085780976464,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004960583441181124,
      "loss": 5.4549,
      "mean_token_accuracy": 0.16408660262823105,
      "num_tokens": 14944117.0,
      "step": 8610
    },
    {
      "entropy": 5.582907962799072,
      "epoch": 0.670297607469364,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004960531773644992,
      "loss": 5.4265,
      "mean_token_accuracy": 0.16474046409130097,
      "num_tokens": 14952957.0,
      "step": 8615
    },
    {
      "entropy": 5.647277021408081,
      "epoch": 0.6706866368410815,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004960480072567557,
      "loss": 5.5077,
      "mean_token_accuracy": 0.15832734555006028,
      "num_tokens": 14960942.0,
      "step": 8620
    },
    {
      "entropy": 5.688439846038818,
      "epoch": 0.671075666212799,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004960428337949604,
      "loss": 5.6049,
      "mean_token_accuracy": 0.1537842944264412,
      "num_tokens": 14970579.0,
      "step": 8625
    },
    {
      "entropy": 5.594122982025146,
      "epoch": 0.6714646955845166,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004960376569791921,
      "loss": 5.4374,
      "mean_token_accuracy": 0.15769007354974746,
      "num_tokens": 14979353.0,
      "step": 8630
    },
    {
      "entropy": 5.67335877418518,
      "epoch": 0.6718537249562342,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004960324768095291,
      "loss": 5.6024,
      "mean_token_accuracy": 0.16516705453395844,
      "num_tokens": 14988435.0,
      "step": 8635
    },
    {
      "entropy": 5.647039937973022,
      "epoch": 0.6722427543279518,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004960272932860502,
      "loss": 5.5194,
      "mean_token_accuracy": 0.15829846411943435,
      "num_tokens": 14996834.0,
      "step": 8640
    },
    {
      "entropy": 5.670257425308227,
      "epoch": 0.6726317836996694,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004960221064088338,
      "loss": 5.5703,
      "mean_token_accuracy": 0.15692353248596191,
      "num_tokens": 15004988.0,
      "step": 8645
    },
    {
      "entropy": 5.666182184219361,
      "epoch": 0.6730208130713868,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004960169161779588,
      "loss": 5.5264,
      "mean_token_accuracy": 0.1628352239727974,
      "num_tokens": 15013482.0,
      "step": 8650
    },
    {
      "entropy": 5.591054677963257,
      "epoch": 0.6734098424431044,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004960117225935038,
      "loss": 5.4371,
      "mean_token_accuracy": 0.16409590989351272,
      "num_tokens": 15021506.0,
      "step": 8655
    },
    {
      "entropy": 5.586325979232788,
      "epoch": 0.673798871814822,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004960065256555477,
      "loss": 5.4772,
      "mean_token_accuracy": 0.16757019460201264,
      "num_tokens": 15029880.0,
      "step": 8660
    },
    {
      "entropy": 5.585851573944092,
      "epoch": 0.6741879011865396,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004960013253641695,
      "loss": 5.4672,
      "mean_token_accuracy": 0.16203958690166473,
      "num_tokens": 15038080.0,
      "step": 8665
    },
    {
      "entropy": 5.597144031524659,
      "epoch": 0.6745769305582572,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004959961217194477,
      "loss": 5.4829,
      "mean_token_accuracy": 0.1624331921339035,
      "num_tokens": 15046779.0,
      "step": 8670
    },
    {
      "entropy": 5.656364631652832,
      "epoch": 0.6749659599299748,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004959909147214615,
      "loss": 5.5129,
      "mean_token_accuracy": 0.15865243524312972,
      "num_tokens": 15055557.0,
      "step": 8675
    },
    {
      "entropy": 5.579842853546142,
      "epoch": 0.6753549893016922,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004959857043702902,
      "loss": 5.3801,
      "mean_token_accuracy": 0.1640066534280777,
      "num_tokens": 15063054.0,
      "step": 8680
    },
    {
      "entropy": 5.585133504867554,
      "epoch": 0.6757440186734098,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004959804906660124,
      "loss": 5.4688,
      "mean_token_accuracy": 0.16473619192838668,
      "num_tokens": 15071510.0,
      "step": 8685
    },
    {
      "entropy": 5.563712930679321,
      "epoch": 0.6761330480451274,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004959752736087073,
      "loss": 5.4888,
      "mean_token_accuracy": 0.1553959906101227,
      "num_tokens": 15080524.0,
      "step": 8690
    },
    {
      "entropy": 5.669127082824707,
      "epoch": 0.676522077416845,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004959700531984543,
      "loss": 5.5237,
      "mean_token_accuracy": 0.15884414985775946,
      "num_tokens": 15089689.0,
      "step": 8695
    },
    {
      "entropy": 5.576257514953613,
      "epoch": 0.6769111067885626,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004959648294353324,
      "loss": 5.4073,
      "mean_token_accuracy": 0.16998513787984848,
      "num_tokens": 15097970.0,
      "step": 8700
    },
    {
      "entropy": 5.54570426940918,
      "epoch": 0.6773001361602801,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004959596023194208,
      "loss": 5.509,
      "mean_token_accuracy": 0.15639949291944505,
      "num_tokens": 15107786.0,
      "step": 8705
    },
    {
      "entropy": 5.655686950683593,
      "epoch": 0.6776891655319977,
      "grad_norm": 1.7265625,
      "learning_rate": 0.000495954371850799,
      "loss": 5.4862,
      "mean_token_accuracy": 0.16529091447591782,
      "num_tokens": 15116370.0,
      "step": 8710
    },
    {
      "entropy": 5.68412218093872,
      "epoch": 0.6780781949037152,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004959491380295463,
      "loss": 5.5175,
      "mean_token_accuracy": 0.15937916487455367,
      "num_tokens": 15124961.0,
      "step": 8715
    },
    {
      "entropy": 5.645760822296142,
      "epoch": 0.6784672242754328,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004959439008557422,
      "loss": 5.5278,
      "mean_token_accuracy": 0.15579134672880174,
      "num_tokens": 15134230.0,
      "step": 8720
    },
    {
      "entropy": 5.597600364685059,
      "epoch": 0.6788562536471504,
      "grad_norm": 1.53125,
      "learning_rate": 0.000495938660329466,
      "loss": 5.4762,
      "mean_token_accuracy": 0.15846306532621385,
      "num_tokens": 15143585.0,
      "step": 8725
    },
    {
      "entropy": 5.5997742176055905,
      "epoch": 0.6792452830188679,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004959334164507973,
      "loss": 5.3996,
      "mean_token_accuracy": 0.15709985047578812,
      "num_tokens": 15151834.0,
      "step": 8730
    },
    {
      "entropy": 5.526013803482056,
      "epoch": 0.6796343123905855,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004959281692198155,
      "loss": 5.459,
      "mean_token_accuracy": 0.15873907804489135,
      "num_tokens": 15160630.0,
      "step": 8735
    },
    {
      "entropy": 5.591261196136474,
      "epoch": 0.6800233417623031,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004959229186366007,
      "loss": 5.4203,
      "mean_token_accuracy": 0.16888799667358398,
      "num_tokens": 15168999.0,
      "step": 8740
    },
    {
      "entropy": 5.640165138244629,
      "epoch": 0.6804123711340206,
      "grad_norm": 1.5703125,
      "learning_rate": 0.000495917664701232,
      "loss": 5.4996,
      "mean_token_accuracy": 0.16761998534202577,
      "num_tokens": 15177507.0,
      "step": 8745
    },
    {
      "entropy": 5.663268566131592,
      "epoch": 0.6808014005057382,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004959124074137894,
      "loss": 5.5497,
      "mean_token_accuracy": 0.15307750552892685,
      "num_tokens": 15186843.0,
      "step": 8750
    },
    {
      "entropy": 5.590552043914795,
      "epoch": 0.6811904298774557,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004959071467743526,
      "loss": 5.4358,
      "mean_token_accuracy": 0.16405264288187027,
      "num_tokens": 15195081.0,
      "step": 8755
    },
    {
      "entropy": 5.632944202423095,
      "epoch": 0.6815794592491733,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004959018827830016,
      "loss": 5.5331,
      "mean_token_accuracy": 0.16543604135513307,
      "num_tokens": 15203538.0,
      "step": 8760
    },
    {
      "entropy": 5.714319372177124,
      "epoch": 0.6819684886208909,
      "grad_norm": 1.75,
      "learning_rate": 0.000495896615439816,
      "loss": 5.5638,
      "mean_token_accuracy": 0.15726466104388237,
      "num_tokens": 15211845.0,
      "step": 8765
    },
    {
      "entropy": 5.64163293838501,
      "epoch": 0.6823575179926085,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004958913447448759,
      "loss": 5.5604,
      "mean_token_accuracy": 0.16496984660625458,
      "num_tokens": 15220128.0,
      "step": 8770
    },
    {
      "entropy": 5.609681749343872,
      "epoch": 0.682746547364326,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004958860706982613,
      "loss": 5.5371,
      "mean_token_accuracy": 0.1546408787369728,
      "num_tokens": 15228741.0,
      "step": 8775
    },
    {
      "entropy": 5.650665616989135,
      "epoch": 0.6831355767360435,
      "grad_norm": 1.8046875,
      "learning_rate": 0.000495880793300052,
      "loss": 5.5375,
      "mean_token_accuracy": 0.1545580267906189,
      "num_tokens": 15237674.0,
      "step": 8780
    },
    {
      "entropy": 5.608680534362793,
      "epoch": 0.6835246061077611,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004958755125503284,
      "loss": 5.4188,
      "mean_token_accuracy": 0.1603718653321266,
      "num_tokens": 15245727.0,
      "step": 8785
    },
    {
      "entropy": 5.636897563934326,
      "epoch": 0.6839136354794787,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004958702284491703,
      "loss": 5.5221,
      "mean_token_accuracy": 0.1570213332772255,
      "num_tokens": 15254511.0,
      "step": 8790
    },
    {
      "entropy": 5.611495447158814,
      "epoch": 0.6843026648511963,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004958649409966581,
      "loss": 5.4847,
      "mean_token_accuracy": 0.160319522023201,
      "num_tokens": 15263500.0,
      "step": 8795
    },
    {
      "entropy": 5.60971474647522,
      "epoch": 0.6846916942229139,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000495859650192872,
      "loss": 5.4799,
      "mean_token_accuracy": 0.16392674893140793,
      "num_tokens": 15271960.0,
      "step": 8800
    },
    {
      "entropy": 5.703287553787232,
      "epoch": 0.6850807235946313,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004958543560378922,
      "loss": 5.6659,
      "mean_token_accuracy": 0.15225871056318283,
      "num_tokens": 15280576.0,
      "step": 8805
    },
    {
      "entropy": 5.663115358352661,
      "epoch": 0.6854697529663489,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004958490585317991,
      "loss": 5.5318,
      "mean_token_accuracy": 0.1648571327328682,
      "num_tokens": 15288674.0,
      "step": 8810
    },
    {
      "entropy": 5.625303268432617,
      "epoch": 0.6858587823380665,
      "grad_norm": 1.828125,
      "learning_rate": 0.000495843757674673,
      "loss": 5.4181,
      "mean_token_accuracy": 0.16334908306598664,
      "num_tokens": 15296881.0,
      "step": 8815
    },
    {
      "entropy": 5.608910036087036,
      "epoch": 0.6862478117097841,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004958384534665945,
      "loss": 5.4782,
      "mean_token_accuracy": 0.16224911957979202,
      "num_tokens": 15305379.0,
      "step": 8820
    },
    {
      "entropy": 5.6033713817596436,
      "epoch": 0.6866368410815017,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004958331459076438,
      "loss": 5.5691,
      "mean_token_accuracy": 0.15493535324931146,
      "num_tokens": 15314572.0,
      "step": 8825
    },
    {
      "entropy": 5.63705039024353,
      "epoch": 0.6870258704532192,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004958278349979018,
      "loss": 5.5418,
      "mean_token_accuracy": 0.15657844468951226,
      "num_tokens": 15323407.0,
      "step": 8830
    },
    {
      "entropy": 5.627943754196167,
      "epoch": 0.6874148998249368,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004958225207374488,
      "loss": 5.4151,
      "mean_token_accuracy": 0.15650082528591155,
      "num_tokens": 15332208.0,
      "step": 8835
    },
    {
      "entropy": 5.583755350112915,
      "epoch": 0.6878039291966543,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004958172031263655,
      "loss": 5.4595,
      "mean_token_accuracy": 0.1620849221944809,
      "num_tokens": 15340972.0,
      "step": 8840
    },
    {
      "entropy": 5.57933497428894,
      "epoch": 0.6881929585683719,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004958118821647326,
      "loss": 5.4426,
      "mean_token_accuracy": 0.1666816309094429,
      "num_tokens": 15349991.0,
      "step": 8845
    },
    {
      "entropy": 5.530833578109741,
      "epoch": 0.6885819879400895,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004958065578526308,
      "loss": 5.3791,
      "mean_token_accuracy": 0.17523722350597382,
      "num_tokens": 15358194.0,
      "step": 8850
    },
    {
      "entropy": 5.619156074523926,
      "epoch": 0.688971017311807,
      "grad_norm": 1.7109375,
      "learning_rate": 0.000495801230190141,
      "loss": 5.4257,
      "mean_token_accuracy": 0.163530196249485,
      "num_tokens": 15366954.0,
      "step": 8855
    },
    {
      "entropy": 5.704611015319824,
      "epoch": 0.6893600466835246,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004957958991773441,
      "loss": 5.6093,
      "mean_token_accuracy": 0.15734027922153473,
      "num_tokens": 15376107.0,
      "step": 8860
    },
    {
      "entropy": 5.599389457702637,
      "epoch": 0.6897490760552422,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004957905648143206,
      "loss": 5.4884,
      "mean_token_accuracy": 0.16068901717662812,
      "num_tokens": 15384517.0,
      "step": 8865
    },
    {
      "entropy": 5.707489633560181,
      "epoch": 0.6901381054269597,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004957852271011519,
      "loss": 5.6396,
      "mean_token_accuracy": 0.15333401635289193,
      "num_tokens": 15393544.0,
      "step": 8870
    },
    {
      "entropy": 5.63929591178894,
      "epoch": 0.6905271347986773,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004957798860379187,
      "loss": 5.4442,
      "mean_token_accuracy": 0.1683732807636261,
      "num_tokens": 15402165.0,
      "step": 8875
    },
    {
      "entropy": 5.622973155975342,
      "epoch": 0.6909161641703949,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004957745416247022,
      "loss": 5.4916,
      "mean_token_accuracy": 0.1617909163236618,
      "num_tokens": 15411430.0,
      "step": 8880
    },
    {
      "entropy": 5.629455804824829,
      "epoch": 0.6913051935421124,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004957691938615833,
      "loss": 5.4046,
      "mean_token_accuracy": 0.16690257638692857,
      "num_tokens": 15420346.0,
      "step": 8885
    },
    {
      "entropy": 5.520572233200073,
      "epoch": 0.69169422291383,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004957638427486434,
      "loss": 5.3353,
      "mean_token_accuracy": 0.17267585545778275,
      "num_tokens": 15428747.0,
      "step": 8890
    },
    {
      "entropy": 5.414216136932373,
      "epoch": 0.6920832522855476,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004957584882859636,
      "loss": 5.3316,
      "mean_token_accuracy": 0.16797439754009247,
      "num_tokens": 15437110.0,
      "step": 8895
    },
    {
      "entropy": 5.571062803268433,
      "epoch": 0.6924722816572652,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004957531304736251,
      "loss": 5.4768,
      "mean_token_accuracy": 0.16320954859256745,
      "num_tokens": 15446524.0,
      "step": 8900
    },
    {
      "entropy": 5.65091347694397,
      "epoch": 0.6928613110289827,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004957477693117091,
      "loss": 5.5152,
      "mean_token_accuracy": 0.1618165910243988,
      "num_tokens": 15455372.0,
      "step": 8905
    },
    {
      "entropy": 5.650473690032959,
      "epoch": 0.6932503404007002,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004957424048002971,
      "loss": 5.4078,
      "mean_token_accuracy": 0.16240484565496444,
      "num_tokens": 15463060.0,
      "step": 8910
    },
    {
      "entropy": 5.531956577301026,
      "epoch": 0.6936393697724178,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004957370369394706,
      "loss": 5.5246,
      "mean_token_accuracy": 0.1603606067597866,
      "num_tokens": 15471836.0,
      "step": 8915
    },
    {
      "entropy": 5.662734794616699,
      "epoch": 0.6940283991441354,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004957316657293107,
      "loss": 5.5207,
      "mean_token_accuracy": 0.16369528323411942,
      "num_tokens": 15479819.0,
      "step": 8920
    },
    {
      "entropy": 5.639108276367187,
      "epoch": 0.694417428515853,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004957262911698992,
      "loss": 5.5088,
      "mean_token_accuracy": 0.16640071272850038,
      "num_tokens": 15488562.0,
      "step": 8925
    },
    {
      "entropy": 5.617952966690064,
      "epoch": 0.6948064578875706,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004957209132613175,
      "loss": 5.5899,
      "mean_token_accuracy": 0.15694498494267464,
      "num_tokens": 15497822.0,
      "step": 8930
    },
    {
      "entropy": 5.66477222442627,
      "epoch": 0.695195487259288,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004957155320036473,
      "loss": 5.5297,
      "mean_token_accuracy": 0.15420645251870155,
      "num_tokens": 15506913.0,
      "step": 8935
    },
    {
      "entropy": 5.650326061248779,
      "epoch": 0.6955845166310056,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004957101473969702,
      "loss": 5.4725,
      "mean_token_accuracy": 0.1661776825785637,
      "num_tokens": 15516053.0,
      "step": 8940
    },
    {
      "entropy": 5.649831771850586,
      "epoch": 0.6959735460027232,
      "grad_norm": 1.453125,
      "learning_rate": 0.000495704759441368,
      "loss": 5.5135,
      "mean_token_accuracy": 0.15613418817520142,
      "num_tokens": 15524174.0,
      "step": 8945
    },
    {
      "entropy": 5.603764915466309,
      "epoch": 0.6963625753744408,
      "grad_norm": 1.5,
      "learning_rate": 0.0004956993681369221,
      "loss": 5.4317,
      "mean_token_accuracy": 0.16271826028823852,
      "num_tokens": 15533652.0,
      "step": 8950
    },
    {
      "entropy": 5.661756324768066,
      "epoch": 0.6967516047461584,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004956939734837148,
      "loss": 5.4747,
      "mean_token_accuracy": 0.16325116157531738,
      "num_tokens": 15541644.0,
      "step": 8955
    },
    {
      "entropy": 5.509464406967163,
      "epoch": 0.6971406341178759,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004956885754818277,
      "loss": 5.432,
      "mean_token_accuracy": 0.17381385564804078,
      "num_tokens": 15550298.0,
      "step": 8960
    },
    {
      "entropy": 5.629638576507569,
      "epoch": 0.6975296634895934,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004956831741313427,
      "loss": 5.5018,
      "mean_token_accuracy": 0.1647512845695019,
      "num_tokens": 15559388.0,
      "step": 8965
    },
    {
      "entropy": 5.500730657577515,
      "epoch": 0.697918692861311,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004956777694323418,
      "loss": 5.3335,
      "mean_token_accuracy": 0.16650238931179046,
      "num_tokens": 15567543.0,
      "step": 8970
    },
    {
      "entropy": 5.505261754989624,
      "epoch": 0.6983077222330286,
      "grad_norm": 1.5625,
      "learning_rate": 0.000495672361384907,
      "loss": 5.4324,
      "mean_token_accuracy": 0.1692634642124176,
      "num_tokens": 15575365.0,
      "step": 8975
    },
    {
      "entropy": 5.681888818740845,
      "epoch": 0.6986967516047462,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004956669499891202,
      "loss": 5.4978,
      "mean_token_accuracy": 0.16197374761104583,
      "num_tokens": 15584295.0,
      "step": 8980
    },
    {
      "entropy": 5.666158962249756,
      "epoch": 0.6990857809764637,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004956615352450639,
      "loss": 5.5156,
      "mean_token_accuracy": 0.17047316581010818,
      "num_tokens": 15592906.0,
      "step": 8985
    },
    {
      "entropy": 5.530295705795288,
      "epoch": 0.6994748103481813,
      "grad_norm": 1.875,
      "learning_rate": 0.0004956561171528198,
      "loss": 5.4542,
      "mean_token_accuracy": 0.15547731518745422,
      "num_tokens": 15601938.0,
      "step": 8990
    },
    {
      "entropy": 5.546193027496338,
      "epoch": 0.6998638397198989,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004956506957124704,
      "loss": 5.4481,
      "mean_token_accuracy": 0.16723582297563552,
      "num_tokens": 15610173.0,
      "step": 8995
    },
    {
      "entropy": 5.643675851821899,
      "epoch": 0.7002528690916164,
      "grad_norm": 1.5,
      "learning_rate": 0.000495645270924098,
      "loss": 5.5277,
      "mean_token_accuracy": 0.15238175243139268,
      "num_tokens": 15618896.0,
      "step": 9000
    },
    {
      "epoch": 0.7002528690916164,
      "eval_entropy": 5.48892997332326,
      "eval_loss": 5.537880897521973,
      "eval_mean_token_accuracy": 0.16768242414104928,
      "eval_num_tokens": 15618896.0,
      "eval_runtime": 28.6719,
      "eval_samples_per_second": 1449.435,
      "eval_steps_per_second": 181.188,
      "step": 9000
    },
    {
      "entropy": 5.598692321777344,
      "epoch": 0.700641898463334,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004956398427877847,
      "loss": 5.446,
      "mean_token_accuracy": 0.16387993842363358,
      "num_tokens": 15627890.0,
      "step": 9005
    },
    {
      "entropy": 5.573201608657837,
      "epoch": 0.7010309278350515,
      "grad_norm": 1.609375,
      "learning_rate": 0.000495634411303613,
      "loss": 5.3828,
      "mean_token_accuracy": 0.1656647354364395,
      "num_tokens": 15636290.0,
      "step": 9010
    },
    {
      "entropy": 5.572251892089843,
      "epoch": 0.7014199572067691,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004956289764716654,
      "loss": 5.4455,
      "mean_token_accuracy": 0.16509263068437577,
      "num_tokens": 15644777.0,
      "step": 9015
    },
    {
      "entropy": 5.600341558456421,
      "epoch": 0.7018089865784867,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004956235382920241,
      "loss": 5.4793,
      "mean_token_accuracy": 0.15838624760508538,
      "num_tokens": 15653333.0,
      "step": 9020
    },
    {
      "entropy": 5.6746580600738525,
      "epoch": 0.7021980159502043,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004956180967647717,
      "loss": 5.4714,
      "mean_token_accuracy": 0.16225895434617996,
      "num_tokens": 15662140.0,
      "step": 9025
    },
    {
      "entropy": 5.617575931549072,
      "epoch": 0.7025870453219218,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004956126518899911,
      "loss": 5.5609,
      "mean_token_accuracy": 0.1655264288187027,
      "num_tokens": 15671387.0,
      "step": 9030
    },
    {
      "entropy": 5.625704574584961,
      "epoch": 0.7029760746936393,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004956072036677644,
      "loss": 5.4991,
      "mean_token_accuracy": 0.1554996997117996,
      "num_tokens": 15680884.0,
      "step": 9035
    },
    {
      "entropy": 5.684255933761596,
      "epoch": 0.7033651040653569,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004956017520981746,
      "loss": 5.4485,
      "mean_token_accuracy": 0.158163383603096,
      "num_tokens": 15690030.0,
      "step": 9040
    },
    {
      "entropy": 5.610305118560791,
      "epoch": 0.7037541334370745,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004955962971813044,
      "loss": 5.467,
      "mean_token_accuracy": 0.1622978761792183,
      "num_tokens": 15698260.0,
      "step": 9045
    },
    {
      "entropy": 5.518650770187378,
      "epoch": 0.7041431628087921,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004955908389172364,
      "loss": 5.3716,
      "mean_token_accuracy": 0.17074191123247145,
      "num_tokens": 15706275.0,
      "step": 9050
    },
    {
      "entropy": 5.656791067123413,
      "epoch": 0.7045321921805097,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004955853773060536,
      "loss": 5.6008,
      "mean_token_accuracy": 0.16207863092422486,
      "num_tokens": 15715086.0,
      "step": 9055
    },
    {
      "entropy": 5.6516862392425535,
      "epoch": 0.7049212215522271,
      "grad_norm": 1.75,
      "learning_rate": 0.0004955799123478388,
      "loss": 5.4052,
      "mean_token_accuracy": 0.16449855417013168,
      "num_tokens": 15723931.0,
      "step": 9060
    },
    {
      "entropy": 5.590332937240601,
      "epoch": 0.7053102509239447,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004955744440426748,
      "loss": 5.5517,
      "mean_token_accuracy": 0.16194690018892288,
      "num_tokens": 15732905.0,
      "step": 9065
    },
    {
      "entropy": 5.602451610565185,
      "epoch": 0.7056992802956623,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004955689723906448,
      "loss": 5.4469,
      "mean_token_accuracy": 0.1638907566666603,
      "num_tokens": 15742355.0,
      "step": 9070
    },
    {
      "entropy": 5.709460067749023,
      "epoch": 0.7060883096673799,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004955634973918318,
      "loss": 5.521,
      "mean_token_accuracy": 0.16189123094081878,
      "num_tokens": 15750924.0,
      "step": 9075
    },
    {
      "entropy": 5.617863130569458,
      "epoch": 0.7064773390390975,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004955580190463186,
      "loss": 5.4484,
      "mean_token_accuracy": 0.16356792151927949,
      "num_tokens": 15759677.0,
      "step": 9080
    },
    {
      "entropy": 5.595055437088012,
      "epoch": 0.7068663684108151,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004955525373541886,
      "loss": 5.4445,
      "mean_token_accuracy": 0.15860881954431533,
      "num_tokens": 15767875.0,
      "step": 9085
    },
    {
      "entropy": 5.545309209823609,
      "epoch": 0.7072553977825325,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004955470523155249,
      "loss": 5.4341,
      "mean_token_accuracy": 0.16337229758501054,
      "num_tokens": 15776194.0,
      "step": 9090
    },
    {
      "entropy": 5.632459306716919,
      "epoch": 0.7076444271542501,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004955415639304107,
      "loss": 5.5201,
      "mean_token_accuracy": 0.16406060010194778,
      "num_tokens": 15784774.0,
      "step": 9095
    },
    {
      "entropy": 5.611529779434204,
      "epoch": 0.7080334565259677,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004955360721989292,
      "loss": 5.3971,
      "mean_token_accuracy": 0.170825731754303,
      "num_tokens": 15793121.0,
      "step": 9100
    },
    {
      "entropy": 5.703958511352539,
      "epoch": 0.7084224858976853,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004955305771211641,
      "loss": 5.6047,
      "mean_token_accuracy": 0.1560478091239929,
      "num_tokens": 15801518.0,
      "step": 9105
    },
    {
      "entropy": 5.6174711227417,
      "epoch": 0.7088115152694029,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004955250786971982,
      "loss": 5.3588,
      "mean_token_accuracy": 0.1709561228752136,
      "num_tokens": 15809602.0,
      "step": 9110
    },
    {
      "entropy": 5.60283932685852,
      "epoch": 0.7092005446411204,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004955195769271154,
      "loss": 5.5686,
      "mean_token_accuracy": 0.15899789780378343,
      "num_tokens": 15818529.0,
      "step": 9115
    },
    {
      "entropy": 5.648442792892456,
      "epoch": 0.709589574012838,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004955140718109991,
      "loss": 5.4401,
      "mean_token_accuracy": 0.173095266520977,
      "num_tokens": 15827176.0,
      "step": 9120
    },
    {
      "entropy": 5.634540843963623,
      "epoch": 0.7099786033845555,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004955085633489326,
      "loss": 5.521,
      "mean_token_accuracy": 0.1553187772631645,
      "num_tokens": 15837095.0,
      "step": 9125
    },
    {
      "entropy": 5.575646877288818,
      "epoch": 0.7103676327562731,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004955030515409997,
      "loss": 5.4907,
      "mean_token_accuracy": 0.16527149379253386,
      "num_tokens": 15845285.0,
      "step": 9130
    },
    {
      "entropy": 5.5941849708557125,
      "epoch": 0.7107566621279907,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004954975363872838,
      "loss": 5.4942,
      "mean_token_accuracy": 0.16929271966218948,
      "num_tokens": 15853468.0,
      "step": 9135
    },
    {
      "entropy": 5.637126636505127,
      "epoch": 0.7111456914997082,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004954920178878689,
      "loss": 5.431,
      "mean_token_accuracy": 0.16908954679965973,
      "num_tokens": 15861699.0,
      "step": 9140
    },
    {
      "entropy": 5.6686560153961185,
      "epoch": 0.7115347208714258,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004954864960428385,
      "loss": 5.5156,
      "mean_token_accuracy": 0.15655331686139107,
      "num_tokens": 15870401.0,
      "step": 9145
    },
    {
      "entropy": 5.5921454429626465,
      "epoch": 0.7119237502431434,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004954809708522765,
      "loss": 5.3192,
      "mean_token_accuracy": 0.17602779418230058,
      "num_tokens": 15878214.0,
      "step": 9150
    },
    {
      "entropy": 5.578209257125854,
      "epoch": 0.712312779614861,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004954754423162667,
      "loss": 5.5048,
      "mean_token_accuracy": 0.1651099443435669,
      "num_tokens": 15886581.0,
      "step": 9155
    },
    {
      "entropy": 5.6807318210601805,
      "epoch": 0.7127018089865785,
      "grad_norm": 1.375,
      "learning_rate": 0.000495469910434893,
      "loss": 5.6188,
      "mean_token_accuracy": 0.15459591150283813,
      "num_tokens": 15895748.0,
      "step": 9160
    },
    {
      "entropy": 5.697744464874267,
      "epoch": 0.713090838358296,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004954643752082392,
      "loss": 5.4706,
      "mean_token_accuracy": 0.15997869968414308,
      "num_tokens": 15905030.0,
      "step": 9165
    },
    {
      "entropy": 5.5733599185943605,
      "epoch": 0.7134798677300136,
      "grad_norm": 1.5,
      "learning_rate": 0.0004954588366363896,
      "loss": 5.453,
      "mean_token_accuracy": 0.15960024297237396,
      "num_tokens": 15913430.0,
      "step": 9170
    },
    {
      "entropy": 5.546750068664551,
      "epoch": 0.7138688971017312,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004954532947194279,
      "loss": 5.3884,
      "mean_token_accuracy": 0.16921509355306624,
      "num_tokens": 15921432.0,
      "step": 9175
    },
    {
      "entropy": 5.661558675765991,
      "epoch": 0.7142579264734488,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004954477494574384,
      "loss": 5.4398,
      "mean_token_accuracy": 0.1639270529150963,
      "num_tokens": 15930159.0,
      "step": 9180
    },
    {
      "entropy": 5.588769388198853,
      "epoch": 0.7146469558451664,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004954422008505052,
      "loss": 5.5109,
      "mean_token_accuracy": 0.15652605295181274,
      "num_tokens": 15939911.0,
      "step": 9185
    },
    {
      "entropy": 5.568434810638427,
      "epoch": 0.7150359852168838,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004954366488987125,
      "loss": 5.4305,
      "mean_token_accuracy": 0.16211689561605452,
      "num_tokens": 15948586.0,
      "step": 9190
    },
    {
      "entropy": 5.506918573379517,
      "epoch": 0.7154250145886014,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004954310936021444,
      "loss": 5.3097,
      "mean_token_accuracy": 0.1671778976917267,
      "num_tokens": 15956818.0,
      "step": 9195
    },
    {
      "entropy": 5.614367055892944,
      "epoch": 0.715814043960319,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004954255349608853,
      "loss": 5.5179,
      "mean_token_accuracy": 0.15673997849225998,
      "num_tokens": 15966027.0,
      "step": 9200
    },
    {
      "entropy": 5.669200229644775,
      "epoch": 0.7162030733320366,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004954199729750198,
      "loss": 5.5671,
      "mean_token_accuracy": 0.1588587701320648,
      "num_tokens": 15975186.0,
      "step": 9205
    },
    {
      "entropy": 5.546558856964111,
      "epoch": 0.7165921027037542,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004954144076446318,
      "loss": 5.4239,
      "mean_token_accuracy": 0.1685514986515045,
      "num_tokens": 15983924.0,
      "step": 9210
    },
    {
      "entropy": 5.585919570922852,
      "epoch": 0.7169811320754716,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004954088389698061,
      "loss": 5.4512,
      "mean_token_accuracy": 0.17108927965164183,
      "num_tokens": 15992906.0,
      "step": 9215
    },
    {
      "entropy": 5.5795739650726315,
      "epoch": 0.7173701614471892,
      "grad_norm": 1.5625,
      "learning_rate": 0.000495403266950627,
      "loss": 5.4739,
      "mean_token_accuracy": 0.16261557191610337,
      "num_tokens": 16001378.0,
      "step": 9220
    },
    {
      "entropy": 5.598306322097779,
      "epoch": 0.7177591908189068,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004953976915871792,
      "loss": 5.4686,
      "mean_token_accuracy": 0.16207181066274642,
      "num_tokens": 16010704.0,
      "step": 9225
    },
    {
      "entropy": 5.566759204864502,
      "epoch": 0.7181482201906244,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004953921128795472,
      "loss": 5.4508,
      "mean_token_accuracy": 0.1655542552471161,
      "num_tokens": 16018936.0,
      "step": 9230
    },
    {
      "entropy": 5.593916034698486,
      "epoch": 0.718537249562342,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004953865308278156,
      "loss": 5.3588,
      "mean_token_accuracy": 0.1663544625043869,
      "num_tokens": 16027295.0,
      "step": 9235
    },
    {
      "entropy": 5.501459980010987,
      "epoch": 0.7189262789340595,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004953809454320693,
      "loss": 5.3808,
      "mean_token_accuracy": 0.16815916448831558,
      "num_tokens": 16036493.0,
      "step": 9240
    },
    {
      "entropy": 5.560643815994263,
      "epoch": 0.7193153083057771,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004953753566923929,
      "loss": 5.5038,
      "mean_token_accuracy": 0.1660083144903183,
      "num_tokens": 16045406.0,
      "step": 9245
    },
    {
      "entropy": 5.623446416854859,
      "epoch": 0.7197043376774946,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004953697646088712,
      "loss": 5.4826,
      "mean_token_accuracy": 0.16748983412981033,
      "num_tokens": 16054126.0,
      "step": 9250
    },
    {
      "entropy": 5.632719039916992,
      "epoch": 0.7200933670492122,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004953641691815891,
      "loss": 5.5064,
      "mean_token_accuracy": 0.15853000283241273,
      "num_tokens": 16063921.0,
      "step": 9255
    },
    {
      "entropy": 5.6339634418487545,
      "epoch": 0.7204823964209298,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004953585704106315,
      "loss": 5.5061,
      "mean_token_accuracy": 0.16534079760313034,
      "num_tokens": 16072160.0,
      "step": 9260
    },
    {
      "entropy": 5.63057131767273,
      "epoch": 0.7208714257926473,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004953529682960832,
      "loss": 5.4452,
      "mean_token_accuracy": 0.1626740127801895,
      "num_tokens": 16080794.0,
      "step": 9265
    },
    {
      "entropy": 5.553777599334717,
      "epoch": 0.7212604551643649,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004953473628380295,
      "loss": 5.4747,
      "mean_token_accuracy": 0.1672894313931465,
      "num_tokens": 16089615.0,
      "step": 9270
    },
    {
      "entropy": 5.552600383758545,
      "epoch": 0.7216494845360825,
      "grad_norm": 1.625,
      "learning_rate": 0.0004953417540365553,
      "loss": 5.394,
      "mean_token_accuracy": 0.15982172787189483,
      "num_tokens": 16098363.0,
      "step": 9275
    },
    {
      "entropy": 5.596477413177491,
      "epoch": 0.7220385139078,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004953361418917455,
      "loss": 5.4838,
      "mean_token_accuracy": 0.15724485367536545,
      "num_tokens": 16106837.0,
      "step": 9280
    },
    {
      "entropy": 5.656517791748047,
      "epoch": 0.7224275432795176,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004953305264036856,
      "loss": 5.4247,
      "mean_token_accuracy": 0.1660174235701561,
      "num_tokens": 16114923.0,
      "step": 9285
    },
    {
      "entropy": 5.614937114715576,
      "epoch": 0.7228165726512352,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004953249075724607,
      "loss": 5.4609,
      "mean_token_accuracy": 0.15922097563743592,
      "num_tokens": 16123698.0,
      "step": 9290
    },
    {
      "entropy": 5.586376905441284,
      "epoch": 0.7232056020229527,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004953192853981559,
      "loss": 5.4395,
      "mean_token_accuracy": 0.1639003649353981,
      "num_tokens": 16132361.0,
      "step": 9295
    },
    {
      "entropy": 5.663545083999634,
      "epoch": 0.7235946313946703,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004953136598808566,
      "loss": 5.5057,
      "mean_token_accuracy": 0.1602555587887764,
      "num_tokens": 16141554.0,
      "step": 9300
    },
    {
      "entropy": 5.642953014373779,
      "epoch": 0.7239836607663879,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004953080310206481,
      "loss": 5.4305,
      "mean_token_accuracy": 0.16263717859983445,
      "num_tokens": 16150206.0,
      "step": 9305
    },
    {
      "entropy": 5.631220960617066,
      "epoch": 0.7243726901381055,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004953023988176162,
      "loss": 5.5605,
      "mean_token_accuracy": 0.1562460854649544,
      "num_tokens": 16158995.0,
      "step": 9310
    },
    {
      "entropy": 5.60384202003479,
      "epoch": 0.724761719509823,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004952967632718458,
      "loss": 5.52,
      "mean_token_accuracy": 0.15501880943775176,
      "num_tokens": 16167809.0,
      "step": 9315
    },
    {
      "entropy": 5.793039560317993,
      "epoch": 0.7251507488815405,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004952911243834227,
      "loss": 5.6074,
      "mean_token_accuracy": 0.15796300321817397,
      "num_tokens": 16176257.0,
      "step": 9320
    },
    {
      "entropy": 5.660436630249023,
      "epoch": 0.7255397782532581,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004952854821524324,
      "loss": 5.5777,
      "mean_token_accuracy": 0.15514086186885834,
      "num_tokens": 16185597.0,
      "step": 9325
    },
    {
      "entropy": 5.703607749938965,
      "epoch": 0.7259288076249757,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004952798365789606,
      "loss": 5.5843,
      "mean_token_accuracy": 0.15821659713983535,
      "num_tokens": 16194591.0,
      "step": 9330
    },
    {
      "entropy": 5.749100112915039,
      "epoch": 0.7263178369966933,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004952741876630928,
      "loss": 5.5342,
      "mean_token_accuracy": 0.1550159201025963,
      "num_tokens": 16203073.0,
      "step": 9335
    },
    {
      "entropy": 5.5745586395263675,
      "epoch": 0.7267068663684109,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004952685354049148,
      "loss": 5.497,
      "mean_token_accuracy": 0.1630001738667488,
      "num_tokens": 16211208.0,
      "step": 9340
    },
    {
      "entropy": 5.6307062149047855,
      "epoch": 0.7270958957401283,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004952628798045124,
      "loss": 5.439,
      "mean_token_accuracy": 0.1698139950633049,
      "num_tokens": 16219687.0,
      "step": 9345
    },
    {
      "entropy": 5.658086538314819,
      "epoch": 0.7274849251118459,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004952572208619714,
      "loss": 5.539,
      "mean_token_accuracy": 0.1642148092389107,
      "num_tokens": 16228214.0,
      "step": 9350
    },
    {
      "entropy": 5.654706954956055,
      "epoch": 0.7278739544835635,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004952515585773778,
      "loss": 5.6282,
      "mean_token_accuracy": 0.15758585929870605,
      "num_tokens": 16237765.0,
      "step": 9355
    },
    {
      "entropy": 5.650431346893311,
      "epoch": 0.7282629838552811,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004952458929508171,
      "loss": 5.4818,
      "mean_token_accuracy": 0.17496364265680314,
      "num_tokens": 16245339.0,
      "step": 9360
    },
    {
      "entropy": 5.55873589515686,
      "epoch": 0.7286520132269987,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004952402239823757,
      "loss": 5.4573,
      "mean_token_accuracy": 0.15955497622489928,
      "num_tokens": 16253966.0,
      "step": 9365
    },
    {
      "entropy": 5.583206415176392,
      "epoch": 0.7290410425987162,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004952345516721393,
      "loss": 5.4502,
      "mean_token_accuracy": 0.16356470733880996,
      "num_tokens": 16262859.0,
      "step": 9370
    },
    {
      "entropy": 5.610286617279053,
      "epoch": 0.7294300719704337,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004952288760201942,
      "loss": 5.5051,
      "mean_token_accuracy": 0.16029743254184722,
      "num_tokens": 16271646.0,
      "step": 9375
    },
    {
      "entropy": 5.547354507446289,
      "epoch": 0.7298191013421513,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004952231970266265,
      "loss": 5.4811,
      "mean_token_accuracy": 0.15630216598510743,
      "num_tokens": 16281058.0,
      "step": 9380
    },
    {
      "entropy": 5.652565431594849,
      "epoch": 0.7302081307138689,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004952175146915223,
      "loss": 5.4287,
      "mean_token_accuracy": 0.1669951409101486,
      "num_tokens": 16289039.0,
      "step": 9385
    },
    {
      "entropy": 5.510778188705444,
      "epoch": 0.7305971600855865,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004952118290149679,
      "loss": 5.3194,
      "mean_token_accuracy": 0.17808689475059508,
      "num_tokens": 16297373.0,
      "step": 9390
    },
    {
      "entropy": 5.546924877166748,
      "epoch": 0.730986189457304,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004952061399970492,
      "loss": 5.5022,
      "mean_token_accuracy": 0.15341666340827942,
      "num_tokens": 16306066.0,
      "step": 9395
    },
    {
      "entropy": 5.642647647857666,
      "epoch": 0.7313752188290216,
      "grad_norm": 1.6640625,
      "learning_rate": 0.000495200447637853,
      "loss": 5.5518,
      "mean_token_accuracy": 0.15412714183330536,
      "num_tokens": 16315282.0,
      "step": 9400
    },
    {
      "entropy": 5.785994386672973,
      "epoch": 0.7317642482007392,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004951947519374655,
      "loss": 5.6386,
      "mean_token_accuracy": 0.15836541950702668,
      "num_tokens": 16324830.0,
      "step": 9405
    },
    {
      "entropy": 5.633684873580933,
      "epoch": 0.7321532775724567,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004951890528959731,
      "loss": 5.4955,
      "mean_token_accuracy": 0.16180311292409896,
      "num_tokens": 16332990.0,
      "step": 9410
    },
    {
      "entropy": 5.587968397140503,
      "epoch": 0.7325423069441743,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004951833505134623,
      "loss": 5.4687,
      "mean_token_accuracy": 0.1543179526925087,
      "num_tokens": 16342115.0,
      "step": 9415
    },
    {
      "entropy": 5.665604782104492,
      "epoch": 0.7329313363158918,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004951776447900196,
      "loss": 5.4552,
      "mean_token_accuracy": 0.16402145326137543,
      "num_tokens": 16350329.0,
      "step": 9420
    },
    {
      "entropy": 5.574960565567016,
      "epoch": 0.7333203656876094,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004951719357257317,
      "loss": 5.5541,
      "mean_token_accuracy": 0.16425096988677979,
      "num_tokens": 16358890.0,
      "step": 9425
    },
    {
      "entropy": 5.708090877532959,
      "epoch": 0.733709395059327,
      "grad_norm": 1.6171875,
      "learning_rate": 0.000495166223320685,
      "loss": 5.5926,
      "mean_token_accuracy": 0.1554354265332222,
      "num_tokens": 16367338.0,
      "step": 9430
    },
    {
      "entropy": 5.721589708328247,
      "epoch": 0.7340984244310446,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004951605075749662,
      "loss": 5.5067,
      "mean_token_accuracy": 0.16801688820123672,
      "num_tokens": 16375472.0,
      "step": 9435
    },
    {
      "entropy": 5.642661428451538,
      "epoch": 0.7344874538027621,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004951547884886623,
      "loss": 5.4899,
      "mean_token_accuracy": 0.1639014407992363,
      "num_tokens": 16384212.0,
      "step": 9440
    },
    {
      "entropy": 5.571622610092163,
      "epoch": 0.7348764831744796,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004951490660618598,
      "loss": 5.388,
      "mean_token_accuracy": 0.1667047306895256,
      "num_tokens": 16393447.0,
      "step": 9445
    },
    {
      "entropy": 5.558136415481568,
      "epoch": 0.7352655125461972,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004951433402946457,
      "loss": 5.3829,
      "mean_token_accuracy": 0.1678649365901947,
      "num_tokens": 16401876.0,
      "step": 9450
    },
    {
      "entropy": 5.6018385887146,
      "epoch": 0.7356545419179148,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004951376111871068,
      "loss": 5.4963,
      "mean_token_accuracy": 0.17034075856208802,
      "num_tokens": 16411020.0,
      "step": 9455
    },
    {
      "entropy": 5.632222080230713,
      "epoch": 0.7360435712896324,
      "grad_norm": 1.5,
      "learning_rate": 0.0004951318787393299,
      "loss": 5.5339,
      "mean_token_accuracy": 0.15947003960609435,
      "num_tokens": 16420492.0,
      "step": 9460
    },
    {
      "entropy": 5.598445320129395,
      "epoch": 0.73643260066135,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004951261429514023,
      "loss": 5.4316,
      "mean_token_accuracy": 0.16751165241003035,
      "num_tokens": 16429277.0,
      "step": 9465
    },
    {
      "entropy": 5.616234636306762,
      "epoch": 0.7368216300330674,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004951204038234106,
      "loss": 5.4709,
      "mean_token_accuracy": 0.16523509919643403,
      "num_tokens": 16438008.0,
      "step": 9470
    },
    {
      "entropy": 5.654847097396851,
      "epoch": 0.737210659404785,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004951146613554424,
      "loss": 5.4806,
      "mean_token_accuracy": 0.16680918782949447,
      "num_tokens": 16446791.0,
      "step": 9475
    },
    {
      "entropy": 5.593355751037597,
      "epoch": 0.7375996887765026,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004951089155475843,
      "loss": 5.5062,
      "mean_token_accuracy": 0.16251966655254363,
      "num_tokens": 16455277.0,
      "step": 9480
    },
    {
      "entropy": 5.593828105926514,
      "epoch": 0.7379887181482202,
      "grad_norm": 1.765625,
      "learning_rate": 0.000495103166399924,
      "loss": 5.4303,
      "mean_token_accuracy": 0.16394807398319244,
      "num_tokens": 16463310.0,
      "step": 9485
    },
    {
      "entropy": 5.653588676452637,
      "epoch": 0.7383777475199378,
      "grad_norm": 1.625,
      "learning_rate": 0.0004950974139125484,
      "loss": 5.4723,
      "mean_token_accuracy": 0.1603311374783516,
      "num_tokens": 16472421.0,
      "step": 9490
    },
    {
      "entropy": 5.643746042251587,
      "epoch": 0.7387667768916554,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004950916580855448,
      "loss": 5.5334,
      "mean_token_accuracy": 0.16023790836334229,
      "num_tokens": 16481814.0,
      "step": 9495
    },
    {
      "entropy": 5.6236348152160645,
      "epoch": 0.7391558062633728,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004950858989190007,
      "loss": 5.5019,
      "mean_token_accuracy": 0.15711402893066406,
      "num_tokens": 16490902.0,
      "step": 9500
    },
    {
      "entropy": 5.670229244232178,
      "epoch": 0.7395448356350904,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004950801364130032,
      "loss": 5.4574,
      "mean_token_accuracy": 0.16394625455141068,
      "num_tokens": 16499418.0,
      "step": 9505
    },
    {
      "entropy": 5.550073432922363,
      "epoch": 0.739933865006808,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004950743705676401,
      "loss": 5.3998,
      "mean_token_accuracy": 0.16792995035648345,
      "num_tokens": 16508578.0,
      "step": 9510
    },
    {
      "entropy": 5.5597185611724855,
      "epoch": 0.7403228943785256,
      "grad_norm": 1.375,
      "learning_rate": 0.0004950686013829987,
      "loss": 5.4663,
      "mean_token_accuracy": 0.16282816976308823,
      "num_tokens": 16517356.0,
      "step": 9515
    },
    {
      "entropy": 5.569721984863281,
      "epoch": 0.7407119237502432,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004950628288591665,
      "loss": 5.4441,
      "mean_token_accuracy": 0.15830240696668624,
      "num_tokens": 16525834.0,
      "step": 9520
    },
    {
      "entropy": 5.61745057106018,
      "epoch": 0.7411009531219607,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004950570529962311,
      "loss": 5.4867,
      "mean_token_accuracy": 0.16102957129478454,
      "num_tokens": 16534513.0,
      "step": 9525
    },
    {
      "entropy": 5.634679746627808,
      "epoch": 0.7414899824936783,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004950512737942803,
      "loss": 5.4562,
      "mean_token_accuracy": 0.1676692172884941,
      "num_tokens": 16543326.0,
      "step": 9530
    },
    {
      "entropy": 5.561607074737549,
      "epoch": 0.7418790118653958,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004950454912534015,
      "loss": 5.4452,
      "mean_token_accuracy": 0.16409103125333785,
      "num_tokens": 16551589.0,
      "step": 9535
    },
    {
      "entropy": 5.591177320480346,
      "epoch": 0.7422680412371134,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004950397053736827,
      "loss": 5.4473,
      "mean_token_accuracy": 0.1670309916138649,
      "num_tokens": 16560261.0,
      "step": 9540
    },
    {
      "entropy": 5.590814447402954,
      "epoch": 0.742657070608831,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004950339161552117,
      "loss": 5.4316,
      "mean_token_accuracy": 0.16776785999536514,
      "num_tokens": 16568645.0,
      "step": 9545
    },
    {
      "entropy": 5.56022047996521,
      "epoch": 0.7430460999805485,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004950281235980761,
      "loss": 5.4474,
      "mean_token_accuracy": 0.16762655973434448,
      "num_tokens": 16577206.0,
      "step": 9550
    },
    {
      "entropy": 5.632214689254761,
      "epoch": 0.7434351293522661,
      "grad_norm": 1.46875,
      "learning_rate": 0.000495022327702364,
      "loss": 5.4363,
      "mean_token_accuracy": 0.16455136835575104,
      "num_tokens": 16585011.0,
      "step": 9555
    },
    {
      "entropy": 5.6199352741241455,
      "epoch": 0.7438241587239837,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004950165284681631,
      "loss": 5.4862,
      "mean_token_accuracy": 0.1609924092888832,
      "num_tokens": 16593574.0,
      "step": 9560
    },
    {
      "entropy": 5.594768810272217,
      "epoch": 0.7442131880957013,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004950107258955618,
      "loss": 5.4256,
      "mean_token_accuracy": 0.16947073340415955,
      "num_tokens": 16603023.0,
      "step": 9565
    },
    {
      "entropy": 5.702660179138183,
      "epoch": 0.7446022174674188,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004950049199846479,
      "loss": 5.5135,
      "mean_token_accuracy": 0.16235789135098458,
      "num_tokens": 16611963.0,
      "step": 9570
    },
    {
      "entropy": 5.64899435043335,
      "epoch": 0.7449912468391363,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004949991107355095,
      "loss": 5.4564,
      "mean_token_accuracy": 0.16407136619091034,
      "num_tokens": 16619942.0,
      "step": 9575
    },
    {
      "entropy": 5.508935070037841,
      "epoch": 0.7453802762108539,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004949932981482347,
      "loss": 5.3456,
      "mean_token_accuracy": 0.16973289400339125,
      "num_tokens": 16628261.0,
      "step": 9580
    },
    {
      "entropy": 5.684511280059814,
      "epoch": 0.7457693055825715,
      "grad_norm": 1.625,
      "learning_rate": 0.0004949874822229118,
      "loss": 5.4888,
      "mean_token_accuracy": 0.1607719212770462,
      "num_tokens": 16637290.0,
      "step": 9585
    },
    {
      "entropy": 5.614342832565308,
      "epoch": 0.7461583349542891,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004949816629596288,
      "loss": 5.4721,
      "mean_token_accuracy": 0.1669323280453682,
      "num_tokens": 16646147.0,
      "step": 9590
    },
    {
      "entropy": 5.5784093856811525,
      "epoch": 0.7465473643260067,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004949758403584746,
      "loss": 5.5734,
      "mean_token_accuracy": 0.1605289027094841,
      "num_tokens": 16655299.0,
      "step": 9595
    },
    {
      "entropy": 5.6553062915802,
      "epoch": 0.7469363936977241,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004949700144195368,
      "loss": 5.4556,
      "mean_token_accuracy": 0.16472315341234206,
      "num_tokens": 16664091.0,
      "step": 9600
    },
    {
      "entropy": 5.74340591430664,
      "epoch": 0.7473254230694417,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004949641851429043,
      "loss": 5.5141,
      "mean_token_accuracy": 0.16007463932037352,
      "num_tokens": 16672665.0,
      "step": 9605
    },
    {
      "entropy": 5.585549449920654,
      "epoch": 0.7477144524411593,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004949583525286654,
      "loss": 5.4367,
      "mean_token_accuracy": 0.1651037484407425,
      "num_tokens": 16680735.0,
      "step": 9610
    },
    {
      "entropy": 5.609591007232666,
      "epoch": 0.7481034818128769,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004949525165769085,
      "loss": 5.5232,
      "mean_token_accuracy": 0.1619318887591362,
      "num_tokens": 16689560.0,
      "step": 9615
    },
    {
      "entropy": 5.675543212890625,
      "epoch": 0.7484925111845945,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004949466772877224,
      "loss": 5.4672,
      "mean_token_accuracy": 0.16306936591863633,
      "num_tokens": 16698953.0,
      "step": 9620
    },
    {
      "entropy": 5.625356245040893,
      "epoch": 0.748881540556312,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004949408346611955,
      "loss": 5.5121,
      "mean_token_accuracy": 0.16117900460958481,
      "num_tokens": 16707601.0,
      "step": 9625
    },
    {
      "entropy": 5.564679431915283,
      "epoch": 0.7492705699280295,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004949349886974165,
      "loss": 5.4926,
      "mean_token_accuracy": 0.1600220173597336,
      "num_tokens": 16716660.0,
      "step": 9630
    },
    {
      "entropy": 5.647576713562012,
      "epoch": 0.7496595992997471,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004949291393964741,
      "loss": 5.4463,
      "mean_token_accuracy": 0.16327247470617295,
      "num_tokens": 16725274.0,
      "step": 9635
    },
    {
      "entropy": 5.619313144683838,
      "epoch": 0.7500486286714647,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000494923286758457,
      "loss": 5.4217,
      "mean_token_accuracy": 0.16573825627565383,
      "num_tokens": 16733301.0,
      "step": 9640
    },
    {
      "entropy": 5.66570897102356,
      "epoch": 0.7504376580431823,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004949174307834541,
      "loss": 5.4743,
      "mean_token_accuracy": 0.16345229148864746,
      "num_tokens": 16742072.0,
      "step": 9645
    },
    {
      "entropy": 5.661173248291016,
      "epoch": 0.7508266874148998,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004949115714715543,
      "loss": 5.48,
      "mean_token_accuracy": 0.16001084744930266,
      "num_tokens": 16751497.0,
      "step": 9650
    },
    {
      "entropy": 5.544694662094116,
      "epoch": 0.7512157167866174,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004949057088228465,
      "loss": 5.4181,
      "mean_token_accuracy": 0.16923647224903107,
      "num_tokens": 16760518.0,
      "step": 9655
    },
    {
      "entropy": 5.656236028671264,
      "epoch": 0.751604746158335,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004948998428374194,
      "loss": 5.54,
      "mean_token_accuracy": 0.15598669797182083,
      "num_tokens": 16768689.0,
      "step": 9660
    },
    {
      "entropy": 5.614003610610962,
      "epoch": 0.7519937755300525,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004948939735153622,
      "loss": 5.3853,
      "mean_token_accuracy": 0.17029613703489305,
      "num_tokens": 16777047.0,
      "step": 9665
    },
    {
      "entropy": 5.555197858810425,
      "epoch": 0.7523828049017701,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004948881008567641,
      "loss": 5.3616,
      "mean_token_accuracy": 0.17010919004678726,
      "num_tokens": 16785402.0,
      "step": 9670
    },
    {
      "entropy": 5.5848133087158205,
      "epoch": 0.7527718342734876,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004948822248617139,
      "loss": 5.455,
      "mean_token_accuracy": 0.1636451430618763,
      "num_tokens": 16794687.0,
      "step": 9675
    },
    {
      "entropy": 5.694547653198242,
      "epoch": 0.7531608636452052,
      "grad_norm": 1.640625,
      "learning_rate": 0.000494876345530301,
      "loss": 5.491,
      "mean_token_accuracy": 0.16249312162399293,
      "num_tokens": 16803762.0,
      "step": 9680
    },
    {
      "entropy": 5.608018636703491,
      "epoch": 0.7535498930169228,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004948704628626145,
      "loss": 5.51,
      "mean_token_accuracy": 0.1589452415704727,
      "num_tokens": 16812872.0,
      "step": 9685
    },
    {
      "entropy": 5.654308128356933,
      "epoch": 0.7539389223886404,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004948645768587437,
      "loss": 5.5666,
      "mean_token_accuracy": 0.15174276679754256,
      "num_tokens": 16821270.0,
      "step": 9690
    },
    {
      "entropy": 5.632787799835205,
      "epoch": 0.7543279517603579,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004948586875187778,
      "loss": 5.4284,
      "mean_token_accuracy": 0.16506413072347642,
      "num_tokens": 16829508.0,
      "step": 9695
    },
    {
      "entropy": 5.59177532196045,
      "epoch": 0.7547169811320755,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004948527948428064,
      "loss": 5.421,
      "mean_token_accuracy": 0.16233740672469138,
      "num_tokens": 16838327.0,
      "step": 9700
    },
    {
      "entropy": 5.598222398757935,
      "epoch": 0.755106010503793,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004948468988309187,
      "loss": 5.4128,
      "mean_token_accuracy": 0.16744261384010314,
      "num_tokens": 16846530.0,
      "step": 9705
    },
    {
      "entropy": 5.502956771850586,
      "epoch": 0.7554950398755106,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004948409994832043,
      "loss": 5.3434,
      "mean_token_accuracy": 0.16664833426475525,
      "num_tokens": 16854825.0,
      "step": 9710
    },
    {
      "entropy": 5.598453044891357,
      "epoch": 0.7558840692472282,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004948350967997526,
      "loss": 5.5003,
      "mean_token_accuracy": 0.1536581963300705,
      "num_tokens": 16864096.0,
      "step": 9715
    },
    {
      "entropy": 5.674967241287232,
      "epoch": 0.7562730986189458,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004948291907806532,
      "loss": 5.5619,
      "mean_token_accuracy": 0.15005003958940505,
      "num_tokens": 16873000.0,
      "step": 9720
    },
    {
      "entropy": 5.726628828048706,
      "epoch": 0.7566621279906633,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004948232814259957,
      "loss": 5.5606,
      "mean_token_accuracy": 0.16236240267753602,
      "num_tokens": 16881374.0,
      "step": 9725
    },
    {
      "entropy": 5.637835550308227,
      "epoch": 0.7570511573623808,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004948173687358699,
      "loss": 5.5264,
      "mean_token_accuracy": 0.15957718193531037,
      "num_tokens": 16889192.0,
      "step": 9730
    },
    {
      "entropy": 5.582265377044678,
      "epoch": 0.7574401867340984,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004948114527103652,
      "loss": 5.5184,
      "mean_token_accuracy": 0.1577650398015976,
      "num_tokens": 16898538.0,
      "step": 9735
    },
    {
      "entropy": 5.630859899520874,
      "epoch": 0.757829216105816,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004948055333495717,
      "loss": 5.5183,
      "mean_token_accuracy": 0.15783605873584747,
      "num_tokens": 16907543.0,
      "step": 9740
    },
    {
      "entropy": 5.654094886779785,
      "epoch": 0.7582182454775336,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004947996106535791,
      "loss": 5.4257,
      "mean_token_accuracy": 0.17062853574752807,
      "num_tokens": 16916140.0,
      "step": 9745
    },
    {
      "entropy": 5.5783079147338865,
      "epoch": 0.7586072748492512,
      "grad_norm": 2.375,
      "learning_rate": 0.0004947936846224773,
      "loss": 5.4182,
      "mean_token_accuracy": 0.1665373921394348,
      "num_tokens": 16925433.0,
      "step": 9750
    },
    {
      "entropy": 5.532951974868775,
      "epoch": 0.7589963042209686,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004947877552563562,
      "loss": 5.4507,
      "mean_token_accuracy": 0.16356708705425263,
      "num_tokens": 16933598.0,
      "step": 9755
    },
    {
      "entropy": 5.639309597015381,
      "epoch": 0.7593853335926862,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004947818225553056,
      "loss": 5.5459,
      "mean_token_accuracy": 0.1587366744875908,
      "num_tokens": 16942270.0,
      "step": 9760
    },
    {
      "entropy": 5.602426338195801,
      "epoch": 0.7597743629644038,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004947758865194156,
      "loss": 5.3833,
      "mean_token_accuracy": 0.16099860072135924,
      "num_tokens": 16951044.0,
      "step": 9765
    },
    {
      "entropy": 5.673260307312011,
      "epoch": 0.7601633923361214,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004947699471487766,
      "loss": 5.5603,
      "mean_token_accuracy": 0.1531473256647587,
      "num_tokens": 16960215.0,
      "step": 9770
    },
    {
      "entropy": 5.716788911819458,
      "epoch": 0.760552421707839,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004947640044434782,
      "loss": 5.5952,
      "mean_token_accuracy": 0.15658744424581528,
      "num_tokens": 16968515.0,
      "step": 9775
    },
    {
      "entropy": 5.679812908172607,
      "epoch": 0.7609414510795565,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004947580584036109,
      "loss": 5.5651,
      "mean_token_accuracy": 0.15306783169507981,
      "num_tokens": 16977597.0,
      "step": 9780
    },
    {
      "entropy": 5.585596179962158,
      "epoch": 0.761330480451274,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004947521090292649,
      "loss": 5.4231,
      "mean_token_accuracy": 0.1596038371324539,
      "num_tokens": 16987376.0,
      "step": 9785
    },
    {
      "entropy": 5.5853808403015135,
      "epoch": 0.7617195098229916,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004947461563205304,
      "loss": 5.4246,
      "mean_token_accuracy": 0.16407533586025239,
      "num_tokens": 16995976.0,
      "step": 9790
    },
    {
      "entropy": 5.5631688117980955,
      "epoch": 0.7621085391947092,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004947402002774977,
      "loss": 5.4284,
      "mean_token_accuracy": 0.16948177963495253,
      "num_tokens": 17003819.0,
      "step": 9795
    },
    {
      "entropy": 5.549920654296875,
      "epoch": 0.7624975685664268,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004947342409002572,
      "loss": 5.4126,
      "mean_token_accuracy": 0.1728143498301506,
      "num_tokens": 17011632.0,
      "step": 9800
    },
    {
      "entropy": 5.55021185874939,
      "epoch": 0.7628865979381443,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004947282781888994,
      "loss": 5.3461,
      "mean_token_accuracy": 0.1698859602212906,
      "num_tokens": 17019737.0,
      "step": 9805
    },
    {
      "entropy": 5.623792028427124,
      "epoch": 0.7632756273098619,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004947223121435147,
      "loss": 5.5057,
      "mean_token_accuracy": 0.16593750119209288,
      "num_tokens": 17028803.0,
      "step": 9810
    },
    {
      "entropy": 5.5954584121704105,
      "epoch": 0.7636646566815795,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004947163427641936,
      "loss": 5.3402,
      "mean_token_accuracy": 0.16582608968019485,
      "num_tokens": 17036847.0,
      "step": 9815
    },
    {
      "entropy": 5.676225996017456,
      "epoch": 0.764053686053297,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004947103700510268,
      "loss": 5.4853,
      "mean_token_accuracy": 0.1590005040168762,
      "num_tokens": 17045665.0,
      "step": 9820
    },
    {
      "entropy": 5.690594434738159,
      "epoch": 0.7644427154250146,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004947043940041047,
      "loss": 5.4871,
      "mean_token_accuracy": 0.16544217467308045,
      "num_tokens": 17053842.0,
      "step": 9825
    },
    {
      "entropy": 5.657180213928223,
      "epoch": 0.7648317447967321,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004946984146235183,
      "loss": 5.5492,
      "mean_token_accuracy": 0.15866318196058274,
      "num_tokens": 17062287.0,
      "step": 9830
    },
    {
      "entropy": 5.671737289428711,
      "epoch": 0.7652207741684497,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004946924319093579,
      "loss": 5.5675,
      "mean_token_accuracy": 0.16470288038253783,
      "num_tokens": 17071495.0,
      "step": 9835
    },
    {
      "entropy": 5.566572856903076,
      "epoch": 0.7656098035401673,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004946864458617148,
      "loss": 5.495,
      "mean_token_accuracy": 0.1597565859556198,
      "num_tokens": 17080734.0,
      "step": 9840
    },
    {
      "entropy": 5.645231056213379,
      "epoch": 0.7659988329118849,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004946804564806793,
      "loss": 5.5168,
      "mean_token_accuracy": 0.16035535037517548,
      "num_tokens": 17089683.0,
      "step": 9845
    },
    {
      "entropy": 5.651087903976441,
      "epoch": 0.7663878622836025,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004946744637663427,
      "loss": 5.5363,
      "mean_token_accuracy": 0.15783768147230148,
      "num_tokens": 17098203.0,
      "step": 9850
    },
    {
      "entropy": 5.565146541595459,
      "epoch": 0.7667768916553199,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004946684677187956,
      "loss": 5.3894,
      "mean_token_accuracy": 0.16356709450483323,
      "num_tokens": 17107185.0,
      "step": 9855
    },
    {
      "entropy": 5.602199363708496,
      "epoch": 0.7671659210270375,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004946624683381292,
      "loss": 5.4465,
      "mean_token_accuracy": 0.16764336079359055,
      "num_tokens": 17115168.0,
      "step": 9860
    },
    {
      "entropy": 5.595751810073852,
      "epoch": 0.7675549503987551,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004946564656244345,
      "loss": 5.4376,
      "mean_token_accuracy": 0.16738105267286302,
      "num_tokens": 17124037.0,
      "step": 9865
    },
    {
      "entropy": 5.572898054122925,
      "epoch": 0.7679439797704727,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004946504595778026,
      "loss": 5.3668,
      "mean_token_accuracy": 0.16299717128276825,
      "num_tokens": 17132238.0,
      "step": 9870
    },
    {
      "entropy": 5.559570550918579,
      "epoch": 0.7683330091421903,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004946444501983246,
      "loss": 5.5043,
      "mean_token_accuracy": 0.15664197653532028,
      "num_tokens": 17140405.0,
      "step": 9875
    },
    {
      "entropy": 5.5682971477508545,
      "epoch": 0.7687220385139077,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004946384374860916,
      "loss": 5.42,
      "mean_token_accuracy": 0.1662541151046753,
      "num_tokens": 17149207.0,
      "step": 9880
    },
    {
      "entropy": 5.61993441581726,
      "epoch": 0.7691110678856253,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004946324214411949,
      "loss": 5.4076,
      "mean_token_accuracy": 0.15628940016031265,
      "num_tokens": 17157680.0,
      "step": 9885
    },
    {
      "entropy": 5.643505573272705,
      "epoch": 0.7695000972573429,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004946264020637259,
      "loss": 5.4463,
      "mean_token_accuracy": 0.16137076020240784,
      "num_tokens": 17166907.0,
      "step": 9890
    },
    {
      "entropy": 5.586683464050293,
      "epoch": 0.7698891266290605,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004946203793537757,
      "loss": 5.4843,
      "mean_token_accuracy": 0.16788864582777024,
      "num_tokens": 17176318.0,
      "step": 9895
    },
    {
      "entropy": 5.6048585891723635,
      "epoch": 0.7702781560007781,
      "grad_norm": 1.625,
      "learning_rate": 0.0004946143533114358,
      "loss": 5.5217,
      "mean_token_accuracy": 0.15332429856061935,
      "num_tokens": 17185380.0,
      "step": 9900
    },
    {
      "entropy": 5.660828351974487,
      "epoch": 0.7706671853724957,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004946083239367976,
      "loss": 5.4994,
      "mean_token_accuracy": 0.1601640537381172,
      "num_tokens": 17193177.0,
      "step": 9905
    },
    {
      "entropy": 5.5617344856262205,
      "epoch": 0.7710562147442132,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004946022912299527,
      "loss": 5.4757,
      "mean_token_accuracy": 0.15949599891901017,
      "num_tokens": 17201071.0,
      "step": 9910
    },
    {
      "entropy": 5.543600606918335,
      "epoch": 0.7714452441159307,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004945962551909926,
      "loss": 5.3834,
      "mean_token_accuracy": 0.16966844946146012,
      "num_tokens": 17209375.0,
      "step": 9915
    },
    {
      "entropy": 5.601286697387695,
      "epoch": 0.7718342734876483,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004945902158200089,
      "loss": 5.337,
      "mean_token_accuracy": 0.17217303067445755,
      "num_tokens": 17217453.0,
      "step": 9920
    },
    {
      "entropy": 5.601703071594239,
      "epoch": 0.7722233028593659,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004945841731170931,
      "loss": 5.4188,
      "mean_token_accuracy": 0.16042266339063643,
      "num_tokens": 17225671.0,
      "step": 9925
    },
    {
      "entropy": 5.519887161254883,
      "epoch": 0.7726123322310835,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004945781270823369,
      "loss": 5.3674,
      "mean_token_accuracy": 0.16603152453899384,
      "num_tokens": 17234396.0,
      "step": 9930
    },
    {
      "entropy": 5.570225620269776,
      "epoch": 0.773001361602801,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004945720777158323,
      "loss": 5.4813,
      "mean_token_accuracy": 0.1608129024505615,
      "num_tokens": 17243509.0,
      "step": 9935
    },
    {
      "entropy": 5.619941139221192,
      "epoch": 0.7733903909745186,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004945660250176708,
      "loss": 5.4021,
      "mean_token_accuracy": 0.17428109794855118,
      "num_tokens": 17251447.0,
      "step": 9940
    },
    {
      "entropy": 5.535233879089356,
      "epoch": 0.7737794203462361,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004945599689879443,
      "loss": 5.4664,
      "mean_token_accuracy": 0.16355775743722917,
      "num_tokens": 17260240.0,
      "step": 9945
    },
    {
      "entropy": 5.602076053619385,
      "epoch": 0.7741684497179537,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004945539096267448,
      "loss": 5.467,
      "mean_token_accuracy": 0.16678984612226486,
      "num_tokens": 17267835.0,
      "step": 9950
    },
    {
      "entropy": 5.57802038192749,
      "epoch": 0.7745574790896713,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004945478469341642,
      "loss": 5.4616,
      "mean_token_accuracy": 0.16033925563097,
      "num_tokens": 17277069.0,
      "step": 9955
    },
    {
      "entropy": 5.602939796447754,
      "epoch": 0.7749465084613888,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004945417809102944,
      "loss": 5.4513,
      "mean_token_accuracy": 0.16086394041776658,
      "num_tokens": 17285014.0,
      "step": 9960
    },
    {
      "entropy": 5.531858968734741,
      "epoch": 0.7753355378331064,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004945357115552275,
      "loss": 5.3995,
      "mean_token_accuracy": 0.16076785176992417,
      "num_tokens": 17292649.0,
      "step": 9965
    },
    {
      "entropy": 5.668289852142334,
      "epoch": 0.775724567204824,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004945296388690557,
      "loss": 5.5467,
      "mean_token_accuracy": 0.15805940628051757,
      "num_tokens": 17301792.0,
      "step": 9970
    },
    {
      "entropy": 5.5542267799377445,
      "epoch": 0.7761135965765416,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000494523562851871,
      "loss": 5.2994,
      "mean_token_accuracy": 0.16803679764270782,
      "num_tokens": 17310377.0,
      "step": 9975
    },
    {
      "entropy": 5.576317358016968,
      "epoch": 0.7765026259482591,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004945174835037655,
      "loss": 5.5252,
      "mean_token_accuracy": 0.15372655987739564,
      "num_tokens": 17319779.0,
      "step": 9980
    },
    {
      "entropy": 5.755231428146362,
      "epoch": 0.7768916553199766,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004945114008248319,
      "loss": 5.6449,
      "mean_token_accuracy": 0.14884610027074813,
      "num_tokens": 17328210.0,
      "step": 9985
    },
    {
      "entropy": 5.584620618820191,
      "epoch": 0.7772806846916942,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004945053148151619,
      "loss": 5.3651,
      "mean_token_accuracy": 0.1722089931368828,
      "num_tokens": 17336287.0,
      "step": 9990
    },
    {
      "entropy": 5.564840316772461,
      "epoch": 0.7776697140634118,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004944992254748482,
      "loss": 5.4267,
      "mean_token_accuracy": 0.16362099945545197,
      "num_tokens": 17344171.0,
      "step": 9995
    },
    {
      "entropy": 5.535340070724487,
      "epoch": 0.7780587434351294,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004944931328039832,
      "loss": 5.4545,
      "mean_token_accuracy": 0.169051456451416,
      "num_tokens": 17352786.0,
      "step": 10000
    },
    {
      "entropy": 5.541292333602906,
      "epoch": 0.778447772806847,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004944870368026592,
      "loss": 5.4307,
      "mean_token_accuracy": 0.16499633044004441,
      "num_tokens": 17361824.0,
      "step": 10005
    },
    {
      "entropy": 5.592077016830444,
      "epoch": 0.7788368021785644,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004944809374709689,
      "loss": 5.4585,
      "mean_token_accuracy": 0.16586715281009673,
      "num_tokens": 17370249.0,
      "step": 10010
    },
    {
      "entropy": 5.572728633880615,
      "epoch": 0.779225831550282,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004944748348090045,
      "loss": 5.4421,
      "mean_token_accuracy": 0.16339399069547653,
      "num_tokens": 17379650.0,
      "step": 10015
    },
    {
      "entropy": 5.5525439262390135,
      "epoch": 0.7796148609219996,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004944687288168589,
      "loss": 5.4564,
      "mean_token_accuracy": 0.1660699188709259,
      "num_tokens": 17388046.0,
      "step": 10020
    },
    {
      "entropy": 5.638512372970581,
      "epoch": 0.7800038902937172,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004944626194946246,
      "loss": 5.4271,
      "mean_token_accuracy": 0.17610761970281602,
      "num_tokens": 17395851.0,
      "step": 10025
    },
    {
      "entropy": 5.612651348114014,
      "epoch": 0.7803929196654348,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004944565068423945,
      "loss": 5.5025,
      "mean_token_accuracy": 0.1645132765173912,
      "num_tokens": 17404273.0,
      "step": 10030
    },
    {
      "entropy": 5.528199911117554,
      "epoch": 0.7807819490371523,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004944503908602612,
      "loss": 5.4391,
      "mean_token_accuracy": 0.16655797511339188,
      "num_tokens": 17412884.0,
      "step": 10035
    },
    {
      "entropy": 5.565637111663818,
      "epoch": 0.7811709784088698,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004944442715483174,
      "loss": 5.4246,
      "mean_token_accuracy": 0.16185757219791413,
      "num_tokens": 17420939.0,
      "step": 10040
    },
    {
      "entropy": 5.574648046493531,
      "epoch": 0.7815600077805874,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004944381489066561,
      "loss": 5.4762,
      "mean_token_accuracy": 0.16504090428352355,
      "num_tokens": 17429838.0,
      "step": 10045
    },
    {
      "entropy": 5.630850172042846,
      "epoch": 0.781949037152305,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004944320229353702,
      "loss": 5.4576,
      "mean_token_accuracy": 0.16243240088224412,
      "num_tokens": 17439164.0,
      "step": 10050
    },
    {
      "entropy": 5.634826183319092,
      "epoch": 0.7823380665240226,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004944258936345526,
      "loss": 5.4317,
      "mean_token_accuracy": 0.16340447217226028,
      "num_tokens": 17447544.0,
      "step": 10055
    },
    {
      "entropy": 5.478040027618408,
      "epoch": 0.7827270958957401,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004944197610042963,
      "loss": 5.3042,
      "mean_token_accuracy": 0.1662735790014267,
      "num_tokens": 17456270.0,
      "step": 10060
    },
    {
      "entropy": 5.567981147766114,
      "epoch": 0.7831161252674577,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004944136250446944,
      "loss": 5.4864,
      "mean_token_accuracy": 0.16388331800699235,
      "num_tokens": 17465241.0,
      "step": 10065
    },
    {
      "entropy": 5.686744737625122,
      "epoch": 0.7835051546391752,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004944074857558399,
      "loss": 5.4825,
      "mean_token_accuracy": 0.1565874397754669,
      "num_tokens": 17474460.0,
      "step": 10070
    },
    {
      "entropy": 5.599651908874511,
      "epoch": 0.7838941840108928,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004944013431378261,
      "loss": 5.4503,
      "mean_token_accuracy": 0.16428371965885163,
      "num_tokens": 17483256.0,
      "step": 10075
    },
    {
      "entropy": 5.623570966720581,
      "epoch": 0.7842832133826104,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004943951971907461,
      "loss": 5.4704,
      "mean_token_accuracy": 0.16218913793563844,
      "num_tokens": 17491922.0,
      "step": 10080
    },
    {
      "entropy": 5.608122491836548,
      "epoch": 0.7846722427543279,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004943890479146933,
      "loss": 5.4693,
      "mean_token_accuracy": 0.16983896046876906,
      "num_tokens": 17499763.0,
      "step": 10085
    },
    {
      "entropy": 5.619894170761109,
      "epoch": 0.7850612721260455,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004943828953097607,
      "loss": 5.4396,
      "mean_token_accuracy": 0.16675328761339187,
      "num_tokens": 17508682.0,
      "step": 10090
    },
    {
      "entropy": 5.518069171905518,
      "epoch": 0.7854503014977631,
      "grad_norm": 1.46875,
      "learning_rate": 0.000494376739376042,
      "loss": 5.3658,
      "mean_token_accuracy": 0.16787486970424653,
      "num_tokens": 17517269.0,
      "step": 10095
    },
    {
      "entropy": 5.444378519058228,
      "epoch": 0.7858393308694807,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004943705801136304,
      "loss": 5.3881,
      "mean_token_accuracy": 0.1697043627500534,
      "num_tokens": 17525135.0,
      "step": 10100
    },
    {
      "entropy": 5.537582588195801,
      "epoch": 0.7862283602411982,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004943644175226193,
      "loss": 5.4714,
      "mean_token_accuracy": 0.1688419073820114,
      "num_tokens": 17534195.0,
      "step": 10105
    },
    {
      "entropy": 5.574254274368286,
      "epoch": 0.7866173896129158,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004943582516031025,
      "loss": 5.3923,
      "mean_token_accuracy": 0.16879581660032272,
      "num_tokens": 17543016.0,
      "step": 10110
    },
    {
      "entropy": 5.642699337005615,
      "epoch": 0.7870064189846333,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004943520823551732,
      "loss": 5.4769,
      "mean_token_accuracy": 0.1671273186802864,
      "num_tokens": 17551854.0,
      "step": 10115
    },
    {
      "entropy": 5.58419017791748,
      "epoch": 0.7873954483563509,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004943459097789252,
      "loss": 5.3816,
      "mean_token_accuracy": 0.1663597121834755,
      "num_tokens": 17560739.0,
      "step": 10120
    },
    {
      "entropy": 5.572810363769531,
      "epoch": 0.7877844777280685,
      "grad_norm": 1.375,
      "learning_rate": 0.0004943397338744523,
      "loss": 5.4077,
      "mean_token_accuracy": 0.1652447521686554,
      "num_tokens": 17569784.0,
      "step": 10125
    },
    {
      "entropy": 5.651245927810669,
      "epoch": 0.7881735070997861,
      "grad_norm": 1.5625,
      "learning_rate": 0.000494333554641848,
      "loss": 5.5803,
      "mean_token_accuracy": 0.15896444022655487,
      "num_tokens": 17579262.0,
      "step": 10130
    },
    {
      "entropy": 5.56685996055603,
      "epoch": 0.7885625364715037,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000494327372081206,
      "loss": 5.4249,
      "mean_token_accuracy": 0.16502940356731416,
      "num_tokens": 17587984.0,
      "step": 10135
    },
    {
      "entropy": 5.54715485572815,
      "epoch": 0.7889515658432211,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004943211861926204,
      "loss": 5.3763,
      "mean_token_accuracy": 0.17046990543603896,
      "num_tokens": 17596598.0,
      "step": 10140
    },
    {
      "entropy": 5.5723248481750485,
      "epoch": 0.7893405952149387,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004943149969761848,
      "loss": 5.5089,
      "mean_token_accuracy": 0.16314789652824402,
      "num_tokens": 17605768.0,
      "step": 10145
    },
    {
      "entropy": 5.589661455154419,
      "epoch": 0.7897296245866563,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004943088044319932,
      "loss": 5.4277,
      "mean_token_accuracy": 0.16240018904209136,
      "num_tokens": 17614412.0,
      "step": 10150
    },
    {
      "entropy": 5.651829719543457,
      "epoch": 0.7901186539583739,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004943026085601395,
      "loss": 5.4399,
      "mean_token_accuracy": 0.1601569950580597,
      "num_tokens": 17622515.0,
      "step": 10155
    },
    {
      "entropy": 5.587910747528076,
      "epoch": 0.7905076833300915,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004942964093607178,
      "loss": 5.365,
      "mean_token_accuracy": 0.16747316867113113,
      "num_tokens": 17631256.0,
      "step": 10160
    },
    {
      "entropy": 5.6510865688323975,
      "epoch": 0.790896712701809,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004942902068338222,
      "loss": 5.5021,
      "mean_token_accuracy": 0.1550629422068596,
      "num_tokens": 17639648.0,
      "step": 10165
    },
    {
      "entropy": 5.615673446655274,
      "epoch": 0.7912857420735265,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004942840009795466,
      "loss": 5.4277,
      "mean_token_accuracy": 0.1639723613858223,
      "num_tokens": 17648225.0,
      "step": 10170
    },
    {
      "entropy": 5.598308277130127,
      "epoch": 0.7916747714452441,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004942777917979855,
      "loss": 5.4785,
      "mean_token_accuracy": 0.16637687385082245,
      "num_tokens": 17656764.0,
      "step": 10175
    },
    {
      "entropy": 5.592406749725342,
      "epoch": 0.7920638008169617,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004942715792892328,
      "loss": 5.4409,
      "mean_token_accuracy": 0.15991899520158767,
      "num_tokens": 17665127.0,
      "step": 10180
    },
    {
      "entropy": 5.562086486816407,
      "epoch": 0.7924528301886793,
      "grad_norm": 1.453125,
      "learning_rate": 0.000494265363453383,
      "loss": 5.3483,
      "mean_token_accuracy": 0.1689964562654495,
      "num_tokens": 17673297.0,
      "step": 10185
    },
    {
      "entropy": 5.594065761566162,
      "epoch": 0.7928418595603968,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004942591442905302,
      "loss": 5.4877,
      "mean_token_accuracy": 0.1662108927965164,
      "num_tokens": 17681816.0,
      "step": 10190
    },
    {
      "entropy": 5.619212293624878,
      "epoch": 0.7932308889321144,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004942529218007689,
      "loss": 5.4946,
      "mean_token_accuracy": 0.16034965813159943,
      "num_tokens": 17690588.0,
      "step": 10195
    },
    {
      "entropy": 5.606663799285888,
      "epoch": 0.7936199183038319,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004942466959841936,
      "loss": 5.4009,
      "mean_token_accuracy": 0.16643059998750687,
      "num_tokens": 17699196.0,
      "step": 10200
    },
    {
      "entropy": 5.684493541717529,
      "epoch": 0.7940089476755495,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004942404668408985,
      "loss": 5.4966,
      "mean_token_accuracy": 0.16475944668054582,
      "num_tokens": 17708928.0,
      "step": 10205
    },
    {
      "entropy": 5.64884467124939,
      "epoch": 0.7943979770472671,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004942342343709783,
      "loss": 5.3999,
      "mean_token_accuracy": 0.16832922548055648,
      "num_tokens": 17717888.0,
      "step": 10210
    },
    {
      "entropy": 5.6147425174713135,
      "epoch": 0.7947870064189846,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004942279985745275,
      "loss": 5.4201,
      "mean_token_accuracy": 0.1606283888220787,
      "num_tokens": 17726022.0,
      "step": 10215
    },
    {
      "entropy": 5.52122130393982,
      "epoch": 0.7951760357907022,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004942217594516409,
      "loss": 5.3901,
      "mean_token_accuracy": 0.16497442424297332,
      "num_tokens": 17733998.0,
      "step": 10220
    },
    {
      "entropy": 5.584072780609131,
      "epoch": 0.7955650651624198,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000494215517002413,
      "loss": 5.4461,
      "mean_token_accuracy": 0.1670831933617592,
      "num_tokens": 17742039.0,
      "step": 10225
    },
    {
      "entropy": 5.565745782852173,
      "epoch": 0.7959540945341373,
      "grad_norm": 1.375,
      "learning_rate": 0.0004942092712269384,
      "loss": 5.3595,
      "mean_token_accuracy": 0.17128196358680725,
      "num_tokens": 17751474.0,
      "step": 10230
    },
    {
      "entropy": 5.531363391876221,
      "epoch": 0.7963431239058549,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004942030221253122,
      "loss": 5.3525,
      "mean_token_accuracy": 0.16915419846773147,
      "num_tokens": 17759483.0,
      "step": 10235
    },
    {
      "entropy": 5.5520336627960205,
      "epoch": 0.7967321532775724,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004941967696976289,
      "loss": 5.4772,
      "mean_token_accuracy": 0.16447118073701858,
      "num_tokens": 17768025.0,
      "step": 10240
    },
    {
      "entropy": 5.5067604064941404,
      "epoch": 0.79712118264929,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004941905139439835,
      "loss": 5.3213,
      "mean_token_accuracy": 0.16718516796827315,
      "num_tokens": 17776861.0,
      "step": 10245
    },
    {
      "entropy": 5.572161340713501,
      "epoch": 0.7975102120210076,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004941842548644712,
      "loss": 5.4757,
      "mean_token_accuracy": 0.165050645172596,
      "num_tokens": 17786221.0,
      "step": 10250
    },
    {
      "entropy": 5.631086683273315,
      "epoch": 0.7978992413927252,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004941779924591864,
      "loss": 5.4619,
      "mean_token_accuracy": 0.15975369215011598,
      "num_tokens": 17794619.0,
      "step": 10255
    },
    {
      "entropy": 5.614037752151489,
      "epoch": 0.7982882707644428,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004941717267282244,
      "loss": 5.4449,
      "mean_token_accuracy": 0.16635915637016296,
      "num_tokens": 17803288.0,
      "step": 10260
    },
    {
      "entropy": 5.569530868530274,
      "epoch": 0.7986773001361602,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004941654576716806,
      "loss": 5.513,
      "mean_token_accuracy": 0.1579871207475662,
      "num_tokens": 17812648.0,
      "step": 10265
    },
    {
      "entropy": 5.544220733642578,
      "epoch": 0.7990663295078778,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004941591852896495,
      "loss": 5.4183,
      "mean_token_accuracy": 0.168813855946064,
      "num_tokens": 17821955.0,
      "step": 10270
    },
    {
      "entropy": 5.671693468093872,
      "epoch": 0.7994553588795954,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004941529095822268,
      "loss": 5.4531,
      "mean_token_accuracy": 0.16473222225904466,
      "num_tokens": 17829987.0,
      "step": 10275
    },
    {
      "entropy": 5.563750839233398,
      "epoch": 0.799844388251313,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004941466305495074,
      "loss": 5.4015,
      "mean_token_accuracy": 0.17213702201843262,
      "num_tokens": 17839298.0,
      "step": 10280
    },
    {
      "entropy": 5.580596446990967,
      "epoch": 0.8002334176230306,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004941403481915867,
      "loss": 5.4335,
      "mean_token_accuracy": 0.16103956699371338,
      "num_tokens": 17847950.0,
      "step": 10285
    },
    {
      "entropy": 5.6678954601287845,
      "epoch": 0.800622446994748,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004941340625085601,
      "loss": 5.5517,
      "mean_token_accuracy": 0.15696440488100052,
      "num_tokens": 17857416.0,
      "step": 10290
    },
    {
      "entropy": 5.685216808319092,
      "epoch": 0.8010114763664656,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004941277735005228,
      "loss": 5.4051,
      "mean_token_accuracy": 0.17213125079870223,
      "num_tokens": 17866516.0,
      "step": 10295
    },
    {
      "entropy": 5.651232957839966,
      "epoch": 0.8014005057381832,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004941214811675702,
      "loss": 5.5713,
      "mean_token_accuracy": 0.1570074513554573,
      "num_tokens": 17875823.0,
      "step": 10300
    },
    {
      "entropy": 5.529204177856445,
      "epoch": 0.8017895351099008,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004941151855097982,
      "loss": 5.4453,
      "mean_token_accuracy": 0.16270828992128372,
      "num_tokens": 17883597.0,
      "step": 10305
    },
    {
      "entropy": 5.655588626861572,
      "epoch": 0.8021785644816184,
      "grad_norm": 1.5,
      "learning_rate": 0.0004941088865273018,
      "loss": 5.4267,
      "mean_token_accuracy": 0.15925383567810059,
      "num_tokens": 17892053.0,
      "step": 10310
    },
    {
      "entropy": 5.571537494659424,
      "epoch": 0.802567593853336,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004941025842201769,
      "loss": 5.34,
      "mean_token_accuracy": 0.1758824184536934,
      "num_tokens": 17900187.0,
      "step": 10315
    },
    {
      "entropy": 5.566791629791259,
      "epoch": 0.8029566232250535,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004940962785885189,
      "loss": 5.5103,
      "mean_token_accuracy": 0.16310321539640427,
      "num_tokens": 17908608.0,
      "step": 10320
    },
    {
      "entropy": 5.621613883972168,
      "epoch": 0.803345652596771,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004940899696324237,
      "loss": 5.3931,
      "mean_token_accuracy": 0.16399924606084823,
      "num_tokens": 17917331.0,
      "step": 10325
    },
    {
      "entropy": 5.639385461807251,
      "epoch": 0.8037346819684886,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004940836573519868,
      "loss": 5.5026,
      "mean_token_accuracy": 0.16221323907375335,
      "num_tokens": 17925228.0,
      "step": 10330
    },
    {
      "entropy": 5.588077640533447,
      "epoch": 0.8041237113402062,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004940773417473043,
      "loss": 5.386,
      "mean_token_accuracy": 0.16964919567108155,
      "num_tokens": 17934110.0,
      "step": 10335
    },
    {
      "entropy": 5.571057844161987,
      "epoch": 0.8045127407119238,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004940710228184717,
      "loss": 5.4002,
      "mean_token_accuracy": 0.16821952313184738,
      "num_tokens": 17942237.0,
      "step": 10340
    },
    {
      "entropy": 5.536181497573852,
      "epoch": 0.8049017700836413,
      "grad_norm": 1.296875,
      "learning_rate": 0.000494064700565585,
      "loss": 5.2754,
      "mean_token_accuracy": 0.17488184869289397,
      "num_tokens": 17951288.0,
      "step": 10345
    },
    {
      "entropy": 5.613046884536743,
      "epoch": 0.8052907994553589,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004940583749887403,
      "loss": 5.5177,
      "mean_token_accuracy": 0.1582699842751026,
      "num_tokens": 17960300.0,
      "step": 10350
    },
    {
      "entropy": 5.569222831726075,
      "epoch": 0.8056798288270764,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004940520460880333,
      "loss": 5.388,
      "mean_token_accuracy": 0.16975254267454148,
      "num_tokens": 17968745.0,
      "step": 10355
    },
    {
      "entropy": 5.536905717849732,
      "epoch": 0.806068858198794,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004940457138635601,
      "loss": 5.3868,
      "mean_token_accuracy": 0.17145797312259675,
      "num_tokens": 17976614.0,
      "step": 10360
    },
    {
      "entropy": 5.601132726669311,
      "epoch": 0.8064578875705116,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004940393783154169,
      "loss": 5.5609,
      "mean_token_accuracy": 0.15555154234170915,
      "num_tokens": 17984769.0,
      "step": 10365
    },
    {
      "entropy": 5.637803840637207,
      "epoch": 0.8068469169422291,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004940330394436999,
      "loss": 5.4096,
      "mean_token_accuracy": 0.16131651550531387,
      "num_tokens": 17992885.0,
      "step": 10370
    },
    {
      "entropy": 5.6023236274719235,
      "epoch": 0.8072359463139467,
      "grad_norm": 1.5625,
      "learning_rate": 0.000494026697248505,
      "loss": 5.4427,
      "mean_token_accuracy": 0.16060977727174758,
      "num_tokens": 18001752.0,
      "step": 10375
    },
    {
      "entropy": 5.539025354385376,
      "epoch": 0.8076249756856643,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004940203517299287,
      "loss": 5.4514,
      "mean_token_accuracy": 0.16818694472312928,
      "num_tokens": 18009939.0,
      "step": 10380
    },
    {
      "entropy": 5.642893218994141,
      "epoch": 0.8080140050573819,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000494014002888067,
      "loss": 5.4256,
      "mean_token_accuracy": 0.17641784995794296,
      "num_tokens": 18017993.0,
      "step": 10385
    },
    {
      "entropy": 5.685582017898559,
      "epoch": 0.8084030344290994,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004940076507230166,
      "loss": 5.5275,
      "mean_token_accuracy": 0.1595136985182762,
      "num_tokens": 18027564.0,
      "step": 10390
    },
    {
      "entropy": 5.601905393600464,
      "epoch": 0.8087920638008169,
      "grad_norm": 1.5,
      "learning_rate": 0.0004940012952348735,
      "loss": 5.4925,
      "mean_token_accuracy": 0.1663194105029106,
      "num_tokens": 18036174.0,
      "step": 10395
    },
    {
      "entropy": 5.584610557556152,
      "epoch": 0.8091810931725345,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004939949364237345,
      "loss": 5.4184,
      "mean_token_accuracy": 0.16337654888629913,
      "num_tokens": 18044994.0,
      "step": 10400
    },
    {
      "entropy": 5.556111240386963,
      "epoch": 0.8095701225442521,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004939885742896958,
      "loss": 5.414,
      "mean_token_accuracy": 0.17001941949129104,
      "num_tokens": 18053789.0,
      "step": 10405
    },
    {
      "entropy": 5.656559991836548,
      "epoch": 0.8099591519159697,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000493982208832854,
      "loss": 5.5409,
      "mean_token_accuracy": 0.15735476613044738,
      "num_tokens": 18063483.0,
      "step": 10410
    },
    {
      "entropy": 5.597215843200684,
      "epoch": 0.8103481812876873,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004939758400533058,
      "loss": 5.3369,
      "mean_token_accuracy": 0.16720130443572997,
      "num_tokens": 18071952.0,
      "step": 10415
    },
    {
      "entropy": 5.521363878250122,
      "epoch": 0.8107372106594047,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004939694679511478,
      "loss": 5.3825,
      "mean_token_accuracy": 0.16801090091466903,
      "num_tokens": 18080323.0,
      "step": 10420
    },
    {
      "entropy": 5.6019659519195555,
      "epoch": 0.8111262400311223,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004939630925264765,
      "loss": 5.4981,
      "mean_token_accuracy": 0.16568824797868728,
      "num_tokens": 18089458.0,
      "step": 10425
    },
    {
      "entropy": 5.640874862670898,
      "epoch": 0.8115152694028399,
      "grad_norm": 1.953125,
      "learning_rate": 0.000493956713779389,
      "loss": 5.4191,
      "mean_token_accuracy": 0.16081434786319732,
      "num_tokens": 18098553.0,
      "step": 10430
    },
    {
      "entropy": 5.577661180496216,
      "epoch": 0.8119042987745575,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004939503317099817,
      "loss": 5.5074,
      "mean_token_accuracy": 0.16271842867136002,
      "num_tokens": 18106794.0,
      "step": 10435
    },
    {
      "entropy": 5.592539739608765,
      "epoch": 0.8122933281462751,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004939439463183517,
      "loss": 5.3803,
      "mean_token_accuracy": 0.1721673861145973,
      "num_tokens": 18115509.0,
      "step": 10440
    },
    {
      "entropy": 5.6062883853912355,
      "epoch": 0.8126823575179926,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004939375576045958,
      "loss": 5.5061,
      "mean_token_accuracy": 0.16080288887023925,
      "num_tokens": 18124548.0,
      "step": 10445
    },
    {
      "entropy": 5.59943413734436,
      "epoch": 0.8130713868897101,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004939311655688109,
      "loss": 5.4267,
      "mean_token_accuracy": 0.16586086750030518,
      "num_tokens": 18132295.0,
      "step": 10450
    },
    {
      "entropy": 5.616068983078003,
      "epoch": 0.8134604162614277,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004939247702110941,
      "loss": 5.5311,
      "mean_token_accuracy": 0.16209654808044432,
      "num_tokens": 18140731.0,
      "step": 10455
    },
    {
      "entropy": 5.633013343811035,
      "epoch": 0.8138494456331453,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004939183715315423,
      "loss": 5.4039,
      "mean_token_accuracy": 0.16486435532569885,
      "num_tokens": 18149834.0,
      "step": 10460
    },
    {
      "entropy": 5.60955114364624,
      "epoch": 0.8142384750048629,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004939119695302526,
      "loss": 5.3614,
      "mean_token_accuracy": 0.17106925398111344,
      "num_tokens": 18158745.0,
      "step": 10465
    },
    {
      "entropy": 5.645635509490967,
      "epoch": 0.8146275043765804,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004939055642073224,
      "loss": 5.5618,
      "mean_token_accuracy": 0.15414967238903046,
      "num_tokens": 18168292.0,
      "step": 10470
    },
    {
      "entropy": 5.646367979049683,
      "epoch": 0.815016533748298,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004938991555628484,
      "loss": 5.4561,
      "mean_token_accuracy": 0.15823084786534308,
      "num_tokens": 18177592.0,
      "step": 10475
    },
    {
      "entropy": 5.6279990673065186,
      "epoch": 0.8154055631200156,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004938927435969283,
      "loss": 5.4272,
      "mean_token_accuracy": 0.16151497662067413,
      "num_tokens": 18186845.0,
      "step": 10480
    },
    {
      "entropy": 5.614060544967652,
      "epoch": 0.8157945924917331,
      "grad_norm": 1.375,
      "learning_rate": 0.0004938863283096592,
      "loss": 5.5079,
      "mean_token_accuracy": 0.15635852068662642,
      "num_tokens": 18195417.0,
      "step": 10485
    },
    {
      "entropy": 5.601277017593384,
      "epoch": 0.8161836218634507,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004938799097011384,
      "loss": 5.4367,
      "mean_token_accuracy": 0.16019178926944733,
      "num_tokens": 18204009.0,
      "step": 10490
    },
    {
      "entropy": 5.568668603897095,
      "epoch": 0.8165726512351682,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004938734877714634,
      "loss": 5.4354,
      "mean_token_accuracy": 0.1707957848906517,
      "num_tokens": 18211927.0,
      "step": 10495
    },
    {
      "entropy": 5.516373252868652,
      "epoch": 0.8169616806068858,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004938670625207317,
      "loss": 5.3591,
      "mean_token_accuracy": 0.17173811048269272,
      "num_tokens": 18220394.0,
      "step": 10500
    },
    {
      "entropy": 5.631072282791138,
      "epoch": 0.8173507099786034,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004938606339490406,
      "loss": 5.4695,
      "mean_token_accuracy": 0.16214495599269868,
      "num_tokens": 18229803.0,
      "step": 10505
    },
    {
      "entropy": 5.617847442626953,
      "epoch": 0.817739739350321,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004938542020564877,
      "loss": 5.4333,
      "mean_token_accuracy": 0.16107070446014404,
      "num_tokens": 18238431.0,
      "step": 10510
    },
    {
      "entropy": 5.549394273757935,
      "epoch": 0.8181287687220385,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004938477668431706,
      "loss": 5.4465,
      "mean_token_accuracy": 0.16725078821182252,
      "num_tokens": 18246745.0,
      "step": 10515
    },
    {
      "entropy": 5.530304431915283,
      "epoch": 0.8185177980937561,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004938413283091871,
      "loss": 5.4062,
      "mean_token_accuracy": 0.1659283459186554,
      "num_tokens": 18255569.0,
      "step": 10520
    },
    {
      "entropy": 5.569682216644287,
      "epoch": 0.8189068274654736,
      "grad_norm": 1.625,
      "learning_rate": 0.0004938348864546347,
      "loss": 5.3463,
      "mean_token_accuracy": 0.16907418817281722,
      "num_tokens": 18263711.0,
      "step": 10525
    },
    {
      "entropy": 5.705848264694214,
      "epoch": 0.8192958568371912,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004938284412796111,
      "loss": 5.5241,
      "mean_token_accuracy": 0.16439357846975328,
      "num_tokens": 18272220.0,
      "step": 10530
    },
    {
      "entropy": 5.573349475860596,
      "epoch": 0.8196848862089088,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004938219927842144,
      "loss": 5.4329,
      "mean_token_accuracy": 0.163949453830719,
      "num_tokens": 18281107.0,
      "step": 10535
    },
    {
      "entropy": 5.4945001125335695,
      "epoch": 0.8200739155806264,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004938155409685422,
      "loss": 5.3947,
      "mean_token_accuracy": 0.16415611356496812,
      "num_tokens": 18290196.0,
      "step": 10540
    },
    {
      "entropy": 5.641933822631836,
      "epoch": 0.820462944952344,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004938090858326923,
      "loss": 5.5487,
      "mean_token_accuracy": 0.16089368164539336,
      "num_tokens": 18299418.0,
      "step": 10545
    },
    {
      "entropy": 5.534397268295288,
      "epoch": 0.8208519743240614,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000493802627376763,
      "loss": 5.2681,
      "mean_token_accuracy": 0.17511842250823975,
      "num_tokens": 18308171.0,
      "step": 10550
    },
    {
      "entropy": 5.6119359016418455,
      "epoch": 0.821241003695779,
      "grad_norm": 2.5,
      "learning_rate": 0.0004937961656008518,
      "loss": 5.469,
      "mean_token_accuracy": 0.16383302211761475,
      "num_tokens": 18317144.0,
      "step": 10555
    },
    {
      "entropy": 5.598818588256836,
      "epoch": 0.8216300330674966,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004937897005050573,
      "loss": 5.5389,
      "mean_token_accuracy": 0.16111208945512773,
      "num_tokens": 18326648.0,
      "step": 10560
    },
    {
      "entropy": 5.663065338134766,
      "epoch": 0.8220190624392142,
      "grad_norm": 1.625,
      "learning_rate": 0.0004937832320894772,
      "loss": 5.4837,
      "mean_token_accuracy": 0.16640617996454238,
      "num_tokens": 18335901.0,
      "step": 10565
    },
    {
      "entropy": 5.592900085449219,
      "epoch": 0.8224080918109318,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004937767603542098,
      "loss": 5.4167,
      "mean_token_accuracy": 0.16778741925954818,
      "num_tokens": 18344905.0,
      "step": 10570
    },
    {
      "entropy": 5.480745840072632,
      "epoch": 0.8227971211826492,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004937702852993534,
      "loss": 5.3637,
      "mean_token_accuracy": 0.17065452486276628,
      "num_tokens": 18353422.0,
      "step": 10575
    },
    {
      "entropy": 5.719949340820312,
      "epoch": 0.8231861505543668,
      "grad_norm": 1.484375,
      "learning_rate": 0.000493763806925006,
      "loss": 5.5222,
      "mean_token_accuracy": 0.1527490124106407,
      "num_tokens": 18361632.0,
      "step": 10580
    },
    {
      "entropy": 5.584030675888061,
      "epoch": 0.8235751799260844,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004937573252312661,
      "loss": 5.3702,
      "mean_token_accuracy": 0.17391784489154816,
      "num_tokens": 18370142.0,
      "step": 10585
    },
    {
      "entropy": 5.516370964050293,
      "epoch": 0.823964209297802,
      "grad_norm": 1.8125,
      "learning_rate": 0.000493750840218232,
      "loss": 5.3987,
      "mean_token_accuracy": 0.16997681856155394,
      "num_tokens": 18378245.0,
      "step": 10590
    },
    {
      "entropy": 5.530230712890625,
      "epoch": 0.8243532386695196,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004937443518860021,
      "loss": 5.371,
      "mean_token_accuracy": 0.16971075683832168,
      "num_tokens": 18386814.0,
      "step": 10595
    },
    {
      "entropy": 5.557858180999756,
      "epoch": 0.8247422680412371,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004937378602346747,
      "loss": 5.3934,
      "mean_token_accuracy": 0.16131658852100372,
      "num_tokens": 18396276.0,
      "step": 10600
    },
    {
      "entropy": 5.602482557296753,
      "epoch": 0.8251312974129547,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004937313652643485,
      "loss": 5.4063,
      "mean_token_accuracy": 0.16615637689828872,
      "num_tokens": 18404445.0,
      "step": 10605
    },
    {
      "entropy": 5.520392036437988,
      "epoch": 0.8255203267846722,
      "grad_norm": 2.296875,
      "learning_rate": 0.000493724866975122,
      "loss": 5.3031,
      "mean_token_accuracy": 0.1748577743768692,
      "num_tokens": 18412870.0,
      "step": 10610
    },
    {
      "entropy": 5.508503103256226,
      "epoch": 0.8259093561563898,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004937183653670937,
      "loss": 5.4393,
      "mean_token_accuracy": 0.16420634537935258,
      "num_tokens": 18421084.0,
      "step": 10615
    },
    {
      "entropy": 5.564705753326416,
      "epoch": 0.8262983855281074,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004937118604403623,
      "loss": 5.4678,
      "mean_token_accuracy": 0.16587617248296738,
      "num_tokens": 18429869.0,
      "step": 10620
    },
    {
      "entropy": 5.573436832427978,
      "epoch": 0.8266874148998249,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004937053521950266,
      "loss": 5.3284,
      "mean_token_accuracy": 0.16766274869441986,
      "num_tokens": 18438747.0,
      "step": 10625
    },
    {
      "entropy": 5.665083026885986,
      "epoch": 0.8270764442715425,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004936988406311854,
      "loss": 5.5365,
      "mean_token_accuracy": 0.15890450030565262,
      "num_tokens": 18447241.0,
      "step": 10630
    },
    {
      "entropy": 5.7015656471252445,
      "epoch": 0.8274654736432601,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004936923257489373,
      "loss": 5.5596,
      "mean_token_accuracy": 0.15821138247847558,
      "num_tokens": 18455757.0,
      "step": 10635
    },
    {
      "entropy": 5.544709491729736,
      "epoch": 0.8278545030149776,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004936858075483811,
      "loss": 5.3769,
      "mean_token_accuracy": 0.17397935390472413,
      "num_tokens": 18463863.0,
      "step": 10640
    },
    {
      "entropy": 5.556622886657715,
      "epoch": 0.8282435323866952,
      "grad_norm": 3.0625,
      "learning_rate": 0.000493679286029616,
      "loss": 5.557,
      "mean_token_accuracy": 0.1569342777132988,
      "num_tokens": 18472396.0,
      "step": 10645
    },
    {
      "entropy": 5.580170726776123,
      "epoch": 0.8286325617584127,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004936727611927407,
      "loss": 5.4125,
      "mean_token_accuracy": 0.1605270177125931,
      "num_tokens": 18480625.0,
      "step": 10650
    },
    {
      "entropy": 5.61102328300476,
      "epoch": 0.8290215911301303,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004936662330378546,
      "loss": 5.4834,
      "mean_token_accuracy": 0.1617751434445381,
      "num_tokens": 18489878.0,
      "step": 10655
    },
    {
      "entropy": 5.580025768280029,
      "epoch": 0.8294106205018479,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004936597015650561,
      "loss": 5.3825,
      "mean_token_accuracy": 0.16149143129587173,
      "num_tokens": 18498873.0,
      "step": 10660
    },
    {
      "entropy": 5.667516469955444,
      "epoch": 0.8297996498735655,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004936531667744448,
      "loss": 5.514,
      "mean_token_accuracy": 0.15458784252405167,
      "num_tokens": 18507716.0,
      "step": 10665
    },
    {
      "entropy": 5.58152208328247,
      "epoch": 0.8301886792452831,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004936466286661197,
      "loss": 5.3541,
      "mean_token_accuracy": 0.1737774208188057,
      "num_tokens": 18516547.0,
      "step": 10670
    },
    {
      "entropy": 5.597664213180542,
      "epoch": 0.8305777086170005,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00049364008724018,
      "loss": 5.4877,
      "mean_token_accuracy": 0.16016840189695358,
      "num_tokens": 18525714.0,
      "step": 10675
    },
    {
      "entropy": 5.611033296585083,
      "epoch": 0.8309667379887181,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000493633542496725,
      "loss": 5.5175,
      "mean_token_accuracy": 0.16490766108036042,
      "num_tokens": 18534981.0,
      "step": 10680
    },
    {
      "entropy": 5.605623769760132,
      "epoch": 0.8313557673604357,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004936269944358539,
      "loss": 5.4396,
      "mean_token_accuracy": 0.16582519561052322,
      "num_tokens": 18543599.0,
      "step": 10685
    },
    {
      "entropy": 5.6430579662323,
      "epoch": 0.8317447967321533,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004936204430576664,
      "loss": 5.4097,
      "mean_token_accuracy": 0.17386528998613357,
      "num_tokens": 18552207.0,
      "step": 10690
    },
    {
      "entropy": 5.586783695220947,
      "epoch": 0.8321338261038709,
      "grad_norm": 1.375,
      "learning_rate": 0.0004936138883622614,
      "loss": 5.4901,
      "mean_token_accuracy": 0.16271726936101913,
      "num_tokens": 18560421.0,
      "step": 10695
    },
    {
      "entropy": 5.592414093017578,
      "epoch": 0.8325228554755884,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004936073303497387,
      "loss": 5.3776,
      "mean_token_accuracy": 0.17105771005153655,
      "num_tokens": 18568812.0,
      "step": 10700
    },
    {
      "entropy": 5.595654630661011,
      "epoch": 0.8329118848473059,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004936007690201976,
      "loss": 5.3693,
      "mean_token_accuracy": 0.16839257776737213,
      "num_tokens": 18577077.0,
      "step": 10705
    },
    {
      "entropy": 5.580293750762939,
      "epoch": 0.8333009142190235,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000493594204373738,
      "loss": 5.4792,
      "mean_token_accuracy": 0.16489122807979584,
      "num_tokens": 18585754.0,
      "step": 10710
    },
    {
      "entropy": 5.522100019454956,
      "epoch": 0.8336899435907411,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004935876364104591,
      "loss": 5.3631,
      "mean_token_accuracy": 0.16696490049362184,
      "num_tokens": 18594578.0,
      "step": 10715
    },
    {
      "entropy": 5.617912912368775,
      "epoch": 0.8340789729624587,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004935810651304608,
      "loss": 5.4762,
      "mean_token_accuracy": 0.1588385820388794,
      "num_tokens": 18603810.0,
      "step": 10720
    },
    {
      "entropy": 5.6939843654632565,
      "epoch": 0.8344680023341763,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004935744905338427,
      "loss": 5.4456,
      "mean_token_accuracy": 0.161747607588768,
      "num_tokens": 18612870.0,
      "step": 10725
    },
    {
      "entropy": 5.620265007019043,
      "epoch": 0.8348570317058938,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004935679126207046,
      "loss": 5.5157,
      "mean_token_accuracy": 0.1638943523168564,
      "num_tokens": 18621353.0,
      "step": 10730
    },
    {
      "entropy": 5.502186822891235,
      "epoch": 0.8352460610776113,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004935613313911463,
      "loss": 5.3857,
      "mean_token_accuracy": 0.1673603981733322,
      "num_tokens": 18630152.0,
      "step": 10735
    },
    {
      "entropy": 5.562418603897095,
      "epoch": 0.8356350904493289,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004935547468452678,
      "loss": 5.4313,
      "mean_token_accuracy": 0.16592019349336623,
      "num_tokens": 18639278.0,
      "step": 10740
    },
    {
      "entropy": 5.610720300674439,
      "epoch": 0.8360241198210465,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004935481589831688,
      "loss": 5.4182,
      "mean_token_accuracy": 0.1697460889816284,
      "num_tokens": 18647436.0,
      "step": 10745
    },
    {
      "entropy": 5.701122903823853,
      "epoch": 0.8364131491927641,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004935415678049492,
      "loss": 5.442,
      "mean_token_accuracy": 0.15971136838197708,
      "num_tokens": 18655532.0,
      "step": 10750
    },
    {
      "entropy": 5.636491298675537,
      "epoch": 0.8368021785644816,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004935349733107094,
      "loss": 5.4711,
      "mean_token_accuracy": 0.16443484872579575,
      "num_tokens": 18664827.0,
      "step": 10755
    },
    {
      "entropy": 5.548076486587524,
      "epoch": 0.8371912079361992,
      "grad_norm": 1.5,
      "learning_rate": 0.000493528375500549,
      "loss": 5.3793,
      "mean_token_accuracy": 0.1734402894973755,
      "num_tokens": 18673535.0,
      "step": 10760
    },
    {
      "entropy": 5.615402460098267,
      "epoch": 0.8375802373079168,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004935217743745685,
      "loss": 5.5589,
      "mean_token_accuracy": 0.15757714211940765,
      "num_tokens": 18682223.0,
      "step": 10765
    },
    {
      "entropy": 5.717745971679688,
      "epoch": 0.8379692666796343,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004935151699328677,
      "loss": 5.4631,
      "mean_token_accuracy": 0.16686998009681703,
      "num_tokens": 18690847.0,
      "step": 10770
    },
    {
      "entropy": 5.659970808029175,
      "epoch": 0.8383582960513519,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004935085621755471,
      "loss": 5.4589,
      "mean_token_accuracy": 0.1638115555047989,
      "num_tokens": 18699658.0,
      "step": 10775
    },
    {
      "entropy": 5.596157026290894,
      "epoch": 0.8387473254230694,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000493501951102707,
      "loss": 5.4399,
      "mean_token_accuracy": 0.1618892252445221,
      "num_tokens": 18708396.0,
      "step": 10780
    },
    {
      "entropy": 5.562104082107544,
      "epoch": 0.839136354794787,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004934953367144474,
      "loss": 5.4614,
      "mean_token_accuracy": 0.16248811781406403,
      "num_tokens": 18717185.0,
      "step": 10785
    },
    {
      "entropy": 5.659436798095703,
      "epoch": 0.8395253841665046,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004934887190108688,
      "loss": 5.3989,
      "mean_token_accuracy": 0.16952570229768754,
      "num_tokens": 18725493.0,
      "step": 10790
    },
    {
      "entropy": 5.608885335922241,
      "epoch": 0.8399144135382222,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004934820979920719,
      "loss": 5.3988,
      "mean_token_accuracy": 0.16731245517730714,
      "num_tokens": 18733843.0,
      "step": 10795
    },
    {
      "entropy": 5.575446748733521,
      "epoch": 0.8403034429099397,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004934754736581567,
      "loss": 5.4595,
      "mean_token_accuracy": 0.16522205770015716,
      "num_tokens": 18742226.0,
      "step": 10800
    },
    {
      "entropy": 5.669210577011109,
      "epoch": 0.8406924722816572,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000493468846009224,
      "loss": 5.5991,
      "mean_token_accuracy": 0.15872693657875062,
      "num_tokens": 18751117.0,
      "step": 10805
    },
    {
      "entropy": 5.614519691467285,
      "epoch": 0.8410815016533748,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004934622150453742,
      "loss": 5.4192,
      "mean_token_accuracy": 0.16686538308858873,
      "num_tokens": 18759244.0,
      "step": 10810
    },
    {
      "entropy": 5.567561769485474,
      "epoch": 0.8414705310250924,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000493455580766708,
      "loss": 5.4834,
      "mean_token_accuracy": 0.16303838342428206,
      "num_tokens": 18768107.0,
      "step": 10815
    },
    {
      "entropy": 5.531203556060791,
      "epoch": 0.84185956039681,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004934489431733262,
      "loss": 5.323,
      "mean_token_accuracy": 0.16933274567127227,
      "num_tokens": 18777419.0,
      "step": 10820
    },
    {
      "entropy": 5.629980516433716,
      "epoch": 0.8422485897685276,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004934423022653293,
      "loss": 5.4494,
      "mean_token_accuracy": 0.16474525183439254,
      "num_tokens": 18786410.0,
      "step": 10825
    },
    {
      "entropy": 5.597207403182983,
      "epoch": 0.842637619140245,
      "grad_norm": 1.5,
      "learning_rate": 0.0004934356580428182,
      "loss": 5.4088,
      "mean_token_accuracy": 0.1639324113726616,
      "num_tokens": 18794409.0,
      "step": 10830
    },
    {
      "entropy": 5.546269083023072,
      "epoch": 0.8430266485119626,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004934290105058937,
      "loss": 5.4686,
      "mean_token_accuracy": 0.16208438575267792,
      "num_tokens": 18803561.0,
      "step": 10835
    },
    {
      "entropy": 5.5684432029724125,
      "epoch": 0.8434156778836802,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004934223596546565,
      "loss": 5.4267,
      "mean_token_accuracy": 0.16030320823192595,
      "num_tokens": 18812096.0,
      "step": 10840
    },
    {
      "entropy": 5.600976467132568,
      "epoch": 0.8438047072553978,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004934157054892077,
      "loss": 5.4528,
      "mean_token_accuracy": 0.1602882444858551,
      "num_tokens": 18820506.0,
      "step": 10845
    },
    {
      "entropy": 5.616394281387329,
      "epoch": 0.8441937366271154,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004934090480096482,
      "loss": 5.4516,
      "mean_token_accuracy": 0.16456348299980164,
      "num_tokens": 18829437.0,
      "step": 10850
    },
    {
      "entropy": 5.699944686889649,
      "epoch": 0.8445827659988329,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004934023872160791,
      "loss": 5.5665,
      "mean_token_accuracy": 0.16061313152313234,
      "num_tokens": 18838759.0,
      "step": 10855
    },
    {
      "entropy": 5.597001600265503,
      "epoch": 0.8449717953705504,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004933957231086014,
      "loss": 5.4612,
      "mean_token_accuracy": 0.16601901054382323,
      "num_tokens": 18847000.0,
      "step": 10860
    },
    {
      "entropy": 5.601649618148803,
      "epoch": 0.845360824742268,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004933890556873161,
      "loss": 5.5343,
      "mean_token_accuracy": 0.1564929887652397,
      "num_tokens": 18855370.0,
      "step": 10865
    },
    {
      "entropy": 5.698432302474975,
      "epoch": 0.8457498541139856,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004933823849523246,
      "loss": 5.5949,
      "mean_token_accuracy": 0.15831229239702224,
      "num_tokens": 18863825.0,
      "step": 10870
    },
    {
      "entropy": 5.620156240463257,
      "epoch": 0.8461388834857032,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004933757109037279,
      "loss": 5.3282,
      "mean_token_accuracy": 0.1642865091562271,
      "num_tokens": 18872088.0,
      "step": 10875
    },
    {
      "entropy": 5.5633057117462155,
      "epoch": 0.8465279128574207,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004933690335416274,
      "loss": 5.3486,
      "mean_token_accuracy": 0.1773277997970581,
      "num_tokens": 18880472.0,
      "step": 10880
    },
    {
      "entropy": 5.661197471618652,
      "epoch": 0.8469169422291383,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004933623528661245,
      "loss": 5.5364,
      "mean_token_accuracy": 0.16062142550945283,
      "num_tokens": 18889822.0,
      "step": 10885
    },
    {
      "entropy": 5.6680803298950195,
      "epoch": 0.8473059716008559,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004933556688773203,
      "loss": 5.461,
      "mean_token_accuracy": 0.1654386967420578,
      "num_tokens": 18897542.0,
      "step": 10890
    },
    {
      "entropy": 5.52772364616394,
      "epoch": 0.8476950009725734,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004933489815753165,
      "loss": 5.5129,
      "mean_token_accuracy": 0.1636799842119217,
      "num_tokens": 18906330.0,
      "step": 10895
    },
    {
      "entropy": 5.554571342468262,
      "epoch": 0.848084030344291,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004933422909602143,
      "loss": 5.3405,
      "mean_token_accuracy": 0.17273216545581818,
      "num_tokens": 18914176.0,
      "step": 10900
    },
    {
      "entropy": 5.5777268409729,
      "epoch": 0.8484730597160085,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004933355970321155,
      "loss": 5.3792,
      "mean_token_accuracy": 0.17397787570953369,
      "num_tokens": 18922233.0,
      "step": 10905
    },
    {
      "entropy": 5.512351560592651,
      "epoch": 0.8488620890877261,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004933288997911215,
      "loss": 5.4074,
      "mean_token_accuracy": 0.17171715646982194,
      "num_tokens": 18930742.0,
      "step": 10910
    },
    {
      "entropy": 5.58294711112976,
      "epoch": 0.8492511184594437,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004933221992373338,
      "loss": 5.3429,
      "mean_token_accuracy": 0.1723746880888939,
      "num_tokens": 18938879.0,
      "step": 10915
    },
    {
      "entropy": 5.702480030059815,
      "epoch": 0.8496401478311613,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004933154953708544,
      "loss": 5.5466,
      "mean_token_accuracy": 0.1529100403189659,
      "num_tokens": 18947753.0,
      "step": 10920
    },
    {
      "entropy": 5.628855419158936,
      "epoch": 0.8500291772028788,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004933087881917848,
      "loss": 5.4423,
      "mean_token_accuracy": 0.16019981801509858,
      "num_tokens": 18956805.0,
      "step": 10925
    },
    {
      "entropy": 5.513213014602661,
      "epoch": 0.8504182065745964,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004933020777002268,
      "loss": 5.3946,
      "mean_token_accuracy": 0.1699616253376007,
      "num_tokens": 18965683.0,
      "step": 10930
    },
    {
      "entropy": 5.545962190628051,
      "epoch": 0.8508072359463139,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004932953638962823,
      "loss": 5.3836,
      "mean_token_accuracy": 0.1677742287516594,
      "num_tokens": 18973873.0,
      "step": 10935
    },
    {
      "entropy": 5.533621215820313,
      "epoch": 0.8511962653180315,
      "grad_norm": 1.46875,
      "learning_rate": 0.000493288646780053,
      "loss": 5.3252,
      "mean_token_accuracy": 0.1759396269917488,
      "num_tokens": 18982097.0,
      "step": 10940
    },
    {
      "entropy": 5.569990348815918,
      "epoch": 0.8515852946897491,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004932819263516409,
      "loss": 5.3822,
      "mean_token_accuracy": 0.16632181257009507,
      "num_tokens": 18991019.0,
      "step": 10945
    },
    {
      "entropy": 5.584324312210083,
      "epoch": 0.8519743240614667,
      "grad_norm": 1.375,
      "learning_rate": 0.0004932752026111481,
      "loss": 5.3909,
      "mean_token_accuracy": 0.16592037975788115,
      "num_tokens": 18999231.0,
      "step": 10950
    },
    {
      "entropy": 5.627434206008911,
      "epoch": 0.8523633534331843,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004932684755586765,
      "loss": 5.4736,
      "mean_token_accuracy": 0.16317906975746155,
      "num_tokens": 19007807.0,
      "step": 10955
    },
    {
      "entropy": 5.536687231063842,
      "epoch": 0.8527523828049017,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004932617451943282,
      "loss": 5.4578,
      "mean_token_accuracy": 0.16746993213891984,
      "num_tokens": 19017908.0,
      "step": 10960
    },
    {
      "entropy": 5.55510458946228,
      "epoch": 0.8531414121766193,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004932550115182053,
      "loss": 5.4541,
      "mean_token_accuracy": 0.16267437040805816,
      "num_tokens": 19026999.0,
      "step": 10965
    },
    {
      "entropy": 5.516574144363403,
      "epoch": 0.8535304415483369,
      "grad_norm": 1.515625,
      "learning_rate": 0.00049324827453041,
      "loss": 5.3652,
      "mean_token_accuracy": 0.17764215916395187,
      "num_tokens": 19035738.0,
      "step": 10970
    },
    {
      "entropy": 5.5748766422271725,
      "epoch": 0.8539194709200545,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004932415342310446,
      "loss": 5.4544,
      "mean_token_accuracy": 0.16744358092546463,
      "num_tokens": 19044530.0,
      "step": 10975
    },
    {
      "entropy": 5.563995313644409,
      "epoch": 0.8543085002917721,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004932347906202111,
      "loss": 5.4462,
      "mean_token_accuracy": 0.16135307550430297,
      "num_tokens": 19053306.0,
      "step": 10980
    },
    {
      "entropy": 5.668405628204345,
      "epoch": 0.8546975296634896,
      "grad_norm": 1.46875,
      "learning_rate": 0.000493228043698012,
      "loss": 5.444,
      "mean_token_accuracy": 0.1676616132259369,
      "num_tokens": 19062172.0,
      "step": 10985
    },
    {
      "entropy": 5.448013257980347,
      "epoch": 0.8550865590352071,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004932212934645498,
      "loss": 5.2317,
      "mean_token_accuracy": 0.17657214850187303,
      "num_tokens": 19070804.0,
      "step": 10990
    },
    {
      "entropy": 5.558322095870972,
      "epoch": 0.8554755884069247,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004932145399199268,
      "loss": 5.3942,
      "mean_token_accuracy": 0.17129886597394944,
      "num_tokens": 19079055.0,
      "step": 10995
    },
    {
      "entropy": 5.553512859344482,
      "epoch": 0.8558646177786423,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004932077830642455,
      "loss": 5.3717,
      "mean_token_accuracy": 0.16821828484535217,
      "num_tokens": 19087459.0,
      "step": 11000
    },
    {
      "entropy": 5.458574676513672,
      "epoch": 0.8562536471503599,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004932010228976084,
      "loss": 5.2555,
      "mean_token_accuracy": 0.1769370749592781,
      "num_tokens": 19096271.0,
      "step": 11005
    },
    {
      "entropy": 5.552489280700684,
      "epoch": 0.8566426765220774,
      "grad_norm": 1.59375,
      "learning_rate": 0.000493194259420118,
      "loss": 5.4576,
      "mean_token_accuracy": 0.1654369756579399,
      "num_tokens": 19105257.0,
      "step": 11010
    },
    {
      "entropy": 5.599417304992675,
      "epoch": 0.857031705893795,
      "grad_norm": 1.6015625,
      "learning_rate": 0.000493187492631877,
      "loss": 5.4055,
      "mean_token_accuracy": 0.17078562825918198,
      "num_tokens": 19113673.0,
      "step": 11015
    },
    {
      "entropy": 5.586803674697876,
      "epoch": 0.8574207352655125,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004931807225329882,
      "loss": 5.3628,
      "mean_token_accuracy": 0.17019713819026946,
      "num_tokens": 19121635.0,
      "step": 11020
    },
    {
      "entropy": 5.576512002944947,
      "epoch": 0.8578097646372301,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004931739491235541,
      "loss": 5.4334,
      "mean_token_accuracy": 0.16907898634672164,
      "num_tokens": 19130097.0,
      "step": 11025
    },
    {
      "entropy": 5.505575513839721,
      "epoch": 0.8581987940089477,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004931671724036777,
      "loss": 5.3411,
      "mean_token_accuracy": 0.17085011452436447,
      "num_tokens": 19138349.0,
      "step": 11030
    },
    {
      "entropy": 5.521160554885864,
      "epoch": 0.8585878233806652,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004931603923734616,
      "loss": 5.431,
      "mean_token_accuracy": 0.16136478036642074,
      "num_tokens": 19146948.0,
      "step": 11035
    },
    {
      "entropy": 5.5303449630737305,
      "epoch": 0.8589768527523828,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004931536090330088,
      "loss": 5.3903,
      "mean_token_accuracy": 0.16915357261896133,
      "num_tokens": 19155372.0,
      "step": 11040
    },
    {
      "entropy": 5.659363651275635,
      "epoch": 0.8593658821241004,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004931468223824222,
      "loss": 5.4744,
      "mean_token_accuracy": 0.16891405135393142,
      "num_tokens": 19163634.0,
      "step": 11045
    },
    {
      "entropy": 5.687005615234375,
      "epoch": 0.859754911495818,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004931400324218048,
      "loss": 5.4874,
      "mean_token_accuracy": 0.16447663456201553,
      "num_tokens": 19171582.0,
      "step": 11050
    },
    {
      "entropy": 5.593333196640015,
      "epoch": 0.8601439408675355,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004931332391512597,
      "loss": 5.4967,
      "mean_token_accuracy": 0.16020538061857223,
      "num_tokens": 19181136.0,
      "step": 11055
    },
    {
      "entropy": 5.567076349258423,
      "epoch": 0.860532970239253,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004931264425708897,
      "loss": 5.4898,
      "mean_token_accuracy": 0.15613535791635513,
      "num_tokens": 19189389.0,
      "step": 11060
    },
    {
      "entropy": 5.629894351959228,
      "epoch": 0.8609219996109706,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004931196426807983,
      "loss": 5.462,
      "mean_token_accuracy": 0.171858774125576,
      "num_tokens": 19198315.0,
      "step": 11065
    },
    {
      "entropy": 5.690217018127441,
      "epoch": 0.8613110289826882,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004931128394810884,
      "loss": 5.4957,
      "mean_token_accuracy": 0.1585194043815136,
      "num_tokens": 19207234.0,
      "step": 11070
    },
    {
      "entropy": 5.594514226913452,
      "epoch": 0.8617000583544058,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004931060329718634,
      "loss": 5.4816,
      "mean_token_accuracy": 0.15357260406017303,
      "num_tokens": 19216552.0,
      "step": 11075
    },
    {
      "entropy": 5.605134344100952,
      "epoch": 0.8620890877261234,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004930992231532265,
      "loss": 5.4304,
      "mean_token_accuracy": 0.15944818556308746,
      "num_tokens": 19225237.0,
      "step": 11080
    },
    {
      "entropy": 5.532908391952515,
      "epoch": 0.8624781170978408,
      "grad_norm": 1.9609375,
      "learning_rate": 0.000493092410025281,
      "loss": 5.3591,
      "mean_token_accuracy": 0.17342484295368193,
      "num_tokens": 19233246.0,
      "step": 11085
    },
    {
      "entropy": 5.534931993484497,
      "epoch": 0.8628671464695584,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004930855935881302,
      "loss": 5.374,
      "mean_token_accuracy": 0.16980006992816926,
      "num_tokens": 19241748.0,
      "step": 11090
    },
    {
      "entropy": 5.624445199966431,
      "epoch": 0.863256175841276,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004930787738418777,
      "loss": 5.5339,
      "mean_token_accuracy": 0.16268638968467714,
      "num_tokens": 19250259.0,
      "step": 11095
    },
    {
      "entropy": 5.631997871398926,
      "epoch": 0.8636452052129936,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004930719507866269,
      "loss": 5.607,
      "mean_token_accuracy": 0.16095600426197051,
      "num_tokens": 19259266.0,
      "step": 11100
    },
    {
      "entropy": 5.629003381729126,
      "epoch": 0.8640342345847112,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004930651244224814,
      "loss": 5.4365,
      "mean_token_accuracy": 0.16444171816110612,
      "num_tokens": 19268099.0,
      "step": 11105
    },
    {
      "entropy": 5.575818014144898,
      "epoch": 0.8644232639564287,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004930582947495448,
      "loss": 5.3094,
      "mean_token_accuracy": 0.17518702149391174,
      "num_tokens": 19276894.0,
      "step": 11110
    },
    {
      "entropy": 5.506184387207031,
      "epoch": 0.8648122933281462,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004930514617679206,
      "loss": 5.4067,
      "mean_token_accuracy": 0.16996466964483262,
      "num_tokens": 19285724.0,
      "step": 11115
    },
    {
      "entropy": 5.595422697067261,
      "epoch": 0.8652013226998638,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004930446254777125,
      "loss": 5.3213,
      "mean_token_accuracy": 0.17358873635530472,
      "num_tokens": 19294468.0,
      "step": 11120
    },
    {
      "entropy": 5.595965957641601,
      "epoch": 0.8655903520715814,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004930377858790242,
      "loss": 5.3934,
      "mean_token_accuracy": 0.16887759119272233,
      "num_tokens": 19303575.0,
      "step": 11125
    },
    {
      "entropy": 5.632161092758179,
      "epoch": 0.865979381443299,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004930309429719595,
      "loss": 5.5483,
      "mean_token_accuracy": 0.1665432259440422,
      "num_tokens": 19312778.0,
      "step": 11130
    },
    {
      "entropy": 5.628087949752808,
      "epoch": 0.8663684108150166,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004930240967566224,
      "loss": 5.4051,
      "mean_token_accuracy": 0.16356466412544252,
      "num_tokens": 19320725.0,
      "step": 11135
    },
    {
      "entropy": 5.617732286453247,
      "epoch": 0.8667574401867341,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004930172472331167,
      "loss": 5.4315,
      "mean_token_accuracy": 0.17228960543870925,
      "num_tokens": 19329842.0,
      "step": 11140
    },
    {
      "entropy": 5.508480453491211,
      "epoch": 0.8671464695584516,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004930103944015463,
      "loss": 5.4311,
      "mean_token_accuracy": 0.1688416540622711,
      "num_tokens": 19338136.0,
      "step": 11145
    },
    {
      "entropy": 5.53710036277771,
      "epoch": 0.8675354989301692,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004930035382620149,
      "loss": 5.3925,
      "mean_token_accuracy": 0.1666646420955658,
      "num_tokens": 19346597.0,
      "step": 11150
    },
    {
      "entropy": 5.604273986816406,
      "epoch": 0.8679245283018868,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000492996678814627,
      "loss": 5.3847,
      "mean_token_accuracy": 0.16759981960058212,
      "num_tokens": 19355595.0,
      "step": 11155
    },
    {
      "entropy": 5.591334676742553,
      "epoch": 0.8683135576736044,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004929898160594865,
      "loss": 5.4532,
      "mean_token_accuracy": 0.17089799791574478,
      "num_tokens": 19364133.0,
      "step": 11160
    },
    {
      "entropy": 5.59797625541687,
      "epoch": 0.8687025870453219,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004929829499966974,
      "loss": 5.5101,
      "mean_token_accuracy": 0.15949829816818237,
      "num_tokens": 19373628.0,
      "step": 11165
    },
    {
      "entropy": 5.588370513916016,
      "epoch": 0.8690916164170395,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004929760806263641,
      "loss": 5.4369,
      "mean_token_accuracy": 0.16313839703798294,
      "num_tokens": 19381901.0,
      "step": 11170
    },
    {
      "entropy": 5.536091613769531,
      "epoch": 0.8694806457887571,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004929692079485906,
      "loss": 5.527,
      "mean_token_accuracy": 0.1661791443824768,
      "num_tokens": 19390390.0,
      "step": 11175
    },
    {
      "entropy": 5.6095479965209964,
      "epoch": 0.8698696751604746,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004929623319634814,
      "loss": 5.4858,
      "mean_token_accuracy": 0.1574648216366768,
      "num_tokens": 19399858.0,
      "step": 11180
    },
    {
      "entropy": 5.576297378540039,
      "epoch": 0.8702587045321922,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004929554526711407,
      "loss": 5.4277,
      "mean_token_accuracy": 0.16378122717142105,
      "num_tokens": 19409062.0,
      "step": 11185
    },
    {
      "entropy": 5.563224458694458,
      "epoch": 0.8706477339039097,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004929485700716729,
      "loss": 5.4028,
      "mean_token_accuracy": 0.1680426374077797,
      "num_tokens": 19417228.0,
      "step": 11190
    },
    {
      "entropy": 5.610040950775146,
      "epoch": 0.8710367632756273,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004929416841651824,
      "loss": 5.509,
      "mean_token_accuracy": 0.1611345425248146,
      "num_tokens": 19426086.0,
      "step": 11195
    },
    {
      "entropy": 5.659562110900879,
      "epoch": 0.8714257926473449,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004929347949517739,
      "loss": 5.5053,
      "mean_token_accuracy": 0.1610720545053482,
      "num_tokens": 19435363.0,
      "step": 11200
    },
    {
      "entropy": 5.569135046005249,
      "epoch": 0.8718148220190625,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004929279024315516,
      "loss": 5.305,
      "mean_token_accuracy": 0.17181036323308946,
      "num_tokens": 19443355.0,
      "step": 11205
    },
    {
      "entropy": 5.524402189254761,
      "epoch": 0.87220385139078,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004929210066046204,
      "loss": 5.3494,
      "mean_token_accuracy": 0.16735538244247436,
      "num_tokens": 19452217.0,
      "step": 11210
    },
    {
      "entropy": 5.613103628158569,
      "epoch": 0.8725928807624975,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004929141074710847,
      "loss": 5.5384,
      "mean_token_accuracy": 0.156987027823925,
      "num_tokens": 19460034.0,
      "step": 11215
    },
    {
      "entropy": 5.593009805679321,
      "epoch": 0.8729819101342151,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004929072050310491,
      "loss": 5.4874,
      "mean_token_accuracy": 0.1655286356806755,
      "num_tokens": 19468403.0,
      "step": 11220
    },
    {
      "entropy": 5.5963386535644535,
      "epoch": 0.8733709395059327,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004929002992846188,
      "loss": 5.5069,
      "mean_token_accuracy": 0.16459607928991318,
      "num_tokens": 19477551.0,
      "step": 11225
    },
    {
      "entropy": 5.642898416519165,
      "epoch": 0.8737599688776503,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004928933902318981,
      "loss": 5.4011,
      "mean_token_accuracy": 0.16497209072113037,
      "num_tokens": 19486131.0,
      "step": 11230
    },
    {
      "entropy": 5.639529895782471,
      "epoch": 0.8741489982493679,
      "grad_norm": 1.546875,
      "learning_rate": 0.000492886477872992,
      "loss": 5.5351,
      "mean_token_accuracy": 0.15575776249170303,
      "num_tokens": 19496212.0,
      "step": 11235
    },
    {
      "entropy": 5.561580181121826,
      "epoch": 0.8745380276210853,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004928795622080054,
      "loss": 5.372,
      "mean_token_accuracy": 0.1667099341750145,
      "num_tokens": 19504556.0,
      "step": 11240
    },
    {
      "entropy": 5.589718866348266,
      "epoch": 0.8749270569928029,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004928726432370434,
      "loss": 5.4418,
      "mean_token_accuracy": 0.1647622615098953,
      "num_tokens": 19512832.0,
      "step": 11245
    },
    {
      "entropy": 5.621614551544189,
      "epoch": 0.8753160863645205,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004928657209602107,
      "loss": 5.4634,
      "mean_token_accuracy": 0.16495981961488723,
      "num_tokens": 19521410.0,
      "step": 11250
    },
    {
      "entropy": 5.576944732666016,
      "epoch": 0.8757051157362381,
      "grad_norm": 1.375,
      "learning_rate": 0.0004928587953776125,
      "loss": 5.4316,
      "mean_token_accuracy": 0.16559310704469682,
      "num_tokens": 19529935.0,
      "step": 11255
    },
    {
      "entropy": 5.463144540786743,
      "epoch": 0.8760941451079557,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004928518664893538,
      "loss": 5.3266,
      "mean_token_accuracy": 0.17452301383018493,
      "num_tokens": 19538330.0,
      "step": 11260
    },
    {
      "entropy": 5.523600101470947,
      "epoch": 0.8764831744796732,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004928449342955397,
      "loss": 5.3399,
      "mean_token_accuracy": 0.16870104670524597,
      "num_tokens": 19546860.0,
      "step": 11265
    },
    {
      "entropy": 5.629951238632202,
      "epoch": 0.8768722038513908,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004928379987962756,
      "loss": 5.4226,
      "mean_token_accuracy": 0.16777731627225875,
      "num_tokens": 19554945.0,
      "step": 11270
    },
    {
      "entropy": 5.602519750595093,
      "epoch": 0.8772612332231083,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004928310599916666,
      "loss": 5.4509,
      "mean_token_accuracy": 0.16262111216783523,
      "num_tokens": 19563182.0,
      "step": 11275
    },
    {
      "entropy": 5.526263952255249,
      "epoch": 0.8776502625948259,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004928241178818178,
      "loss": 5.4305,
      "mean_token_accuracy": 0.16654721945524215,
      "num_tokens": 19571405.0,
      "step": 11280
    },
    {
      "entropy": 5.725631141662598,
      "epoch": 0.8780392919665435,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004928171724668349,
      "loss": 5.5185,
      "mean_token_accuracy": 0.16668618023395537,
      "num_tokens": 19580053.0,
      "step": 11285
    },
    {
      "entropy": 5.691386985778808,
      "epoch": 0.878428321338261,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004928102237468229,
      "loss": 5.5353,
      "mean_token_accuracy": 0.16114630401134492,
      "num_tokens": 19588457.0,
      "step": 11290
    },
    {
      "entropy": 5.520010614395142,
      "epoch": 0.8788173507099786,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004928032717218876,
      "loss": 5.3756,
      "mean_token_accuracy": 0.16722139418125154,
      "num_tokens": 19597608.0,
      "step": 11295
    },
    {
      "entropy": 5.5498902797698975,
      "epoch": 0.8792063800816962,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004927963163921343,
      "loss": 5.3957,
      "mean_token_accuracy": 0.17279530167579651,
      "num_tokens": 19605904.0,
      "step": 11300
    },
    {
      "entropy": 5.604985475540161,
      "epoch": 0.8795954094534137,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004927893577576685,
      "loss": 5.3813,
      "mean_token_accuracy": 0.16346621215343476,
      "num_tokens": 19614660.0,
      "step": 11305
    },
    {
      "entropy": 5.506232166290284,
      "epoch": 0.8799844388251313,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004927823958185959,
      "loss": 5.3655,
      "mean_token_accuracy": 0.17348123341798782,
      "num_tokens": 19623210.0,
      "step": 11310
    },
    {
      "entropy": 5.561007976531982,
      "epoch": 0.8803734681968488,
      "grad_norm": 1.5703125,
      "learning_rate": 0.000492775430575022,
      "loss": 5.4444,
      "mean_token_accuracy": 0.16233074367046357,
      "num_tokens": 19631747.0,
      "step": 11315
    },
    {
      "entropy": 5.64417781829834,
      "epoch": 0.8807624975685664,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004927684620270527,
      "loss": 5.5251,
      "mean_token_accuracy": 0.15691719055175782,
      "num_tokens": 19639902.0,
      "step": 11320
    },
    {
      "entropy": 5.673055934906006,
      "epoch": 0.881151526940284,
      "grad_norm": 1.625,
      "learning_rate": 0.0004927614901747935,
      "loss": 5.5495,
      "mean_token_accuracy": 0.15953097343444825,
      "num_tokens": 19648478.0,
      "step": 11325
    },
    {
      "entropy": 5.701460456848144,
      "epoch": 0.8815405563120016,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004927545150183503,
      "loss": 5.5502,
      "mean_token_accuracy": 0.16400340497493743,
      "num_tokens": 19657467.0,
      "step": 11330
    },
    {
      "entropy": 5.691887903213501,
      "epoch": 0.8819295856837192,
      "grad_norm": 1.6640625,
      "learning_rate": 0.000492747536557829,
      "loss": 5.5182,
      "mean_token_accuracy": 0.1579813688993454,
      "num_tokens": 19665382.0,
      "step": 11335
    },
    {
      "entropy": 5.561461162567139,
      "epoch": 0.8823186150554367,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004927405547933353,
      "loss": 5.3391,
      "mean_token_accuracy": 0.16862319111824037,
      "num_tokens": 19674486.0,
      "step": 11340
    },
    {
      "entropy": 5.607643938064575,
      "epoch": 0.8827076444271542,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004927335697249753,
      "loss": 5.3618,
      "mean_token_accuracy": 0.16630233973264694,
      "num_tokens": 19683060.0,
      "step": 11345
    },
    {
      "entropy": 5.49778938293457,
      "epoch": 0.8830966737988718,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004927265813528549,
      "loss": 5.3447,
      "mean_token_accuracy": 0.1723550483584404,
      "num_tokens": 19691634.0,
      "step": 11350
    },
    {
      "entropy": 5.5745201587677,
      "epoch": 0.8834857031705894,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004927195896770804,
      "loss": 5.43,
      "mean_token_accuracy": 0.1592835769057274,
      "num_tokens": 19699975.0,
      "step": 11355
    },
    {
      "entropy": 5.609763813018799,
      "epoch": 0.883874732542307,
      "grad_norm": 1.625,
      "learning_rate": 0.0004927125946977574,
      "loss": 5.3904,
      "mean_token_accuracy": 0.16351550817489624,
      "num_tokens": 19708431.0,
      "step": 11360
    },
    {
      "entropy": 5.54373517036438,
      "epoch": 0.8842637619140246,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004927055964149923,
      "loss": 5.4088,
      "mean_token_accuracy": 0.16205148696899413,
      "num_tokens": 19717189.0,
      "step": 11365
    },
    {
      "entropy": 5.623295307159424,
      "epoch": 0.884652791285742,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004926985948288914,
      "loss": 5.4746,
      "mean_token_accuracy": 0.1615665078163147,
      "num_tokens": 19725855.0,
      "step": 11370
    },
    {
      "entropy": 5.572314357757568,
      "epoch": 0.8850418206574596,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004926915899395608,
      "loss": 5.373,
      "mean_token_accuracy": 0.1688873991370201,
      "num_tokens": 19734617.0,
      "step": 11375
    },
    {
      "entropy": 5.653919458389282,
      "epoch": 0.8854308500291772,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004926845817471068,
      "loss": 5.5065,
      "mean_token_accuracy": 0.16308294236660004,
      "num_tokens": 19743731.0,
      "step": 11380
    },
    {
      "entropy": 5.674256706237793,
      "epoch": 0.8858198794008948,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004926775702516358,
      "loss": 5.4546,
      "mean_token_accuracy": 0.16395652443170547,
      "num_tokens": 19752336.0,
      "step": 11385
    },
    {
      "entropy": 5.597683382034302,
      "epoch": 0.8862089087726124,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004926705554532541,
      "loss": 5.4435,
      "mean_token_accuracy": 0.16464806348085403,
      "num_tokens": 19761488.0,
      "step": 11390
    },
    {
      "entropy": 5.5336109638214115,
      "epoch": 0.8865979381443299,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004926635373520682,
      "loss": 5.3615,
      "mean_token_accuracy": 0.16801237165927888,
      "num_tokens": 19770330.0,
      "step": 11395
    },
    {
      "entropy": 5.606996393203735,
      "epoch": 0.8869869675160474,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004926565159481845,
      "loss": 5.4215,
      "mean_token_accuracy": 0.1609994277358055,
      "num_tokens": 19779037.0,
      "step": 11400
    },
    {
      "entropy": 5.6447957992553714,
      "epoch": 0.887375996887765,
      "grad_norm": 1.5,
      "learning_rate": 0.0004926494912417097,
      "loss": 5.4884,
      "mean_token_accuracy": 0.16048842221498488,
      "num_tokens": 19787661.0,
      "step": 11405
    },
    {
      "entropy": 5.582346391677857,
      "epoch": 0.8877650262594826,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004926424632327503,
      "loss": 5.3951,
      "mean_token_accuracy": 0.16768926084041597,
      "num_tokens": 19796421.0,
      "step": 11410
    },
    {
      "entropy": 5.584974193572998,
      "epoch": 0.8881540556312002,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004926354319214129,
      "loss": 5.3981,
      "mean_token_accuracy": 0.16604495644569398,
      "num_tokens": 19805842.0,
      "step": 11415
    },
    {
      "entropy": 5.655823516845703,
      "epoch": 0.8885430850029177,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004926283973078041,
      "loss": 5.4218,
      "mean_token_accuracy": 0.16787973940372466,
      "num_tokens": 19814179.0,
      "step": 11420
    },
    {
      "entropy": 5.613330984115601,
      "epoch": 0.8889321143746353,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004926213593920309,
      "loss": 5.4923,
      "mean_token_accuracy": 0.16426144540309906,
      "num_tokens": 19822435.0,
      "step": 11425
    },
    {
      "entropy": 5.588300609588623,
      "epoch": 0.8893211437463528,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004926143181742,
      "loss": 5.4055,
      "mean_token_accuracy": 0.1616242453455925,
      "num_tokens": 19830300.0,
      "step": 11430
    },
    {
      "entropy": 5.58564510345459,
      "epoch": 0.8897101731180704,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004926072736544181,
      "loss": 5.5159,
      "mean_token_accuracy": 0.16276822090148926,
      "num_tokens": 19839264.0,
      "step": 11435
    },
    {
      "entropy": 5.621684646606445,
      "epoch": 0.890099202489788,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004926002258327922,
      "loss": 5.4597,
      "mean_token_accuracy": 0.16228195577859877,
      "num_tokens": 19848086.0,
      "step": 11440
    },
    {
      "entropy": 5.6392816543579105,
      "epoch": 0.8904882318615055,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004925931747094292,
      "loss": 5.4315,
      "mean_token_accuracy": 0.15673258900642395,
      "num_tokens": 19856571.0,
      "step": 11445
    },
    {
      "entropy": 5.5869261741638185,
      "epoch": 0.8908772612332231,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004925861202844361,
      "loss": 5.4554,
      "mean_token_accuracy": 0.16131344437599182,
      "num_tokens": 19865272.0,
      "step": 11450
    },
    {
      "entropy": 5.5566261291503904,
      "epoch": 0.8912662906049407,
      "grad_norm": 1.609375,
      "learning_rate": 0.00049257906255792,
      "loss": 5.4055,
      "mean_token_accuracy": 0.1614033855497837,
      "num_tokens": 19874307.0,
      "step": 11455
    },
    {
      "entropy": 5.58982629776001,
      "epoch": 0.8916553199766583,
      "grad_norm": 1.6015625,
      "learning_rate": 0.000492572001529988,
      "loss": 5.4526,
      "mean_token_accuracy": 0.16618746519088745,
      "num_tokens": 19882851.0,
      "step": 11460
    },
    {
      "entropy": 5.654384803771973,
      "epoch": 0.8920443493483758,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004925649372007469,
      "loss": 5.3495,
      "mean_token_accuracy": 0.1758921191096306,
      "num_tokens": 19891276.0,
      "step": 11465
    },
    {
      "entropy": 5.620083808898926,
      "epoch": 0.8924333787200933,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004925578695703045,
      "loss": 5.4917,
      "mean_token_accuracy": 0.15829311907291413,
      "num_tokens": 19900551.0,
      "step": 11470
    },
    {
      "entropy": 5.639655208587646,
      "epoch": 0.8928224080918109,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004925507986387676,
      "loss": 5.5149,
      "mean_token_accuracy": 0.16138267070055007,
      "num_tokens": 19909412.0,
      "step": 11475
    },
    {
      "entropy": 5.615196323394775,
      "epoch": 0.8932114374635285,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004925437244062436,
      "loss": 5.3594,
      "mean_token_accuracy": 0.16976200491189958,
      "num_tokens": 19918073.0,
      "step": 11480
    },
    {
      "entropy": 5.482284116744995,
      "epoch": 0.8936004668352461,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004925366468728397,
      "loss": 5.3073,
      "mean_token_accuracy": 0.1825845032930374,
      "num_tokens": 19926040.0,
      "step": 11485
    },
    {
      "entropy": 5.53025107383728,
      "epoch": 0.8939894962069637,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004925295660386635,
      "loss": 5.4712,
      "mean_token_accuracy": 0.16550873219966888,
      "num_tokens": 19934543.0,
      "step": 11490
    },
    {
      "entropy": 5.575025939941407,
      "epoch": 0.8943785255786811,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004925224819038224,
      "loss": 5.4113,
      "mean_token_accuracy": 0.1682252526283264,
      "num_tokens": 19943198.0,
      "step": 11495
    },
    {
      "entropy": 5.554923725128174,
      "epoch": 0.8947675549503987,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004925153944684239,
      "loss": 5.3523,
      "mean_token_accuracy": 0.17520031929016114,
      "num_tokens": 19952319.0,
      "step": 11500
    },
    {
      "entropy": 5.547123098373413,
      "epoch": 0.8951565843221163,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004925083037325754,
      "loss": 5.3438,
      "mean_token_accuracy": 0.16900049149990082,
      "num_tokens": 19961982.0,
      "step": 11505
    },
    {
      "entropy": 5.5912707328796385,
      "epoch": 0.8955456136938339,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004925012096963847,
      "loss": 5.3775,
      "mean_token_accuracy": 0.16102658361196517,
      "num_tokens": 19971055.0,
      "step": 11510
    },
    {
      "entropy": 5.426222467422486,
      "epoch": 0.8959346430655515,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004924941123599593,
      "loss": 5.2642,
      "mean_token_accuracy": 0.1792299598455429,
      "num_tokens": 19979850.0,
      "step": 11515
    },
    {
      "entropy": 5.569857549667359,
      "epoch": 0.896323672437269,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004924870117234069,
      "loss": 5.4525,
      "mean_token_accuracy": 0.16205697506666183,
      "num_tokens": 19989261.0,
      "step": 11520
    },
    {
      "entropy": 5.60758605003357,
      "epoch": 0.8967127018089865,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004924799077868353,
      "loss": 5.3745,
      "mean_token_accuracy": 0.16446672976017,
      "num_tokens": 19998321.0,
      "step": 11525
    },
    {
      "entropy": 5.539510774612427,
      "epoch": 0.8971017311807041,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004924728005503522,
      "loss": 5.4442,
      "mean_token_accuracy": 0.16351263523101806,
      "num_tokens": 20007191.0,
      "step": 11530
    },
    {
      "entropy": 5.566989707946777,
      "epoch": 0.8974907605524217,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004924656900140655,
      "loss": 5.4821,
      "mean_token_accuracy": 0.16563700139522552,
      "num_tokens": 20015805.0,
      "step": 11535
    },
    {
      "entropy": 5.563309526443481,
      "epoch": 0.8978797899241393,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004924585761780831,
      "loss": 5.4314,
      "mean_token_accuracy": 0.1693728670477867,
      "num_tokens": 20024861.0,
      "step": 11540
    },
    {
      "entropy": 5.718994665145874,
      "epoch": 0.8982688192958569,
      "grad_norm": 1.40625,
      "learning_rate": 0.000492451459042513,
      "loss": 5.5048,
      "mean_token_accuracy": 0.16540486142039298,
      "num_tokens": 20034208.0,
      "step": 11545
    },
    {
      "entropy": 5.63351469039917,
      "epoch": 0.8986578486675744,
      "grad_norm": 1.6953125,
      "learning_rate": 0.000492444338607463,
      "loss": 5.4469,
      "mean_token_accuracy": 0.16422081738710403,
      "num_tokens": 20042786.0,
      "step": 11550
    },
    {
      "entropy": 5.641626787185669,
      "epoch": 0.899046878039292,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004924372148730413,
      "loss": 5.5006,
      "mean_token_accuracy": 0.1618044413626194,
      "num_tokens": 20051697.0,
      "step": 11555
    },
    {
      "entropy": 5.640315103530884,
      "epoch": 0.8994359074110095,
      "grad_norm": 1.671875,
      "learning_rate": 0.000492430087839356,
      "loss": 5.4725,
      "mean_token_accuracy": 0.1687396615743637,
      "num_tokens": 20059814.0,
      "step": 11560
    },
    {
      "entropy": 5.575307369232178,
      "epoch": 0.8998249367827271,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004924229575065152,
      "loss": 5.4575,
      "mean_token_accuracy": 0.16698689609766007,
      "num_tokens": 20068329.0,
      "step": 11565
    },
    {
      "entropy": 5.614092016220093,
      "epoch": 0.9002139661544447,
      "grad_norm": 1.75,
      "learning_rate": 0.000492415823874627,
      "loss": 5.4582,
      "mean_token_accuracy": 0.1686816081404686,
      "num_tokens": 20076925.0,
      "step": 11570
    },
    {
      "entropy": 5.618636560440064,
      "epoch": 0.9006029955261622,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004924086869437998,
      "loss": 5.477,
      "mean_token_accuracy": 0.15615838915109634,
      "num_tokens": 20085248.0,
      "step": 11575
    },
    {
      "entropy": 5.653695392608642,
      "epoch": 0.9009920248978798,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004924015467141417,
      "loss": 5.4472,
      "mean_token_accuracy": 0.16372430920600892,
      "num_tokens": 20093929.0,
      "step": 11580
    },
    {
      "entropy": 5.575537443161011,
      "epoch": 0.9013810542695974,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004923944031857613,
      "loss": 5.5273,
      "mean_token_accuracy": 0.1625973328948021,
      "num_tokens": 20103141.0,
      "step": 11585
    },
    {
      "entropy": 5.583071327209472,
      "epoch": 0.901770083641315,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004923872563587668,
      "loss": 5.3516,
      "mean_token_accuracy": 0.17516722828149794,
      "num_tokens": 20111938.0,
      "step": 11590
    },
    {
      "entropy": 5.563779306411743,
      "epoch": 0.9021591130130325,
      "grad_norm": 1.625,
      "learning_rate": 0.0004923801062332666,
      "loss": 5.3575,
      "mean_token_accuracy": 0.17270515710115433,
      "num_tokens": 20120795.0,
      "step": 11595
    },
    {
      "entropy": 5.5637914657592775,
      "epoch": 0.90254814238475,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004923729528093694,
      "loss": 5.4399,
      "mean_token_accuracy": 0.1667784184217453,
      "num_tokens": 20129816.0,
      "step": 11600
    },
    {
      "entropy": 5.681299018859863,
      "epoch": 0.9029371717564676,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004923657960871836,
      "loss": 5.5512,
      "mean_token_accuracy": 0.15675133019685744,
      "num_tokens": 20138380.0,
      "step": 11605
    },
    {
      "entropy": 5.657552194595337,
      "epoch": 0.9033262011281852,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004923586360668178,
      "loss": 5.5556,
      "mean_token_accuracy": 0.16022806912660598,
      "num_tokens": 20147201.0,
      "step": 11610
    },
    {
      "entropy": 5.5744353294372555,
      "epoch": 0.9037152304999028,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004923514727483807,
      "loss": 5.3958,
      "mean_token_accuracy": 0.16239219754934311,
      "num_tokens": 20156744.0,
      "step": 11615
    },
    {
      "entropy": 5.643081378936768,
      "epoch": 0.9041042598716204,
      "grad_norm": 1.5,
      "learning_rate": 0.000492344306131981,
      "loss": 5.4098,
      "mean_token_accuracy": 0.168283312022686,
      "num_tokens": 20165081.0,
      "step": 11620
    },
    {
      "entropy": 5.673436975479126,
      "epoch": 0.9044932892433378,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004923371362177272,
      "loss": 5.4099,
      "mean_token_accuracy": 0.16650529354810714,
      "num_tokens": 20172893.0,
      "step": 11625
    },
    {
      "entropy": 5.595490980148315,
      "epoch": 0.9048823186150554,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004923299630057284,
      "loss": 5.4815,
      "mean_token_accuracy": 0.16391136050224303,
      "num_tokens": 20181351.0,
      "step": 11630
    },
    {
      "entropy": 5.583002185821533,
      "epoch": 0.905271347986773,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004923227864960934,
      "loss": 5.4898,
      "mean_token_accuracy": 0.15725645869970323,
      "num_tokens": 20189650.0,
      "step": 11635
    },
    {
      "entropy": 5.686719799041748,
      "epoch": 0.9056603773584906,
      "grad_norm": 1.796875,
      "learning_rate": 0.000492315606688931,
      "loss": 5.4394,
      "mean_token_accuracy": 0.15606636852025985,
      "num_tokens": 20198184.0,
      "step": 11640
    },
    {
      "entropy": 5.691554021835327,
      "epoch": 0.9060494067302082,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004923084235843501,
      "loss": 5.3587,
      "mean_token_accuracy": 0.172410349547863,
      "num_tokens": 20206357.0,
      "step": 11645
    },
    {
      "entropy": 5.539533615112305,
      "epoch": 0.9064384361019256,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004923012371824598,
      "loss": 5.4629,
      "mean_token_accuracy": 0.17097723335027695,
      "num_tokens": 20214384.0,
      "step": 11650
    },
    {
      "entropy": 5.616372919082641,
      "epoch": 0.9068274654736432,
      "grad_norm": 1.5546875,
      "learning_rate": 0.000492294047483369,
      "loss": 5.4746,
      "mean_token_accuracy": 0.16025118231773378,
      "num_tokens": 20224103.0,
      "step": 11655
    },
    {
      "entropy": 5.634894609451294,
      "epoch": 0.9072164948453608,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004922868544871869,
      "loss": 5.4794,
      "mean_token_accuracy": 0.165157687664032,
      "num_tokens": 20233414.0,
      "step": 11660
    },
    {
      "entropy": 5.620892858505249,
      "epoch": 0.9076055242170784,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004922796581940227,
      "loss": 5.3324,
      "mean_token_accuracy": 0.17546915560960769,
      "num_tokens": 20240988.0,
      "step": 11665
    },
    {
      "entropy": 5.528362703323364,
      "epoch": 0.907994553588796,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004922724586039855,
      "loss": 5.3643,
      "mean_token_accuracy": 0.1678197294473648,
      "num_tokens": 20250066.0,
      "step": 11670
    },
    {
      "entropy": 5.513184261322022,
      "epoch": 0.9083835829605135,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004922652557171846,
      "loss": 5.3023,
      "mean_token_accuracy": 0.17026133388280867,
      "num_tokens": 20258435.0,
      "step": 11675
    },
    {
      "entropy": 5.593416833877564,
      "epoch": 0.9087726123322311,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004922580495337292,
      "loss": 5.4836,
      "mean_token_accuracy": 0.1679489091038704,
      "num_tokens": 20267345.0,
      "step": 11680
    },
    {
      "entropy": 5.65469651222229,
      "epoch": 0.9091616417039486,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004922508400537288,
      "loss": 5.5081,
      "mean_token_accuracy": 0.16376658529043198,
      "num_tokens": 20275644.0,
      "step": 11685
    },
    {
      "entropy": 5.640007209777832,
      "epoch": 0.9095506710756662,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004922436272772926,
      "loss": 5.476,
      "mean_token_accuracy": 0.16782302558422088,
      "num_tokens": 20284459.0,
      "step": 11690
    },
    {
      "entropy": 5.6208737850189205,
      "epoch": 0.9099397004473838,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004922364112045301,
      "loss": 5.3894,
      "mean_token_accuracy": 0.1607832819223404,
      "num_tokens": 20293088.0,
      "step": 11695
    },
    {
      "entropy": 5.504998254776001,
      "epoch": 0.9103287298191013,
      "grad_norm": 2.484375,
      "learning_rate": 0.000492229191835551,
      "loss": 5.4,
      "mean_token_accuracy": 0.16904088854789734,
      "num_tokens": 20301811.0,
      "step": 11700
    },
    {
      "entropy": 5.640305519104004,
      "epoch": 0.9107177591908189,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004922219691704645,
      "loss": 5.5325,
      "mean_token_accuracy": 0.15783090144395828,
      "num_tokens": 20310365.0,
      "step": 11705
    },
    {
      "entropy": 5.680600500106811,
      "epoch": 0.9111067885625365,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004922147432093805,
      "loss": 5.4597,
      "mean_token_accuracy": 0.1692796379327774,
      "num_tokens": 20318891.0,
      "step": 11710
    },
    {
      "entropy": 5.569218397140503,
      "epoch": 0.911495817934254,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004922075139524083,
      "loss": 5.3474,
      "mean_token_accuracy": 0.17611400932073593,
      "num_tokens": 20327313.0,
      "step": 11715
    },
    {
      "entropy": 5.541305351257324,
      "epoch": 0.9118848473059716,
      "grad_norm": 1.984375,
      "learning_rate": 0.000492200281399658,
      "loss": 5.3301,
      "mean_token_accuracy": 0.17205900996923446,
      "num_tokens": 20335511.0,
      "step": 11720
    },
    {
      "entropy": 5.546508932113648,
      "epoch": 0.9122738766776891,
      "grad_norm": 1.75,
      "learning_rate": 0.000492193045551239,
      "loss": 5.3346,
      "mean_token_accuracy": 0.17597660124301912,
      "num_tokens": 20343588.0,
      "step": 11725
    },
    {
      "entropy": 5.595340919494629,
      "epoch": 0.9126629060494067,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004921858064072612,
      "loss": 5.4424,
      "mean_token_accuracy": 0.16616809666156768,
      "num_tokens": 20352793.0,
      "step": 11730
    },
    {
      "entropy": 5.558742332458496,
      "epoch": 0.9130519354211243,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004921785639678347,
      "loss": 5.4892,
      "mean_token_accuracy": 0.16771798133850097,
      "num_tokens": 20362480.0,
      "step": 11735
    },
    {
      "entropy": 5.562952852249145,
      "epoch": 0.9134409647928419,
      "grad_norm": 2.03125,
      "learning_rate": 0.000492171318233069,
      "loss": 5.363,
      "mean_token_accuracy": 0.17176808118820192,
      "num_tokens": 20371192.0,
      "step": 11740
    },
    {
      "entropy": 5.651773405075073,
      "epoch": 0.9138299941645595,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004921640692030742,
      "loss": 5.4804,
      "mean_token_accuracy": 0.16320589035749436,
      "num_tokens": 20379322.0,
      "step": 11745
    },
    {
      "entropy": 5.657601451873779,
      "epoch": 0.914219023536277,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004921568168779603,
      "loss": 5.4271,
      "mean_token_accuracy": 0.1589912310242653,
      "num_tokens": 20387976.0,
      "step": 11750
    },
    {
      "entropy": 5.621883726119995,
      "epoch": 0.9146080529079945,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004921495612578374,
      "loss": 5.5132,
      "mean_token_accuracy": 0.1600305199623108,
      "num_tokens": 20397398.0,
      "step": 11755
    },
    {
      "entropy": 5.622820043563843,
      "epoch": 0.9149970822797121,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004921423023428156,
      "loss": 5.424,
      "mean_token_accuracy": 0.16491100639104844,
      "num_tokens": 20406503.0,
      "step": 11760
    },
    {
      "entropy": 5.540057563781739,
      "epoch": 0.9153861116514297,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004921350401330049,
      "loss": 5.3406,
      "mean_token_accuracy": 0.17000339776277543,
      "num_tokens": 20415117.0,
      "step": 11765
    },
    {
      "entropy": 5.543955755233765,
      "epoch": 0.9157751410231473,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004921277746285155,
      "loss": 5.4153,
      "mean_token_accuracy": 0.16678078174591066,
      "num_tokens": 20423519.0,
      "step": 11770
    },
    {
      "entropy": 5.573290872573852,
      "epoch": 0.9161641703948649,
      "grad_norm": 1.5,
      "learning_rate": 0.0004921205058294579,
      "loss": 5.3659,
      "mean_token_accuracy": 0.16791573017835618,
      "num_tokens": 20431831.0,
      "step": 11775
    },
    {
      "entropy": 5.66247525215149,
      "epoch": 0.9165531997665823,
      "grad_norm": 1.609375,
      "learning_rate": 0.000492113233735942,
      "loss": 5.5059,
      "mean_token_accuracy": 0.15884622186422348,
      "num_tokens": 20440927.0,
      "step": 11780
    },
    {
      "entropy": 5.5654441833496096,
      "epoch": 0.9169422291382999,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004921059583480785,
      "loss": 5.3926,
      "mean_token_accuracy": 0.16572857946157454,
      "num_tokens": 20448974.0,
      "step": 11785
    },
    {
      "entropy": 5.515251779556275,
      "epoch": 0.9173312585100175,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004920986796659775,
      "loss": 5.3082,
      "mean_token_accuracy": 0.17805312275886537,
      "num_tokens": 20457955.0,
      "step": 11790
    },
    {
      "entropy": 5.5523580551147464,
      "epoch": 0.9177202878817351,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004920913976897498,
      "loss": 5.3972,
      "mean_token_accuracy": 0.16891333162784578,
      "num_tokens": 20465942.0,
      "step": 11795
    },
    {
      "entropy": 5.5653056621551515,
      "epoch": 0.9181093172534527,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004920841124195055,
      "loss": 5.4712,
      "mean_token_accuracy": 0.1669631913304329,
      "num_tokens": 20474726.0,
      "step": 11800
    },
    {
      "entropy": 5.561779737472534,
      "epoch": 0.9184983466251702,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004920768238553554,
      "loss": 5.416,
      "mean_token_accuracy": 0.16652749478816986,
      "num_tokens": 20483442.0,
      "step": 11805
    },
    {
      "entropy": 5.547149133682251,
      "epoch": 0.9188873759968877,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004920695319974099,
      "loss": 5.3741,
      "mean_token_accuracy": 0.17090764045715331,
      "num_tokens": 20492950.0,
      "step": 11810
    },
    {
      "entropy": 5.61188268661499,
      "epoch": 0.9192764053686053,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004920622368457798,
      "loss": 5.4666,
      "mean_token_accuracy": 0.1626928448677063,
      "num_tokens": 20502035.0,
      "step": 11815
    },
    {
      "entropy": 5.617907381057739,
      "epoch": 0.9196654347403229,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004920549384005759,
      "loss": 5.3446,
      "mean_token_accuracy": 0.17139900624752044,
      "num_tokens": 20510140.0,
      "step": 11820
    },
    {
      "entropy": 5.583600568771362,
      "epoch": 0.9200544641120405,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004920476366619086,
      "loss": 5.4569,
      "mean_token_accuracy": 0.1582721382379532,
      "num_tokens": 20519391.0,
      "step": 11825
    },
    {
      "entropy": 5.664299488067627,
      "epoch": 0.920443493483758,
      "grad_norm": 1.9453125,
      "learning_rate": 0.000492040331629889,
      "loss": 5.4511,
      "mean_token_accuracy": 0.16522680073976517,
      "num_tokens": 20527635.0,
      "step": 11830
    },
    {
      "entropy": 5.6060832977294925,
      "epoch": 0.9208325228554756,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004920330233046277,
      "loss": 5.4075,
      "mean_token_accuracy": 0.16618909686803818,
      "num_tokens": 20536455.0,
      "step": 11835
    },
    {
      "entropy": 5.636282634735108,
      "epoch": 0.9212215522271932,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004920257116862359,
      "loss": 5.546,
      "mean_token_accuracy": 0.1542813390493393,
      "num_tokens": 20545887.0,
      "step": 11840
    },
    {
      "entropy": 5.57595739364624,
      "epoch": 0.9216105815989107,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004920183967748241,
      "loss": 5.381,
      "mean_token_accuracy": 0.16687051951885223,
      "num_tokens": 20554415.0,
      "step": 11845
    },
    {
      "entropy": 5.616524982452392,
      "epoch": 0.9219996109706283,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004920110785705037,
      "loss": 5.4938,
      "mean_token_accuracy": 0.16545359045267105,
      "num_tokens": 20563081.0,
      "step": 11850
    },
    {
      "entropy": 5.513118267059326,
      "epoch": 0.9223886403423458,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004920037570733857,
      "loss": 5.3445,
      "mean_token_accuracy": 0.17197140008211137,
      "num_tokens": 20571800.0,
      "step": 11855
    },
    {
      "entropy": 5.633137226104736,
      "epoch": 0.9227776697140634,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004919964322835809,
      "loss": 5.4673,
      "mean_token_accuracy": 0.1649052232503891,
      "num_tokens": 20580481.0,
      "step": 11860
    },
    {
      "entropy": 5.6207211971282955,
      "epoch": 0.923166699085781,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004919891042012006,
      "loss": 5.3584,
      "mean_token_accuracy": 0.1680561199784279,
      "num_tokens": 20589575.0,
      "step": 11865
    },
    {
      "entropy": 5.512158679962158,
      "epoch": 0.9235557284574986,
      "grad_norm": 1.5,
      "learning_rate": 0.0004919817728263562,
      "loss": 5.4403,
      "mean_token_accuracy": 0.1664802387356758,
      "num_tokens": 20598277.0,
      "step": 11870
    },
    {
      "entropy": 5.693507719039917,
      "epoch": 0.9239447578292161,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004919744381591586,
      "loss": 5.5295,
      "mean_token_accuracy": 0.16748165041208268,
      "num_tokens": 20607013.0,
      "step": 11875
    },
    {
      "entropy": 5.640137624740601,
      "epoch": 0.9243337872009336,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004919671001997193,
      "loss": 5.4994,
      "mean_token_accuracy": 0.16177828460931779,
      "num_tokens": 20615605.0,
      "step": 11880
    },
    {
      "entropy": 5.675401878356934,
      "epoch": 0.9247228165726512,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004919597589481497,
      "loss": 5.4499,
      "mean_token_accuracy": 0.15949563086032867,
      "num_tokens": 20624275.0,
      "step": 11885
    },
    {
      "entropy": 5.6085638999938965,
      "epoch": 0.9251118459443688,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000491952414404561,
      "loss": 5.3888,
      "mean_token_accuracy": 0.1679642006754875,
      "num_tokens": 20632681.0,
      "step": 11890
    },
    {
      "entropy": 5.659155607223511,
      "epoch": 0.9255008753160864,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004919450665690646,
      "loss": 5.4556,
      "mean_token_accuracy": 0.16983082741498948,
      "num_tokens": 20641432.0,
      "step": 11895
    },
    {
      "entropy": 5.585770940780639,
      "epoch": 0.925889904687804,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004919377154417724,
      "loss": 5.4649,
      "mean_token_accuracy": 0.16276452094316482,
      "num_tokens": 20650120.0,
      "step": 11900
    },
    {
      "entropy": 5.545875215530396,
      "epoch": 0.9262789340595214,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004919303610227954,
      "loss": 5.359,
      "mean_token_accuracy": 0.16669993549585344,
      "num_tokens": 20658188.0,
      "step": 11905
    },
    {
      "entropy": 5.514937448501587,
      "epoch": 0.926667963431239,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004919230033122457,
      "loss": 5.2875,
      "mean_token_accuracy": 0.1804579585790634,
      "num_tokens": 20666157.0,
      "step": 11910
    },
    {
      "entropy": 5.547408819198608,
      "epoch": 0.9270569928029566,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004919156423102345,
      "loss": 5.4928,
      "mean_token_accuracy": 0.16427554711699485,
      "num_tokens": 20674995.0,
      "step": 11915
    },
    {
      "entropy": 5.641458177566529,
      "epoch": 0.9274460221746742,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004919082780168736,
      "loss": 5.4405,
      "mean_token_accuracy": 0.16236664950847626,
      "num_tokens": 20683786.0,
      "step": 11920
    },
    {
      "entropy": 5.585165119171142,
      "epoch": 0.9278350515463918,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004919009104322751,
      "loss": 5.4596,
      "mean_token_accuracy": 0.16301724314689636,
      "num_tokens": 20692582.0,
      "step": 11925
    },
    {
      "entropy": 5.576903438568115,
      "epoch": 0.9282240809181093,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004918935395565503,
      "loss": 5.4632,
      "mean_token_accuracy": 0.16221937090158461,
      "num_tokens": 20700892.0,
      "step": 11930
    },
    {
      "entropy": 5.551387596130371,
      "epoch": 0.9286131102898268,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004918861653898113,
      "loss": 5.3381,
      "mean_token_accuracy": 0.17207940220832824,
      "num_tokens": 20709521.0,
      "step": 11935
    },
    {
      "entropy": 5.562303304672241,
      "epoch": 0.9290021396615444,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004918787879321701,
      "loss": 5.4336,
      "mean_token_accuracy": 0.16194119304418564,
      "num_tokens": 20717614.0,
      "step": 11940
    },
    {
      "entropy": 5.563432693481445,
      "epoch": 0.929391169033262,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004918714071837384,
      "loss": 5.3715,
      "mean_token_accuracy": 0.1687655344605446,
      "num_tokens": 20726624.0,
      "step": 11945
    },
    {
      "entropy": 5.61021842956543,
      "epoch": 0.9297801984049796,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004918640231446282,
      "loss": 5.4828,
      "mean_token_accuracy": 0.16277274787425994,
      "num_tokens": 20734738.0,
      "step": 11950
    },
    {
      "entropy": 5.605218696594238,
      "epoch": 0.9301692277766972,
      "grad_norm": 1.5,
      "learning_rate": 0.0004918566358149517,
      "loss": 5.3988,
      "mean_token_accuracy": 0.16683424711227418,
      "num_tokens": 20742825.0,
      "step": 11955
    },
    {
      "entropy": 5.578117942810058,
      "epoch": 0.9305582571484147,
      "grad_norm": 1.6015625,
      "learning_rate": 0.000491849245194821,
      "loss": 5.3779,
      "mean_token_accuracy": 0.1739774078130722,
      "num_tokens": 20752132.0,
      "step": 11960
    },
    {
      "entropy": 5.637444067001343,
      "epoch": 0.9309472865201323,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004918418512843479,
      "loss": 5.5075,
      "mean_token_accuracy": 0.1613238826394081,
      "num_tokens": 20761838.0,
      "step": 11965
    },
    {
      "entropy": 5.56450719833374,
      "epoch": 0.9313363158918498,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004918344540836451,
      "loss": 5.3957,
      "mean_token_accuracy": 0.16996693909168242,
      "num_tokens": 20770405.0,
      "step": 11970
    },
    {
      "entropy": 5.715441370010376,
      "epoch": 0.9317253452635674,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004918270535928244,
      "loss": 5.5864,
      "mean_token_accuracy": 0.15619297772645951,
      "num_tokens": 20780130.0,
      "step": 11975
    },
    {
      "entropy": 5.594107055664063,
      "epoch": 0.932114374635285,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004918196498119984,
      "loss": 5.4561,
      "mean_token_accuracy": 0.17283804416656495,
      "num_tokens": 20788852.0,
      "step": 11980
    },
    {
      "entropy": 5.553250312805176,
      "epoch": 0.9325034040070025,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004918122427412793,
      "loss": 5.3715,
      "mean_token_accuracy": 0.1689470276236534,
      "num_tokens": 20796944.0,
      "step": 11985
    },
    {
      "entropy": 5.54477891921997,
      "epoch": 0.9328924333787201,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004918048323807795,
      "loss": 5.5106,
      "mean_token_accuracy": 0.1635565549135208,
      "num_tokens": 20806118.0,
      "step": 11990
    },
    {
      "entropy": 5.649374771118164,
      "epoch": 0.9332814627504377,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004917974187306114,
      "loss": 5.3038,
      "mean_token_accuracy": 0.17418421804904938,
      "num_tokens": 20814643.0,
      "step": 11995
    },
    {
      "entropy": 5.6086163997650145,
      "epoch": 0.9336704921221552,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004917900017908875,
      "loss": 5.3672,
      "mean_token_accuracy": 0.1686716139316559,
      "num_tokens": 20823315.0,
      "step": 12000
    },
    {
      "epoch": 0.9336704921221552,
      "eval_entropy": 5.371572630719799,
      "eval_loss": 5.466134548187256,
      "eval_mean_token_accuracy": 0.1723012616067293,
      "eval_num_tokens": 20823315.0,
      "eval_runtime": 28.5605,
      "eval_samples_per_second": 1455.088,
      "eval_steps_per_second": 181.895,
      "step": 12000
    },
    {
      "entropy": 5.507457256317139,
      "epoch": 0.9340595214938728,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004917825815617205,
      "loss": 5.3884,
      "mean_token_accuracy": 0.17000421285629272,
      "num_tokens": 20832811.0,
      "step": 12005
    },
    {
      "entropy": 5.613448762893677,
      "epoch": 0.9344485508655903,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004917751580432229,
      "loss": 5.4398,
      "mean_token_accuracy": 0.16664963215589523,
      "num_tokens": 20841824.0,
      "step": 12010
    },
    {
      "entropy": 5.599476003646851,
      "epoch": 0.9348375802373079,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004917677312355072,
      "loss": 5.3926,
      "mean_token_accuracy": 0.17509149014949799,
      "num_tokens": 20850220.0,
      "step": 12015
    },
    {
      "entropy": 5.488557815551758,
      "epoch": 0.9352266096090255,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004917603011386863,
      "loss": 5.3347,
      "mean_token_accuracy": 0.17387936115264893,
      "num_tokens": 20859193.0,
      "step": 12020
    },
    {
      "entropy": 5.54458532333374,
      "epoch": 0.9356156389807431,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004917528677528727,
      "loss": 5.3549,
      "mean_token_accuracy": 0.16718203574419022,
      "num_tokens": 20867610.0,
      "step": 12025
    },
    {
      "entropy": 5.615087413787842,
      "epoch": 0.9360046683524607,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004917454310781795,
      "loss": 5.454,
      "mean_token_accuracy": 0.16309396028518677,
      "num_tokens": 20876991.0,
      "step": 12030
    },
    {
      "entropy": 5.541305351257324,
      "epoch": 0.9363936977241781,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004917379911147193,
      "loss": 5.3173,
      "mean_token_accuracy": 0.16801157742738723,
      "num_tokens": 20885794.0,
      "step": 12035
    },
    {
      "entropy": 5.619901990890503,
      "epoch": 0.9367827270958957,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004917305478626049,
      "loss": 5.5104,
      "mean_token_accuracy": 0.16595079600811005,
      "num_tokens": 20893487.0,
      "step": 12040
    },
    {
      "entropy": 5.6008594036102295,
      "epoch": 0.9371717564676133,
      "grad_norm": 1.625,
      "learning_rate": 0.0004917231013219495,
      "loss": 5.4598,
      "mean_token_accuracy": 0.1607996016740799,
      "num_tokens": 20902587.0,
      "step": 12045
    },
    {
      "entropy": 5.596264743804932,
      "epoch": 0.9375607858393309,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000491715651492866,
      "loss": 5.3903,
      "mean_token_accuracy": 0.17339229583740234,
      "num_tokens": 20911062.0,
      "step": 12050
    },
    {
      "entropy": 5.620618295669556,
      "epoch": 0.9379498152110485,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004917081983754675,
      "loss": 5.4123,
      "mean_token_accuracy": 0.16720927357673646,
      "num_tokens": 20919369.0,
      "step": 12055
    },
    {
      "entropy": 5.613325119018555,
      "epoch": 0.938338844582766,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004917007419698669,
      "loss": 5.4666,
      "mean_token_accuracy": 0.1663628563284874,
      "num_tokens": 20927514.0,
      "step": 12060
    },
    {
      "entropy": 5.677690505981445,
      "epoch": 0.9387278739544835,
      "grad_norm": 5.875,
      "learning_rate": 0.0004916932822761776,
      "loss": 5.5023,
      "mean_token_accuracy": 0.16759129613637924,
      "num_tokens": 20935750.0,
      "step": 12065
    },
    {
      "entropy": 5.5808741569519045,
      "epoch": 0.9391169033262011,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004916858192945126,
      "loss": 5.3986,
      "mean_token_accuracy": 0.16836540699005126,
      "num_tokens": 20944089.0,
      "step": 12070
    },
    {
      "entropy": 5.581806325912476,
      "epoch": 0.9395059326979187,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004916783530249852,
      "loss": 5.5052,
      "mean_token_accuracy": 0.1604120537638664,
      "num_tokens": 20952704.0,
      "step": 12075
    },
    {
      "entropy": 5.61510329246521,
      "epoch": 0.9398949620696363,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004916708834677086,
      "loss": 5.3592,
      "mean_token_accuracy": 0.16789056211709977,
      "num_tokens": 20960741.0,
      "step": 12080
    },
    {
      "entropy": 5.604568719863892,
      "epoch": 0.9402839914413538,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004916634106227962,
      "loss": 5.3752,
      "mean_token_accuracy": 0.17388292998075486,
      "num_tokens": 20968778.0,
      "step": 12085
    },
    {
      "entropy": 5.603709936141968,
      "epoch": 0.9406730208130714,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004916559344903614,
      "loss": 5.5405,
      "mean_token_accuracy": 0.1607699140906334,
      "num_tokens": 20977993.0,
      "step": 12090
    },
    {
      "entropy": 5.612804269790649,
      "epoch": 0.941062050184789,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004916484550705176,
      "loss": 5.3795,
      "mean_token_accuracy": 0.17117612361907958,
      "num_tokens": 20986974.0,
      "step": 12095
    },
    {
      "entropy": 5.604335594177246,
      "epoch": 0.9414510795565065,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004916409723633785,
      "loss": 5.4724,
      "mean_token_accuracy": 0.16735209971666337,
      "num_tokens": 20995224.0,
      "step": 12100
    },
    {
      "entropy": 5.6226420402526855,
      "epoch": 0.9418401089282241,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004916334863690573,
      "loss": 5.4697,
      "mean_token_accuracy": 0.16305154114961623,
      "num_tokens": 21003270.0,
      "step": 12105
    },
    {
      "entropy": 5.595528697967529,
      "epoch": 0.9422291382999416,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004916259970876678,
      "loss": 5.4472,
      "mean_token_accuracy": 0.16190327256917952,
      "num_tokens": 21012665.0,
      "step": 12110
    },
    {
      "entropy": 5.630600214004517,
      "epoch": 0.9426181676716592,
      "grad_norm": 1.875,
      "learning_rate": 0.0004916185045193237,
      "loss": 5.4708,
      "mean_token_accuracy": 0.1663910672068596,
      "num_tokens": 21021043.0,
      "step": 12115
    },
    {
      "entropy": 5.621525621414184,
      "epoch": 0.9430071970433768,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004916110086641384,
      "loss": 5.3871,
      "mean_token_accuracy": 0.16917871534824372,
      "num_tokens": 21029525.0,
      "step": 12120
    },
    {
      "entropy": 5.529054307937622,
      "epoch": 0.9433962264150944,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004916035095222259,
      "loss": 5.3924,
      "mean_token_accuracy": 0.16976384520530702,
      "num_tokens": 21037687.0,
      "step": 12125
    },
    {
      "entropy": 5.611918115615845,
      "epoch": 0.9437852557868119,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004915960070936999,
      "loss": 5.4768,
      "mean_token_accuracy": 0.16243709176778792,
      "num_tokens": 21046095.0,
      "step": 12130
    },
    {
      "entropy": 5.659605216979981,
      "epoch": 0.9441742851585294,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004915885013786742,
      "loss": 5.433,
      "mean_token_accuracy": 0.16504660099744797,
      "num_tokens": 21054565.0,
      "step": 12135
    },
    {
      "entropy": 5.49802794456482,
      "epoch": 0.944563314530247,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004915809923772628,
      "loss": 5.3443,
      "mean_token_accuracy": 0.17390815615653993,
      "num_tokens": 21062942.0,
      "step": 12140
    },
    {
      "entropy": 5.5965657234191895,
      "epoch": 0.9449523439019646,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004915734800895796,
      "loss": 5.396,
      "mean_token_accuracy": 0.1641937553882599,
      "num_tokens": 21070642.0,
      "step": 12145
    },
    {
      "entropy": 5.623910331726075,
      "epoch": 0.9453413732736822,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004915659645157383,
      "loss": 5.3635,
      "mean_token_accuracy": 0.16894838064908982,
      "num_tokens": 21079007.0,
      "step": 12150
    },
    {
      "entropy": 5.605703544616699,
      "epoch": 0.9457304026453998,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004915584456558535,
      "loss": 5.4377,
      "mean_token_accuracy": 0.16110673993825914,
      "num_tokens": 21087672.0,
      "step": 12155
    },
    {
      "entropy": 5.686621379852295,
      "epoch": 0.9461194320171173,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004915509235100388,
      "loss": 5.376,
      "mean_token_accuracy": 0.16856016516685485,
      "num_tokens": 21095954.0,
      "step": 12160
    },
    {
      "entropy": 5.604067611694336,
      "epoch": 0.9465084613888348,
      "grad_norm": 1.375,
      "learning_rate": 0.0004915433980784086,
      "loss": 5.3702,
      "mean_token_accuracy": 0.16448871940374374,
      "num_tokens": 21104439.0,
      "step": 12165
    },
    {
      "entropy": 5.524645709991455,
      "epoch": 0.9468974907605524,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004915358693610769,
      "loss": 5.4059,
      "mean_token_accuracy": 0.15946614742279053,
      "num_tokens": 21112933.0,
      "step": 12170
    },
    {
      "entropy": 5.538733863830567,
      "epoch": 0.94728652013227,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004915283373581581,
      "loss": 5.402,
      "mean_token_accuracy": 0.1738529697060585,
      "num_tokens": 21122085.0,
      "step": 12175
    },
    {
      "entropy": 5.634131717681885,
      "epoch": 0.9476755495039876,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004915208020697664,
      "loss": 5.4597,
      "mean_token_accuracy": 0.16466081142425537,
      "num_tokens": 21130994.0,
      "step": 12180
    },
    {
      "entropy": 5.645436811447143,
      "epoch": 0.9480645788757052,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004915132634960162,
      "loss": 5.3834,
      "mean_token_accuracy": 0.17394153475761415,
      "num_tokens": 21139744.0,
      "step": 12185
    },
    {
      "entropy": 5.595538473129272,
      "epoch": 0.9484536082474226,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004915057216370218,
      "loss": 5.3965,
      "mean_token_accuracy": 0.1731519356369972,
      "num_tokens": 21148333.0,
      "step": 12190
    },
    {
      "entropy": 5.519620656967163,
      "epoch": 0.9488426376191402,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004914981764928977,
      "loss": 5.326,
      "mean_token_accuracy": 0.17691762298345565,
      "num_tokens": 21157450.0,
      "step": 12195
    },
    {
      "entropy": 5.648521041870117,
      "epoch": 0.9492316669908578,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004914906280637584,
      "loss": 5.5001,
      "mean_token_accuracy": 0.16335393339395524,
      "num_tokens": 21166410.0,
      "step": 12200
    },
    {
      "entropy": 5.604200220108032,
      "epoch": 0.9496206963625754,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004914830763497183,
      "loss": 5.435,
      "mean_token_accuracy": 0.16810693889856337,
      "num_tokens": 21175104.0,
      "step": 12205
    },
    {
      "entropy": 5.597310638427734,
      "epoch": 0.950009725734293,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004914755213508922,
      "loss": 5.3788,
      "mean_token_accuracy": 0.16749434620141984,
      "num_tokens": 21182597.0,
      "step": 12210
    },
    {
      "entropy": 5.5915443897247314,
      "epoch": 0.9503987551060105,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004914679630673945,
      "loss": 5.4896,
      "mean_token_accuracy": 0.16872426271438598,
      "num_tokens": 21192399.0,
      "step": 12215
    },
    {
      "entropy": 5.56868987083435,
      "epoch": 0.950787784477728,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004914604014993401,
      "loss": 5.3472,
      "mean_token_accuracy": 0.17401702851057052,
      "num_tokens": 21201275.0,
      "step": 12220
    },
    {
      "entropy": 5.5518927574157715,
      "epoch": 0.9511768138494456,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004914528366468436,
      "loss": 5.4565,
      "mean_token_accuracy": 0.16536201536655426,
      "num_tokens": 21210491.0,
      "step": 12225
    },
    {
      "entropy": 5.564200925827026,
      "epoch": 0.9515658432211632,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004914452685100199,
      "loss": 5.4495,
      "mean_token_accuracy": 0.1683812901377678,
      "num_tokens": 21218899.0,
      "step": 12230
    },
    {
      "entropy": 5.501306676864624,
      "epoch": 0.9519548725928808,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004914376970889836,
      "loss": 5.3524,
      "mean_token_accuracy": 0.16297265887260437,
      "num_tokens": 21227640.0,
      "step": 12235
    },
    {
      "entropy": 5.575056791305542,
      "epoch": 0.9523439019645983,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004914301223838497,
      "loss": 5.3197,
      "mean_token_accuracy": 0.16264411211013793,
      "num_tokens": 21236422.0,
      "step": 12240
    },
    {
      "entropy": 5.630202198028565,
      "epoch": 0.9527329313363159,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004914225443947334,
      "loss": 5.4947,
      "mean_token_accuracy": 0.16732407957315446,
      "num_tokens": 21245820.0,
      "step": 12245
    },
    {
      "entropy": 5.647228384017945,
      "epoch": 0.9531219607080335,
      "grad_norm": 3.90625,
      "learning_rate": 0.0004914149631217494,
      "loss": 5.4399,
      "mean_token_accuracy": 0.16611770391464234,
      "num_tokens": 21255570.0,
      "step": 12250
    },
    {
      "entropy": 5.662381029129028,
      "epoch": 0.953510990079751,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004914073785650127,
      "loss": 5.5614,
      "mean_token_accuracy": 0.16004973351955415,
      "num_tokens": 21263712.0,
      "step": 12255
    },
    {
      "entropy": 5.6631018161773685,
      "epoch": 0.9539000194514686,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004913997907246385,
      "loss": 5.436,
      "mean_token_accuracy": 0.17167474776506425,
      "num_tokens": 21272309.0,
      "step": 12260
    },
    {
      "entropy": 5.506390190124511,
      "epoch": 0.9542890488231861,
      "grad_norm": 1.453125,
      "learning_rate": 0.000491392199600742,
      "loss": 5.359,
      "mean_token_accuracy": 0.1665118530392647,
      "num_tokens": 21281711.0,
      "step": 12265
    },
    {
      "entropy": 5.51539626121521,
      "epoch": 0.9546780781949037,
      "grad_norm": 3.734375,
      "learning_rate": 0.0004913846051934382,
      "loss": 5.3196,
      "mean_token_accuracy": 0.17208221256732942,
      "num_tokens": 21289524.0,
      "step": 12270
    },
    {
      "entropy": 5.513137531280518,
      "epoch": 0.9550671075666213,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004913770075028425,
      "loss": 5.2626,
      "mean_token_accuracy": 0.17546534538269043,
      "num_tokens": 21298225.0,
      "step": 12275
    },
    {
      "entropy": 5.542500305175781,
      "epoch": 0.9554561369383389,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004913694065290701,
      "loss": 5.2958,
      "mean_token_accuracy": 0.178589104115963,
      "num_tokens": 21306819.0,
      "step": 12280
    },
    {
      "entropy": 5.576265525817871,
      "epoch": 0.9558451663100564,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004913618022722363,
      "loss": 5.4454,
      "mean_token_accuracy": 0.16563066840171814,
      "num_tokens": 21315517.0,
      "step": 12285
    },
    {
      "entropy": 5.614781618118286,
      "epoch": 0.9562341956817739,
      "grad_norm": 1.5,
      "learning_rate": 0.0004913541947324566,
      "loss": 5.3488,
      "mean_token_accuracy": 0.16864724308252335,
      "num_tokens": 21324144.0,
      "step": 12290
    },
    {
      "entropy": 5.605503416061401,
      "epoch": 0.9566232250534915,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004913465839098463,
      "loss": 5.4197,
      "mean_token_accuracy": 0.17107993364334106,
      "num_tokens": 21333329.0,
      "step": 12295
    },
    {
      "entropy": 5.522213315963745,
      "epoch": 0.9570122544252091,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004913389698045209,
      "loss": 5.3781,
      "mean_token_accuracy": 0.1652178406715393,
      "num_tokens": 21341955.0,
      "step": 12300
    },
    {
      "entropy": 5.586077404022217,
      "epoch": 0.9574012837969267,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000491331352416596,
      "loss": 5.3473,
      "mean_token_accuracy": 0.16738866716623307,
      "num_tokens": 21350593.0,
      "step": 12305
    },
    {
      "entropy": 5.535979175567627,
      "epoch": 0.9577903131686443,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004913237317461872,
      "loss": 5.4292,
      "mean_token_accuracy": 0.16674810349941255,
      "num_tokens": 21359007.0,
      "step": 12310
    },
    {
      "entropy": 5.549071025848389,
      "epoch": 0.9581793425403617,
      "grad_norm": 1.359375,
      "learning_rate": 0.00049131610779341,
      "loss": 5.4487,
      "mean_token_accuracy": 0.16577992141246795,
      "num_tokens": 21367645.0,
      "step": 12315
    },
    {
      "entropy": 5.510221481323242,
      "epoch": 0.9585683719120793,
      "grad_norm": 1.375,
      "learning_rate": 0.0004913084805583803,
      "loss": 5.263,
      "mean_token_accuracy": 0.17191761881113052,
      "num_tokens": 21376040.0,
      "step": 12320
    },
    {
      "entropy": 5.58285551071167,
      "epoch": 0.9589574012837969,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004913008500412136,
      "loss": 5.4068,
      "mean_token_accuracy": 0.17451067715883256,
      "num_tokens": 21385305.0,
      "step": 12325
    },
    {
      "entropy": 5.545878791809082,
      "epoch": 0.9593464306555145,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004912932162420259,
      "loss": 5.3043,
      "mean_token_accuracy": 0.18116566985845567,
      "num_tokens": 21393705.0,
      "step": 12330
    },
    {
      "entropy": 5.582909774780274,
      "epoch": 0.9597354600272321,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004912855791609327,
      "loss": 5.428,
      "mean_token_accuracy": 0.16888475716114043,
      "num_tokens": 21401830.0,
      "step": 12335
    },
    {
      "entropy": 5.569549655914306,
      "epoch": 0.9601244893989496,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004912779387980502,
      "loss": 5.4796,
      "mean_token_accuracy": 0.16422994881868364,
      "num_tokens": 21410398.0,
      "step": 12340
    },
    {
      "entropy": 5.64050612449646,
      "epoch": 0.9605135187706672,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004912702951534942,
      "loss": 5.4314,
      "mean_token_accuracy": 0.16885158866643907,
      "num_tokens": 21419082.0,
      "step": 12345
    },
    {
      "entropy": 5.648242807388305,
      "epoch": 0.9609025481423847,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004912626482273809,
      "loss": 5.427,
      "mean_token_accuracy": 0.1632491961121559,
      "num_tokens": 21427992.0,
      "step": 12350
    },
    {
      "entropy": 5.555550384521484,
      "epoch": 0.9612915775141023,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004912549980198258,
      "loss": 5.3935,
      "mean_token_accuracy": 0.16734340637922288,
      "num_tokens": 21436933.0,
      "step": 12355
    },
    {
      "entropy": 5.5616072654724125,
      "epoch": 0.9616806068858199,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004912473445309455,
      "loss": 5.3882,
      "mean_token_accuracy": 0.16933085173368453,
      "num_tokens": 21445640.0,
      "step": 12360
    },
    {
      "entropy": 5.698657846450805,
      "epoch": 0.9620696362575375,
      "grad_norm": 1.4375,
      "learning_rate": 0.000491239687760856,
      "loss": 5.5502,
      "mean_token_accuracy": 0.15852493792772293,
      "num_tokens": 21454511.0,
      "step": 12365
    },
    {
      "entropy": 5.60389199256897,
      "epoch": 0.962458665629255,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004912320277096732,
      "loss": 5.5412,
      "mean_token_accuracy": 0.1576765350997448,
      "num_tokens": 21463851.0,
      "step": 12370
    },
    {
      "entropy": 5.536577939987183,
      "epoch": 0.9628476950009726,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004912243643775137,
      "loss": 5.3283,
      "mean_token_accuracy": 0.17465019077062607,
      "num_tokens": 21471620.0,
      "step": 12375
    },
    {
      "entropy": 5.570523643493653,
      "epoch": 0.9632367243726901,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004912166977644936,
      "loss": 5.3446,
      "mean_token_accuracy": 0.1683275133371353,
      "num_tokens": 21479944.0,
      "step": 12380
    },
    {
      "entropy": 5.631869506835938,
      "epoch": 0.9636257537444077,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004912090278707293,
      "loss": 5.5166,
      "mean_token_accuracy": 0.15817615985870362,
      "num_tokens": 21489175.0,
      "step": 12385
    },
    {
      "entropy": 5.595966625213623,
      "epoch": 0.9640147831161253,
      "grad_norm": 1.65625,
      "learning_rate": 0.000491201354696337,
      "loss": 5.4373,
      "mean_token_accuracy": 0.17099682986736298,
      "num_tokens": 21498533.0,
      "step": 12390
    },
    {
      "entropy": 5.68024845123291,
      "epoch": 0.9644038124878428,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004911936782414334,
      "loss": 5.5038,
      "mean_token_accuracy": 0.15963890701532363,
      "num_tokens": 21506771.0,
      "step": 12395
    },
    {
      "entropy": 5.534885454177856,
      "epoch": 0.9647928418595604,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004911859985061348,
      "loss": 5.3038,
      "mean_token_accuracy": 0.1820862114429474,
      "num_tokens": 21515359.0,
      "step": 12400
    },
    {
      "entropy": 5.566308212280274,
      "epoch": 0.965181871231278,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004911783154905577,
      "loss": 5.3759,
      "mean_token_accuracy": 0.16567621976137162,
      "num_tokens": 21524313.0,
      "step": 12405
    },
    {
      "entropy": 5.529404783248902,
      "epoch": 0.9655709006029956,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004911706291948188,
      "loss": 5.3766,
      "mean_token_accuracy": 0.1746726393699646,
      "num_tokens": 21533211.0,
      "step": 12410
    },
    {
      "entropy": 5.5782609462738035,
      "epoch": 0.9659599299747131,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004911629396190348,
      "loss": 5.3304,
      "mean_token_accuracy": 0.17233569473028182,
      "num_tokens": 21541575.0,
      "step": 12415
    },
    {
      "entropy": 5.597880029678345,
      "epoch": 0.9663489593464306,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004911552467633221,
      "loss": 5.4257,
      "mean_token_accuracy": 0.1635533392429352,
      "num_tokens": 21549569.0,
      "step": 12420
    },
    {
      "entropy": 5.5263899803161625,
      "epoch": 0.9667379887181482,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004911475506277976,
      "loss": 5.3771,
      "mean_token_accuracy": 0.16541996747255325,
      "num_tokens": 21557277.0,
      "step": 12425
    },
    {
      "entropy": 5.5775562763214115,
      "epoch": 0.9671270180898658,
      "grad_norm": 1.6015625,
      "learning_rate": 0.000491139851212578,
      "loss": 5.4123,
      "mean_token_accuracy": 0.16224730163812637,
      "num_tokens": 21566590.0,
      "step": 12430
    },
    {
      "entropy": 5.653618907928466,
      "epoch": 0.9675160474615834,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004911321485177802,
      "loss": 5.479,
      "mean_token_accuracy": 0.15981924384832383,
      "num_tokens": 21575140.0,
      "step": 12435
    },
    {
      "entropy": 5.5106744289398195,
      "epoch": 0.967905076833301,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004911244425435212,
      "loss": 5.4187,
      "mean_token_accuracy": 0.16120911836624147,
      "num_tokens": 21584647.0,
      "step": 12440
    },
    {
      "entropy": 5.612622928619385,
      "epoch": 0.9682941062050184,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004911167332899176,
      "loss": 5.3749,
      "mean_token_accuracy": 0.16861445009708403,
      "num_tokens": 21593609.0,
      "step": 12445
    },
    {
      "entropy": 5.710604286193847,
      "epoch": 0.968683135576736,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004911090207570867,
      "loss": 5.4977,
      "mean_token_accuracy": 0.16604768186807634,
      "num_tokens": 21602584.0,
      "step": 12450
    },
    {
      "entropy": 5.497505331039429,
      "epoch": 0.9690721649484536,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004911013049451453,
      "loss": 5.4567,
      "mean_token_accuracy": 0.1718402773141861,
      "num_tokens": 21611109.0,
      "step": 12455
    },
    {
      "entropy": 5.647250127792359,
      "epoch": 0.9694611943201712,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004910935858542106,
      "loss": 5.4644,
      "mean_token_accuracy": 0.1647668480873108,
      "num_tokens": 21620173.0,
      "step": 12460
    },
    {
      "entropy": 5.57645320892334,
      "epoch": 0.9698502236918888,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004910858634843997,
      "loss": 5.2968,
      "mean_token_accuracy": 0.17491580843925475,
      "num_tokens": 21628318.0,
      "step": 12465
    },
    {
      "entropy": 5.589544105529785,
      "epoch": 0.9702392530636063,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004910781378358297,
      "loss": 5.5078,
      "mean_token_accuracy": 0.164412023127079,
      "num_tokens": 21637102.0,
      "step": 12470
    },
    {
      "entropy": 5.590525102615357,
      "epoch": 0.9706282824353238,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004910704089086178,
      "loss": 5.3554,
      "mean_token_accuracy": 0.17072507441043855,
      "num_tokens": 21645055.0,
      "step": 12475
    },
    {
      "entropy": 5.6012671947479244,
      "epoch": 0.9710173118070414,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004910626767028815,
      "loss": 5.4189,
      "mean_token_accuracy": 0.16050756871700286,
      "num_tokens": 21654106.0,
      "step": 12480
    },
    {
      "entropy": 5.5423140048980715,
      "epoch": 0.971406341178759,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004910549412187379,
      "loss": 5.4081,
      "mean_token_accuracy": 0.17627297341823578,
      "num_tokens": 21663081.0,
      "step": 12485
    },
    {
      "entropy": 5.586370897293091,
      "epoch": 0.9717953705504766,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004910472024563045,
      "loss": 5.3733,
      "mean_token_accuracy": 0.1717690572142601,
      "num_tokens": 21671033.0,
      "step": 12490
    },
    {
      "entropy": 5.61957197189331,
      "epoch": 0.9721843999221941,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004910394604156987,
      "loss": 5.3861,
      "mean_token_accuracy": 0.17137907147407533,
      "num_tokens": 21679167.0,
      "step": 12495
    },
    {
      "entropy": 5.644653844833374,
      "epoch": 0.9725734292939117,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004910317150970379,
      "loss": 5.5172,
      "mean_token_accuracy": 0.16495581418275834,
      "num_tokens": 21688365.0,
      "step": 12500
    },
    {
      "entropy": 5.595451211929321,
      "epoch": 0.9729624586656292,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004910239665004397,
      "loss": 5.3996,
      "mean_token_accuracy": 0.16991671025753022,
      "num_tokens": 21697003.0,
      "step": 12505
    },
    {
      "entropy": 5.587980651855469,
      "epoch": 0.9733514880373468,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004910162146260216,
      "loss": 5.408,
      "mean_token_accuracy": 0.16433726847171784,
      "num_tokens": 21705947.0,
      "step": 12510
    },
    {
      "entropy": 5.568564796447754,
      "epoch": 0.9737405174090644,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004910084594739013,
      "loss": 5.4325,
      "mean_token_accuracy": 0.16390192359685898,
      "num_tokens": 21714153.0,
      "step": 12515
    },
    {
      "entropy": 5.630647468566894,
      "epoch": 0.9741295467807819,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004910007010441964,
      "loss": 5.5104,
      "mean_token_accuracy": 0.16128549575805665,
      "num_tokens": 21723480.0,
      "step": 12520
    },
    {
      "entropy": 5.6597191333770756,
      "epoch": 0.9745185761524995,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004909929393370248,
      "loss": 5.3811,
      "mean_token_accuracy": 0.1717036947607994,
      "num_tokens": 21731297.0,
      "step": 12525
    },
    {
      "entropy": 5.553610706329346,
      "epoch": 0.9749076055242171,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004909851743525041,
      "loss": 5.3617,
      "mean_token_accuracy": 0.16348666846752166,
      "num_tokens": 21740975.0,
      "step": 12530
    },
    {
      "entropy": 5.5200341701507565,
      "epoch": 0.9752966348959347,
      "grad_norm": 1.59375,
      "learning_rate": 0.000490977406090752,
      "loss": 5.3565,
      "mean_token_accuracy": 0.16492999494075775,
      "num_tokens": 21749642.0,
      "step": 12535
    },
    {
      "entropy": 5.591656494140625,
      "epoch": 0.9756856642676522,
      "grad_norm": 1.625,
      "learning_rate": 0.0004909696345518867,
      "loss": 5.3955,
      "mean_token_accuracy": 0.16482315063476563,
      "num_tokens": 21757962.0,
      "step": 12540
    },
    {
      "entropy": 5.629759502410889,
      "epoch": 0.9760746936393697,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004909618597360258,
      "loss": 5.3771,
      "mean_token_accuracy": 0.17003940045833588,
      "num_tokens": 21766053.0,
      "step": 12545
    },
    {
      "entropy": 5.507699871063233,
      "epoch": 0.9764637230110873,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004909540816432876,
      "loss": 5.3511,
      "mean_token_accuracy": 0.16111702620983123,
      "num_tokens": 21774280.0,
      "step": 12550
    },
    {
      "entropy": 5.573996067047119,
      "epoch": 0.9768527523828049,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004909463002737897,
      "loss": 5.3712,
      "mean_token_accuracy": 0.1694996878504753,
      "num_tokens": 21782879.0,
      "step": 12555
    },
    {
      "entropy": 5.654839944839478,
      "epoch": 0.9772417817545225,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004909385156276506,
      "loss": 5.4579,
      "mean_token_accuracy": 0.16924386769533156,
      "num_tokens": 21791294.0,
      "step": 12560
    },
    {
      "entropy": 5.591029214859009,
      "epoch": 0.9776308111262401,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004909307277049881,
      "loss": 5.2873,
      "mean_token_accuracy": 0.17874531745910643,
      "num_tokens": 21799304.0,
      "step": 12565
    },
    {
      "entropy": 5.599150848388672,
      "epoch": 0.9780198404979576,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004909229365059205,
      "loss": 5.5029,
      "mean_token_accuracy": 0.15985802412033082,
      "num_tokens": 21809024.0,
      "step": 12570
    },
    {
      "entropy": 5.582186079025268,
      "epoch": 0.9784088698696751,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004909151420305661,
      "loss": 5.2952,
      "mean_token_accuracy": 0.17661263346672057,
      "num_tokens": 21817412.0,
      "step": 12575
    },
    {
      "entropy": 5.504785776138306,
      "epoch": 0.9787978992413927,
      "grad_norm": 1.578125,
      "learning_rate": 0.000490907344279043,
      "loss": 5.3278,
      "mean_token_accuracy": 0.17064658403396607,
      "num_tokens": 21825261.0,
      "step": 12580
    },
    {
      "entropy": 5.61702446937561,
      "epoch": 0.9791869286131103,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004908995432514698,
      "loss": 5.4527,
      "mean_token_accuracy": 0.17164114564657212,
      "num_tokens": 21833318.0,
      "step": 12585
    },
    {
      "entropy": 5.565180969238281,
      "epoch": 0.9795759579848279,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004908917389479645,
      "loss": 5.4065,
      "mean_token_accuracy": 0.17269199788570405,
      "num_tokens": 21841377.0,
      "step": 12590
    },
    {
      "entropy": 5.559814596176148,
      "epoch": 0.9799649873565455,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004908839313686456,
      "loss": 5.3641,
      "mean_token_accuracy": 0.16850048750638963,
      "num_tokens": 21850183.0,
      "step": 12595
    },
    {
      "entropy": 5.67911696434021,
      "epoch": 0.980354016728263,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004908761205136319,
      "loss": 5.4554,
      "mean_token_accuracy": 0.17185364663600922,
      "num_tokens": 21858108.0,
      "step": 12600
    },
    {
      "entropy": 5.575015640258789,
      "epoch": 0.9807430460999805,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004908683063830414,
      "loss": 5.3452,
      "mean_token_accuracy": 0.16938232034444808,
      "num_tokens": 21866330.0,
      "step": 12605
    },
    {
      "entropy": 5.568648910522461,
      "epoch": 0.9811320754716981,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004908604889769932,
      "loss": 5.4134,
      "mean_token_accuracy": 0.16798173189163207,
      "num_tokens": 21875254.0,
      "step": 12610
    },
    {
      "entropy": 5.618528223037719,
      "epoch": 0.9815211048434157,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004908526682956054,
      "loss": 5.4826,
      "mean_token_accuracy": 0.16005044877529145,
      "num_tokens": 21884009.0,
      "step": 12615
    },
    {
      "entropy": 5.646177816390991,
      "epoch": 0.9819101342151333,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004908448443389972,
      "loss": 5.4054,
      "mean_token_accuracy": 0.16468650102615356,
      "num_tokens": 21892428.0,
      "step": 12620
    },
    {
      "entropy": 5.6400901794433596,
      "epoch": 0.9822991635868508,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004908370171072869,
      "loss": 5.4489,
      "mean_token_accuracy": 0.1713656887412071,
      "num_tokens": 21901140.0,
      "step": 12625
    },
    {
      "entropy": 5.524168920516968,
      "epoch": 0.9826881929585684,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004908291866005933,
      "loss": 5.3428,
      "mean_token_accuracy": 0.1827618509531021,
      "num_tokens": 21909072.0,
      "step": 12630
    },
    {
      "entropy": 5.499984264373779,
      "epoch": 0.9830772223302859,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004908213528190355,
      "loss": 5.3942,
      "mean_token_accuracy": 0.16729398667812348,
      "num_tokens": 21917873.0,
      "step": 12635
    },
    {
      "entropy": 5.642496728897095,
      "epoch": 0.9834662517020035,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004908135157627321,
      "loss": 5.5039,
      "mean_token_accuracy": 0.16090389639139174,
      "num_tokens": 21927302.0,
      "step": 12640
    },
    {
      "entropy": 5.616642904281616,
      "epoch": 0.9838552810737211,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004908056754318022,
      "loss": 5.4414,
      "mean_token_accuracy": 0.1711207300424576,
      "num_tokens": 21936440.0,
      "step": 12645
    },
    {
      "entropy": 5.565749645233154,
      "epoch": 0.9842443104454386,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004907978318263646,
      "loss": 5.3079,
      "mean_token_accuracy": 0.16842927783727646,
      "num_tokens": 21944655.0,
      "step": 12650
    },
    {
      "entropy": 5.664102029800415,
      "epoch": 0.9846333398171562,
      "grad_norm": 1.625,
      "learning_rate": 0.0004907899849465383,
      "loss": 5.5415,
      "mean_token_accuracy": 0.16499512493610383,
      "num_tokens": 21953202.0,
      "step": 12655
    },
    {
      "entropy": 5.607707691192627,
      "epoch": 0.9850223691888738,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004907821347924424,
      "loss": 5.4077,
      "mean_token_accuracy": 0.16860270798206328,
      "num_tokens": 21961889.0,
      "step": 12660
    },
    {
      "entropy": 5.5387133121490475,
      "epoch": 0.9854113985605913,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004907742813641963,
      "loss": 5.2683,
      "mean_token_accuracy": 0.17518474757671357,
      "num_tokens": 21970864.0,
      "step": 12665
    },
    {
      "entropy": 5.5290234088897705,
      "epoch": 0.9858004279323089,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004907664246619187,
      "loss": 5.3742,
      "mean_token_accuracy": 0.16862202137708665,
      "num_tokens": 21979271.0,
      "step": 12670
    },
    {
      "entropy": 5.556199789047241,
      "epoch": 0.9861894573040264,
      "grad_norm": 1.5,
      "learning_rate": 0.0004907585646857293,
      "loss": 5.2483,
      "mean_token_accuracy": 0.17392691522836684,
      "num_tokens": 21987321.0,
      "step": 12675
    },
    {
      "entropy": 5.612700462341309,
      "epoch": 0.986578486675744,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004907507014357467,
      "loss": 5.5053,
      "mean_token_accuracy": 0.16294969171285628,
      "num_tokens": 21996094.0,
      "step": 12680
    },
    {
      "entropy": 5.541552543640137,
      "epoch": 0.9869675160474616,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004907428349120909,
      "loss": 5.3535,
      "mean_token_accuracy": 0.16713285595178604,
      "num_tokens": 22004095.0,
      "step": 12685
    },
    {
      "entropy": 5.634124612808227,
      "epoch": 0.9873565454191792,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004907349651148809,
      "loss": 5.4669,
      "mean_token_accuracy": 0.1649239957332611,
      "num_tokens": 22013371.0,
      "step": 12690
    },
    {
      "entropy": 5.548127555847168,
      "epoch": 0.9877455747908968,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004907270920442361,
      "loss": 5.3348,
      "mean_token_accuracy": 0.16820223927497863,
      "num_tokens": 22022313.0,
      "step": 12695
    },
    {
      "entropy": 5.640724039077758,
      "epoch": 0.9881346041626142,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004907192157002762,
      "loss": 5.5035,
      "mean_token_accuracy": 0.16597417145967483,
      "num_tokens": 22031690.0,
      "step": 12700
    },
    {
      "entropy": 5.720732975006103,
      "epoch": 0.9885236335343318,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004907113360831204,
      "loss": 5.5498,
      "mean_token_accuracy": 0.15509618520736695,
      "num_tokens": 22040957.0,
      "step": 12705
    },
    {
      "entropy": 5.640105438232422,
      "epoch": 0.9889126629060494,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004907034531928886,
      "loss": 5.384,
      "mean_token_accuracy": 0.16889772564172745,
      "num_tokens": 22049623.0,
      "step": 12710
    },
    {
      "entropy": 5.541781282424926,
      "epoch": 0.989301692277767,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004906955670297001,
      "loss": 5.3755,
      "mean_token_accuracy": 0.16839928478002547,
      "num_tokens": 22058205.0,
      "step": 12715
    },
    {
      "entropy": 5.555943965911865,
      "epoch": 0.9896907216494846,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004906876775936746,
      "loss": 5.2751,
      "mean_token_accuracy": 0.17056004852056503,
      "num_tokens": 22067375.0,
      "step": 12720
    },
    {
      "entropy": 5.523751258850098,
      "epoch": 0.990079751021202,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004906797848849321,
      "loss": 5.3152,
      "mean_token_accuracy": 0.1756708651781082,
      "num_tokens": 22075670.0,
      "step": 12725
    },
    {
      "entropy": 5.517127323150635,
      "epoch": 0.9904687803929196,
      "grad_norm": 1.7265625,
      "learning_rate": 0.000490671888903592,
      "loss": 5.3986,
      "mean_token_accuracy": 0.16902980208396912,
      "num_tokens": 22084360.0,
      "step": 12730
    },
    {
      "entropy": 5.547106170654297,
      "epoch": 0.9908578097646372,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004906639896497744,
      "loss": 5.4079,
      "mean_token_accuracy": 0.1749236837029457,
      "num_tokens": 22092666.0,
      "step": 12735
    },
    {
      "entropy": 5.639094638824463,
      "epoch": 0.9912468391363548,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004906560871235988,
      "loss": 5.5243,
      "mean_token_accuracy": 0.16201450526714326,
      "num_tokens": 22101311.0,
      "step": 12740
    },
    {
      "entropy": 5.6404026508331295,
      "epoch": 0.9916358685080724,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004906481813251854,
      "loss": 5.3512,
      "mean_token_accuracy": 0.17302687615156173,
      "num_tokens": 22109640.0,
      "step": 12745
    },
    {
      "entropy": 5.558566522598267,
      "epoch": 0.9920248978797899,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004906402722546542,
      "loss": 5.3578,
      "mean_token_accuracy": 0.16336167603731155,
      "num_tokens": 22118721.0,
      "step": 12750
    },
    {
      "entropy": 5.5309693813323975,
      "epoch": 0.9924139272515075,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000490632359912125,
      "loss": 5.3549,
      "mean_token_accuracy": 0.16795935928821565,
      "num_tokens": 22126745.0,
      "step": 12755
    },
    {
      "entropy": 5.620975685119629,
      "epoch": 0.992802956623225,
      "grad_norm": 1.625,
      "learning_rate": 0.000490624444297718,
      "loss": 5.4324,
      "mean_token_accuracy": 0.16674798727035522,
      "num_tokens": 22135010.0,
      "step": 12760
    },
    {
      "entropy": 5.562928485870361,
      "epoch": 0.9931919859949426,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004906165254115533,
      "loss": 5.3876,
      "mean_token_accuracy": 0.1693186119198799,
      "num_tokens": 22144339.0,
      "step": 12765
    },
    {
      "entropy": 5.504650115966797,
      "epoch": 0.9935810153666602,
      "grad_norm": 1.5625,
      "learning_rate": 0.000490608603253751,
      "loss": 5.3311,
      "mean_token_accuracy": 0.17818619459867477,
      "num_tokens": 22153405.0,
      "step": 12770
    },
    {
      "entropy": 5.613975858688354,
      "epoch": 0.9939700447383778,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004906006778244312,
      "loss": 5.3972,
      "mean_token_accuracy": 0.16567289531230928,
      "num_tokens": 22161392.0,
      "step": 12775
    },
    {
      "entropy": 5.626464748382569,
      "epoch": 0.9943590741100953,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004905927491237145,
      "loss": 5.3751,
      "mean_token_accuracy": 0.16952494531869888,
      "num_tokens": 22170635.0,
      "step": 12780
    },
    {
      "entropy": 5.496331119537354,
      "epoch": 0.9947481034818129,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004905848171517208,
      "loss": 5.2583,
      "mean_token_accuracy": 0.17667841017246247,
      "num_tokens": 22178629.0,
      "step": 12785
    },
    {
      "entropy": 5.4876610279083256,
      "epoch": 0.9951371328535304,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004905768819085706,
      "loss": 5.3403,
      "mean_token_accuracy": 0.16736946552991866,
      "num_tokens": 22187142.0,
      "step": 12790
    },
    {
      "entropy": 5.61784143447876,
      "epoch": 0.995526162225248,
      "grad_norm": 1.375,
      "learning_rate": 0.0004905689433943846,
      "loss": 5.4147,
      "mean_token_accuracy": 0.16835962533950805,
      "num_tokens": 22195249.0,
      "step": 12795
    },
    {
      "entropy": 5.604678726196289,
      "epoch": 0.9959151915969656,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004905610016092828,
      "loss": 5.4356,
      "mean_token_accuracy": 0.17299420088529588,
      "num_tokens": 22203621.0,
      "step": 12800
    },
    {
      "entropy": 5.513427066802978,
      "epoch": 0.9963042209686831,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004905530565533859,
      "loss": 5.3902,
      "mean_token_accuracy": 0.17429412454366683,
      "num_tokens": 22212547.0,
      "step": 12805
    },
    {
      "entropy": 5.576586627960205,
      "epoch": 0.9966932503404007,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004905451082268146,
      "loss": 5.3753,
      "mean_token_accuracy": 0.16535618603229524,
      "num_tokens": 22221642.0,
      "step": 12810
    },
    {
      "entropy": 5.57655897140503,
      "epoch": 0.9970822797121183,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004905371566296891,
      "loss": 5.4221,
      "mean_token_accuracy": 0.16553240567445754,
      "num_tokens": 22230227.0,
      "step": 12815
    },
    {
      "entropy": 5.546317768096924,
      "epoch": 0.9974713090838359,
      "grad_norm": 1.375,
      "learning_rate": 0.0004905292017621305,
      "loss": 5.379,
      "mean_token_accuracy": 0.17151281237602234,
      "num_tokens": 22238973.0,
      "step": 12820
    },
    {
      "entropy": 5.574538230895996,
      "epoch": 0.9978603384555534,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004905212436242593,
      "loss": 5.3664,
      "mean_token_accuracy": 0.17034893482923508,
      "num_tokens": 22247613.0,
      "step": 12825
    },
    {
      "entropy": 5.531944751739502,
      "epoch": 0.9982493678272709,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004905132822161962,
      "loss": 5.4013,
      "mean_token_accuracy": 0.1644030660390854,
      "num_tokens": 22255707.0,
      "step": 12830
    },
    {
      "entropy": 5.529854869842529,
      "epoch": 0.9986383971989885,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004905053175380621,
      "loss": 5.2829,
      "mean_token_accuracy": 0.17318745702505112,
      "num_tokens": 22264524.0,
      "step": 12835
    },
    {
      "entropy": 5.589191246032715,
      "epoch": 0.9990274265707061,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004904973495899778,
      "loss": 5.4015,
      "mean_token_accuracy": 0.1655420944094658,
      "num_tokens": 22272973.0,
      "step": 12840
    },
    {
      "entropy": 5.55372052192688,
      "epoch": 0.9994164559424237,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004904893783720642,
      "loss": 5.4276,
      "mean_token_accuracy": 0.16893602758646012,
      "num_tokens": 22281545.0,
      "step": 12845
    },
    {
      "entropy": 5.563304615020752,
      "epoch": 0.9998054853141413,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004904814038844424,
      "loss": 5.4281,
      "mean_token_accuracy": 0.17418748140335083,
      "num_tokens": 22289835.0,
      "step": 12850
    },
    {
      "entropy": 5.521205902099609,
      "epoch": 1.000155611748687,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004904734261272332,
      "loss": 5.3891,
      "mean_token_accuracy": 0.16933360861407387,
      "num_tokens": 22297794.0,
      "step": 12855
    },
    {
      "entropy": 5.524118947982788,
      "epoch": 1.0005446411204046,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004904654451005576,
      "loss": 5.3053,
      "mean_token_accuracy": 0.1667938694357872,
      "num_tokens": 22305637.0,
      "step": 12860
    },
    {
      "entropy": 5.5762371063232425,
      "epoch": 1.0009336704921221,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004904574608045369,
      "loss": 5.3809,
      "mean_token_accuracy": 0.16399641185998917,
      "num_tokens": 22314796.0,
      "step": 12865
    },
    {
      "entropy": 5.551935291290283,
      "epoch": 1.0013226998638398,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004904494732392923,
      "loss": 5.3254,
      "mean_token_accuracy": 0.16489121317863464,
      "num_tokens": 22324522.0,
      "step": 12870
    },
    {
      "entropy": 5.583127498626709,
      "epoch": 1.0017117292355573,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004904414824049448,
      "loss": 5.3107,
      "mean_token_accuracy": 0.17289249002933502,
      "num_tokens": 22333404.0,
      "step": 12875
    },
    {
      "entropy": 5.531590461730957,
      "epoch": 1.0021007586072748,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004904334883016156,
      "loss": 5.4215,
      "mean_token_accuracy": 0.16096602827310563,
      "num_tokens": 22343364.0,
      "step": 12880
    },
    {
      "entropy": 5.633839654922485,
      "epoch": 1.0024897879789925,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004904254909294261,
      "loss": 5.3302,
      "mean_token_accuracy": 0.1673327311873436,
      "num_tokens": 22352854.0,
      "step": 12885
    },
    {
      "entropy": 5.585352230072021,
      "epoch": 1.00287881735071,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004904174902884978,
      "loss": 5.2469,
      "mean_token_accuracy": 0.17581924498081208,
      "num_tokens": 22361916.0,
      "step": 12890
    },
    {
      "entropy": 5.492735099792481,
      "epoch": 1.0032678467224276,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004904094863789519,
      "loss": 5.3255,
      "mean_token_accuracy": 0.167428807169199,
      "num_tokens": 22370848.0,
      "step": 12895
    },
    {
      "entropy": 5.497122097015381,
      "epoch": 1.0036568760941451,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00049040147920091,
      "loss": 5.2529,
      "mean_token_accuracy": 0.1785206213593483,
      "num_tokens": 22379945.0,
      "step": 12900
    },
    {
      "entropy": 5.555041217803955,
      "epoch": 1.0040459054658626,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004903934687544933,
      "loss": 5.2859,
      "mean_token_accuracy": 0.1729058489203453,
      "num_tokens": 22388795.0,
      "step": 12905
    },
    {
      "entropy": 5.58409013748169,
      "epoch": 1.0044349348375803,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004903854550398237,
      "loss": 5.3286,
      "mean_token_accuracy": 0.17441962510347367,
      "num_tokens": 22396960.0,
      "step": 12910
    },
    {
      "entropy": 5.5396336078643795,
      "epoch": 1.0048239642092978,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004903774380570226,
      "loss": 5.2166,
      "mean_token_accuracy": 0.17924095094203948,
      "num_tokens": 22404698.0,
      "step": 12915
    },
    {
      "entropy": 5.519770288467408,
      "epoch": 1.0052129935810155,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004903694178062117,
      "loss": 5.2649,
      "mean_token_accuracy": 0.16885225474834442,
      "num_tokens": 22413391.0,
      "step": 12920
    },
    {
      "entropy": 5.59088659286499,
      "epoch": 1.005602022952733,
      "grad_norm": 1.5,
      "learning_rate": 0.0004903613942875126,
      "loss": 5.3689,
      "mean_token_accuracy": 0.16756330281496049,
      "num_tokens": 22421357.0,
      "step": 12925
    },
    {
      "entropy": 5.522199535369873,
      "epoch": 1.0059910523244504,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004903533675010472,
      "loss": 5.2402,
      "mean_token_accuracy": 0.17148982286453246,
      "num_tokens": 22430286.0,
      "step": 12930
    },
    {
      "entropy": 5.498510885238647,
      "epoch": 1.006380081696168,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004903453374469373,
      "loss": 5.3072,
      "mean_token_accuracy": 0.17554282248020173,
      "num_tokens": 22439128.0,
      "step": 12935
    },
    {
      "entropy": 5.548002052307129,
      "epoch": 1.0067691110678856,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004903373041253045,
      "loss": 5.2589,
      "mean_token_accuracy": 0.1688343971967697,
      "num_tokens": 22447724.0,
      "step": 12940
    },
    {
      "entropy": 5.5518271923065186,
      "epoch": 1.0071581404396033,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004903292675362709,
      "loss": 5.3047,
      "mean_token_accuracy": 0.17320431768894196,
      "num_tokens": 22456792.0,
      "step": 12945
    },
    {
      "entropy": 5.446905088424683,
      "epoch": 1.0075471698113208,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004903212276799584,
      "loss": 5.2342,
      "mean_token_accuracy": 0.17724946439266204,
      "num_tokens": 22465465.0,
      "step": 12950
    },
    {
      "entropy": 5.501704168319702,
      "epoch": 1.0079361991830382,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004903131845564889,
      "loss": 5.2831,
      "mean_token_accuracy": 0.1737855315208435,
      "num_tokens": 22474498.0,
      "step": 12955
    },
    {
      "entropy": 5.480178499221802,
      "epoch": 1.008325228554756,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004903051381659847,
      "loss": 5.2168,
      "mean_token_accuracy": 0.17675953656435012,
      "num_tokens": 22482926.0,
      "step": 12960
    },
    {
      "entropy": 5.51214280128479,
      "epoch": 1.0087142579264734,
      "grad_norm": 1.875,
      "learning_rate": 0.0004902970885085677,
      "loss": 5.2527,
      "mean_token_accuracy": 0.17354887425899507,
      "num_tokens": 22491984.0,
      "step": 12965
    },
    {
      "entropy": 5.5587738990783695,
      "epoch": 1.009103287298191,
      "grad_norm": 1.46875,
      "learning_rate": 0.00049028903558436,
      "loss": 5.2774,
      "mean_token_accuracy": 0.1709383174777031,
      "num_tokens": 22501411.0,
      "step": 12970
    },
    {
      "entropy": 5.599527645111084,
      "epoch": 1.0094923166699086,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004902809793934838,
      "loss": 5.3921,
      "mean_token_accuracy": 0.17097476199269296,
      "num_tokens": 22510289.0,
      "step": 12975
    },
    {
      "entropy": 5.573763227462768,
      "epoch": 1.009881346041626,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004902729199360615,
      "loss": 5.4092,
      "mean_token_accuracy": 0.16954686790704726,
      "num_tokens": 22519061.0,
      "step": 12980
    },
    {
      "entropy": 5.623443698883056,
      "epoch": 1.0102703754133437,
      "grad_norm": 1.625,
      "learning_rate": 0.0004902648572122153,
      "loss": 5.4159,
      "mean_token_accuracy": 0.16341166198253632,
      "num_tokens": 22528207.0,
      "step": 12985
    },
    {
      "entropy": 5.678278112411499,
      "epoch": 1.0106594047850612,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004902567912220674,
      "loss": 5.3627,
      "mean_token_accuracy": 0.17135710418224334,
      "num_tokens": 22537055.0,
      "step": 12990
    },
    {
      "entropy": 5.484136247634888,
      "epoch": 1.011048434156779,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004902487219657404,
      "loss": 5.3185,
      "mean_token_accuracy": 0.16931733042001723,
      "num_tokens": 22545239.0,
      "step": 12995
    },
    {
      "entropy": 5.575659942626953,
      "epoch": 1.0114374635284964,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004902406494433566,
      "loss": 5.4656,
      "mean_token_accuracy": 0.15858107060194016,
      "num_tokens": 22554850.0,
      "step": 13000
    },
    {
      "entropy": 5.607913684844971,
      "epoch": 1.0118264929002139,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004902325736550386,
      "loss": 5.3069,
      "mean_token_accuracy": 0.168249773979187,
      "num_tokens": 22563799.0,
      "step": 13005
    },
    {
      "entropy": 5.5935163497924805,
      "epoch": 1.0122155222719316,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004902244946009088,
      "loss": 5.4982,
      "mean_token_accuracy": 0.16486130356788636,
      "num_tokens": 22572699.0,
      "step": 13010
    },
    {
      "entropy": 5.699872589111328,
      "epoch": 1.012604551643649,
      "grad_norm": 1.625,
      "learning_rate": 0.0004902164122810899,
      "loss": 5.4254,
      "mean_token_accuracy": 0.16156801134347915,
      "num_tokens": 22580916.0,
      "step": 13015
    },
    {
      "entropy": 5.599102592468261,
      "epoch": 1.0129935810153667,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004902083266957045,
      "loss": 5.3041,
      "mean_token_accuracy": 0.17652194797992707,
      "num_tokens": 22589525.0,
      "step": 13020
    },
    {
      "entropy": 5.562756013870239,
      "epoch": 1.0133826103870842,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004902002378448753,
      "loss": 5.3147,
      "mean_token_accuracy": 0.1701074570417404,
      "num_tokens": 22597929.0,
      "step": 13025
    },
    {
      "entropy": 5.572381925582886,
      "epoch": 1.0137716397588017,
      "grad_norm": 1.546875,
      "learning_rate": 0.000490192145728725,
      "loss": 5.3423,
      "mean_token_accuracy": 0.16986422687768937,
      "num_tokens": 22606500.0,
      "step": 13030
    },
    {
      "entropy": 5.5856163024902346,
      "epoch": 1.0141606691305194,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004901840503473764,
      "loss": 5.3023,
      "mean_token_accuracy": 0.1720694214105606,
      "num_tokens": 22614626.0,
      "step": 13035
    },
    {
      "entropy": 5.555132865905762,
      "epoch": 1.0145496985022369,
      "grad_norm": 1.5,
      "learning_rate": 0.0004901759517009522,
      "loss": 5.3449,
      "mean_token_accuracy": 0.16404235512018203,
      "num_tokens": 22623488.0,
      "step": 13040
    },
    {
      "entropy": 5.514724254608154,
      "epoch": 1.0149387278739546,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004901678497895755,
      "loss": 5.2677,
      "mean_token_accuracy": 0.17093850672245026,
      "num_tokens": 22632413.0,
      "step": 13045
    },
    {
      "entropy": 5.612932395935059,
      "epoch": 1.015327757245672,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004901597446133692,
      "loss": 5.4009,
      "mean_token_accuracy": 0.16236702352762222,
      "num_tokens": 22640982.0,
      "step": 13050
    },
    {
      "entropy": 5.627354383468628,
      "epoch": 1.0157167866173895,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004901516361724564,
      "loss": 5.3796,
      "mean_token_accuracy": 0.16859890520572662,
      "num_tokens": 22649809.0,
      "step": 13055
    },
    {
      "entropy": 5.535728788375854,
      "epoch": 1.0161058159891072,
      "grad_norm": 1.625,
      "learning_rate": 0.0004901435244669597,
      "loss": 5.2795,
      "mean_token_accuracy": 0.16824382692575454,
      "num_tokens": 22658292.0,
      "step": 13060
    },
    {
      "entropy": 5.60081639289856,
      "epoch": 1.0164948453608247,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004901354094970025,
      "loss": 5.3647,
      "mean_token_accuracy": 0.17014574706554414,
      "num_tokens": 22667063.0,
      "step": 13065
    },
    {
      "entropy": 5.50766077041626,
      "epoch": 1.0168838747325424,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004901272912627081,
      "loss": 5.2963,
      "mean_token_accuracy": 0.1683926209807396,
      "num_tokens": 22676213.0,
      "step": 13070
    },
    {
      "entropy": 5.399009561538696,
      "epoch": 1.0172729041042599,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004901191697641992,
      "loss": 5.2223,
      "mean_token_accuracy": 0.17431732416152954,
      "num_tokens": 22684948.0,
      "step": 13075
    },
    {
      "entropy": 5.559270000457763,
      "epoch": 1.0176619334759773,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004901110450015994,
      "loss": 5.3226,
      "mean_token_accuracy": 0.16503900289535522,
      "num_tokens": 22693218.0,
      "step": 13080
    },
    {
      "entropy": 5.565805530548095,
      "epoch": 1.018050962847695,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004901029169750319,
      "loss": 5.1915,
      "mean_token_accuracy": 0.17868994325399398,
      "num_tokens": 22701572.0,
      "step": 13085
    },
    {
      "entropy": 5.513815641403198,
      "epoch": 1.0184399922194125,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004900947856846201,
      "loss": 5.2641,
      "mean_token_accuracy": 0.17290471345186234,
      "num_tokens": 22709498.0,
      "step": 13090
    },
    {
      "entropy": 5.539499950408936,
      "epoch": 1.0188290215911302,
      "grad_norm": 1.78125,
      "learning_rate": 0.000490086651130487,
      "loss": 5.2725,
      "mean_token_accuracy": 0.17217059284448624,
      "num_tokens": 22717922.0,
      "step": 13095
    },
    {
      "entropy": 5.584228277206421,
      "epoch": 1.0192180509628477,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004900785133127566,
      "loss": 5.298,
      "mean_token_accuracy": 0.174782095849514,
      "num_tokens": 22725933.0,
      "step": 13100
    },
    {
      "entropy": 5.56052885055542,
      "epoch": 1.0196070803345652,
      "grad_norm": 1.6171875,
      "learning_rate": 0.000490070372231552,
      "loss": 5.2871,
      "mean_token_accuracy": 0.17098522782325745,
      "num_tokens": 22734198.0,
      "step": 13105
    },
    {
      "entropy": 5.549863290786743,
      "epoch": 1.0199961097062828,
      "grad_norm": 1.5,
      "learning_rate": 0.0004900622278869968,
      "loss": 5.3158,
      "mean_token_accuracy": 0.17081442326307297,
      "num_tokens": 22742419.0,
      "step": 13110
    },
    {
      "entropy": 5.590565347671509,
      "epoch": 1.0203851390780003,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004900540802792146,
      "loss": 5.2721,
      "mean_token_accuracy": 0.17368832975625992,
      "num_tokens": 22750576.0,
      "step": 13115
    },
    {
      "entropy": 5.571694278717041,
      "epoch": 1.020774168449718,
      "grad_norm": 1.6640625,
      "learning_rate": 0.000490045929408329,
      "loss": 5.443,
      "mean_token_accuracy": 0.17148976922035217,
      "num_tokens": 22759604.0,
      "step": 13120
    },
    {
      "entropy": 5.519921159744262,
      "epoch": 1.0211631978214355,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004900377752744637,
      "loss": 5.2,
      "mean_token_accuracy": 0.1760471448302269,
      "num_tokens": 22768305.0,
      "step": 13125
    },
    {
      "entropy": 5.598498058319092,
      "epoch": 1.021552227193153,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004900296178777426,
      "loss": 5.3585,
      "mean_token_accuracy": 0.16258263885974883,
      "num_tokens": 22777364.0,
      "step": 13130
    },
    {
      "entropy": 5.523081064224243,
      "epoch": 1.0219412565648707,
      "grad_norm": 1.53125,
      "learning_rate": 0.000490021457218289,
      "loss": 5.2833,
      "mean_token_accuracy": 0.17148469686508178,
      "num_tokens": 22786192.0,
      "step": 13135
    },
    {
      "entropy": 5.619490909576416,
      "epoch": 1.0223302859365881,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004900132932962272,
      "loss": 5.3412,
      "mean_token_accuracy": 0.17192437499761581,
      "num_tokens": 22794244.0,
      "step": 13140
    },
    {
      "entropy": 5.561174011230468,
      "epoch": 1.0227193153083058,
      "grad_norm": 1.59375,
      "learning_rate": 0.000490005126111681,
      "loss": 5.3224,
      "mean_token_accuracy": 0.17580875903367996,
      "num_tokens": 22802543.0,
      "step": 13145
    },
    {
      "entropy": 5.4827522277832035,
      "epoch": 1.0231083446800233,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000489996955664774,
      "loss": 5.3307,
      "mean_token_accuracy": 0.1704048439860344,
      "num_tokens": 22811267.0,
      "step": 13150
    },
    {
      "entropy": 5.540749597549438,
      "epoch": 1.023497374051741,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004899887819556306,
      "loss": 5.2402,
      "mean_token_accuracy": 0.17342166900634765,
      "num_tokens": 22820196.0,
      "step": 13155
    },
    {
      "entropy": 5.611469745635986,
      "epoch": 1.0238864034234585,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004899806049843745,
      "loss": 5.4058,
      "mean_token_accuracy": 0.16959395557641982,
      "num_tokens": 22828498.0,
      "step": 13160
    },
    {
      "entropy": 5.590411138534546,
      "epoch": 1.024275432795176,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004899724247511299,
      "loss": 5.3153,
      "mean_token_accuracy": 0.17874199599027635,
      "num_tokens": 22837411.0,
      "step": 13165
    },
    {
      "entropy": 5.43924765586853,
      "epoch": 1.0246644621668937,
      "grad_norm": 1.640625,
      "learning_rate": 0.000489964241256021,
      "loss": 5.2425,
      "mean_token_accuracy": 0.174375019967556,
      "num_tokens": 22846265.0,
      "step": 13170
    },
    {
      "entropy": 5.445515918731689,
      "epoch": 1.0250534915386111,
      "grad_norm": 1.75,
      "learning_rate": 0.0004899560544991718,
      "loss": 5.2115,
      "mean_token_accuracy": 0.17506559789180756,
      "num_tokens": 22854489.0,
      "step": 13175
    },
    {
      "entropy": 5.581643438339233,
      "epoch": 1.0254425209103288,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004899478644807068,
      "loss": 5.4031,
      "mean_token_accuracy": 0.1649957537651062,
      "num_tokens": 22863269.0,
      "step": 13180
    },
    {
      "entropy": 5.599470853805542,
      "epoch": 1.0258315502820463,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004899396712007498,
      "loss": 5.3845,
      "mean_token_accuracy": 0.1650130942463875,
      "num_tokens": 22872304.0,
      "step": 13185
    },
    {
      "entropy": 5.527854061126709,
      "epoch": 1.0262205796537638,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004899314746594256,
      "loss": 5.1749,
      "mean_token_accuracy": 0.17417426705360411,
      "num_tokens": 22881409.0,
      "step": 13190
    },
    {
      "entropy": 5.547877550125122,
      "epoch": 1.0266096090254815,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004899232748568584,
      "loss": 5.2728,
      "mean_token_accuracy": 0.17250572741031647,
      "num_tokens": 22889560.0,
      "step": 13195
    },
    {
      "entropy": 5.515940761566162,
      "epoch": 1.026998638397199,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004899150717931724,
      "loss": 5.3383,
      "mean_token_accuracy": 0.16536664813756943,
      "num_tokens": 22897434.0,
      "step": 13200
    },
    {
      "entropy": 5.558090734481811,
      "epoch": 1.0273876677689167,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004899068654684924,
      "loss": 5.3265,
      "mean_token_accuracy": 0.17122836261987687,
      "num_tokens": 22906050.0,
      "step": 13205
    },
    {
      "entropy": 5.543594074249268,
      "epoch": 1.0277766971406341,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004898986558829428,
      "loss": 5.2867,
      "mean_token_accuracy": 0.17371345311403275,
      "num_tokens": 22914118.0,
      "step": 13210
    },
    {
      "entropy": 5.535507440567017,
      "epoch": 1.0281657265123516,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004898904430366479,
      "loss": 5.3026,
      "mean_token_accuracy": 0.17229761481285094,
      "num_tokens": 22922539.0,
      "step": 13215
    },
    {
      "entropy": 5.513501453399658,
      "epoch": 1.0285547558840693,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004898822269297328,
      "loss": 5.3049,
      "mean_token_accuracy": 0.16814896166324617,
      "num_tokens": 22932371.0,
      "step": 13220
    },
    {
      "entropy": 5.560471725463867,
      "epoch": 1.0289437852557868,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004898740075623218,
      "loss": 5.3368,
      "mean_token_accuracy": 0.17443976998329164,
      "num_tokens": 22940608.0,
      "step": 13225
    },
    {
      "entropy": 5.570888376235962,
      "epoch": 1.0293328146275045,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004898657849345399,
      "loss": 5.3154,
      "mean_token_accuracy": 0.16587201207876207,
      "num_tokens": 22948905.0,
      "step": 13230
    },
    {
      "entropy": 5.490921688079834,
      "epoch": 1.029721843999222,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004898575590465116,
      "loss": 5.2805,
      "mean_token_accuracy": 0.17339199185371398,
      "num_tokens": 22957218.0,
      "step": 13235
    },
    {
      "entropy": 5.541296672821045,
      "epoch": 1.0301108733709394,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004898493298983619,
      "loss": 5.4322,
      "mean_token_accuracy": 0.17029115557670593,
      "num_tokens": 22965700.0,
      "step": 13240
    },
    {
      "entropy": 5.665392017364502,
      "epoch": 1.0304999027426571,
      "grad_norm": 1.625,
      "learning_rate": 0.0004898410974902155,
      "loss": 5.3734,
      "mean_token_accuracy": 0.1702515408396721,
      "num_tokens": 22973996.0,
      "step": 13245
    },
    {
      "entropy": 5.611515378952026,
      "epoch": 1.0308889321143746,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004898328618221974,
      "loss": 5.3472,
      "mean_token_accuracy": 0.16921236664056777,
      "num_tokens": 22983142.0,
      "step": 13250
    },
    {
      "entropy": 5.511696481704712,
      "epoch": 1.0312779614860923,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004898246228944327,
      "loss": 5.3042,
      "mean_token_accuracy": 0.17110096514225007,
      "num_tokens": 22991263.0,
      "step": 13255
    },
    {
      "entropy": 5.589747524261474,
      "epoch": 1.0316669908578098,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004898163807070461,
      "loss": 5.3915,
      "mean_token_accuracy": 0.16521840393543244,
      "num_tokens": 23000068.0,
      "step": 13260
    },
    {
      "entropy": 5.590217065811157,
      "epoch": 1.0320560202295272,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004898081352601629,
      "loss": 5.2952,
      "mean_token_accuracy": 0.1719805970788002,
      "num_tokens": 23007935.0,
      "step": 13265
    },
    {
      "entropy": 5.508534669876099,
      "epoch": 1.032445049601245,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004897998865539082,
      "loss": 5.3762,
      "mean_token_accuracy": 0.1628126934170723,
      "num_tokens": 23017462.0,
      "step": 13270
    },
    {
      "entropy": 5.587372779846191,
      "epoch": 1.0328340789729624,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004897916345884069,
      "loss": 5.3703,
      "mean_token_accuracy": 0.1638731062412262,
      "num_tokens": 23026681.0,
      "step": 13275
    },
    {
      "entropy": 5.562488794326782,
      "epoch": 1.0332231083446801,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004897833793637847,
      "loss": 5.333,
      "mean_token_accuracy": 0.16925524473190307,
      "num_tokens": 23035720.0,
      "step": 13280
    },
    {
      "entropy": 5.599439382553101,
      "epoch": 1.0336121377163976,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004897751208801665,
      "loss": 5.3498,
      "mean_token_accuracy": 0.16761697828769684,
      "num_tokens": 23044491.0,
      "step": 13285
    },
    {
      "entropy": 5.585770988464356,
      "epoch": 1.034001167088115,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004897668591376777,
      "loss": 5.3603,
      "mean_token_accuracy": 0.1694803535938263,
      "num_tokens": 23054034.0,
      "step": 13290
    },
    {
      "entropy": 5.564758253097534,
      "epoch": 1.0343901964598328,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004897585941364436,
      "loss": 5.2971,
      "mean_token_accuracy": 0.1661220148205757,
      "num_tokens": 23062478.0,
      "step": 13295
    },
    {
      "entropy": 5.516941356658935,
      "epoch": 1.0347792258315502,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004897503258765896,
      "loss": 5.3332,
      "mean_token_accuracy": 0.16616709232330323,
      "num_tokens": 23070776.0,
      "step": 13300
    },
    {
      "entropy": 5.580062818527222,
      "epoch": 1.035168255203268,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004897420543582413,
      "loss": 5.3791,
      "mean_token_accuracy": 0.17137093096971512,
      "num_tokens": 23078459.0,
      "step": 13305
    },
    {
      "entropy": 5.574888563156128,
      "epoch": 1.0355572845749854,
      "grad_norm": 1.8046875,
      "learning_rate": 0.000489733779581524,
      "loss": 5.3411,
      "mean_token_accuracy": 0.17564865797758103,
      "num_tokens": 23087113.0,
      "step": 13310
    },
    {
      "entropy": 5.4958878517150875,
      "epoch": 1.035946313946703,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004897255015465635,
      "loss": 5.2687,
      "mean_token_accuracy": 0.17245230823755264,
      "num_tokens": 23095989.0,
      "step": 13315
    },
    {
      "entropy": 5.522052526473999,
      "epoch": 1.0363353433184206,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004897172202534853,
      "loss": 5.2964,
      "mean_token_accuracy": 0.17018729597330093,
      "num_tokens": 23104578.0,
      "step": 13320
    },
    {
      "entropy": 5.564993953704834,
      "epoch": 1.036724372690138,
      "grad_norm": 1.6015625,
      "learning_rate": 0.000489708935702415,
      "loss": 5.3517,
      "mean_token_accuracy": 0.16749211847782136,
      "num_tokens": 23114390.0,
      "step": 13325
    },
    {
      "entropy": 5.672108793258667,
      "epoch": 1.0371134020618558,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004897006478934784,
      "loss": 5.2382,
      "mean_token_accuracy": 0.1764823541045189,
      "num_tokens": 23122411.0,
      "step": 13330
    },
    {
      "entropy": 5.504877090454102,
      "epoch": 1.0375024314335732,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004896923568268011,
      "loss": 5.2102,
      "mean_token_accuracy": 0.17106251716613768,
      "num_tokens": 23130801.0,
      "step": 13335
    },
    {
      "entropy": 5.494357776641846,
      "epoch": 1.0378914608052907,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000489684062502509,
      "loss": 5.2545,
      "mean_token_accuracy": 0.17517763674259185,
      "num_tokens": 23138578.0,
      "step": 13340
    },
    {
      "entropy": 5.535081005096435,
      "epoch": 1.0382804901770084,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004896757649207281,
      "loss": 5.3263,
      "mean_token_accuracy": 0.1695863649249077,
      "num_tokens": 23147135.0,
      "step": 13345
    },
    {
      "entropy": 5.62409987449646,
      "epoch": 1.0386695195487259,
      "grad_norm": 1.53125,
      "learning_rate": 0.000489667464081584,
      "loss": 5.3969,
      "mean_token_accuracy": 0.16792264580726624,
      "num_tokens": 23155791.0,
      "step": 13350
    },
    {
      "entropy": 5.559703826904297,
      "epoch": 1.0390585489204436,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004896591599852029,
      "loss": 5.2531,
      "mean_token_accuracy": 0.17504205256700517,
      "num_tokens": 23163674.0,
      "step": 13355
    },
    {
      "entropy": 5.526022291183471,
      "epoch": 1.039447578292161,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004896508526317107,
      "loss": 5.3217,
      "mean_token_accuracy": 0.16219640225172044,
      "num_tokens": 23171931.0,
      "step": 13360
    },
    {
      "entropy": 5.580244541168213,
      "epoch": 1.0398366076638785,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004896425420212334,
      "loss": 5.333,
      "mean_token_accuracy": 0.1642221838235855,
      "num_tokens": 23180195.0,
      "step": 13365
    },
    {
      "entropy": 5.52488956451416,
      "epoch": 1.0402256370355962,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004896342281538973,
      "loss": 5.2704,
      "mean_token_accuracy": 0.1739368423819542,
      "num_tokens": 23189486.0,
      "step": 13370
    },
    {
      "entropy": 5.542613410949707,
      "epoch": 1.0406146664073137,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004896259110298283,
      "loss": 5.3894,
      "mean_token_accuracy": 0.16582002192735673,
      "num_tokens": 23198055.0,
      "step": 13375
    },
    {
      "entropy": 5.539638423919678,
      "epoch": 1.0410036957790314,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004896175906491528,
      "loss": 5.2664,
      "mean_token_accuracy": 0.17826088666915893,
      "num_tokens": 23206092.0,
      "step": 13380
    },
    {
      "entropy": 5.595705842971801,
      "epoch": 1.0413927251507489,
      "grad_norm": 1.625,
      "learning_rate": 0.0004896092670119968,
      "loss": 5.3396,
      "mean_token_accuracy": 0.17463093847036362,
      "num_tokens": 23214820.0,
      "step": 13385
    },
    {
      "entropy": 5.502323818206787,
      "epoch": 1.0417817545224664,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004896009401184869,
      "loss": 5.3084,
      "mean_token_accuracy": 0.17374697476625442,
      "num_tokens": 23224033.0,
      "step": 13390
    },
    {
      "entropy": 5.497268962860107,
      "epoch": 1.042170783894184,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004895926099687493,
      "loss": 5.2872,
      "mean_token_accuracy": 0.1754576474428177,
      "num_tokens": 23233074.0,
      "step": 13395
    },
    {
      "entropy": 5.507416200637818,
      "epoch": 1.0425598132659015,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004895842765629104,
      "loss": 5.2462,
      "mean_token_accuracy": 0.17499990612268448,
      "num_tokens": 23241167.0,
      "step": 13400
    },
    {
      "entropy": 5.596942138671875,
      "epoch": 1.0429488426376192,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004895759399010966,
      "loss": 5.2753,
      "mean_token_accuracy": 0.16960269212722778,
      "num_tokens": 23249601.0,
      "step": 13405
    },
    {
      "entropy": 5.550609159469604,
      "epoch": 1.0433378720093367,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004895675999834345,
      "loss": 5.368,
      "mean_token_accuracy": 0.1588814824819565,
      "num_tokens": 23257868.0,
      "step": 13410
    },
    {
      "entropy": 5.58900089263916,
      "epoch": 1.0437269013810542,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004895592568100505,
      "loss": 5.3533,
      "mean_token_accuracy": 0.16595737785100936,
      "num_tokens": 23266586.0,
      "step": 13415
    },
    {
      "entropy": 5.614063453674317,
      "epoch": 1.0441159307527719,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004895509103810714,
      "loss": 5.3321,
      "mean_token_accuracy": 0.1688019946217537,
      "num_tokens": 23275114.0,
      "step": 13420
    },
    {
      "entropy": 5.413854789733887,
      "epoch": 1.0445049601244893,
      "grad_norm": 1.375,
      "learning_rate": 0.0004895425606966237,
      "loss": 5.2248,
      "mean_token_accuracy": 0.17890949547290802,
      "num_tokens": 23283200.0,
      "step": 13425
    },
    {
      "entropy": 5.542884397506714,
      "epoch": 1.044893989496207,
      "grad_norm": 1.5,
      "learning_rate": 0.000489534207756834,
      "loss": 5.4307,
      "mean_token_accuracy": 0.16368771940469742,
      "num_tokens": 23291999.0,
      "step": 13430
    },
    {
      "entropy": 5.607825660705567,
      "epoch": 1.0452830188679245,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004895258515618293,
      "loss": 5.3143,
      "mean_token_accuracy": 0.16850415319204332,
      "num_tokens": 23300514.0,
      "step": 13435
    },
    {
      "entropy": 5.53573431968689,
      "epoch": 1.045672048239642,
      "grad_norm": 1.625,
      "learning_rate": 0.0004895174921117362,
      "loss": 5.2721,
      "mean_token_accuracy": 0.17205916494131088,
      "num_tokens": 23309543.0,
      "step": 13440
    },
    {
      "entropy": 5.466213941574097,
      "epoch": 1.0460610776113597,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004895091294066816,
      "loss": 5.2948,
      "mean_token_accuracy": 0.1734081506729126,
      "num_tokens": 23319002.0,
      "step": 13445
    },
    {
      "entropy": 5.568212270736694,
      "epoch": 1.0464501069830772,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004895007634467924,
      "loss": 5.346,
      "mean_token_accuracy": 0.1708262339234352,
      "num_tokens": 23326604.0,
      "step": 13450
    },
    {
      "entropy": 5.503519582748413,
      "epoch": 1.0468391363547949,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004894923942321955,
      "loss": 5.2818,
      "mean_token_accuracy": 0.17666701823472977,
      "num_tokens": 23335492.0,
      "step": 13455
    },
    {
      "entropy": 5.5053855895996096,
      "epoch": 1.0472281657265123,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004894840217630179,
      "loss": 5.3065,
      "mean_token_accuracy": 0.17429608553647996,
      "num_tokens": 23343693.0,
      "step": 13460
    },
    {
      "entropy": 5.522336721420288,
      "epoch": 1.0476171950982298,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004894756460393868,
      "loss": 5.2667,
      "mean_token_accuracy": 0.17938611954450606,
      "num_tokens": 23352587.0,
      "step": 13465
    },
    {
      "entropy": 5.547571134567261,
      "epoch": 1.0480062244699475,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000489467267061429,
      "loss": 5.3455,
      "mean_token_accuracy": 0.1718888908624649,
      "num_tokens": 23361293.0,
      "step": 13470
    },
    {
      "entropy": 5.5730455875396725,
      "epoch": 1.048395253841665,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004894588848292718,
      "loss": 5.3243,
      "mean_token_accuracy": 0.17050171196460723,
      "num_tokens": 23369663.0,
      "step": 13475
    },
    {
      "entropy": 5.535304594039917,
      "epoch": 1.0487842832133827,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004894504993430425,
      "loss": 5.2367,
      "mean_token_accuracy": 0.17066449224948882,
      "num_tokens": 23378224.0,
      "step": 13480
    },
    {
      "entropy": 5.581699514389038,
      "epoch": 1.0491733125851002,
      "grad_norm": 1.4375,
      "learning_rate": 0.000489442110602868,
      "loss": 5.4122,
      "mean_token_accuracy": 0.16622019559144974,
      "num_tokens": 23386693.0,
      "step": 13485
    },
    {
      "entropy": 5.533706712722778,
      "epoch": 1.0495623419568176,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004894337186088759,
      "loss": 5.2712,
      "mean_token_accuracy": 0.1727204993367195,
      "num_tokens": 23395226.0,
      "step": 13490
    },
    {
      "entropy": 5.527824020385742,
      "epoch": 1.0499513713285353,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004894253233611934,
      "loss": 5.2939,
      "mean_token_accuracy": 0.16925432831048964,
      "num_tokens": 23403779.0,
      "step": 13495
    },
    {
      "entropy": 5.560998725891113,
      "epoch": 1.0503404007002528,
      "grad_norm": 1.6015625,
      "learning_rate": 0.000489416924859948,
      "loss": 5.3993,
      "mean_token_accuracy": 0.17176871448755265,
      "num_tokens": 23411807.0,
      "step": 13500
    },
    {
      "entropy": 5.492605113983155,
      "epoch": 1.0507294300719705,
      "grad_norm": 1.484375,
      "learning_rate": 0.000489408523105267,
      "loss": 5.2718,
      "mean_token_accuracy": 0.16818220168352127,
      "num_tokens": 23420323.0,
      "step": 13505
    },
    {
      "entropy": 5.528667259216308,
      "epoch": 1.051118459443688,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004894001180972779,
      "loss": 5.2788,
      "mean_token_accuracy": 0.1729546681046486,
      "num_tokens": 23429220.0,
      "step": 13510
    },
    {
      "entropy": 5.546587419509888,
      "epoch": 1.0515074888154055,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004893917098361082,
      "loss": 5.3048,
      "mean_token_accuracy": 0.17642121762037277,
      "num_tokens": 23438471.0,
      "step": 13515
    },
    {
      "entropy": 5.576458263397217,
      "epoch": 1.0518965181871232,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004893832983218856,
      "loss": 5.3738,
      "mean_token_accuracy": 0.17581699937582015,
      "num_tokens": 23447504.0,
      "step": 13520
    },
    {
      "entropy": 5.450729417800903,
      "epoch": 1.0522855475588406,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004893748835547377,
      "loss": 5.113,
      "mean_token_accuracy": 0.185771644115448,
      "num_tokens": 23456065.0,
      "step": 13525
    },
    {
      "entropy": 5.535451936721802,
      "epoch": 1.0526745769305583,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004893664655347921,
      "loss": 5.2961,
      "mean_token_accuracy": 0.17600812017917633,
      "num_tokens": 23463826.0,
      "step": 13530
    },
    {
      "entropy": 5.418683910369873,
      "epoch": 1.0530636063022758,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004893580442621767,
      "loss": 5.2655,
      "mean_token_accuracy": 0.1793366938829422,
      "num_tokens": 23472092.0,
      "step": 13535
    },
    {
      "entropy": 5.565783071517944,
      "epoch": 1.0534526356739935,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000489349619737019,
      "loss": 5.3779,
      "mean_token_accuracy": 0.16387729495763778,
      "num_tokens": 23480934.0,
      "step": 13540
    },
    {
      "entropy": 5.618014240264893,
      "epoch": 1.053841665045711,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000489341191959447,
      "loss": 5.325,
      "mean_token_accuracy": 0.1696184165775776,
      "num_tokens": 23490187.0,
      "step": 13545
    },
    {
      "entropy": 5.5481956005096436,
      "epoch": 1.0542306944174284,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004893327609295887,
      "loss": 5.2614,
      "mean_token_accuracy": 0.17242091447114943,
      "num_tokens": 23498330.0,
      "step": 13550
    },
    {
      "entropy": 5.435535097122193,
      "epoch": 1.0546197237891461,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004893243266475719,
      "loss": 5.188,
      "mean_token_accuracy": 0.17514382153749466,
      "num_tokens": 23506219.0,
      "step": 13555
    },
    {
      "entropy": 5.4826188564300535,
      "epoch": 1.0550087531608636,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004893158891135245,
      "loss": 5.2779,
      "mean_token_accuracy": 0.173899045586586,
      "num_tokens": 23514585.0,
      "step": 13560
    },
    {
      "entropy": 5.568767881393432,
      "epoch": 1.055397782532581,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004893074483275746,
      "loss": 5.2323,
      "mean_token_accuracy": 0.1731153279542923,
      "num_tokens": 23522919.0,
      "step": 13565
    },
    {
      "entropy": 5.561643838882446,
      "epoch": 1.0557868119042988,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004892990042898503,
      "loss": 5.4482,
      "mean_token_accuracy": 0.1605902597308159,
      "num_tokens": 23531613.0,
      "step": 13570
    },
    {
      "entropy": 5.564041614532471,
      "epoch": 1.0561758412760163,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004892905570004798,
      "loss": 5.3891,
      "mean_token_accuracy": 0.16505015641450882,
      "num_tokens": 23539865.0,
      "step": 13575
    },
    {
      "entropy": 5.478564691543579,
      "epoch": 1.056564870647734,
      "grad_norm": 1.6484375,
      "learning_rate": 0.000489282106459591,
      "loss": 5.2238,
      "mean_token_accuracy": 0.18216872662305833,
      "num_tokens": 23548424.0,
      "step": 13580
    },
    {
      "entropy": 5.562547302246093,
      "epoch": 1.0569539000194514,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004892736526673124,
      "loss": 5.3177,
      "mean_token_accuracy": 0.17252922505140306,
      "num_tokens": 23557199.0,
      "step": 13585
    },
    {
      "entropy": 5.565737009048462,
      "epoch": 1.0573429293911691,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004892651956237721,
      "loss": 5.4448,
      "mean_token_accuracy": 0.15989054441452027,
      "num_tokens": 23565979.0,
      "step": 13590
    },
    {
      "entropy": 5.564665126800537,
      "epoch": 1.0577319587628866,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004892567353290986,
      "loss": 5.3022,
      "mean_token_accuracy": 0.16925350278615953,
      "num_tokens": 23575405.0,
      "step": 13595
    },
    {
      "entropy": 5.64054651260376,
      "epoch": 1.058120988134604,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004892482717834201,
      "loss": 5.4364,
      "mean_token_accuracy": 0.15500097647309302,
      "num_tokens": 23584589.0,
      "step": 13600
    },
    {
      "entropy": 5.524608135223389,
      "epoch": 1.0585100175063218,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004892398049868651,
      "loss": 5.2872,
      "mean_token_accuracy": 0.1721488878130913,
      "num_tokens": 23593079.0,
      "step": 13605
    },
    {
      "entropy": 5.564929008483887,
      "epoch": 1.0588990468780393,
      "grad_norm": 1.5,
      "learning_rate": 0.000489231334939562,
      "loss": 5.357,
      "mean_token_accuracy": 0.1696854367852211,
      "num_tokens": 23602010.0,
      "step": 13610
    },
    {
      "entropy": 5.610072231292724,
      "epoch": 1.059288076249757,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004892228616416395,
      "loss": 5.3134,
      "mean_token_accuracy": 0.17270972430706025,
      "num_tokens": 23610707.0,
      "step": 13615
    },
    {
      "entropy": 5.460940933227539,
      "epoch": 1.0596771056214744,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004892143850932259,
      "loss": 5.2961,
      "mean_token_accuracy": 0.16677080243825912,
      "num_tokens": 23619579.0,
      "step": 13620
    },
    {
      "entropy": 5.498000049591065,
      "epoch": 1.060066134993192,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00048920590529445,
      "loss": 5.2719,
      "mean_token_accuracy": 0.18019049018621444,
      "num_tokens": 23628407.0,
      "step": 13625
    },
    {
      "entropy": 5.489911699295044,
      "epoch": 1.0604551643649096,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004891974222454406,
      "loss": 5.1994,
      "mean_token_accuracy": 0.17532860189676286,
      "num_tokens": 23637094.0,
      "step": 13630
    },
    {
      "entropy": 5.492452001571655,
      "epoch": 1.060844193736627,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004891889359463261,
      "loss": 5.2958,
      "mean_token_accuracy": 0.17481526583433152,
      "num_tokens": 23645823.0,
      "step": 13635
    },
    {
      "entropy": 5.570481586456299,
      "epoch": 1.0612332231083448,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004891804463972354,
      "loss": 5.4054,
      "mean_token_accuracy": 0.16421720683574675,
      "num_tokens": 23655027.0,
      "step": 13640
    },
    {
      "entropy": 5.657866048812866,
      "epoch": 1.0616222524800623,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004891719535982974,
      "loss": 5.4441,
      "mean_token_accuracy": 0.16156262755393982,
      "num_tokens": 23663840.0,
      "step": 13645
    },
    {
      "entropy": 5.62884030342102,
      "epoch": 1.0620112818517797,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004891634575496408,
      "loss": 5.3922,
      "mean_token_accuracy": 0.16663903295993804,
      "num_tokens": 23672795.0,
      "step": 13650
    },
    {
      "entropy": 5.578080129623413,
      "epoch": 1.0624003112234974,
      "grad_norm": 1.5,
      "learning_rate": 0.0004891549582513946,
      "loss": 5.3172,
      "mean_token_accuracy": 0.16714794635772706,
      "num_tokens": 23681833.0,
      "step": 13655
    },
    {
      "entropy": 5.574814081192017,
      "epoch": 1.062789340595215,
      "grad_norm": 1.6484375,
      "learning_rate": 0.000489146455703688,
      "loss": 5.2916,
      "mean_token_accuracy": 0.17374734282493592,
      "num_tokens": 23689939.0,
      "step": 13660
    },
    {
      "entropy": 5.514884662628174,
      "epoch": 1.0631783699669326,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004891379499066495,
      "loss": 5.3363,
      "mean_token_accuracy": 0.17417766749858857,
      "num_tokens": 23698535.0,
      "step": 13665
    },
    {
      "entropy": 5.5378015518188475,
      "epoch": 1.06356739933865,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004891294408604087,
      "loss": 5.2683,
      "mean_token_accuracy": 0.17150928527116777,
      "num_tokens": 23707749.0,
      "step": 13670
    },
    {
      "entropy": 5.612474966049194,
      "epoch": 1.0639564287103676,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004891209285650942,
      "loss": 5.414,
      "mean_token_accuracy": 0.1681921124458313,
      "num_tokens": 23716503.0,
      "step": 13675
    },
    {
      "entropy": 5.620874881744385,
      "epoch": 1.0643454580820852,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004891124130208355,
      "loss": 5.353,
      "mean_token_accuracy": 0.17228189632296562,
      "num_tokens": 23724882.0,
      "step": 13680
    },
    {
      "entropy": 5.450166463851929,
      "epoch": 1.0647344874538027,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004891038942277618,
      "loss": 5.3544,
      "mean_token_accuracy": 0.17434402406215668,
      "num_tokens": 23733680.0,
      "step": 13685
    },
    {
      "entropy": 5.5328607082366945,
      "epoch": 1.0651235168255204,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004890953721860022,
      "loss": 5.2802,
      "mean_token_accuracy": 0.1807577818632126,
      "num_tokens": 23742066.0,
      "step": 13690
    },
    {
      "entropy": 5.522846984863281,
      "epoch": 1.065512546197238,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004890868468956862,
      "loss": 5.2014,
      "mean_token_accuracy": 0.1826232448220253,
      "num_tokens": 23750558.0,
      "step": 13695
    },
    {
      "entropy": 5.465801191329956,
      "epoch": 1.0659015755689554,
      "grad_norm": 1.640625,
      "learning_rate": 0.000489078318356943,
      "loss": 5.2722,
      "mean_token_accuracy": 0.1692625030875206,
      "num_tokens": 23759120.0,
      "step": 13700
    },
    {
      "entropy": 5.448861455917358,
      "epoch": 1.066290604940673,
      "grad_norm": 5.0,
      "learning_rate": 0.0004890697865699021,
      "loss": 5.3096,
      "mean_token_accuracy": 0.17092404961586,
      "num_tokens": 23768301.0,
      "step": 13705
    },
    {
      "entropy": 5.515306758880615,
      "epoch": 1.0666796343123905,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004890612515346929,
      "loss": 5.2569,
      "mean_token_accuracy": 0.17921047955751418,
      "num_tokens": 23776155.0,
      "step": 13710
    },
    {
      "entropy": 5.566693305969238,
      "epoch": 1.0670686636841082,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004890527132514448,
      "loss": 5.3532,
      "mean_token_accuracy": 0.16514315009117125,
      "num_tokens": 23785363.0,
      "step": 13715
    },
    {
      "entropy": 5.595513248443604,
      "epoch": 1.0674576930558257,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004890441717202876,
      "loss": 5.2999,
      "mean_token_accuracy": 0.17422668337821962,
      "num_tokens": 23793817.0,
      "step": 13720
    },
    {
      "entropy": 5.550647640228272,
      "epoch": 1.0678467224275432,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004890356269413507,
      "loss": 5.356,
      "mean_token_accuracy": 0.16996671706438066,
      "num_tokens": 23802629.0,
      "step": 13725
    },
    {
      "entropy": 5.465928840637207,
      "epoch": 1.068235751799261,
      "grad_norm": 1.4375,
      "learning_rate": 0.000489027078914764,
      "loss": 5.1411,
      "mean_token_accuracy": 0.1822621151804924,
      "num_tokens": 23811178.0,
      "step": 13730
    },
    {
      "entropy": 5.53959698677063,
      "epoch": 1.0686247811709784,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004890185276406569,
      "loss": 5.3693,
      "mean_token_accuracy": 0.167016638815403,
      "num_tokens": 23819685.0,
      "step": 13735
    },
    {
      "entropy": 5.419653844833374,
      "epoch": 1.069013810542696,
      "grad_norm": 1.5,
      "learning_rate": 0.0004890099731191592,
      "loss": 5.239,
      "mean_token_accuracy": 0.17011093348264694,
      "num_tokens": 23828733.0,
      "step": 13740
    },
    {
      "entropy": 5.601557350158691,
      "epoch": 1.0694028399144135,
      "grad_norm": 1.921875,
      "learning_rate": 0.000489001415350401,
      "loss": 5.4134,
      "mean_token_accuracy": 0.16911170482635499,
      "num_tokens": 23837095.0,
      "step": 13745
    },
    {
      "entropy": 5.624904584884644,
      "epoch": 1.069791869286131,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004889928543345118,
      "loss": 5.3659,
      "mean_token_accuracy": 0.16807382553815842,
      "num_tokens": 23846637.0,
      "step": 13750
    },
    {
      "entropy": 5.593227338790894,
      "epoch": 1.0701808986578487,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004889842900716217,
      "loss": 5.4405,
      "mean_token_accuracy": 0.16046826392412186,
      "num_tokens": 23856042.0,
      "step": 13755
    },
    {
      "entropy": 5.581457424163818,
      "epoch": 1.0705699280295662,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004889757225618606,
      "loss": 5.398,
      "mean_token_accuracy": 0.161847585439682,
      "num_tokens": 23865367.0,
      "step": 13760
    },
    {
      "entropy": 5.560931396484375,
      "epoch": 1.0709589574012839,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004889671518053584,
      "loss": 5.3748,
      "mean_token_accuracy": 0.16762278825044633,
      "num_tokens": 23874740.0,
      "step": 13765
    },
    {
      "entropy": 5.559179353713989,
      "epoch": 1.0713479867730014,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004889585778022453,
      "loss": 5.3194,
      "mean_token_accuracy": 0.17093719094991683,
      "num_tokens": 23884229.0,
      "step": 13770
    },
    {
      "entropy": 5.510400342941284,
      "epoch": 1.0717370161447188,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004889500005526514,
      "loss": 5.23,
      "mean_token_accuracy": 0.17975565642118455,
      "num_tokens": 23892149.0,
      "step": 13775
    },
    {
      "entropy": 5.644508028030396,
      "epoch": 1.0721260455164365,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004889414200567067,
      "loss": 5.4889,
      "mean_token_accuracy": 0.15467537343502044,
      "num_tokens": 23901802.0,
      "step": 13780
    },
    {
      "entropy": 5.592928695678711,
      "epoch": 1.072515074888154,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004889328363145415,
      "loss": 5.3306,
      "mean_token_accuracy": 0.17018810361623765,
      "num_tokens": 23910268.0,
      "step": 13785
    },
    {
      "entropy": 5.456088638305664,
      "epoch": 1.0729041042598717,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004889242493262859,
      "loss": 5.1646,
      "mean_token_accuracy": 0.17440763264894485,
      "num_tokens": 23918351.0,
      "step": 13790
    },
    {
      "entropy": 5.511342859268188,
      "epoch": 1.0732931336315892,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004889156590920704,
      "loss": 5.2312,
      "mean_token_accuracy": 0.17307026088237762,
      "num_tokens": 23927367.0,
      "step": 13795
    },
    {
      "entropy": 5.500313711166382,
      "epoch": 1.0736821630033067,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004889070656120253,
      "loss": 5.3071,
      "mean_token_accuracy": 0.17446416169404982,
      "num_tokens": 23936216.0,
      "step": 13800
    },
    {
      "entropy": 5.636788845062256,
      "epoch": 1.0740711923750244,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004888984688862809,
      "loss": 5.3835,
      "mean_token_accuracy": 0.16611143052577973,
      "num_tokens": 23944344.0,
      "step": 13805
    },
    {
      "entropy": 5.575056076049805,
      "epoch": 1.0744602217467418,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004888898689149677,
      "loss": 5.4094,
      "mean_token_accuracy": 0.15442738458514213,
      "num_tokens": 23953390.0,
      "step": 13810
    },
    {
      "entropy": 5.635413694381714,
      "epoch": 1.0748492511184595,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004888812656982162,
      "loss": 5.3982,
      "mean_token_accuracy": 0.1702481508255005,
      "num_tokens": 23962230.0,
      "step": 13815
    },
    {
      "entropy": 5.562955141067505,
      "epoch": 1.075238280490177,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004888726592361569,
      "loss": 5.2811,
      "mean_token_accuracy": 0.1708732083439827,
      "num_tokens": 23971627.0,
      "step": 13820
    },
    {
      "entropy": 5.496527624130249,
      "epoch": 1.0756273098618945,
      "grad_norm": 1.625,
      "learning_rate": 0.0004888640495289204,
      "loss": 5.282,
      "mean_token_accuracy": 0.17555358856916428,
      "num_tokens": 23980222.0,
      "step": 13825
    },
    {
      "entropy": 5.5275712490081785,
      "epoch": 1.0760163392336122,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004888554365766374,
      "loss": 5.3474,
      "mean_token_accuracy": 0.1654688909649849,
      "num_tokens": 23988803.0,
      "step": 13830
    },
    {
      "entropy": 5.544680786132813,
      "epoch": 1.0764053686053296,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004888468203794385,
      "loss": 5.3409,
      "mean_token_accuracy": 0.16459826976060868,
      "num_tokens": 23997664.0,
      "step": 13835
    },
    {
      "entropy": 5.532573986053467,
      "epoch": 1.0767943979770473,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004888382009374545,
      "loss": 5.3239,
      "mean_token_accuracy": 0.17208846807479858,
      "num_tokens": 24006641.0,
      "step": 13840
    },
    {
      "entropy": 5.502842378616333,
      "epoch": 1.0771834273487648,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004888295782508161,
      "loss": 5.2714,
      "mean_token_accuracy": 0.16638420522212982,
      "num_tokens": 24014852.0,
      "step": 13845
    },
    {
      "entropy": 5.573114013671875,
      "epoch": 1.0775724567204823,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004888209523196542,
      "loss": 5.2946,
      "mean_token_accuracy": 0.1803032010793686,
      "num_tokens": 24023441.0,
      "step": 13850
    },
    {
      "entropy": 5.620143985748291,
      "epoch": 1.0779614860922,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004888123231440997,
      "loss": 5.3048,
      "mean_token_accuracy": 0.17385974228382112,
      "num_tokens": 24032120.0,
      "step": 13855
    },
    {
      "entropy": 5.561611080169678,
      "epoch": 1.0783505154639175,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004888036907242834,
      "loss": 5.3607,
      "mean_token_accuracy": 0.16757990270853043,
      "num_tokens": 24040264.0,
      "step": 13860
    },
    {
      "entropy": 5.642135095596314,
      "epoch": 1.0787395448356352,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004887950550603366,
      "loss": 5.4115,
      "mean_token_accuracy": 0.16733597069978715,
      "num_tokens": 24049194.0,
      "step": 13865
    },
    {
      "entropy": 5.55846471786499,
      "epoch": 1.0791285742073526,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004887864161523901,
      "loss": 5.2732,
      "mean_token_accuracy": 0.17170953005552292,
      "num_tokens": 24058004.0,
      "step": 13870
    },
    {
      "entropy": 5.4610028743743895,
      "epoch": 1.0795176035790701,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004887777740005749,
      "loss": 5.2269,
      "mean_token_accuracy": 0.17321476042270662,
      "num_tokens": 24066871.0,
      "step": 13875
    },
    {
      "entropy": 5.558225488662719,
      "epoch": 1.0799066329507878,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004887691286050224,
      "loss": 5.3347,
      "mean_token_accuracy": 0.16767856925725938,
      "num_tokens": 24075409.0,
      "step": 13880
    },
    {
      "entropy": 5.588159608840942,
      "epoch": 1.0802956623225053,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004887604799658635,
      "loss": 5.2789,
      "mean_token_accuracy": 0.16992218941450118,
      "num_tokens": 24083983.0,
      "step": 13885
    },
    {
      "entropy": 5.536437511444092,
      "epoch": 1.080684691694223,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004887518280832295,
      "loss": 5.2984,
      "mean_token_accuracy": 0.17002954483032226,
      "num_tokens": 24092225.0,
      "step": 13890
    },
    {
      "entropy": 5.474335479736328,
      "epoch": 1.0810737210659405,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004887431729572519,
      "loss": 5.2265,
      "mean_token_accuracy": 0.18012141734361647,
      "num_tokens": 24100045.0,
      "step": 13895
    },
    {
      "entropy": 5.488373041152954,
      "epoch": 1.081462750437658,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004887345145880617,
      "loss": 5.2988,
      "mean_token_accuracy": 0.1758161962032318,
      "num_tokens": 24108540.0,
      "step": 13900
    },
    {
      "entropy": 5.590416717529297,
      "epoch": 1.0818517798093756,
      "grad_norm": 1.5,
      "learning_rate": 0.0004887258529757904,
      "loss": 5.3874,
      "mean_token_accuracy": 0.16554994285106658,
      "num_tokens": 24117819.0,
      "step": 13905
    },
    {
      "entropy": 5.5390746116638185,
      "epoch": 1.082240809181093,
      "grad_norm": 1.75,
      "learning_rate": 0.0004887171881205696,
      "loss": 5.1909,
      "mean_token_accuracy": 0.17586574107408523,
      "num_tokens": 24127000.0,
      "step": 13910
    },
    {
      "entropy": 5.528246259689331,
      "epoch": 1.0826298385528108,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004887085200225306,
      "loss": 5.3117,
      "mean_token_accuracy": 0.16844677478075026,
      "num_tokens": 24135404.0,
      "step": 13915
    },
    {
      "entropy": 5.449680376052856,
      "epoch": 1.0830188679245283,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004886998486818049,
      "loss": 5.1416,
      "mean_token_accuracy": 0.1834289848804474,
      "num_tokens": 24143944.0,
      "step": 13920
    },
    {
      "entropy": 5.510674047470093,
      "epoch": 1.083407897296246,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004886911740985242,
      "loss": 5.2696,
      "mean_token_accuracy": 0.17339787483215333,
      "num_tokens": 24153097.0,
      "step": 13925
    },
    {
      "entropy": 5.570836067199707,
      "epoch": 1.0837969266679635,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048868249627282,
      "loss": 5.4283,
      "mean_token_accuracy": 0.16437558084726334,
      "num_tokens": 24162499.0,
      "step": 13930
    },
    {
      "entropy": 5.550439834594727,
      "epoch": 1.084185956039681,
      "grad_norm": 2.109375,
      "learning_rate": 0.000488673815204824,
      "loss": 5.2494,
      "mean_token_accuracy": 0.1778493732213974,
      "num_tokens": 24170543.0,
      "step": 13935
    },
    {
      "entropy": 5.554481840133667,
      "epoch": 1.0845749854113986,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004886651308946681,
      "loss": 5.3223,
      "mean_token_accuracy": 0.16854367554187774,
      "num_tokens": 24180871.0,
      "step": 13940
    },
    {
      "entropy": 5.505338907241821,
      "epoch": 1.084964014783116,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004886564433424839,
      "loss": 5.2427,
      "mean_token_accuracy": 0.17851458340883256,
      "num_tokens": 24189075.0,
      "step": 13945
    },
    {
      "entropy": 5.594527339935302,
      "epoch": 1.0853530441548336,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004886477525484032,
      "loss": 5.3365,
      "mean_token_accuracy": 0.1770291283726692,
      "num_tokens": 24197686.0,
      "step": 13950
    },
    {
      "entropy": 5.437126588821411,
      "epoch": 1.0857420735265513,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004886390585125579,
      "loss": 5.2393,
      "mean_token_accuracy": 0.16652272939682006,
      "num_tokens": 24207093.0,
      "step": 13955
    },
    {
      "entropy": 5.627029943466186,
      "epoch": 1.0861311028982688,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004886303612350799,
      "loss": 5.3385,
      "mean_token_accuracy": 0.16592801958322526,
      "num_tokens": 24216447.0,
      "step": 13960
    },
    {
      "entropy": 5.685516357421875,
      "epoch": 1.0865201322699864,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004886216607161013,
      "loss": 5.3846,
      "mean_token_accuracy": 0.16547441035509108,
      "num_tokens": 24225041.0,
      "step": 13965
    },
    {
      "entropy": 5.608265018463134,
      "epoch": 1.086909161641704,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004886129569557538,
      "loss": 5.323,
      "mean_token_accuracy": 0.1650741770863533,
      "num_tokens": 24233686.0,
      "step": 13970
    },
    {
      "entropy": 5.504385042190552,
      "epoch": 1.0872981910134216,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004886042499541699,
      "loss": 5.3341,
      "mean_token_accuracy": 0.17844736874103545,
      "num_tokens": 24242553.0,
      "step": 13975
    },
    {
      "entropy": 5.5233869552612305,
      "epoch": 1.087687220385139,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004885955397114813,
      "loss": 5.2467,
      "mean_token_accuracy": 0.17321914732456206,
      "num_tokens": 24251658.0,
      "step": 13980
    },
    {
      "entropy": 5.637213277816772,
      "epoch": 1.0880762497568566,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004885868262278205,
      "loss": 5.3671,
      "mean_token_accuracy": 0.16639908850193025,
      "num_tokens": 24260397.0,
      "step": 13985
    },
    {
      "entropy": 5.4556975841522215,
      "epoch": 1.0884652791285743,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004885781095033195,
      "loss": 5.2728,
      "mean_token_accuracy": 0.17130138874053955,
      "num_tokens": 24268789.0,
      "step": 13990
    },
    {
      "entropy": 5.530916452407837,
      "epoch": 1.0888543085002917,
      "grad_norm": 1.625,
      "learning_rate": 0.0004885693895381108,
      "loss": 5.3914,
      "mean_token_accuracy": 0.16509933024644852,
      "num_tokens": 24278050.0,
      "step": 13995
    },
    {
      "entropy": 5.6599335193634035,
      "epoch": 1.0892433378720092,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004885606663323263,
      "loss": 5.3238,
      "mean_token_accuracy": 0.17296088933944703,
      "num_tokens": 24286477.0,
      "step": 14000
    },
    {
      "entropy": 5.442564010620117,
      "epoch": 1.089632367243727,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004885519398860987,
      "loss": 5.195,
      "mean_token_accuracy": 0.17921269088983535,
      "num_tokens": 24295704.0,
      "step": 14005
    },
    {
      "entropy": 5.424733352661133,
      "epoch": 1.0900213966154444,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004885432101995604,
      "loss": 5.2377,
      "mean_token_accuracy": 0.17668494284152986,
      "num_tokens": 24304132.0,
      "step": 14010
    },
    {
      "entropy": 5.582837057113648,
      "epoch": 1.090410425987162,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004885344772728436,
      "loss": 5.3355,
      "mean_token_accuracy": 0.16859273314476014,
      "num_tokens": 24312864.0,
      "step": 14015
    },
    {
      "entropy": 5.586115026473999,
      "epoch": 1.0907994553588796,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004885257411060812,
      "loss": 5.2558,
      "mean_token_accuracy": 0.17344018965959548,
      "num_tokens": 24321745.0,
      "step": 14020
    },
    {
      "entropy": 5.488592529296875,
      "epoch": 1.0911884847305973,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004885170016994053,
      "loss": 5.2762,
      "mean_token_accuracy": 0.16774366945028304,
      "num_tokens": 24331061.0,
      "step": 14025
    },
    {
      "entropy": 5.45854082107544,
      "epoch": 1.0915775141023147,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004885082590529489,
      "loss": 5.2893,
      "mean_token_accuracy": 0.17020780444145203,
      "num_tokens": 24339935.0,
      "step": 14030
    },
    {
      "entropy": 5.5863566398620605,
      "epoch": 1.0919665434740322,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004884995131668445,
      "loss": 5.3168,
      "mean_token_accuracy": 0.17068814039230346,
      "num_tokens": 24348876.0,
      "step": 14035
    },
    {
      "entropy": 5.534871435165405,
      "epoch": 1.09235557284575,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004884907640412248,
      "loss": 5.3285,
      "mean_token_accuracy": 0.1675461784005165,
      "num_tokens": 24357094.0,
      "step": 14040
    },
    {
      "entropy": 5.484497451782227,
      "epoch": 1.0927446022174674,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004884820116762225,
      "loss": 5.2759,
      "mean_token_accuracy": 0.1699037507176399,
      "num_tokens": 24366320.0,
      "step": 14045
    },
    {
      "entropy": 5.483876132965088,
      "epoch": 1.093133631589185,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004884732560719706,
      "loss": 5.1662,
      "mean_token_accuracy": 0.17697618752717972,
      "num_tokens": 24374390.0,
      "step": 14050
    },
    {
      "entropy": 5.5002411842346195,
      "epoch": 1.0935226609609026,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004884644972286017,
      "loss": 5.2996,
      "mean_token_accuracy": 0.17514243274927138,
      "num_tokens": 24382804.0,
      "step": 14055
    },
    {
      "entropy": 5.550761318206787,
      "epoch": 1.09391169033262,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004884557351462488,
      "loss": 5.2963,
      "mean_token_accuracy": 0.17130143493413924,
      "num_tokens": 24391352.0,
      "step": 14060
    },
    {
      "entropy": 5.5826669216156,
      "epoch": 1.0943007197043377,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004884469698250449,
      "loss": 5.3724,
      "mean_token_accuracy": 0.16699749678373338,
      "num_tokens": 24400348.0,
      "step": 14065
    },
    {
      "entropy": 5.508634042739868,
      "epoch": 1.0946897490760552,
      "grad_norm": 1.6640625,
      "learning_rate": 0.000488438201265123,
      "loss": 5.2093,
      "mean_token_accuracy": 0.1748551085591316,
      "num_tokens": 24409628.0,
      "step": 14070
    },
    {
      "entropy": 5.560254621505737,
      "epoch": 1.095078778447773,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004884294294666161,
      "loss": 5.3047,
      "mean_token_accuracy": 0.16498639136552812,
      "num_tokens": 24418379.0,
      "step": 14075
    },
    {
      "entropy": 5.5908811569213865,
      "epoch": 1.0954678078194904,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004884206544296573,
      "loss": 5.4404,
      "mean_token_accuracy": 0.16364881992340088,
      "num_tokens": 24427719.0,
      "step": 14080
    },
    {
      "entropy": 5.611962556838989,
      "epoch": 1.0958568371912079,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004884118761543797,
      "loss": 5.3801,
      "mean_token_accuracy": 0.16736531555652617,
      "num_tokens": 24435938.0,
      "step": 14085
    },
    {
      "entropy": 5.541747426986694,
      "epoch": 1.0962458665629256,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004884030946409167,
      "loss": 5.2742,
      "mean_token_accuracy": 0.17245944291353227,
      "num_tokens": 24444286.0,
      "step": 14090
    },
    {
      "entropy": 5.480231475830078,
      "epoch": 1.096634895934643,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004883943098894013,
      "loss": 5.2212,
      "mean_token_accuracy": 0.1797743782401085,
      "num_tokens": 24452546.0,
      "step": 14095
    },
    {
      "entropy": 5.530523443222046,
      "epoch": 1.0970239253063607,
      "grad_norm": 1.203125,
      "learning_rate": 0.000488385521899967,
      "loss": 5.3538,
      "mean_token_accuracy": 0.16807045191526412,
      "num_tokens": 24460765.0,
      "step": 14100
    },
    {
      "entropy": 5.508094596862793,
      "epoch": 1.0974129546780782,
      "grad_norm": 1.34375,
      "learning_rate": 0.000488376730672747,
      "loss": 5.2161,
      "mean_token_accuracy": 0.1738041266798973,
      "num_tokens": 24470018.0,
      "step": 14105
    },
    {
      "entropy": 5.511989879608154,
      "epoch": 1.0978019840497957,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004883679362078746,
      "loss": 5.1706,
      "mean_token_accuracy": 0.17320188879966736,
      "num_tokens": 24478882.0,
      "step": 14110
    },
    {
      "entropy": 5.533711957931518,
      "epoch": 1.0981910134215134,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004883591385054836,
      "loss": 5.3484,
      "mean_token_accuracy": 0.1637899562716484,
      "num_tokens": 24487801.0,
      "step": 14115
    },
    {
      "entropy": 5.5117205619812015,
      "epoch": 1.0985800427932308,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004883503375657072,
      "loss": 5.3151,
      "mean_token_accuracy": 0.16933534294366837,
      "num_tokens": 24496082.0,
      "step": 14120
    },
    {
      "entropy": 5.497330617904663,
      "epoch": 1.0989690721649485,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004883415333886789,
      "loss": 5.2197,
      "mean_token_accuracy": 0.18041443526744844,
      "num_tokens": 24504664.0,
      "step": 14125
    },
    {
      "entropy": 5.512371015548706,
      "epoch": 1.099358101536666,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004883327259745325,
      "loss": 5.2592,
      "mean_token_accuracy": 0.17645634114742278,
      "num_tokens": 24512797.0,
      "step": 14130
    },
    {
      "entropy": 5.48935284614563,
      "epoch": 1.0997471309083835,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004883239153234015,
      "loss": 5.2479,
      "mean_token_accuracy": 0.1720615178346634,
      "num_tokens": 24520941.0,
      "step": 14135
    },
    {
      "entropy": 5.573987817764282,
      "epoch": 1.1001361602801012,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004883151014354197,
      "loss": 5.3462,
      "mean_token_accuracy": 0.17067505270242692,
      "num_tokens": 24530087.0,
      "step": 14140
    },
    {
      "entropy": 5.512291717529297,
      "epoch": 1.1005251896518187,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004883062843107207,
      "loss": 5.2979,
      "mean_token_accuracy": 0.16665249168872834,
      "num_tokens": 24538593.0,
      "step": 14145
    },
    {
      "entropy": 5.629961347579956,
      "epoch": 1.1009142190235364,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004882974639494383,
      "loss": 5.3658,
      "mean_token_accuracy": 0.1617593139410019,
      "num_tokens": 24547158.0,
      "step": 14150
    },
    {
      "entropy": 5.558168029785156,
      "epoch": 1.1013032483952538,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004882886403517065,
      "loss": 5.3639,
      "mean_token_accuracy": 0.1636526554822922,
      "num_tokens": 24556266.0,
      "step": 14155
    },
    {
      "entropy": 5.560190868377686,
      "epoch": 1.1016922777669713,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004882798135176589,
      "loss": 5.4088,
      "mean_token_accuracy": 0.16728487610816956,
      "num_tokens": 24565122.0,
      "step": 14160
    },
    {
      "entropy": 5.498580121994019,
      "epoch": 1.102081307138689,
      "grad_norm": 1.609375,
      "learning_rate": 0.00048827098344742973,
      "loss": 5.2514,
      "mean_token_accuracy": 0.1830837294459343,
      "num_tokens": 24573944.0,
      "step": 14165
    },
    {
      "entropy": 5.534769296646118,
      "epoch": 1.1024703365104065,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004882621501411528,
      "loss": 5.4202,
      "mean_token_accuracy": 0.16984884440898895,
      "num_tokens": 24582233.0,
      "step": 14170
    },
    {
      "entropy": 5.539715147018432,
      "epoch": 1.1028593658821242,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004882533135989622,
      "loss": 5.2744,
      "mean_token_accuracy": 0.17063359171152115,
      "num_tokens": 24590855.0,
      "step": 14175
    },
    {
      "entropy": 5.675771284103393,
      "epoch": 1.1032483952538417,
      "grad_norm": 2.171875,
      "learning_rate": 0.000488244473820992,
      "loss": 5.3394,
      "mean_token_accuracy": 0.15959832072257996,
      "num_tokens": 24599877.0,
      "step": 14180
    },
    {
      "entropy": 5.5097575187683105,
      "epoch": 1.1036374246255591,
      "grad_norm": 2.09375,
      "learning_rate": 0.00048823563080737634,
      "loss": 5.2142,
      "mean_token_accuracy": 0.18019939064979554,
      "num_tokens": 24608450.0,
      "step": 14185
    },
    {
      "entropy": 5.470316219329834,
      "epoch": 1.1040264539972768,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004882267845582493,
      "loss": 5.3138,
      "mean_token_accuracy": 0.1766979455947876,
      "num_tokens": 24618018.0,
      "step": 14190
    },
    {
      "entropy": 5.555082178115844,
      "epoch": 1.1044154833689943,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00048821793507374526,
      "loss": 5.2955,
      "mean_token_accuracy": 0.16985682994127274,
      "num_tokens": 24626388.0,
      "step": 14195
    },
    {
      "entropy": 5.606828451156616,
      "epoch": 1.104804512740712,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004882090823539984,
      "loss": 5.3944,
      "mean_token_accuracy": 0.16266640722751619,
      "num_tokens": 24635284.0,
      "step": 14200
    },
    {
      "entropy": 5.566130876541138,
      "epoch": 1.1051935421124295,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004882002263991431,
      "loss": 5.2479,
      "mean_token_accuracy": 0.17859185934066774,
      "num_tokens": 24644297.0,
      "step": 14205
    },
    {
      "entropy": 5.407494163513183,
      "epoch": 1.105582571484147,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00048819136720931364,
      "loss": 5.1151,
      "mean_token_accuracy": 0.17962373048067093,
      "num_tokens": 24652965.0,
      "step": 14210
    },
    {
      "entropy": 5.3790207862854,
      "epoch": 1.1059716008558647,
      "grad_norm": 1.59375,
      "learning_rate": 0.00048818250478464457,
      "loss": 5.2001,
      "mean_token_accuracy": 0.17739752382040025,
      "num_tokens": 24661034.0,
      "step": 14215
    },
    {
      "entropy": 5.523409461975097,
      "epoch": 1.1063606302275821,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004881736391252703,
      "loss": 5.2802,
      "mean_token_accuracy": 0.17606231421232224,
      "num_tokens": 24669831.0,
      "step": 14220
    },
    {
      "entropy": 5.578432893753051,
      "epoch": 1.1067496595992998,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004881647702313253,
      "loss": 5.3555,
      "mean_token_accuracy": 0.16432537734508515,
      "num_tokens": 24678491.0,
      "step": 14225
    },
    {
      "entropy": 5.539248991012573,
      "epoch": 1.1071386889710173,
      "grad_norm": 1.859375,
      "learning_rate": 0.00048815589810294427,
      "loss": 5.293,
      "mean_token_accuracy": 0.173760125041008,
      "num_tokens": 24687308.0,
      "step": 14230
    },
    {
      "entropy": 5.497361469268799,
      "epoch": 1.1075277183427348,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00048814702274026174,
      "loss": 5.2635,
      "mean_token_accuracy": 0.17785624116659166,
      "num_tokens": 24696080.0,
      "step": 14235
    },
    {
      "entropy": 5.502005767822266,
      "epoch": 1.1079167477144525,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004881381441434124,
      "loss": 5.3353,
      "mean_token_accuracy": 0.1668131798505783,
      "num_tokens": 24704561.0,
      "step": 14240
    },
    {
      "entropy": 5.609838771820068,
      "epoch": 1.10830577708617,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000488129262312531,
      "loss": 5.3316,
      "mean_token_accuracy": 0.16675823777914048,
      "num_tokens": 24713403.0,
      "step": 14245
    },
    {
      "entropy": 5.606697750091553,
      "epoch": 1.1086948064578876,
      "grad_norm": 1.59375,
      "learning_rate": 0.00048812037724775217,
      "loss": 5.3491,
      "mean_token_accuracy": 0.1648279294371605,
      "num_tokens": 24722031.0,
      "step": 14250
    },
    {
      "entropy": 5.531491279602051,
      "epoch": 1.1090838358296051,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004881114889492109,
      "loss": 5.2941,
      "mean_token_accuracy": 0.17142567038536072,
      "num_tokens": 24731176.0,
      "step": 14255
    },
    {
      "entropy": 5.54642653465271,
      "epoch": 1.1094728652013226,
      "grad_norm": 2.03125,
      "learning_rate": 0.00048810259741704197,
      "loss": 5.344,
      "mean_token_accuracy": 0.16734461337327958,
      "num_tokens": 24740037.0,
      "step": 14260
    },
    {
      "entropy": 5.627961587905884,
      "epoch": 1.1098618945730403,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004880937026513803,
      "loss": 5.3994,
      "mean_token_accuracy": 0.16683503091335297,
      "num_tokens": 24748778.0,
      "step": 14265
    },
    {
      "entropy": 5.589363050460816,
      "epoch": 1.1102509239447578,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004880848046523609,
      "loss": 5.2414,
      "mean_token_accuracy": 0.17778424322605133,
      "num_tokens": 24757226.0,
      "step": 14270
    },
    {
      "entropy": 5.475401449203491,
      "epoch": 1.1106399533164755,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00048807590342011875,
      "loss": 5.2835,
      "mean_token_accuracy": 0.17758438140153884,
      "num_tokens": 24765734.0,
      "step": 14275
    },
    {
      "entropy": 5.456071424484253,
      "epoch": 1.111028982688193,
      "grad_norm": 1.75,
      "learning_rate": 0.00048806699895478895,
      "loss": 5.2599,
      "mean_token_accuracy": 0.1693689614534378,
      "num_tokens": 24773773.0,
      "step": 14280
    },
    {
      "entropy": 5.522355270385742,
      "epoch": 1.1114180120599104,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004880580912565065,
      "loss": 5.29,
      "mean_token_accuracy": 0.17302699387073517,
      "num_tokens": 24782563.0,
      "step": 14285
    },
    {
      "entropy": 5.5858558177947994,
      "epoch": 1.1118070414316281,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004880491803254067,
      "loss": 5.3871,
      "mean_token_accuracy": 0.16795666962862016,
      "num_tokens": 24791492.0,
      "step": 14290
    },
    {
      "entropy": 5.493483591079712,
      "epoch": 1.1121960708033456,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004880402661616246,
      "loss": 5.1918,
      "mean_token_accuracy": 0.17944994568824768,
      "num_tokens": 24800229.0,
      "step": 14295
    },
    {
      "entropy": 5.486127138137817,
      "epoch": 1.1125851001750633,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004880313487652956,
      "loss": 5.2656,
      "mean_token_accuracy": 0.17114198803901673,
      "num_tokens": 24809454.0,
      "step": 14300
    },
    {
      "entropy": 5.483691024780273,
      "epoch": 1.1129741295467808,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00048802242813655503,
      "loss": 5.2698,
      "mean_token_accuracy": 0.17059182226657868,
      "num_tokens": 24817956.0,
      "step": 14305
    },
    {
      "entropy": 5.473423290252685,
      "epoch": 1.1133631589184985,
      "grad_norm": 1.5546875,
      "learning_rate": 0.000488013504275538,
      "loss": 5.1898,
      "mean_token_accuracy": 0.17933583855628968,
      "num_tokens": 24826288.0,
      "step": 14310
    },
    {
      "entropy": 5.65793080329895,
      "epoch": 1.113752188290216,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004880045771823802,
      "loss": 5.3618,
      "mean_token_accuracy": 0.16813699007034302,
      "num_tokens": 24834720.0,
      "step": 14315
    },
    {
      "entropy": 5.522952938079834,
      "epoch": 1.1141412176619334,
      "grad_norm": 1.5,
      "learning_rate": 0.00048799564685721695,
      "loss": 5.3064,
      "mean_token_accuracy": 0.17395334541797638,
      "num_tokens": 24843000.0,
      "step": 14320
    },
    {
      "entropy": 5.525973606109619,
      "epoch": 1.1145302470336511,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004879867133001837,
      "loss": 5.3015,
      "mean_token_accuracy": 0.16755297631025315,
      "num_tokens": 24851060.0,
      "step": 14325
    },
    {
      "entropy": 5.454747104644776,
      "epoch": 1.1149192764053686,
      "grad_norm": 1.75,
      "learning_rate": 0.0004879777765114162,
      "loss": 5.2085,
      "mean_token_accuracy": 0.1771145850419998,
      "num_tokens": 24859913.0,
      "step": 14330
    },
    {
      "entropy": 5.609213256835938,
      "epoch": 1.115308305777086,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00048796883649104996,
      "loss": 5.4089,
      "mean_token_accuracy": 0.16464822739362717,
      "num_tokens": 24869648.0,
      "step": 14335
    },
    {
      "entropy": 5.571947288513184,
      "epoch": 1.1156973351488038,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004879598932392206,
      "loss": 5.3292,
      "mean_token_accuracy": 0.17350474745035172,
      "num_tokens": 24878813.0,
      "step": 14340
    },
    {
      "entropy": 5.5525922775268555,
      "epoch": 1.1160863645205212,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048795094675606364,
      "loss": 5.3082,
      "mean_token_accuracy": 0.16897975355386735,
      "num_tokens": 24887714.0,
      "step": 14345
    },
    {
      "entropy": 5.571045875549316,
      "epoch": 1.116475393892239,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004879419970417152,
      "loss": 5.2703,
      "mean_token_accuracy": 0.172675059735775,
      "num_tokens": 24895961.0,
      "step": 14350
    },
    {
      "entropy": 5.55090298652649,
      "epoch": 1.1168644232639564,
      "grad_norm": 1.375,
      "learning_rate": 0.00048793304409631077,
      "loss": 5.301,
      "mean_token_accuracy": 0.1732762098312378,
      "num_tokens": 24904873.0,
      "step": 14355
    },
    {
      "entropy": 5.455278730392456,
      "epoch": 1.117253452635674,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004879240879199863,
      "loss": 5.1985,
      "mean_token_accuracy": 0.1836355596780777,
      "num_tokens": 24913486.0,
      "step": 14360
    },
    {
      "entropy": 5.500343942642212,
      "epoch": 1.1176424820073916,
      "grad_norm": 1.53125,
      "learning_rate": 0.00048791512851287786,
      "loss": 5.2854,
      "mean_token_accuracy": 0.1700807571411133,
      "num_tokens": 24921943.0,
      "step": 14365
    },
    {
      "entropy": 5.606582069396973,
      "epoch": 1.118031511379109,
      "grad_norm": 1.5625,
      "learning_rate": 0.00048790616587512107,
      "loss": 5.3794,
      "mean_token_accuracy": 0.16794403791427612,
      "num_tokens": 24930081.0,
      "step": 14370
    },
    {
      "entropy": 5.658608436584473,
      "epoch": 1.1184205407508268,
      "grad_norm": 2.0,
      "learning_rate": 0.0004878972000068521,
      "loss": 5.3183,
      "mean_token_accuracy": 0.1709811270236969,
      "num_tokens": 24938985.0,
      "step": 14375
    },
    {
      "entropy": 5.435786533355713,
      "epoch": 1.1188095701225442,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048788823090820707,
      "loss": 5.1481,
      "mean_token_accuracy": 0.18129525780677797,
      "num_tokens": 24947497.0,
      "step": 14380
    },
    {
      "entropy": 5.550786018371582,
      "epoch": 1.1191985994942617,
      "grad_norm": 1.6875,
      "learning_rate": 0.00048787925857932194,
      "loss": 5.3791,
      "mean_token_accuracy": 0.1630413845181465,
      "num_tokens": 24956263.0,
      "step": 14385
    },
    {
      "entropy": 5.653364372253418,
      "epoch": 1.1195876288659794,
      "grad_norm": 1.625,
      "learning_rate": 0.00048787028302033287,
      "loss": 5.4578,
      "mean_token_accuracy": 0.16429930925369263,
      "num_tokens": 24965646.0,
      "step": 14390
    },
    {
      "entropy": 5.579395961761475,
      "epoch": 1.1199766582376969,
      "grad_norm": 1.578125,
      "learning_rate": 0.00048786130423137606,
      "loss": 5.3332,
      "mean_token_accuracy": 0.17410521060228348,
      "num_tokens": 24973318.0,
      "step": 14395
    },
    {
      "entropy": 5.562328958511353,
      "epoch": 1.1203656876094146,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004878523222125879,
      "loss": 5.414,
      "mean_token_accuracy": 0.17157252877950668,
      "num_tokens": 24982600.0,
      "step": 14400
    },
    {
      "entropy": 5.674773550033569,
      "epoch": 1.120754716981132,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004878433369641044,
      "loss": 5.3954,
      "mean_token_accuracy": 0.16661668866872786,
      "num_tokens": 24990386.0,
      "step": 14405
    },
    {
      "entropy": 5.459098386764526,
      "epoch": 1.1211437463528497,
      "grad_norm": 1.5,
      "learning_rate": 0.0004878343484860621,
      "loss": 5.1695,
      "mean_token_accuracy": 0.17870809882879257,
      "num_tokens": 24999064.0,
      "step": 14410
    },
    {
      "entropy": 5.5220627784729,
      "epoch": 1.1215327757245672,
      "grad_norm": 1.75,
      "learning_rate": 0.0004878253567785972,
      "loss": 5.34,
      "mean_token_accuracy": 0.17166408747434617,
      "num_tokens": 25008448.0,
      "step": 14415
    },
    {
      "entropy": 5.417818355560303,
      "epoch": 1.1219218050962847,
      "grad_norm": 1.5625,
      "learning_rate": 0.00048781636184184644,
      "loss": 5.1728,
      "mean_token_accuracy": 0.18554657995700835,
      "num_tokens": 25016722.0,
      "step": 14420
    },
    {
      "entropy": 5.433246850967407,
      "epoch": 1.1223108344680024,
      "grad_norm": 1.53125,
      "learning_rate": 0.000487807363675946,
      "loss": 5.2511,
      "mean_token_accuracy": 0.17665931433439255,
      "num_tokens": 25026263.0,
      "step": 14425
    },
    {
      "entropy": 5.568420124053955,
      "epoch": 1.1226998638397199,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004877983622810326,
      "loss": 5.2876,
      "mean_token_accuracy": 0.17753019481897353,
      "num_tokens": 25035202.0,
      "step": 14430
    },
    {
      "entropy": 5.610867261886597,
      "epoch": 1.1230888932114376,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004877893576572427,
      "loss": 5.3618,
      "mean_token_accuracy": 0.17109027057886123,
      "num_tokens": 25044309.0,
      "step": 14435
    },
    {
      "entropy": 5.577419710159302,
      "epoch": 1.123477922583155,
      "grad_norm": 1.421875,
      "learning_rate": 0.000487780349804713,
      "loss": 5.3391,
      "mean_token_accuracy": 0.17795383781194687,
      "num_tokens": 25053044.0,
      "step": 14440
    },
    {
      "entropy": 5.42353835105896,
      "epoch": 1.1238669519548725,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004877713387235802,
      "loss": 5.2455,
      "mean_token_accuracy": 0.17633127123117448,
      "num_tokens": 25061869.0,
      "step": 14445
    },
    {
      "entropy": 5.548495864868164,
      "epoch": 1.1242559813265902,
      "grad_norm": 1.5625,
      "learning_rate": 0.000487762324413981,
      "loss": 5.4236,
      "mean_token_accuracy": 0.1668712839484215,
      "num_tokens": 25070022.0,
      "step": 14450
    },
    {
      "entropy": 5.605709362030029,
      "epoch": 1.1246450106983077,
      "grad_norm": 1.578125,
      "learning_rate": 0.00048775330687605213,
      "loss": 5.3215,
      "mean_token_accuracy": 0.16896626800298692,
      "num_tokens": 25078604.0,
      "step": 14455
    },
    {
      "entropy": 5.547206926345825,
      "epoch": 1.1250340400700254,
      "grad_norm": 1.75,
      "learning_rate": 0.0004877442861099305,
      "loss": 5.1993,
      "mean_token_accuracy": 0.1766393095254898,
      "num_tokens": 25087256.0,
      "step": 14460
    },
    {
      "entropy": 5.542341804504394,
      "epoch": 1.1254230694417429,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004877352621157529,
      "loss": 5.2768,
      "mean_token_accuracy": 0.17198699563741685,
      "num_tokens": 25095593.0,
      "step": 14465
    },
    {
      "entropy": 5.560934972763062,
      "epoch": 1.1258120988134603,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00048772623489365635,
      "loss": 5.3341,
      "mean_token_accuracy": 0.17415791749954224,
      "num_tokens": 25104852.0,
      "step": 14470
    },
    {
      "entropy": 5.539960718154907,
      "epoch": 1.126201128185178,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004877172044437777,
      "loss": 5.3123,
      "mean_token_accuracy": 0.17148227542638778,
      "num_tokens": 25114453.0,
      "step": 14475
    },
    {
      "entropy": 5.502276039123535,
      "epoch": 1.1265901575568955,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00048770817076625416,
      "loss": 5.3048,
      "mean_token_accuracy": 0.17388371527194976,
      "num_tokens": 25123733.0,
      "step": 14480
    },
    {
      "entropy": 5.490148496627808,
      "epoch": 1.126979186928613,
      "grad_norm": 1.875,
      "learning_rate": 0.00048769913386122253,
      "loss": 5.2261,
      "mean_token_accuracy": 0.17898911088705063,
      "num_tokens": 25132572.0,
      "step": 14485
    },
    {
      "entropy": 5.45403938293457,
      "epoch": 1.1273682163003307,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004876900937288202,
      "loss": 5.2571,
      "mean_token_accuracy": 0.17854359596967698,
      "num_tokens": 25141494.0,
      "step": 14490
    },
    {
      "entropy": 5.433139324188232,
      "epoch": 1.1277572456720482,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00048768105036918426,
      "loss": 5.2129,
      "mean_token_accuracy": 0.1757223427295685,
      "num_tokens": 25150186.0,
      "step": 14495
    },
    {
      "entropy": 5.563412809371949,
      "epoch": 1.1281462750437659,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00048767200378245187,
      "loss": 5.3077,
      "mean_token_accuracy": 0.16997037082910538,
      "num_tokens": 25159251.0,
      "step": 14500
    },
    {
      "entropy": 5.578856515884399,
      "epoch": 1.1285353044154833,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00048766295396876025,
      "loss": 5.3103,
      "mean_token_accuracy": 0.17455729991197586,
      "num_tokens": 25167780.0,
      "step": 14505
    },
    {
      "entropy": 5.528367328643799,
      "epoch": 1.128924333787201,
      "grad_norm": 1.71875,
      "learning_rate": 0.00048765390092824683,
      "loss": 5.2092,
      "mean_token_accuracy": 0.18061490654945372,
      "num_tokens": 25175358.0,
      "step": 14510
    },
    {
      "entropy": 5.4804182052612305,
      "epoch": 1.1293133631589185,
      "grad_norm": 1.65625,
      "learning_rate": 0.000487644844661049,
      "loss": 5.2571,
      "mean_token_accuracy": 0.16878363937139512,
      "num_tokens": 25184247.0,
      "step": 14515
    },
    {
      "entropy": 5.45369668006897,
      "epoch": 1.129702392530636,
      "grad_norm": 1.53125,
      "learning_rate": 0.000487635785167304,
      "loss": 5.1891,
      "mean_token_accuracy": 0.1781545549631119,
      "num_tokens": 25192311.0,
      "step": 14520
    },
    {
      "entropy": 5.438900709152222,
      "epoch": 1.1300914219023537,
      "grad_norm": 2.0,
      "learning_rate": 0.00048762672244714953,
      "loss": 5.2317,
      "mean_token_accuracy": 0.1773073673248291,
      "num_tokens": 25200861.0,
      "step": 14525
    },
    {
      "entropy": 5.553978395462036,
      "epoch": 1.1304804512740712,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004876176565007229,
      "loss": 5.3739,
      "mean_token_accuracy": 0.17362226843833922,
      "num_tokens": 25208938.0,
      "step": 14530
    },
    {
      "entropy": 5.598419666290283,
      "epoch": 1.1308694806457888,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00048760858732816174,
      "loss": 5.2705,
      "mean_token_accuracy": 0.1793088510632515,
      "num_tokens": 25216997.0,
      "step": 14535
    },
    {
      "entropy": 5.50171012878418,
      "epoch": 1.1312585100175063,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004875995149296037,
      "loss": 5.1898,
      "mean_token_accuracy": 0.1773835837841034,
      "num_tokens": 25225205.0,
      "step": 14540
    },
    {
      "entropy": 5.476949024200439,
      "epoch": 1.1316475393892238,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004875904393051864,
      "loss": 5.2522,
      "mean_token_accuracy": 0.1770787462592125,
      "num_tokens": 25233807.0,
      "step": 14545
    },
    {
      "entropy": 5.501559400558472,
      "epoch": 1.1320365687609415,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004875813604550476,
      "loss": 5.2673,
      "mean_token_accuracy": 0.169118233025074,
      "num_tokens": 25241715.0,
      "step": 14550
    },
    {
      "entropy": 5.589974689483642,
      "epoch": 1.132425598132659,
      "grad_norm": 1.515625,
      "learning_rate": 0.00048757227837932494,
      "loss": 5.3951,
      "mean_token_accuracy": 0.1651352420449257,
      "num_tokens": 25250418.0,
      "step": 14555
    },
    {
      "entropy": 5.5561353206634525,
      "epoch": 1.1328146275043767,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004875631930781563,
      "loss": 5.2879,
      "mean_token_accuracy": 0.1675178065896034,
      "num_tokens": 25259164.0,
      "step": 14560
    },
    {
      "entropy": 5.5276085376739506,
      "epoch": 1.1332036568760941,
      "grad_norm": 1.515625,
      "learning_rate": 0.00048755410455167967,
      "loss": 5.3302,
      "mean_token_accuracy": 0.17142942398786545,
      "num_tokens": 25267902.0,
      "step": 14565
    },
    {
      "entropy": 5.522876644134522,
      "epoch": 1.1335926862478116,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004875450128000326,
      "loss": 5.2757,
      "mean_token_accuracy": 0.16805899292230606,
      "num_tokens": 25275991.0,
      "step": 14570
    },
    {
      "entropy": 5.52580041885376,
      "epoch": 1.1339817156195293,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004875359178233534,
      "loss": 5.2765,
      "mean_token_accuracy": 0.17196872234344482,
      "num_tokens": 25284454.0,
      "step": 14575
    },
    {
      "entropy": 5.604035186767578,
      "epoch": 1.1343707449912468,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00048752681962177985,
      "loss": 5.3838,
      "mean_token_accuracy": 0.16469692885875703,
      "num_tokens": 25292867.0,
      "step": 14580
    },
    {
      "entropy": 5.543754291534424,
      "epoch": 1.1347597743629645,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048751771819545013,
      "loss": 5.2724,
      "mean_token_accuracy": 0.1736215054988861,
      "num_tokens": 25301486.0,
      "step": 14585
    },
    {
      "entropy": 5.497570991516113,
      "epoch": 1.135148803734682,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00048750861354450236,
      "loss": 5.2591,
      "mean_token_accuracy": 0.17186922580003738,
      "num_tokens": 25310800.0,
      "step": 14590
    },
    {
      "entropy": 5.563240098953247,
      "epoch": 1.1355378331063994,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004874995056690746,
      "loss": 5.3134,
      "mean_token_accuracy": 0.17563033998012542,
      "num_tokens": 25319063.0,
      "step": 14595
    },
    {
      "entropy": 5.639831590652466,
      "epoch": 1.1359268624781171,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000487490394569305,
      "loss": 5.364,
      "mean_token_accuracy": 0.16975239664316177,
      "num_tokens": 25327817.0,
      "step": 14600
    },
    {
      "entropy": 5.545086622238159,
      "epoch": 1.1363158918498346,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004874812802453319,
      "loss": 5.1958,
      "mean_token_accuracy": 0.1755492463707924,
      "num_tokens": 25335753.0,
      "step": 14605
    },
    {
      "entropy": 5.520526933670044,
      "epoch": 1.1367049212215523,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004874721626972936,
      "loss": 5.3828,
      "mean_token_accuracy": 0.1666047140955925,
      "num_tokens": 25344514.0,
      "step": 14610
    },
    {
      "entropy": 5.554873514175415,
      "epoch": 1.1370939505932698,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004874630419253284,
      "loss": 5.3061,
      "mean_token_accuracy": 0.17575568705797195,
      "num_tokens": 25353003.0,
      "step": 14615
    },
    {
      "entropy": 5.550239753723145,
      "epoch": 1.1374829799649873,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00048745391792957474,
      "loss": 5.2843,
      "mean_token_accuracy": 0.1726512610912323,
      "num_tokens": 25361599.0,
      "step": 14620
    },
    {
      "entropy": 5.561546659469604,
      "epoch": 1.137872009336705,
      "grad_norm": 1.53125,
      "learning_rate": 0.00048744479071017097,
      "loss": 5.3446,
      "mean_token_accuracy": 0.1625458464026451,
      "num_tokens": 25370566.0,
      "step": 14625
    },
    {
      "entropy": 5.575703144073486,
      "epoch": 1.1382610387084224,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004874356602672556,
      "loss": 5.3284,
      "mean_token_accuracy": 0.17115466445684432,
      "num_tokens": 25379589.0,
      "step": 14630
    },
    {
      "entropy": 5.550799322128296,
      "epoch": 1.1386500680801401,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004874265266009673,
      "loss": 5.2631,
      "mean_token_accuracy": 0.17970920950174332,
      "num_tokens": 25387481.0,
      "step": 14635
    },
    {
      "entropy": 5.496348476409912,
      "epoch": 1.1390390974518576,
      "grad_norm": 1.5,
      "learning_rate": 0.0004874173897114445,
      "loss": 5.2754,
      "mean_token_accuracy": 0.1765144497156143,
      "num_tokens": 25395551.0,
      "step": 14640
    },
    {
      "entropy": 5.548038721084595,
      "epoch": 1.1394281268235753,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00048740824959882603,
      "loss": 5.3307,
      "mean_token_accuracy": 0.1669141337275505,
      "num_tokens": 25404303.0,
      "step": 14645
    },
    {
      "entropy": 5.560009288787842,
      "epoch": 1.1398171561952928,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004873991062632504,
      "loss": 5.3259,
      "mean_token_accuracy": 0.16363555788993836,
      "num_tokens": 25412749.0,
      "step": 14650
    },
    {
      "entropy": 5.536481618881226,
      "epoch": 1.1402061855670103,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048738995970485635,
      "loss": 5.2707,
      "mean_token_accuracy": 0.17230881601572037,
      "num_tokens": 25421075.0,
      "step": 14655
    },
    {
      "entropy": 5.568193054199218,
      "epoch": 1.140595214938728,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004873808099237827,
      "loss": 5.3903,
      "mean_token_accuracy": 0.16511864215135574,
      "num_tokens": 25429750.0,
      "step": 14660
    },
    {
      "entropy": 5.561778736114502,
      "epoch": 1.1409842443104454,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004873716569201684,
      "loss": 5.2846,
      "mean_token_accuracy": 0.17441572099924088,
      "num_tokens": 25437686.0,
      "step": 14665
    },
    {
      "entropy": 5.516694164276123,
      "epoch": 1.141373273682163,
      "grad_norm": 1.625,
      "learning_rate": 0.00048736250069415213,
      "loss": 5.2944,
      "mean_token_accuracy": 0.17406139373779297,
      "num_tokens": 25446050.0,
      "step": 14670
    },
    {
      "entropy": 5.4934525966644285,
      "epoch": 1.1417623030538806,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004873533412458729,
      "loss": 5.3043,
      "mean_token_accuracy": 0.1747289478778839,
      "num_tokens": 25455134.0,
      "step": 14675
    },
    {
      "entropy": 5.469335746765137,
      "epoch": 1.142151332425598,
      "grad_norm": 1.875,
      "learning_rate": 0.00048734417857546986,
      "loss": 5.1802,
      "mean_token_accuracy": 0.17695782482624053,
      "num_tokens": 25463258.0,
      "step": 14680
    },
    {
      "entropy": 5.520489025115967,
      "epoch": 1.1425403617973158,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004873350126830818,
      "loss": 5.2667,
      "mean_token_accuracy": 0.1761361226439476,
      "num_tokens": 25472085.0,
      "step": 14685
    },
    {
      "entropy": 5.560411596298218,
      "epoch": 1.1429293911690332,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004873258435688479,
      "loss": 5.278,
      "mean_token_accuracy": 0.17668423801660538,
      "num_tokens": 25481233.0,
      "step": 14690
    },
    {
      "entropy": 5.53919677734375,
      "epoch": 1.143318420540751,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004873166712329073,
      "loss": 5.2557,
      "mean_token_accuracy": 0.18044089078903197,
      "num_tokens": 25490354.0,
      "step": 14695
    },
    {
      "entropy": 5.53734655380249,
      "epoch": 1.1437074499124684,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00048730749567539914,
      "loss": 5.3147,
      "mean_token_accuracy": 0.17236823439598084,
      "num_tokens": 25498467.0,
      "step": 14700
    },
    {
      "entropy": 5.552510499954224,
      "epoch": 1.144096479284186,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048729831689646257,
      "loss": 5.3095,
      "mean_token_accuracy": 0.17018110156059266,
      "num_tokens": 25507543.0,
      "step": 14705
    },
    {
      "entropy": 5.487798404693604,
      "epoch": 1.1444855086559036,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00048728913489623705,
      "loss": 5.1964,
      "mean_token_accuracy": 0.17983233481645583,
      "num_tokens": 25516070.0,
      "step": 14710
    },
    {
      "entropy": 5.534903049468994,
      "epoch": 1.144874538027621,
      "grad_norm": 1.5,
      "learning_rate": 0.0004872799496748618,
      "loss": 5.3293,
      "mean_token_accuracy": 0.17175545543432236,
      "num_tokens": 25525216.0,
      "step": 14715
    },
    {
      "entropy": 5.579160737991333,
      "epoch": 1.1452635673993385,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004872707612324761,
      "loss": 5.3436,
      "mean_token_accuracy": 0.1700536385178566,
      "num_tokens": 25533822.0,
      "step": 14720
    },
    {
      "entropy": 5.660882616043091,
      "epoch": 1.1456525967710562,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004872615695692196,
      "loss": 5.3753,
      "mean_token_accuracy": 0.16802987456321716,
      "num_tokens": 25541580.0,
      "step": 14725
    },
    {
      "entropy": 5.484453248977661,
      "epoch": 1.1460416261427737,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004872523746852315,
      "loss": 5.1773,
      "mean_token_accuracy": 0.18269246816635132,
      "num_tokens": 25549860.0,
      "step": 14730
    },
    {
      "entropy": 5.52522029876709,
      "epoch": 1.1464306555144914,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00048724317658065146,
      "loss": 5.3442,
      "mean_token_accuracy": 0.16894873529672622,
      "num_tokens": 25559060.0,
      "step": 14735
    },
    {
      "entropy": 5.63541898727417,
      "epoch": 1.146819684886209,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00048723397525561916,
      "loss": 5.3874,
      "mean_token_accuracy": 0.16443771123886108,
      "num_tokens": 25567769.0,
      "step": 14740
    },
    {
      "entropy": 5.571277809143067,
      "epoch": 1.1472087142579266,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00048722477071027394,
      "loss": 5.3654,
      "mean_token_accuracy": 0.1711871936917305,
      "num_tokens": 25577443.0,
      "step": 14745
    },
    {
      "entropy": 5.569605827331543,
      "epoch": 1.147597743629644,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00048721556294475574,
      "loss": 5.2731,
      "mean_token_accuracy": 0.18160714358091354,
      "num_tokens": 25586283.0,
      "step": 14750
    },
    {
      "entropy": 5.551212406158447,
      "epoch": 1.1479867730013615,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00048720635195920403,
      "loss": 5.3753,
      "mean_token_accuracy": 0.17017149925231934,
      "num_tokens": 25594881.0,
      "step": 14755
    },
    {
      "entropy": 5.50782527923584,
      "epoch": 1.1483758023730792,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004871971377537588,
      "loss": 5.2436,
      "mean_token_accuracy": 0.1762669488787651,
      "num_tokens": 25603738.0,
      "step": 14760
    },
    {
      "entropy": 5.583714962005615,
      "epoch": 1.1487648317447967,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004871879203285597,
      "loss": 5.298,
      "mean_token_accuracy": 0.17087433785200118,
      "num_tokens": 25611780.0,
      "step": 14765
    },
    {
      "entropy": 5.606764793395996,
      "epoch": 1.1491538611165142,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00048717869968374655,
      "loss": 5.3617,
      "mean_token_accuracy": 0.17412520051002503,
      "num_tokens": 25620489.0,
      "step": 14770
    },
    {
      "entropy": 5.522389936447143,
      "epoch": 1.1495428904882319,
      "grad_norm": 1.65625,
      "learning_rate": 0.00048716947581945954,
      "loss": 5.3226,
      "mean_token_accuracy": 0.1770711585879326,
      "num_tokens": 25629113.0,
      "step": 14775
    },
    {
      "entropy": 5.464048576354981,
      "epoch": 1.1499319198599494,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004871602487358383,
      "loss": 5.2657,
      "mean_token_accuracy": 0.1724536433815956,
      "num_tokens": 25637313.0,
      "step": 14780
    },
    {
      "entropy": 5.5214461326599125,
      "epoch": 1.150320949231667,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000487151018433023,
      "loss": 5.3576,
      "mean_token_accuracy": 0.16645698100328446,
      "num_tokens": 25646522.0,
      "step": 14785
    },
    {
      "entropy": 5.5282148838043215,
      "epoch": 1.1507099786033845,
      "grad_norm": 1.515625,
      "learning_rate": 0.00048714178491115367,
      "loss": 5.2638,
      "mean_token_accuracy": 0.17899245619773865,
      "num_tokens": 25656058.0,
      "step": 14790
    },
    {
      "entropy": 5.586183786392212,
      "epoch": 1.1510990079751022,
      "grad_norm": 1.640625,
      "learning_rate": 0.00048713254817037046,
      "loss": 5.3711,
      "mean_token_accuracy": 0.17049205154180527,
      "num_tokens": 25665072.0,
      "step": 14795
    },
    {
      "entropy": 5.607844257354737,
      "epoch": 1.1514880373468197,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004871233082108135,
      "loss": 5.4,
      "mean_token_accuracy": 0.1711244061589241,
      "num_tokens": 25673877.0,
      "step": 14800
    },
    {
      "entropy": 5.582855701446533,
      "epoch": 1.1518770667185372,
      "grad_norm": 1.5,
      "learning_rate": 0.0004871140650326228,
      "loss": 5.2596,
      "mean_token_accuracy": 0.16818197667598725,
      "num_tokens": 25683238.0,
      "step": 14805
    },
    {
      "entropy": 5.579865217208862,
      "epoch": 1.1522660960902549,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004871048186359389,
      "loss": 5.3595,
      "mean_token_accuracy": 0.16232425570487977,
      "num_tokens": 25692654.0,
      "step": 14810
    },
    {
      "entropy": 5.650218391418457,
      "epoch": 1.1526551254619724,
      "grad_norm": 1.890625,
      "learning_rate": 0.000487095569020902,
      "loss": 5.3455,
      "mean_token_accuracy": 0.16748398393392563,
      "num_tokens": 25701681.0,
      "step": 14815
    },
    {
      "entropy": 5.566237306594848,
      "epoch": 1.1530441548336898,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004870863161876524,
      "loss": 5.26,
      "mean_token_accuracy": 0.17767535746097565,
      "num_tokens": 25709848.0,
      "step": 14820
    },
    {
      "entropy": 5.603987169265747,
      "epoch": 1.1534331842054075,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004870770601363305,
      "loss": 5.4981,
      "mean_token_accuracy": 0.16868632584810256,
      "num_tokens": 25717993.0,
      "step": 14825
    },
    {
      "entropy": 5.587001705169678,
      "epoch": 1.153822213577125,
      "grad_norm": 1.578125,
      "learning_rate": 0.00048706780086707675,
      "loss": 5.3808,
      "mean_token_accuracy": 0.16602419465780258,
      "num_tokens": 25726379.0,
      "step": 14830
    },
    {
      "entropy": 5.568260240554809,
      "epoch": 1.1542112429488427,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004870585383800317,
      "loss": 5.2919,
      "mean_token_accuracy": 0.1783009275794029,
      "num_tokens": 25734968.0,
      "step": 14835
    },
    {
      "entropy": 5.586294174194336,
      "epoch": 1.1546002723205602,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00048704927267533585,
      "loss": 5.3333,
      "mean_token_accuracy": 0.1721921294927597,
      "num_tokens": 25743315.0,
      "step": 14840
    },
    {
      "entropy": 5.529369783401489,
      "epoch": 1.1549893016922779,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00048704000375312975,
      "loss": 5.2378,
      "mean_token_accuracy": 0.17137032002210617,
      "num_tokens": 25752396.0,
      "step": 14845
    },
    {
      "entropy": 5.614004516601563,
      "epoch": 1.1553783310639953,
      "grad_norm": 1.65625,
      "learning_rate": 0.00048703073161355414,
      "loss": 5.3635,
      "mean_token_accuracy": 0.1709185004234314,
      "num_tokens": 25761142.0,
      "step": 14850
    },
    {
      "entropy": 5.493571329116821,
      "epoch": 1.1557673604357128,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004870214562567497,
      "loss": 5.24,
      "mean_token_accuracy": 0.17272724956274033,
      "num_tokens": 25769514.0,
      "step": 14855
    },
    {
      "entropy": 5.521751976013183,
      "epoch": 1.1561563898074305,
      "grad_norm": 1.5,
      "learning_rate": 0.000487012177682857,
      "loss": 5.2379,
      "mean_token_accuracy": 0.17054518312215805,
      "num_tokens": 25777408.0,
      "step": 14860
    },
    {
      "entropy": 5.4967021465301515,
      "epoch": 1.156545419179148,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004870028958920171,
      "loss": 5.259,
      "mean_token_accuracy": 0.1768237456679344,
      "num_tokens": 25785646.0,
      "step": 14865
    },
    {
      "entropy": 5.544138431549072,
      "epoch": 1.1569344485508655,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004869936108843706,
      "loss": 5.3095,
      "mean_token_accuracy": 0.17224012017250062,
      "num_tokens": 25794007.0,
      "step": 14870
    },
    {
      "entropy": 5.52902307510376,
      "epoch": 1.1573234779225832,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00048698432266005854,
      "loss": 5.2489,
      "mean_token_accuracy": 0.17344386130571365,
      "num_tokens": 25802765.0,
      "step": 14875
    },
    {
      "entropy": 5.590705871582031,
      "epoch": 1.1577125072943006,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004869750312192217,
      "loss": 5.3886,
      "mean_token_accuracy": 0.16790701746940612,
      "num_tokens": 25811214.0,
      "step": 14880
    },
    {
      "entropy": 5.591983604431152,
      "epoch": 1.1581015366660183,
      "grad_norm": 1.671875,
      "learning_rate": 0.00048696573656200123,
      "loss": 5.3748,
      "mean_token_accuracy": 0.16566506028175354,
      "num_tokens": 25820350.0,
      "step": 14885
    },
    {
      "entropy": 5.553336143493652,
      "epoch": 1.1584905660377358,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00048695643868853806,
      "loss": 5.2935,
      "mean_token_accuracy": 0.18036726415157317,
      "num_tokens": 25828088.0,
      "step": 14890
    },
    {
      "entropy": 5.5192883014678955,
      "epoch": 1.1588795954094535,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004869471375989733,
      "loss": 5.1599,
      "mean_token_accuracy": 0.181265027821064,
      "num_tokens": 25835909.0,
      "step": 14895
    },
    {
      "entropy": 5.452424430847168,
      "epoch": 1.159268624781171,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004869378332934481,
      "loss": 5.2502,
      "mean_token_accuracy": 0.17498199492692948,
      "num_tokens": 25844360.0,
      "step": 14900
    },
    {
      "entropy": 5.651251745223999,
      "epoch": 1.1596576541528885,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048692852577210356,
      "loss": 5.3665,
      "mean_token_accuracy": 0.16494977325201035,
      "num_tokens": 25853877.0,
      "step": 14905
    },
    {
      "entropy": 5.559357833862305,
      "epoch": 1.1600466835246062,
      "grad_norm": 1.484375,
      "learning_rate": 0.000486919215035081,
      "loss": 5.2664,
      "mean_token_accuracy": 0.175713013112545,
      "num_tokens": 25862200.0,
      "step": 14910
    },
    {
      "entropy": 5.501112365722657,
      "epoch": 1.1604357128963236,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004869099010825217,
      "loss": 5.2218,
      "mean_token_accuracy": 0.17241571992635726,
      "num_tokens": 25871112.0,
      "step": 14915
    },
    {
      "entropy": 5.54647421836853,
      "epoch": 1.1608247422680413,
      "grad_norm": 2.09375,
      "learning_rate": 0.00048690058391456684,
      "loss": 5.3928,
      "mean_token_accuracy": 0.1711567685008049,
      "num_tokens": 25880135.0,
      "step": 14920
    },
    {
      "entropy": 5.55512957572937,
      "epoch": 1.1612137716397588,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004868912635313579,
      "loss": 5.2908,
      "mean_token_accuracy": 0.17410618662834168,
      "num_tokens": 25888609.0,
      "step": 14925
    },
    {
      "entropy": 5.525064182281494,
      "epoch": 1.1616028010114763,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004868819399330364,
      "loss": 5.3393,
      "mean_token_accuracy": 0.16897253096103668,
      "num_tokens": 25897254.0,
      "step": 14930
    },
    {
      "entropy": 5.549260139465332,
      "epoch": 1.161991830383194,
      "grad_norm": 1.5,
      "learning_rate": 0.0004868726131197436,
      "loss": 5.2525,
      "mean_token_accuracy": 0.18024791330099105,
      "num_tokens": 25905705.0,
      "step": 14935
    },
    {
      "entropy": 5.559958553314209,
      "epoch": 1.1623808597549115,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004868632830916212,
      "loss": 5.3027,
      "mean_token_accuracy": 0.17200768291950225,
      "num_tokens": 25914066.0,
      "step": 14940
    },
    {
      "entropy": 5.540140581130982,
      "epoch": 1.1627698891266292,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004868539498488106,
      "loss": 5.3123,
      "mean_token_accuracy": 0.17349241077899932,
      "num_tokens": 25922949.0,
      "step": 14945
    },
    {
      "entropy": 5.418867826461792,
      "epoch": 1.1631589184983466,
      "grad_norm": 1.5,
      "learning_rate": 0.0004868446133914536,
      "loss": 5.1618,
      "mean_token_accuracy": 0.1807738795876503,
      "num_tokens": 25931465.0,
      "step": 14950
    },
    {
      "entropy": 5.532193613052368,
      "epoch": 1.163547947870064,
      "grad_norm": 2.59375,
      "learning_rate": 0.00048683527371969185,
      "loss": 5.3353,
      "mean_token_accuracy": 0.16627979129552842,
      "num_tokens": 25941014.0,
      "step": 14955
    },
    {
      "entropy": 5.4955915927886965,
      "epoch": 1.1639369772417818,
      "grad_norm": 1.65625,
      "learning_rate": 0.00048682593083366685,
      "loss": 5.1995,
      "mean_token_accuracy": 0.17250516265630722,
      "num_tokens": 25950275.0,
      "step": 14960
    },
    {
      "entropy": 5.443646240234375,
      "epoch": 1.1643260066134993,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00048681658473352055,
      "loss": 5.3219,
      "mean_token_accuracy": 0.16984528303146362,
      "num_tokens": 25959138.0,
      "step": 14965
    },
    {
      "entropy": 5.578936243057251,
      "epoch": 1.164715035985217,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004868072354193947,
      "loss": 5.2048,
      "mean_token_accuracy": 0.17416904121637344,
      "num_tokens": 25967221.0,
      "step": 14970
    },
    {
      "entropy": 5.571450471878052,
      "epoch": 1.1651040653569344,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004867978828914312,
      "loss": 5.3198,
      "mean_token_accuracy": 0.1704607143998146,
      "num_tokens": 25976194.0,
      "step": 14975
    },
    {
      "entropy": 5.500711488723755,
      "epoch": 1.165493094728652,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004867885271497719,
      "loss": 5.286,
      "mean_token_accuracy": 0.17450777888298036,
      "num_tokens": 25985493.0,
      "step": 14980
    },
    {
      "entropy": 5.583955335617065,
      "epoch": 1.1658821241003696,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004867791681945588,
      "loss": 5.2399,
      "mean_token_accuracy": 0.17097237408161164,
      "num_tokens": 25994213.0,
      "step": 14985
    },
    {
      "entropy": 5.493848705291748,
      "epoch": 1.166271153472087,
      "grad_norm": 1.546875,
      "learning_rate": 0.00048676980602593395,
      "loss": 5.2218,
      "mean_token_accuracy": 0.1793581262230873,
      "num_tokens": 26003164.0,
      "step": 14990
    },
    {
      "entropy": 5.572969436645508,
      "epoch": 1.1666601828438048,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00048676044064403936,
      "loss": 5.3306,
      "mean_token_accuracy": 0.17016486078500748,
      "num_tokens": 26012049.0,
      "step": 14995
    },
    {
      "entropy": 5.645019149780273,
      "epoch": 1.1670492122155223,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004867510720490171,
      "loss": 5.4245,
      "mean_token_accuracy": 0.16362651586532592,
      "num_tokens": 26020466.0,
      "step": 15000
    },
    {
      "epoch": 1.1670492122155223,
      "eval_entropy": 5.440363345178305,
      "eval_loss": 5.406432151794434,
      "eval_mean_token_accuracy": 0.17719353199062493,
      "eval_num_tokens": 26020466.0,
      "eval_runtime": 28.5321,
      "eval_samples_per_second": 1456.537,
      "eval_steps_per_second": 182.076,
      "step": 15000
    },
    {
      "entropy": 5.528584003448486,
      "epoch": 1.1674382415872397,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004867417002410094,
      "loss": 5.2198,
      "mean_token_accuracy": 0.17334816604852676,
      "num_tokens": 26028815.0,
      "step": 15005
    },
    {
      "entropy": 5.58254861831665,
      "epoch": 1.1678272709589574,
      "grad_norm": 1.59375,
      "learning_rate": 0.00048673232522015845,
      "loss": 5.4157,
      "mean_token_accuracy": 0.16548077017068863,
      "num_tokens": 26037759.0,
      "step": 15010
    },
    {
      "entropy": 5.569189119338989,
      "epoch": 1.168216300330675,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004867229469866064,
      "loss": 5.3363,
      "mean_token_accuracy": 0.16419618874788283,
      "num_tokens": 26046349.0,
      "step": 15015
    },
    {
      "entropy": 5.553538084030151,
      "epoch": 1.1686053297023926,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004867135655404956,
      "loss": 5.3218,
      "mean_token_accuracy": 0.17581884264945985,
      "num_tokens": 26055265.0,
      "step": 15020
    },
    {
      "entropy": 5.570549488067627,
      "epoch": 1.16899435907411,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00048670418088196844,
      "loss": 5.3144,
      "mean_token_accuracy": 0.17261690348386766,
      "num_tokens": 26064335.0,
      "step": 15025
    },
    {
      "entropy": 5.560314559936524,
      "epoch": 1.1693833884458276,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004866947930111674,
      "loss": 5.2445,
      "mean_token_accuracy": 0.1767016112804413,
      "num_tokens": 26073044.0,
      "step": 15030
    },
    {
      "entropy": 5.531310939788819,
      "epoch": 1.1697724178175453,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00048668540192823467,
      "loss": 5.2614,
      "mean_token_accuracy": 0.17961026281118392,
      "num_tokens": 26081361.0,
      "step": 15035
    },
    {
      "entropy": 5.520599699020385,
      "epoch": 1.1701614471892627,
      "grad_norm": 1.7578125,
      "learning_rate": 0.000486676007633313,
      "loss": 5.2636,
      "mean_token_accuracy": 0.17193735688924788,
      "num_tokens": 26089740.0,
      "step": 15040
    },
    {
      "entropy": 5.579999923706055,
      "epoch": 1.1705504765609804,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004866666101265448,
      "loss": 5.3566,
      "mean_token_accuracy": 0.17004507035017014,
      "num_tokens": 26098906.0,
      "step": 15045
    },
    {
      "entropy": 5.565162658691406,
      "epoch": 1.170939505932698,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048665720940807274,
      "loss": 5.3229,
      "mean_token_accuracy": 0.16735642552375793,
      "num_tokens": 26107978.0,
      "step": 15050
    },
    {
      "entropy": 5.539587879180909,
      "epoch": 1.1713285353044154,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048664780547803935,
      "loss": 5.2691,
      "mean_token_accuracy": 0.16775070279836654,
      "num_tokens": 26116756.0,
      "step": 15055
    },
    {
      "entropy": 5.625945043563843,
      "epoch": 1.171717564676133,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004866383983365873,
      "loss": 5.3314,
      "mean_token_accuracy": 0.16616996079683305,
      "num_tokens": 26125078.0,
      "step": 15060
    },
    {
      "entropy": 5.590082883834839,
      "epoch": 1.1721065940478506,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00048662898798385956,
      "loss": 5.3755,
      "mean_token_accuracy": 0.16891691833734512,
      "num_tokens": 26134149.0,
      "step": 15065
    },
    {
      "entropy": 5.55862078666687,
      "epoch": 1.1724956234195683,
      "grad_norm": 1.734375,
      "learning_rate": 0.00048661957441999875,
      "loss": 5.2231,
      "mean_token_accuracy": 0.17921175807714462,
      "num_tokens": 26142789.0,
      "step": 15070
    },
    {
      "entropy": 5.540366888046265,
      "epoch": 1.1728846527912857,
      "grad_norm": 2.34375,
      "learning_rate": 0.00048661015764514765,
      "loss": 5.3997,
      "mean_token_accuracy": 0.16913245618343353,
      "num_tokens": 26151189.0,
      "step": 15075
    },
    {
      "entropy": 5.532864904403686,
      "epoch": 1.1732736821630034,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00048660073765944927,
      "loss": 5.2869,
      "mean_token_accuracy": 0.1758386954665184,
      "num_tokens": 26159495.0,
      "step": 15080
    },
    {
      "entropy": 5.489394569396973,
      "epoch": 1.173662711534721,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004865913144630465,
      "loss": 5.2141,
      "mean_token_accuracy": 0.17278929948806762,
      "num_tokens": 26167804.0,
      "step": 15085
    },
    {
      "entropy": 5.528975915908814,
      "epoch": 1.1740517409064384,
      "grad_norm": 1.515625,
      "learning_rate": 0.00048658188805608227,
      "loss": 5.2723,
      "mean_token_accuracy": 0.17362222373485564,
      "num_tokens": 26176240.0,
      "step": 15090
    },
    {
      "entropy": 5.5410560131072994,
      "epoch": 1.174440770278156,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004865724584386998,
      "loss": 5.2688,
      "mean_token_accuracy": 0.1735546037554741,
      "num_tokens": 26184580.0,
      "step": 15095
    },
    {
      "entropy": 5.5709363460540775,
      "epoch": 1.1748297996498736,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004865630256110418,
      "loss": 5.3421,
      "mean_token_accuracy": 0.17168836742639543,
      "num_tokens": 26193756.0,
      "step": 15100
    },
    {
      "entropy": 5.613444709777832,
      "epoch": 1.175218829021591,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00048655358957325174,
      "loss": 5.3278,
      "mean_token_accuracy": 0.17838720381259918,
      "num_tokens": 26201818.0,
      "step": 15105
    },
    {
      "entropy": 5.6102183818817135,
      "epoch": 1.1756078583933087,
      "grad_norm": 1.703125,
      "learning_rate": 0.00048654415032547277,
      "loss": 5.3963,
      "mean_token_accuracy": 0.15847362875938414,
      "num_tokens": 26211817.0,
      "step": 15110
    },
    {
      "entropy": 5.541627645492554,
      "epoch": 1.1759968877650262,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048653470786784794,
      "loss": 5.2459,
      "mean_token_accuracy": 0.1706738367676735,
      "num_tokens": 26220280.0,
      "step": 15115
    },
    {
      "entropy": 5.454172658920288,
      "epoch": 1.176385917136744,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004865252622005206,
      "loss": 5.2304,
      "mean_token_accuracy": 0.18413861244916915,
      "num_tokens": 26228719.0,
      "step": 15120
    },
    {
      "entropy": 5.595516157150269,
      "epoch": 1.1767749465084614,
      "grad_norm": 1.734375,
      "learning_rate": 0.00048651581332363405,
      "loss": 5.307,
      "mean_token_accuracy": 0.1710882231593132,
      "num_tokens": 26237462.0,
      "step": 15125
    },
    {
      "entropy": 5.616157579421997,
      "epoch": 1.177163975880179,
      "grad_norm": 1.546875,
      "learning_rate": 0.00048650636123733176,
      "loss": 5.4087,
      "mean_token_accuracy": 0.17296858727931977,
      "num_tokens": 26245747.0,
      "step": 15130
    },
    {
      "entropy": 5.611859369277954,
      "epoch": 1.1775530052518965,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004864969059417571,
      "loss": 5.2976,
      "mean_token_accuracy": 0.1728380501270294,
      "num_tokens": 26254977.0,
      "step": 15135
    },
    {
      "entropy": 5.596359348297119,
      "epoch": 1.177942034623614,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004864874474370534,
      "loss": 5.4187,
      "mean_token_accuracy": 0.1656192198395729,
      "num_tokens": 26264298.0,
      "step": 15140
    },
    {
      "entropy": 5.561619901657105,
      "epoch": 1.1783310639953317,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00048647798572336445,
      "loss": 5.2331,
      "mean_token_accuracy": 0.17204071581363678,
      "num_tokens": 26273454.0,
      "step": 15145
    },
    {
      "entropy": 5.590262460708618,
      "epoch": 1.1787200933670492,
      "grad_norm": 1.609375,
      "learning_rate": 0.00048646852080083355,
      "loss": 5.3009,
      "mean_token_accuracy": 0.17017791122198106,
      "num_tokens": 26282584.0,
      "step": 15150
    },
    {
      "entropy": 5.549900007247925,
      "epoch": 1.1791091227387667,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004864590526696045,
      "loss": 5.3298,
      "mean_token_accuracy": 0.17289702147245406,
      "num_tokens": 26291633.0,
      "step": 15155
    },
    {
      "entropy": 5.594415664672852,
      "epoch": 1.1794981521104844,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048644958132982094,
      "loss": 5.3577,
      "mean_token_accuracy": 0.1653517961502075,
      "num_tokens": 26300808.0,
      "step": 15160
    },
    {
      "entropy": 5.53204312324524,
      "epoch": 1.1798871814822018,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048644010678162644,
      "loss": 5.2577,
      "mean_token_accuracy": 0.18163457065820693,
      "num_tokens": 26308976.0,
      "step": 15165
    },
    {
      "entropy": 5.523904848098755,
      "epoch": 1.1802762108539195,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004864306290251649,
      "loss": 5.3142,
      "mean_token_accuracy": 0.16775508150458335,
      "num_tokens": 26316980.0,
      "step": 15170
    },
    {
      "entropy": 5.5156854629516605,
      "epoch": 1.180665240225637,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00048642114806058014,
      "loss": 5.2201,
      "mean_token_accuracy": 0.17746393233537675,
      "num_tokens": 26325723.0,
      "step": 15175
    },
    {
      "entropy": 5.567081308364868,
      "epoch": 1.1810542695973547,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004864116638880159,
      "loss": 5.3091,
      "mean_token_accuracy": 0.17228560298681259,
      "num_tokens": 26334851.0,
      "step": 15180
    },
    {
      "entropy": 5.567082262039184,
      "epoch": 1.1814432989690722,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004864021765076162,
      "loss": 5.3196,
      "mean_token_accuracy": 0.16638908982276918,
      "num_tokens": 26343696.0,
      "step": 15185
    },
    {
      "entropy": 5.6168395519256595,
      "epoch": 1.1818323283407897,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00048639268591952494,
      "loss": 5.4422,
      "mean_token_accuracy": 0.1645281657576561,
      "num_tokens": 26353054.0,
      "step": 15190
    },
    {
      "entropy": 5.564056015014648,
      "epoch": 1.1822213577125074,
      "grad_norm": 1.828125,
      "learning_rate": 0.00048638319212388614,
      "loss": 5.2647,
      "mean_token_accuracy": 0.1795717552304268,
      "num_tokens": 26361484.0,
      "step": 15195
    },
    {
      "entropy": 5.545347452163696,
      "epoch": 1.1826103870842248,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004863736951208438,
      "loss": 5.3686,
      "mean_token_accuracy": 0.16862039715051652,
      "num_tokens": 26370402.0,
      "step": 15200
    },
    {
      "entropy": 5.559129047393799,
      "epoch": 1.1829994164559423,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00048636419491054217,
      "loss": 5.3974,
      "mean_token_accuracy": 0.16546053439378738,
      "num_tokens": 26378746.0,
      "step": 15205
    },
    {
      "entropy": 5.515748929977417,
      "epoch": 1.18338844582766,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004863546914931252,
      "loss": 5.2852,
      "mean_token_accuracy": 0.1783539429306984,
      "num_tokens": 26386984.0,
      "step": 15210
    },
    {
      "entropy": 5.635797166824341,
      "epoch": 1.1837774751993775,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004863451848687373,
      "loss": 5.4063,
      "mean_token_accuracy": 0.1640034019947052,
      "num_tokens": 26395712.0,
      "step": 15215
    },
    {
      "entropy": 5.6407819271087645,
      "epoch": 1.1841665045710952,
      "grad_norm": 4.65625,
      "learning_rate": 0.00048633567503752253,
      "loss": 5.337,
      "mean_token_accuracy": 0.169952854514122,
      "num_tokens": 26406421.0,
      "step": 15220
    },
    {
      "entropy": 5.655284357070923,
      "epoch": 1.1845555339428127,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004863261619996253,
      "loss": 5.2957,
      "mean_token_accuracy": 0.1705295443534851,
      "num_tokens": 26416032.0,
      "step": 15225
    },
    {
      "entropy": 5.508386564254761,
      "epoch": 1.1849445633145304,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004863166457551899,
      "loss": 5.2573,
      "mean_token_accuracy": 0.17223215699195862,
      "num_tokens": 26423903.0,
      "step": 15230
    },
    {
      "entropy": 5.558989000320435,
      "epoch": 1.1853335926862478,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004863071263043608,
      "loss": 5.3597,
      "mean_token_accuracy": 0.16596672236919402,
      "num_tokens": 26432631.0,
      "step": 15235
    },
    {
      "entropy": 5.5874451160430905,
      "epoch": 1.1857226220579653,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004862976036472823,
      "loss": 5.3508,
      "mean_token_accuracy": 0.167239710688591,
      "num_tokens": 26441529.0,
      "step": 15240
    },
    {
      "entropy": 5.567822599411011,
      "epoch": 1.186111651429683,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00048628807778409907,
      "loss": 5.2432,
      "mean_token_accuracy": 0.17956746220588685,
      "num_tokens": 26449928.0,
      "step": 15245
    },
    {
      "entropy": 5.555808877944946,
      "epoch": 1.1865006808014005,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004862785487149555,
      "loss": 5.2176,
      "mean_token_accuracy": 0.18223925828933715,
      "num_tokens": 26458065.0,
      "step": 15250
    },
    {
      "entropy": 5.48214635848999,
      "epoch": 1.186889710173118,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004862690164399963,
      "loss": 5.2736,
      "mean_token_accuracy": 0.1731280580163002,
      "num_tokens": 26466690.0,
      "step": 15255
    },
    {
      "entropy": 5.499134683609009,
      "epoch": 1.1872787395448356,
      "grad_norm": 1.65625,
      "learning_rate": 0.00048625948095936593,
      "loss": 5.3272,
      "mean_token_accuracy": 0.168271504342556,
      "num_tokens": 26476133.0,
      "step": 15260
    },
    {
      "entropy": 5.602310085296631,
      "epoch": 1.1876677689165531,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048624994227320923,
      "loss": 5.2873,
      "mean_token_accuracy": 0.170215705037117,
      "num_tokens": 26485151.0,
      "step": 15265
    },
    {
      "entropy": 5.532072067260742,
      "epoch": 1.1880567982882708,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00048624040038167094,
      "loss": 5.3355,
      "mean_token_accuracy": 0.16652823984622955,
      "num_tokens": 26494178.0,
      "step": 15270
    },
    {
      "entropy": 5.596423387527466,
      "epoch": 1.1884458276599883,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00048623085528489584,
      "loss": 5.3324,
      "mean_token_accuracy": 0.1781013563275337,
      "num_tokens": 26503233.0,
      "step": 15275
    },
    {
      "entropy": 5.4985284328460695,
      "epoch": 1.188834857031706,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00048622130698302863,
      "loss": 5.1961,
      "mean_token_accuracy": 0.17551186680793762,
      "num_tokens": 26512127.0,
      "step": 15280
    },
    {
      "entropy": 5.467780923843383,
      "epoch": 1.1892238864034235,
      "grad_norm": 1.59375,
      "learning_rate": 0.00048621175547621426,
      "loss": 5.1811,
      "mean_token_accuracy": 0.18775192201137542,
      "num_tokens": 26520239.0,
      "step": 15285
    },
    {
      "entropy": 5.492806911468506,
      "epoch": 1.189612915775141,
      "grad_norm": 1.5,
      "learning_rate": 0.0004862022007645978,
      "loss": 5.316,
      "mean_token_accuracy": 0.17370967119932174,
      "num_tokens": 26528946.0,
      "step": 15290
    },
    {
      "entropy": 5.596633815765381,
      "epoch": 1.1900019451468586,
      "grad_norm": 1.5625,
      "learning_rate": 0.00048619264284832403,
      "loss": 5.3054,
      "mean_token_accuracy": 0.17137685269117356,
      "num_tokens": 26536825.0,
      "step": 15295
    },
    {
      "entropy": 5.495219326019287,
      "epoch": 1.1903909745185761,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00048618308172753804,
      "loss": 5.2638,
      "mean_token_accuracy": 0.1775554046034813,
      "num_tokens": 26545208.0,
      "step": 15300
    },
    {
      "entropy": 5.516074848175049,
      "epoch": 1.1907800038902936,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004861735174023849,
      "loss": 5.2621,
      "mean_token_accuracy": 0.1717539519071579,
      "num_tokens": 26553768.0,
      "step": 15305
    },
    {
      "entropy": 5.538155031204224,
      "epoch": 1.1911690332620113,
      "grad_norm": 1.578125,
      "learning_rate": 0.00048616394987300985,
      "loss": 5.3589,
      "mean_token_accuracy": 0.1700994923710823,
      "num_tokens": 26561582.0,
      "step": 15310
    },
    {
      "entropy": 5.477288293838501,
      "epoch": 1.1915580626337288,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004861543791395579,
      "loss": 5.268,
      "mean_token_accuracy": 0.17064569890499115,
      "num_tokens": 26570632.0,
      "step": 15315
    },
    {
      "entropy": 5.496282768249512,
      "epoch": 1.1919470920054465,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004861448052021743,
      "loss": 5.2385,
      "mean_token_accuracy": 0.17956620305776597,
      "num_tokens": 26579507.0,
      "step": 15320
    },
    {
      "entropy": 5.519331216812134,
      "epoch": 1.192336121377164,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004861352280610044,
      "loss": 5.2708,
      "mean_token_accuracy": 0.17697979509830475,
      "num_tokens": 26588765.0,
      "step": 15325
    },
    {
      "entropy": 5.5780798435211185,
      "epoch": 1.1927251507488816,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048612564771619346,
      "loss": 5.3091,
      "mean_token_accuracy": 0.1718684285879135,
      "num_tokens": 26597529.0,
      "step": 15330
    },
    {
      "entropy": 5.517671203613281,
      "epoch": 1.193114180120599,
      "grad_norm": 1.5,
      "learning_rate": 0.00048611606416788686,
      "loss": 5.273,
      "mean_token_accuracy": 0.17543119788169861,
      "num_tokens": 26605960.0,
      "step": 15335
    },
    {
      "entropy": 5.480123376846313,
      "epoch": 1.1935032094923166,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00048610647741622996,
      "loss": 5.2105,
      "mean_token_accuracy": 0.17834984064102172,
      "num_tokens": 26614435.0,
      "step": 15340
    },
    {
      "entropy": 5.498327350616455,
      "epoch": 1.1938922388640343,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004860968874613683,
      "loss": 5.3028,
      "mean_token_accuracy": 0.17556809335947038,
      "num_tokens": 26622873.0,
      "step": 15345
    },
    {
      "entropy": 5.516958522796631,
      "epoch": 1.1942812682357518,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004860872943034474,
      "loss": 5.2521,
      "mean_token_accuracy": 0.18184894621372222,
      "num_tokens": 26631529.0,
      "step": 15350
    },
    {
      "entropy": 5.616390800476074,
      "epoch": 1.1946702976074695,
      "grad_norm": 2.0,
      "learning_rate": 0.0004860776979426128,
      "loss": 5.3387,
      "mean_token_accuracy": 0.1649305284023285,
      "num_tokens": 26639466.0,
      "step": 15355
    },
    {
      "entropy": 5.475955963134766,
      "epoch": 1.195059326979187,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004860680983790101,
      "loss": 5.2721,
      "mean_token_accuracy": 0.17273431569337844,
      "num_tokens": 26648329.0,
      "step": 15360
    },
    {
      "entropy": 5.521997594833374,
      "epoch": 1.1954483563509044,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004860584956127849,
      "loss": 5.2099,
      "mean_token_accuracy": 0.17746111750602722,
      "num_tokens": 26656899.0,
      "step": 15365
    },
    {
      "entropy": 5.599232149124146,
      "epoch": 1.195837385722622,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00048604888964408306,
      "loss": 5.4497,
      "mean_token_accuracy": 0.16418037116527556,
      "num_tokens": 26666155.0,
      "step": 15370
    },
    {
      "entropy": 5.505629587173462,
      "epoch": 1.1962264150943396,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004860392804730502,
      "loss": 5.2305,
      "mean_token_accuracy": 0.17840504944324492,
      "num_tokens": 26675567.0,
      "step": 15375
    },
    {
      "entropy": 5.551540231704712,
      "epoch": 1.1966154444660573,
      "grad_norm": 1.484375,
      "learning_rate": 0.00048602966809983204,
      "loss": 5.2835,
      "mean_token_accuracy": 0.1745466023683548,
      "num_tokens": 26684133.0,
      "step": 15380
    },
    {
      "entropy": 5.596503400802613,
      "epoch": 1.1970044738377748,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004860200525245746,
      "loss": 5.4122,
      "mean_token_accuracy": 0.1621151477098465,
      "num_tokens": 26692746.0,
      "step": 15385
    },
    {
      "entropy": 5.599116277694702,
      "epoch": 1.1973935032094922,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004860104337474239,
      "loss": 5.2925,
      "mean_token_accuracy": 0.17825113534927367,
      "num_tokens": 26701887.0,
      "step": 15390
    },
    {
      "entropy": 5.544364500045776,
      "epoch": 1.19778253258121,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048600081176852555,
      "loss": 5.2699,
      "mean_token_accuracy": 0.173001329600811,
      "num_tokens": 26710346.0,
      "step": 15395
    },
    {
      "entropy": 5.45468282699585,
      "epoch": 1.1981715619529274,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048599118658802575,
      "loss": 5.1874,
      "mean_token_accuracy": 0.18110829144716262,
      "num_tokens": 26718775.0,
      "step": 15400
    },
    {
      "entropy": 5.4572399139404295,
      "epoch": 1.198560591324645,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004859815582060706,
      "loss": 5.2515,
      "mean_token_accuracy": 0.17368060499429702,
      "num_tokens": 26726683.0,
      "step": 15405
    },
    {
      "entropy": 5.513894176483154,
      "epoch": 1.1989496206963626,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004859719266228061,
      "loss": 5.2757,
      "mean_token_accuracy": 0.18025728464126586,
      "num_tokens": 26735450.0,
      "step": 15410
    },
    {
      "entropy": 5.513188743591309,
      "epoch": 1.19933865006808,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004859622918383784,
      "loss": 5.239,
      "mean_token_accuracy": 0.17627025693655013,
      "num_tokens": 26744006.0,
      "step": 15415
    },
    {
      "entropy": 5.434592628479004,
      "epoch": 1.1997276794397977,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004859526538529337,
      "loss": 5.1489,
      "mean_token_accuracy": 0.18480762392282485,
      "num_tokens": 26751964.0,
      "step": 15420
    },
    {
      "entropy": 5.440835046768188,
      "epoch": 1.2001167088115152,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004859430126666182,
      "loss": 5.2328,
      "mean_token_accuracy": 0.17953760772943497,
      "num_tokens": 26760766.0,
      "step": 15425
    },
    {
      "entropy": 5.451300573348999,
      "epoch": 1.200505738183233,
      "grad_norm": 1.734375,
      "learning_rate": 0.00048593336827957824,
      "loss": 5.2157,
      "mean_token_accuracy": 0.17408957779407502,
      "num_tokens": 26769663.0,
      "step": 15430
    },
    {
      "entropy": 5.608166265487671,
      "epoch": 1.2008947675549504,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00048592372069196024,
      "loss": 5.4536,
      "mean_token_accuracy": 0.15993206351995468,
      "num_tokens": 26778693.0,
      "step": 15435
    },
    {
      "entropy": 5.555925989151001,
      "epoch": 1.2012837969266679,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004859140699039104,
      "loss": 5.2388,
      "mean_token_accuracy": 0.1839575320482254,
      "num_tokens": 26787368.0,
      "step": 15440
    },
    {
      "entropy": 5.482231330871582,
      "epoch": 1.2016728262983856,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00048590441591557526,
      "loss": 5.1728,
      "mean_token_accuracy": 0.1869637429714203,
      "num_tokens": 26795559.0,
      "step": 15445
    },
    {
      "entropy": 5.41761155128479,
      "epoch": 1.202061855670103,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00048589475872710134,
      "loss": 5.1875,
      "mean_token_accuracy": 0.18019474893808365,
      "num_tokens": 26803657.0,
      "step": 15450
    },
    {
      "entropy": 5.580560779571533,
      "epoch": 1.2024508850418207,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004858850983386351,
      "loss": 5.3663,
      "mean_token_accuracy": 0.16929758042097093,
      "num_tokens": 26812214.0,
      "step": 15455
    },
    {
      "entropy": 5.5866086959838865,
      "epoch": 1.2028399144135382,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004858754347503231,
      "loss": 5.2159,
      "mean_token_accuracy": 0.18279595375061036,
      "num_tokens": 26820591.0,
      "step": 15460
    },
    {
      "entropy": 5.506124639511109,
      "epoch": 1.203228943785256,
      "grad_norm": 1.546875,
      "learning_rate": 0.00048586576796231216,
      "loss": 5.3095,
      "mean_token_accuracy": 0.17119992822408675,
      "num_tokens": 26829228.0,
      "step": 15465
    },
    {
      "entropy": 5.560684728622436,
      "epoch": 1.2036179731569734,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004858560979747487,
      "loss": 5.444,
      "mean_token_accuracy": 0.16026985198259353,
      "num_tokens": 26838083.0,
      "step": 15470
    },
    {
      "entropy": 5.506431961059571,
      "epoch": 1.2040070025286909,
      "grad_norm": 1.84375,
      "learning_rate": 0.00048584642478777967,
      "loss": 5.145,
      "mean_token_accuracy": 0.17816006392240524,
      "num_tokens": 26846796.0,
      "step": 15475
    },
    {
      "entropy": 5.506115818023682,
      "epoch": 1.2043960319004086,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004858367484015517,
      "loss": 5.2605,
      "mean_token_accuracy": 0.17919433265924453,
      "num_tokens": 26854874.0,
      "step": 15480
    },
    {
      "entropy": 5.477559232711792,
      "epoch": 1.204785061272126,
      "grad_norm": 1.640625,
      "learning_rate": 0.00048582706881621166,
      "loss": 5.2763,
      "mean_token_accuracy": 0.17676855772733688,
      "num_tokens": 26863040.0,
      "step": 15485
    },
    {
      "entropy": 5.5456871509552,
      "epoch": 1.2051740906438435,
      "grad_norm": 1.5,
      "learning_rate": 0.0004858173860319063,
      "loss": 5.2844,
      "mean_token_accuracy": 0.16790855824947357,
      "num_tokens": 26872183.0,
      "step": 15490
    },
    {
      "entropy": 5.567169094085694,
      "epoch": 1.2055631200155612,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00048580770004878277,
      "loss": 5.2502,
      "mean_token_accuracy": 0.17474120408296584,
      "num_tokens": 26880480.0,
      "step": 15495
    },
    {
      "entropy": 5.510172176361084,
      "epoch": 1.2059521493872787,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004857980108669879,
      "loss": 5.2865,
      "mean_token_accuracy": 0.17212100327014923,
      "num_tokens": 26888855.0,
      "step": 15500
    },
    {
      "entropy": 5.485987567901612,
      "epoch": 1.2063411787589964,
      "grad_norm": 2.15625,
      "learning_rate": 0.00048578831848666875,
      "loss": 5.2852,
      "mean_token_accuracy": 0.17178125828504562,
      "num_tokens": 26897692.0,
      "step": 15505
    },
    {
      "entropy": 5.5099524974823,
      "epoch": 1.2067302081307139,
      "grad_norm": 2.09375,
      "learning_rate": 0.00048577862290797235,
      "loss": 5.2627,
      "mean_token_accuracy": 0.17293234765529633,
      "num_tokens": 26906502.0,
      "step": 15510
    },
    {
      "entropy": 5.56928186416626,
      "epoch": 1.2071192375024316,
      "grad_norm": 1.640625,
      "learning_rate": 0.00048576892413104586,
      "loss": 5.2824,
      "mean_token_accuracy": 0.17150285094976425,
      "num_tokens": 26915246.0,
      "step": 15515
    },
    {
      "entropy": 5.599849605560303,
      "epoch": 1.207508266874149,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004857592221560364,
      "loss": 5.394,
      "mean_token_accuracy": 0.17263538390398026,
      "num_tokens": 26923901.0,
      "step": 15520
    },
    {
      "entropy": 5.566528367996216,
      "epoch": 1.2078972962458665,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004857495169830912,
      "loss": 5.4748,
      "mean_token_accuracy": 0.1707826942205429,
      "num_tokens": 26933246.0,
      "step": 15525
    },
    {
      "entropy": 5.5978960514068605,
      "epoch": 1.2082863256175842,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004857398086123575,
      "loss": 5.3148,
      "mean_token_accuracy": 0.17998244166374205,
      "num_tokens": 26941938.0,
      "step": 15530
    },
    {
      "entropy": 5.548754453659058,
      "epoch": 1.2086753549893017,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004857300970439827,
      "loss": 5.2626,
      "mean_token_accuracy": 0.17742191553115844,
      "num_tokens": 26949967.0,
      "step": 15535
    },
    {
      "entropy": 5.47895245552063,
      "epoch": 1.2090643843610192,
      "grad_norm": 1.65625,
      "learning_rate": 0.00048572038227811404,
      "loss": 5.3449,
      "mean_token_accuracy": 0.17146805673837662,
      "num_tokens": 26958743.0,
      "step": 15540
    },
    {
      "entropy": 5.595466423034668,
      "epoch": 1.2094534137327368,
      "grad_norm": 1.78125,
      "learning_rate": 0.000485710664314899,
      "loss": 5.3434,
      "mean_token_accuracy": 0.1736383780837059,
      "num_tokens": 26967338.0,
      "step": 15545
    },
    {
      "entropy": 5.539255046844483,
      "epoch": 1.2098424431044543,
      "grad_norm": 1.7890625,
      "learning_rate": 0.000485700943154485,
      "loss": 5.2201,
      "mean_token_accuracy": 0.17763067781925201,
      "num_tokens": 26976182.0,
      "step": 15550
    },
    {
      "entropy": 5.5409869194030765,
      "epoch": 1.210231472476172,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004856912187970195,
      "loss": 5.3651,
      "mean_token_accuracy": 0.17120445817708968,
      "num_tokens": 26984275.0,
      "step": 15555
    },
    {
      "entropy": 5.618622589111328,
      "epoch": 1.2106205018478895,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004856814912426502,
      "loss": 5.315,
      "mean_token_accuracy": 0.17017755210399627,
      "num_tokens": 26991891.0,
      "step": 15560
    },
    {
      "entropy": 5.5119915962219235,
      "epoch": 1.2110095312196072,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004856717604915245,
      "loss": 5.2254,
      "mean_token_accuracy": 0.1756272718310356,
      "num_tokens": 27000390.0,
      "step": 15565
    },
    {
      "entropy": 5.371413326263427,
      "epoch": 1.2113985605913247,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004856620265437902,
      "loss": 5.2167,
      "mean_token_accuracy": 0.18659651130437852,
      "num_tokens": 27009348.0,
      "step": 15570
    },
    {
      "entropy": 5.507945013046265,
      "epoch": 1.2117875899630421,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000485652289399595,
      "loss": 5.2994,
      "mean_token_accuracy": 0.17704981416463852,
      "num_tokens": 27017529.0,
      "step": 15575
    },
    {
      "entropy": 5.590599632263183,
      "epoch": 1.2121766193347598,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048564254905908655,
      "loss": 5.2962,
      "mean_token_accuracy": 0.1775521606206894,
      "num_tokens": 27026238.0,
      "step": 15580
    },
    {
      "entropy": 5.452028274536133,
      "epoch": 1.2125656487064773,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00048563280552241266,
      "loss": 5.2311,
      "mean_token_accuracy": 0.178492571413517,
      "num_tokens": 27034995.0,
      "step": 15585
    },
    {
      "entropy": 5.636938190460205,
      "epoch": 1.2129546780781948,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004856230587897212,
      "loss": 5.4297,
      "mean_token_accuracy": 0.16370497345924379,
      "num_tokens": 27043631.0,
      "step": 15590
    },
    {
      "entropy": 5.586244440078735,
      "epoch": 1.2133437074499125,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004856133088611602,
      "loss": 5.2814,
      "mean_token_accuracy": 0.1802394911646843,
      "num_tokens": 27052491.0,
      "step": 15595
    },
    {
      "entropy": 5.512923288345337,
      "epoch": 1.21373273682163,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004856035557368773,
      "loss": 5.2502,
      "mean_token_accuracy": 0.17505358457565307,
      "num_tokens": 27061056.0,
      "step": 15600
    },
    {
      "entropy": 5.450316095352173,
      "epoch": 1.2141217661933477,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00048559379941702074,
      "loss": 5.2445,
      "mean_token_accuracy": 0.18168267458677292,
      "num_tokens": 27069562.0,
      "step": 15605
    },
    {
      "entropy": 5.533176374435425,
      "epoch": 1.2145107955650651,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00048558403990173844,
      "loss": 5.2575,
      "mean_token_accuracy": 0.1756133958697319,
      "num_tokens": 27077814.0,
      "step": 15610
    },
    {
      "entropy": 5.523116588592529,
      "epoch": 1.2148998249367828,
      "grad_norm": 2.09375,
      "learning_rate": 0.00048557427719117855,
      "loss": 5.1765,
      "mean_token_accuracy": 0.18126779943704605,
      "num_tokens": 27086515.0,
      "step": 15615
    },
    {
      "entropy": 5.55156512260437,
      "epoch": 1.2152888543085003,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004855645112854891,
      "loss": 5.3001,
      "mean_token_accuracy": 0.17764369249343873,
      "num_tokens": 27095759.0,
      "step": 15620
    },
    {
      "entropy": 5.525755834579468,
      "epoch": 1.2156778836802178,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004855547421848184,
      "loss": 5.3189,
      "mean_token_accuracy": 0.1713688164949417,
      "num_tokens": 27104921.0,
      "step": 15625
    },
    {
      "entropy": 5.56371488571167,
      "epoch": 1.2160669130519355,
      "grad_norm": 2.078125,
      "learning_rate": 0.00048554496988931456,
      "loss": 5.3058,
      "mean_token_accuracy": 0.17698432356119156,
      "num_tokens": 27113876.0,
      "step": 15630
    },
    {
      "entropy": 5.55414400100708,
      "epoch": 1.216455942423653,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00048553519439912597,
      "loss": 5.1844,
      "mean_token_accuracy": 0.17874561995267868,
      "num_tokens": 27121654.0,
      "step": 15635
    },
    {
      "entropy": 5.435419797897339,
      "epoch": 1.2168449717953704,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004855254157144009,
      "loss": 5.2201,
      "mean_token_accuracy": 0.1826067790389061,
      "num_tokens": 27130156.0,
      "step": 15640
    },
    {
      "entropy": 5.513195753097534,
      "epoch": 1.2172340011670881,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004855156338352877,
      "loss": 5.3133,
      "mean_token_accuracy": 0.16902976036071776,
      "num_tokens": 27138646.0,
      "step": 15645
    },
    {
      "entropy": 5.532236242294312,
      "epoch": 1.2176230305388056,
      "grad_norm": 1.53125,
      "learning_rate": 0.00048550584876193493,
      "loss": 5.3127,
      "mean_token_accuracy": 0.17692103683948518,
      "num_tokens": 27147282.0,
      "step": 15650
    },
    {
      "entropy": 5.48130054473877,
      "epoch": 1.2180120599105233,
      "grad_norm": 1.59375,
      "learning_rate": 0.00048549606049449085,
      "loss": 5.2068,
      "mean_token_accuracy": 0.1692296326160431,
      "num_tokens": 27156238.0,
      "step": 15655
    },
    {
      "entropy": 5.534921932220459,
      "epoch": 1.2184010892822408,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004854862690331041,
      "loss": 5.3177,
      "mean_token_accuracy": 0.17091383785009384,
      "num_tokens": 27164936.0,
      "step": 15660
    },
    {
      "entropy": 5.543131589889526,
      "epoch": 1.2187901186539585,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00048547647437792327,
      "loss": 5.2783,
      "mean_token_accuracy": 0.1732231631875038,
      "num_tokens": 27173362.0,
      "step": 15665
    },
    {
      "entropy": 5.475022649765014,
      "epoch": 1.219179148025676,
      "grad_norm": 1.71875,
      "learning_rate": 0.000485466676529097,
      "loss": 5.2085,
      "mean_token_accuracy": 0.18326039016246795,
      "num_tokens": 27181611.0,
      "step": 15670
    },
    {
      "entropy": 5.536137342453003,
      "epoch": 1.2195681773973934,
      "grad_norm": 2.21875,
      "learning_rate": 0.00048545687548677373,
      "loss": 5.2751,
      "mean_token_accuracy": 0.17755033373832702,
      "num_tokens": 27189593.0,
      "step": 15675
    },
    {
      "entropy": 5.541131639480591,
      "epoch": 1.2199572067691111,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004854470712511025,
      "loss": 5.1921,
      "mean_token_accuracy": 0.17833051532506944,
      "num_tokens": 27198006.0,
      "step": 15680
    },
    {
      "entropy": 5.571993207931518,
      "epoch": 1.2203462361408286,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00048543726382223193,
      "loss": 5.3165,
      "mean_token_accuracy": 0.17003822028636933,
      "num_tokens": 27206846.0,
      "step": 15685
    },
    {
      "entropy": 5.518065547943115,
      "epoch": 1.220735265512546,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048542745320031075,
      "loss": 5.2586,
      "mean_token_accuracy": 0.17630101144313812,
      "num_tokens": 27214838.0,
      "step": 15690
    },
    {
      "entropy": 5.508387708663941,
      "epoch": 1.2211242948842638,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00048541763938548795,
      "loss": 5.3418,
      "mean_token_accuracy": 0.16945082545280457,
      "num_tokens": 27224161.0,
      "step": 15695
    },
    {
      "entropy": 5.465720891952515,
      "epoch": 1.2215133242559812,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00048540782237791244,
      "loss": 5.2027,
      "mean_token_accuracy": 0.17481983155012132,
      "num_tokens": 27233084.0,
      "step": 15700
    },
    {
      "entropy": 5.675802373886109,
      "epoch": 1.221902353627699,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00048539800217773293,
      "loss": 5.4057,
      "mean_token_accuracy": 0.16449622958898544,
      "num_tokens": 27242219.0,
      "step": 15705
    },
    {
      "entropy": 5.573157024383545,
      "epoch": 1.2222913829994164,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004853881787850988,
      "loss": 5.3388,
      "mean_token_accuracy": 0.1737979307770729,
      "num_tokens": 27250411.0,
      "step": 15710
    },
    {
      "entropy": 5.55861644744873,
      "epoch": 1.2226804123711341,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00048537835220015886,
      "loss": 5.343,
      "mean_token_accuracy": 0.17057652920484542,
      "num_tokens": 27259472.0,
      "step": 15715
    },
    {
      "entropy": 5.558578681945801,
      "epoch": 1.2230694417428516,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004853685224230623,
      "loss": 5.2428,
      "mean_token_accuracy": 0.17715482264757157,
      "num_tokens": 27267922.0,
      "step": 15720
    },
    {
      "entropy": 5.531707859039306,
      "epoch": 1.223458471114569,
      "grad_norm": 1.84375,
      "learning_rate": 0.00048535868945395825,
      "loss": 5.375,
      "mean_token_accuracy": 0.17383092045783996,
      "num_tokens": 27277129.0,
      "step": 15725
    },
    {
      "entropy": 5.5771197319030765,
      "epoch": 1.2238475004862868,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048534885329299586,
      "loss": 5.3225,
      "mean_token_accuracy": 0.17048002630472184,
      "num_tokens": 27285583.0,
      "step": 15730
    },
    {
      "entropy": 5.515822172164917,
      "epoch": 1.2242365298580042,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004853390139403245,
      "loss": 5.2067,
      "mean_token_accuracy": 0.1759398177266121,
      "num_tokens": 27294292.0,
      "step": 15735
    },
    {
      "entropy": 5.59844741821289,
      "epoch": 1.224625559229722,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048532917139609334,
      "loss": 5.4076,
      "mean_token_accuracy": 0.16913793981075287,
      "num_tokens": 27302606.0,
      "step": 15740
    },
    {
      "entropy": 5.627884387969971,
      "epoch": 1.2250145886014394,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004853193256604518,
      "loss": 5.2607,
      "mean_token_accuracy": 0.18148200511932372,
      "num_tokens": 27311103.0,
      "step": 15745
    },
    {
      "entropy": 5.637231254577637,
      "epoch": 1.2254036179731569,
      "grad_norm": 1.671875,
      "learning_rate": 0.00048530947673354924,
      "loss": 5.2834,
      "mean_token_accuracy": 0.17769556790590285,
      "num_tokens": 27319497.0,
      "step": 15750
    },
    {
      "entropy": 5.523335409164429,
      "epoch": 1.2257926473448746,
      "grad_norm": 1.5,
      "learning_rate": 0.0004852996246155351,
      "loss": 5.3158,
      "mean_token_accuracy": 0.1735464498400688,
      "num_tokens": 27328367.0,
      "step": 15755
    },
    {
      "entropy": 5.507326602935791,
      "epoch": 1.226181676716592,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00048528976930655896,
      "loss": 5.213,
      "mean_token_accuracy": 0.1748963087797165,
      "num_tokens": 27336797.0,
      "step": 15760
    },
    {
      "entropy": 5.462827920913696,
      "epoch": 1.2265707060883098,
      "grad_norm": 1.484375,
      "learning_rate": 0.00048527991080677015,
      "loss": 5.1833,
      "mean_token_accuracy": 0.1751452013850212,
      "num_tokens": 27345052.0,
      "step": 15765
    },
    {
      "entropy": 5.520791006088257,
      "epoch": 1.2269597354600272,
      "grad_norm": 1.625,
      "learning_rate": 0.00048527004911631843,
      "loss": 5.2214,
      "mean_token_accuracy": 0.18662792593240737,
      "num_tokens": 27353062.0,
      "step": 15770
    },
    {
      "entropy": 5.494668340682983,
      "epoch": 1.2273487648317447,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004852601842353534,
      "loss": 5.2634,
      "mean_token_accuracy": 0.17666485607624055,
      "num_tokens": 27361471.0,
      "step": 15775
    },
    {
      "entropy": 5.546452474594116,
      "epoch": 1.2277377942034624,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00048525031616402476,
      "loss": 5.34,
      "mean_token_accuracy": 0.1727466270327568,
      "num_tokens": 27370655.0,
      "step": 15780
    },
    {
      "entropy": 5.552441549301148,
      "epoch": 1.2281268235751799,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00048524044490248227,
      "loss": 5.1714,
      "mean_token_accuracy": 0.1819349005818367,
      "num_tokens": 27379329.0,
      "step": 15785
    },
    {
      "entropy": 5.554012870788574,
      "epoch": 1.2285158529468976,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048523057045087557,
      "loss": 5.3233,
      "mean_token_accuracy": 0.1722200632095337,
      "num_tokens": 27388864.0,
      "step": 15790
    },
    {
      "entropy": 5.5810950756072994,
      "epoch": 1.228904882318615,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00048522069280935466,
      "loss": 5.411,
      "mean_token_accuracy": 0.16008602529764177,
      "num_tokens": 27397804.0,
      "step": 15795
    },
    {
      "entropy": 5.50671911239624,
      "epoch": 1.2292939116903325,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004852108119780693,
      "loss": 5.2768,
      "mean_token_accuracy": 0.17277107387781143,
      "num_tokens": 27406578.0,
      "step": 15800
    },
    {
      "entropy": 5.53170256614685,
      "epoch": 1.2296829410620502,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004852009279571694,
      "loss": 5.3446,
      "mean_token_accuracy": 0.16500737518072128,
      "num_tokens": 27415207.0,
      "step": 15805
    },
    {
      "entropy": 5.552730846405029,
      "epoch": 1.2300719704337677,
      "grad_norm": 1.8203125,
      "learning_rate": 0.000485191040746805,
      "loss": 5.2996,
      "mean_token_accuracy": 0.16809863299131395,
      "num_tokens": 27424201.0,
      "step": 15810
    },
    {
      "entropy": 5.513219118118286,
      "epoch": 1.2304609998054854,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004851811503471262,
      "loss": 5.2975,
      "mean_token_accuracy": 0.17288100570440293,
      "num_tokens": 27433187.0,
      "step": 15815
    },
    {
      "entropy": 5.48877534866333,
      "epoch": 1.2308500291772029,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00048517125675828294,
      "loss": 5.2801,
      "mean_token_accuracy": 0.17049404084682465,
      "num_tokens": 27441976.0,
      "step": 15820
    },
    {
      "entropy": 5.528487777709961,
      "epoch": 1.2312390585489204,
      "grad_norm": 1.59375,
      "learning_rate": 0.00048516135998042536,
      "loss": 5.2056,
      "mean_token_accuracy": 0.1832287698984146,
      "num_tokens": 27450235.0,
      "step": 15825
    },
    {
      "entropy": 5.5694622039794925,
      "epoch": 1.231628087920638,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004851514600137037,
      "loss": 5.4018,
      "mean_token_accuracy": 0.1667611390352249,
      "num_tokens": 27458842.0,
      "step": 15830
    },
    {
      "entropy": 5.632485342025757,
      "epoch": 1.2320171172923555,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00048514155685826807,
      "loss": 5.3476,
      "mean_token_accuracy": 0.1709143579006195,
      "num_tokens": 27467383.0,
      "step": 15835
    },
    {
      "entropy": 5.532180643081665,
      "epoch": 1.2324061466640732,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004851316505142688,
      "loss": 5.3166,
      "mean_token_accuracy": 0.17439887374639512,
      "num_tokens": 27475811.0,
      "step": 15840
    },
    {
      "entropy": 5.632568120956421,
      "epoch": 1.2327951760357907,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048512174098185615,
      "loss": 5.4272,
      "mean_token_accuracy": 0.16330601274967194,
      "num_tokens": 27484032.0,
      "step": 15845
    },
    {
      "entropy": 5.55184383392334,
      "epoch": 1.2331842054075082,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00048511182826118055,
      "loss": 5.1548,
      "mean_token_accuracy": 0.17924610823392867,
      "num_tokens": 27492330.0,
      "step": 15850
    },
    {
      "entropy": 5.467745923995972,
      "epoch": 1.2335732347792259,
      "grad_norm": 1.734375,
      "learning_rate": 0.00048510191235239246,
      "loss": 5.2543,
      "mean_token_accuracy": 0.17472672760486602,
      "num_tokens": 27501042.0,
      "step": 15855
    },
    {
      "entropy": 5.507084846496582,
      "epoch": 1.2339622641509433,
      "grad_norm": 1.515625,
      "learning_rate": 0.00048509199325564217,
      "loss": 5.2061,
      "mean_token_accuracy": 0.17931463569402695,
      "num_tokens": 27509939.0,
      "step": 15860
    },
    {
      "entropy": 5.60815691947937,
      "epoch": 1.234351293522661,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004850820709710803,
      "loss": 5.4009,
      "mean_token_accuracy": 0.16378685384988784,
      "num_tokens": 27518653.0,
      "step": 15865
    },
    {
      "entropy": 5.622842645645141,
      "epoch": 1.2347403228943785,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004850721454988574,
      "loss": 5.2647,
      "mean_token_accuracy": 0.18122659176588057,
      "num_tokens": 27526856.0,
      "step": 15870
    },
    {
      "entropy": 5.4506090641021725,
      "epoch": 1.235129352266096,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00048506221683912405,
      "loss": 5.1987,
      "mean_token_accuracy": 0.18405112624168396,
      "num_tokens": 27535205.0,
      "step": 15875
    },
    {
      "entropy": 5.5352109432220455,
      "epoch": 1.2355183816378137,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004850522849920309,
      "loss": 5.2926,
      "mean_token_accuracy": 0.17213273793458939,
      "num_tokens": 27543516.0,
      "step": 15880
    },
    {
      "entropy": 5.614929294586181,
      "epoch": 1.2359074110095312,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00048504234995772863,
      "loss": 5.351,
      "mean_token_accuracy": 0.17170230001211167,
      "num_tokens": 27552136.0,
      "step": 15885
    },
    {
      "entropy": 5.534029865264893,
      "epoch": 1.2362964403812489,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004850324117363681,
      "loss": 5.3267,
      "mean_token_accuracy": 0.18268054872751235,
      "num_tokens": 27560805.0,
      "step": 15890
    },
    {
      "entropy": 5.523261976242066,
      "epoch": 1.2366854697529663,
      "grad_norm": 3.40625,
      "learning_rate": 0.00048502247032809997,
      "loss": 5.248,
      "mean_token_accuracy": 0.17248456925153732,
      "num_tokens": 27569643.0,
      "step": 15895
    },
    {
      "entropy": 5.5592669486999515,
      "epoch": 1.237074499124684,
      "grad_norm": 1.5,
      "learning_rate": 0.0004850125257330751,
      "loss": 5.2029,
      "mean_token_accuracy": 0.17631499767303466,
      "num_tokens": 27578506.0,
      "step": 15900
    },
    {
      "entropy": 5.4709169387817385,
      "epoch": 1.2374635284964015,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048500257795144446,
      "loss": 5.2974,
      "mean_token_accuracy": 0.17011501491069794,
      "num_tokens": 27586911.0,
      "step": 15905
    },
    {
      "entropy": 5.515015459060669,
      "epoch": 1.237852557868119,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000484992626983359,
      "loss": 5.3283,
      "mean_token_accuracy": 0.17393341809511184,
      "num_tokens": 27595112.0,
      "step": 15910
    },
    {
      "entropy": 5.593227338790894,
      "epoch": 1.2382415872398367,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004849826728289696,
      "loss": 5.3333,
      "mean_token_accuracy": 0.1646855928003788,
      "num_tokens": 27603923.0,
      "step": 15915
    },
    {
      "entropy": 5.441413116455078,
      "epoch": 1.2386306166115542,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00048497271548842737,
      "loss": 5.1339,
      "mean_token_accuracy": 0.18607923090457917,
      "num_tokens": 27612705.0,
      "step": 15920
    },
    {
      "entropy": 5.51904296875,
      "epoch": 1.2390196459832716,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004849627549618834,
      "loss": 5.2743,
      "mean_token_accuracy": 0.1758978247642517,
      "num_tokens": 27621858.0,
      "step": 15925
    },
    {
      "entropy": 5.597062587738037,
      "epoch": 1.2394086753549893,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048495279124948886,
      "loss": 5.3578,
      "mean_token_accuracy": 0.1709895223379135,
      "num_tokens": 27629977.0,
      "step": 15930
    },
    {
      "entropy": 5.603897666931152,
      "epoch": 1.2397977047267068,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004849428243513948,
      "loss": 5.3037,
      "mean_token_accuracy": 0.1735332727432251,
      "num_tokens": 27638433.0,
      "step": 15935
    },
    {
      "entropy": 5.461909389495849,
      "epoch": 1.2401867340984245,
      "grad_norm": 1.53125,
      "learning_rate": 0.00048493285426775255,
      "loss": 5.2284,
      "mean_token_accuracy": 0.17621096074581147,
      "num_tokens": 27647116.0,
      "step": 15940
    },
    {
      "entropy": 5.4775975227355955,
      "epoch": 1.240575763470142,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004849228809987135,
      "loss": 5.3019,
      "mean_token_accuracy": 0.16883142665028572,
      "num_tokens": 27655598.0,
      "step": 15945
    },
    {
      "entropy": 5.652406930923462,
      "epoch": 1.2409647928418597,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004849129045444288,
      "loss": 5.3986,
      "mean_token_accuracy": 0.16651797145605088,
      "num_tokens": 27664308.0,
      "step": 15950
    },
    {
      "entropy": 5.523653602600097,
      "epoch": 1.2413538222135772,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004849029249050498,
      "loss": 5.1381,
      "mean_token_accuracy": 0.18590807169675827,
      "num_tokens": 27672808.0,
      "step": 15955
    },
    {
      "entropy": 5.586093521118164,
      "epoch": 1.2417428515852946,
      "grad_norm": 1.765625,
      "learning_rate": 0.00048489294208072805,
      "loss": 5.3471,
      "mean_token_accuracy": 0.16780729293823243,
      "num_tokens": 27681091.0,
      "step": 15960
    },
    {
      "entropy": 5.5585376739501955,
      "epoch": 1.2421318809570123,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00048488295607161495,
      "loss": 5.2922,
      "mean_token_accuracy": 0.17706929743289948,
      "num_tokens": 27689582.0,
      "step": 15965
    },
    {
      "entropy": 5.5264345645904545,
      "epoch": 1.2425209103287298,
      "grad_norm": 1.75,
      "learning_rate": 0.0004848729668778621,
      "loss": 5.2268,
      "mean_token_accuracy": 0.17582499235868454,
      "num_tokens": 27698509.0,
      "step": 15970
    },
    {
      "entropy": 5.552094411849976,
      "epoch": 1.2429099397004473,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004848629744996211,
      "loss": 5.3387,
      "mean_token_accuracy": 0.16460926830768585,
      "num_tokens": 27708216.0,
      "step": 15975
    },
    {
      "entropy": 5.616294527053833,
      "epoch": 1.243298969072165,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004848529789370434,
      "loss": 5.3306,
      "mean_token_accuracy": 0.1712876006960869,
      "num_tokens": 27716801.0,
      "step": 15980
    },
    {
      "entropy": 5.538486289978027,
      "epoch": 1.2436879984438824,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004848429801902808,
      "loss": 5.2974,
      "mean_token_accuracy": 0.17481158822774887,
      "num_tokens": 27725186.0,
      "step": 15985
    },
    {
      "entropy": 5.6251349449157715,
      "epoch": 1.2440770278156001,
      "grad_norm": 1.5,
      "learning_rate": 0.00048483297825948485,
      "loss": 5.2537,
      "mean_token_accuracy": 0.171275532245636,
      "num_tokens": 27733386.0,
      "step": 15990
    },
    {
      "entropy": 5.57133412361145,
      "epoch": 1.2444660571873176,
      "grad_norm": 1.53125,
      "learning_rate": 0.00048482297314480756,
      "loss": 5.3205,
      "mean_token_accuracy": 0.17034234553575517,
      "num_tokens": 27742253.0,
      "step": 15995
    },
    {
      "entropy": 5.528388786315918,
      "epoch": 1.2448550865590353,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004848129648464006,
      "loss": 5.2425,
      "mean_token_accuracy": 0.1777079001069069,
      "num_tokens": 27751555.0,
      "step": 16000
    },
    {
      "entropy": 5.566807222366333,
      "epoch": 1.2452441159307528,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004848029533644159,
      "loss": 5.3841,
      "mean_token_accuracy": 0.16856226176023484,
      "num_tokens": 27760151.0,
      "step": 16005
    },
    {
      "entropy": 5.586111927032471,
      "epoch": 1.2456331453024703,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004847929386990052,
      "loss": 5.3005,
      "mean_token_accuracy": 0.17904738783836366,
      "num_tokens": 27770017.0,
      "step": 16010
    },
    {
      "entropy": 5.518650197982788,
      "epoch": 1.246022174674188,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048478292085032065,
      "loss": 5.163,
      "mean_token_accuracy": 0.17997154146432875,
      "num_tokens": 27777796.0,
      "step": 16015
    },
    {
      "entropy": 5.4770392894744875,
      "epoch": 1.2464112040459054,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004847728998185142,
      "loss": 5.2616,
      "mean_token_accuracy": 0.17714573293924332,
      "num_tokens": 27785784.0,
      "step": 16020
    },
    {
      "entropy": 5.546535491943359,
      "epoch": 1.246800233417623,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00048476287560373786,
      "loss": 5.2723,
      "mean_token_accuracy": 0.17140836715698243,
      "num_tokens": 27793911.0,
      "step": 16025
    },
    {
      "entropy": 5.446552419662476,
      "epoch": 1.2471892627893406,
      "grad_norm": 2.28125,
      "learning_rate": 0.00048475284820614375,
      "loss": 5.2726,
      "mean_token_accuracy": 0.18089391440153121,
      "num_tokens": 27803125.0,
      "step": 16030
    },
    {
      "entropy": 5.430708980560302,
      "epoch": 1.247578292161058,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048474281762588407,
      "loss": 5.2319,
      "mean_token_accuracy": 0.17744951993227004,
      "num_tokens": 27812008.0,
      "step": 16035
    },
    {
      "entropy": 5.5487429141998295,
      "epoch": 1.2479673215327758,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004847327838631109,
      "loss": 5.3557,
      "mean_token_accuracy": 0.16771409809589385,
      "num_tokens": 27820490.0,
      "step": 16040
    },
    {
      "entropy": 5.603517723083496,
      "epoch": 1.2483563509044933,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004847227469179766,
      "loss": 5.301,
      "mean_token_accuracy": 0.1743951141834259,
      "num_tokens": 27829513.0,
      "step": 16045
    },
    {
      "entropy": 5.511103773117066,
      "epoch": 1.248745380276211,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004847127067906334,
      "loss": 5.2549,
      "mean_token_accuracy": 0.17470561116933822,
      "num_tokens": 27837811.0,
      "step": 16050
    },
    {
      "entropy": 5.579536199569702,
      "epoch": 1.2491344096479284,
      "grad_norm": 1.703125,
      "learning_rate": 0.00048470266348123365,
      "loss": 5.3891,
      "mean_token_accuracy": 0.1684957981109619,
      "num_tokens": 27846449.0,
      "step": 16055
    },
    {
      "entropy": 5.447787570953369,
      "epoch": 1.249523439019646,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004846926169899298,
      "loss": 5.18,
      "mean_token_accuracy": 0.18282399475574493,
      "num_tokens": 27854834.0,
      "step": 16060
    },
    {
      "entropy": 5.5481071949005125,
      "epoch": 1.2499124683913636,
      "grad_norm": 1.515625,
      "learning_rate": 0.00048468256731687426,
      "loss": 5.3233,
      "mean_token_accuracy": 0.17269108444452286,
      "num_tokens": 27862984.0,
      "step": 16065
    },
    {
      "entropy": 5.581409549713134,
      "epoch": 1.250301497763081,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004846725144622194,
      "loss": 5.3548,
      "mean_token_accuracy": 0.16472773104906083,
      "num_tokens": 27872198.0,
      "step": 16070
    },
    {
      "entropy": 5.586838436126709,
      "epoch": 1.2506905271347986,
      "grad_norm": 1.640625,
      "learning_rate": 0.000484662458426118,
      "loss": 5.2592,
      "mean_token_accuracy": 0.17459298968315123,
      "num_tokens": 27880856.0,
      "step": 16075
    },
    {
      "entropy": 5.495987892150879,
      "epoch": 1.2510795565065163,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00048465239920872247,
      "loss": 5.1995,
      "mean_token_accuracy": 0.1802007809281349,
      "num_tokens": 27888677.0,
      "step": 16080
    },
    {
      "entropy": 5.46326208114624,
      "epoch": 1.2514685858782337,
      "grad_norm": 1.484375,
      "learning_rate": 0.00048464233681018545,
      "loss": 5.2571,
      "mean_token_accuracy": 0.17940560579299927,
      "num_tokens": 27896940.0,
      "step": 16085
    },
    {
      "entropy": 5.534651088714599,
      "epoch": 1.2518576152499514,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004846322712306596,
      "loss": 5.3629,
      "mean_token_accuracy": 0.16747429370880126,
      "num_tokens": 27905739.0,
      "step": 16090
    },
    {
      "entropy": 5.588169384002685,
      "epoch": 1.252246644621669,
      "grad_norm": 2.125,
      "learning_rate": 0.00048462220247029783,
      "loss": 5.308,
      "mean_token_accuracy": 0.1772422343492508,
      "num_tokens": 27914029.0,
      "step": 16095
    },
    {
      "entropy": 5.621267461776734,
      "epoch": 1.2526356739933866,
      "grad_norm": 1.546875,
      "learning_rate": 0.00048461213052925265,
      "loss": 5.4159,
      "mean_token_accuracy": 0.16701759546995162,
      "num_tokens": 27923287.0,
      "step": 16100
    },
    {
      "entropy": 5.502909421920776,
      "epoch": 1.253024703365104,
      "grad_norm": 1.625,
      "learning_rate": 0.0004846020554076771,
      "loss": 5.2592,
      "mean_token_accuracy": 0.17615789026021958,
      "num_tokens": 27931727.0,
      "step": 16105
    },
    {
      "entropy": 5.534601783752441,
      "epoch": 1.2534137327368216,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004845919771057239,
      "loss": 5.1926,
      "mean_token_accuracy": 0.1769889771938324,
      "num_tokens": 27939524.0,
      "step": 16110
    },
    {
      "entropy": 5.574967527389527,
      "epoch": 1.2538027621085392,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004845818956235462,
      "loss": 5.2762,
      "mean_token_accuracy": 0.16775871813297272,
      "num_tokens": 27947827.0,
      "step": 16115
    },
    {
      "entropy": 5.494071769714355,
      "epoch": 1.2541917914802567,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00048457181096129664,
      "loss": 5.283,
      "mean_token_accuracy": 0.17416021078824998,
      "num_tokens": 27956447.0,
      "step": 16120
    },
    {
      "entropy": 5.556335926055908,
      "epoch": 1.2545808208519742,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004845617231191285,
      "loss": 5.3239,
      "mean_token_accuracy": 0.16798613965511322,
      "num_tokens": 27965570.0,
      "step": 16125
    },
    {
      "entropy": 5.534073543548584,
      "epoch": 1.254969850223692,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004845516320971948,
      "loss": 5.2915,
      "mean_token_accuracy": 0.1722590446472168,
      "num_tokens": 27974527.0,
      "step": 16130
    },
    {
      "entropy": 5.5188836574554445,
      "epoch": 1.2553588795954094,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004845415378956486,
      "loss": 5.2166,
      "mean_token_accuracy": 0.17618252336978912,
      "num_tokens": 27983190.0,
      "step": 16135
    },
    {
      "entropy": 5.528718614578247,
      "epoch": 1.255747908967127,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000484531440514643,
      "loss": 5.3408,
      "mean_token_accuracy": 0.17083994448184966,
      "num_tokens": 27992203.0,
      "step": 16140
    },
    {
      "entropy": 5.534156799316406,
      "epoch": 1.2561369383388445,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048452133995433136,
      "loss": 5.2954,
      "mean_token_accuracy": 0.17254400253295898,
      "num_tokens": 28001154.0,
      "step": 16145
    },
    {
      "entropy": 5.495703649520874,
      "epoch": 1.2565259677105622,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004845112362148668,
      "loss": 5.2,
      "mean_token_accuracy": 0.17672016322612763,
      "num_tokens": 28009841.0,
      "step": 16150
    },
    {
      "entropy": 5.512458992004395,
      "epoch": 1.2569149970822797,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004845011292964028,
      "loss": 5.2923,
      "mean_token_accuracy": 0.17635370939970016,
      "num_tokens": 28019245.0,
      "step": 16155
    },
    {
      "entropy": 5.579947233200073,
      "epoch": 1.2573040264539972,
      "grad_norm": 2.046875,
      "learning_rate": 0.00048449101919909265,
      "loss": 5.4261,
      "mean_token_accuracy": 0.16537006199359894,
      "num_tokens": 28028342.0,
      "step": 16160
    },
    {
      "entropy": 5.572152137756348,
      "epoch": 1.257693055825715,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00048448090592308955,
      "loss": 5.3589,
      "mean_token_accuracy": 0.171546870470047,
      "num_tokens": 28037308.0,
      "step": 16165
    },
    {
      "entropy": 5.564729070663452,
      "epoch": 1.2580820851974324,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004844707894685473,
      "loss": 5.3209,
      "mean_token_accuracy": 0.17351430505514145,
      "num_tokens": 28046214.0,
      "step": 16170
    },
    {
      "entropy": 5.48063554763794,
      "epoch": 1.2584711145691498,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004844606698356192,
      "loss": 5.159,
      "mean_token_accuracy": 0.17853713184595107,
      "num_tokens": 28054722.0,
      "step": 16175
    },
    {
      "entropy": 5.456463575363159,
      "epoch": 1.2588601439408675,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004844505470244588,
      "loss": 5.1627,
      "mean_token_accuracy": 0.17707505226135253,
      "num_tokens": 28062596.0,
      "step": 16180
    },
    {
      "entropy": 5.396499967575073,
      "epoch": 1.2592491733125852,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004844404210352197,
      "loss": 5.2069,
      "mean_token_accuracy": 0.17821115702390672,
      "num_tokens": 28071183.0,
      "step": 16185
    },
    {
      "entropy": 5.431892681121826,
      "epoch": 1.2596382026843027,
      "grad_norm": 1.875,
      "learning_rate": 0.00048443029186805554,
      "loss": 5.2099,
      "mean_token_accuracy": 0.1791281968355179,
      "num_tokens": 28079194.0,
      "step": 16190
    },
    {
      "entropy": 5.52010555267334,
      "epoch": 1.2600272320560202,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004844201595231201,
      "loss": 5.2124,
      "mean_token_accuracy": 0.1752294510602951,
      "num_tokens": 28087874.0,
      "step": 16195
    },
    {
      "entropy": 5.606992435455322,
      "epoch": 1.2604162614277379,
      "grad_norm": 4.09375,
      "learning_rate": 0.00048441002400056706,
      "loss": 5.3018,
      "mean_token_accuracy": 0.17383768111467363,
      "num_tokens": 28097390.0,
      "step": 16200
    },
    {
      "entropy": 5.4187366485595705,
      "epoch": 1.2608052907994554,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004843998853005502,
      "loss": 5.2,
      "mean_token_accuracy": 0.17888538539409637,
      "num_tokens": 28105445.0,
      "step": 16205
    },
    {
      "entropy": 5.543935775756836,
      "epoch": 1.2611943201711728,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004843897434232233,
      "loss": 5.2495,
      "mean_token_accuracy": 0.1747748598456383,
      "num_tokens": 28113855.0,
      "step": 16210
    },
    {
      "entropy": 5.548325538635254,
      "epoch": 1.2615833495428905,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048437959836874047,
      "loss": 5.2141,
      "mean_token_accuracy": 0.1837356135249138,
      "num_tokens": 28123300.0,
      "step": 16215
    },
    {
      "entropy": 5.47448878288269,
      "epoch": 1.261972378914608,
      "grad_norm": 1.84375,
      "learning_rate": 0.00048436945013725544,
      "loss": 5.1998,
      "mean_token_accuracy": 0.1759326219558716,
      "num_tokens": 28131829.0,
      "step": 16220
    },
    {
      "entropy": 5.526947689056397,
      "epoch": 1.2623614082863255,
      "grad_norm": 2.078125,
      "learning_rate": 0.00048435929872892226,
      "loss": 5.3663,
      "mean_token_accuracy": 0.1652940958738327,
      "num_tokens": 28141060.0,
      "step": 16225
    },
    {
      "entropy": 5.589622354507446,
      "epoch": 1.2627504376580432,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004843491441438949,
      "loss": 5.3689,
      "mean_token_accuracy": 0.16020533740520476,
      "num_tokens": 28149866.0,
      "step": 16230
    },
    {
      "entropy": 5.5559464454650875,
      "epoch": 1.2631394670297609,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00048433898638232755,
      "loss": 5.2884,
      "mean_token_accuracy": 0.1736864283680916,
      "num_tokens": 28158593.0,
      "step": 16235
    },
    {
      "entropy": 5.503639888763428,
      "epoch": 1.2635284964014784,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004843288254443742,
      "loss": 5.2248,
      "mean_token_accuracy": 0.17802777588367463,
      "num_tokens": 28167301.0,
      "step": 16240
    },
    {
      "entropy": 5.524279689788818,
      "epoch": 1.2639175257731958,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004843186613301892,
      "loss": 5.2754,
      "mean_token_accuracy": 0.17966293692588806,
      "num_tokens": 28176092.0,
      "step": 16245
    },
    {
      "entropy": 5.599130153656006,
      "epoch": 1.2643065551449135,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004843084940399266,
      "loss": 5.3472,
      "mean_token_accuracy": 0.17642695009708403,
      "num_tokens": 28184570.0,
      "step": 16250
    },
    {
      "entropy": 5.530882930755615,
      "epoch": 1.264695584516631,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004842983235737408,
      "loss": 5.2617,
      "mean_token_accuracy": 0.1801413506269455,
      "num_tokens": 28192874.0,
      "step": 16255
    },
    {
      "entropy": 5.5411170482635494,
      "epoch": 1.2650846138883485,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004842881499317861,
      "loss": 5.2667,
      "mean_token_accuracy": 0.18436603993177414,
      "num_tokens": 28201546.0,
      "step": 16260
    },
    {
      "entropy": 5.49763593673706,
      "epoch": 1.2654736432600662,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004842779731142168,
      "loss": 5.2584,
      "mean_token_accuracy": 0.17174842804670334,
      "num_tokens": 28210061.0,
      "step": 16265
    },
    {
      "entropy": 5.432314538955689,
      "epoch": 1.2658626726317836,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00048426779312118735,
      "loss": 5.1915,
      "mean_token_accuracy": 0.17794633507728577,
      "num_tokens": 28219334.0,
      "step": 16270
    },
    {
      "entropy": 5.59707465171814,
      "epoch": 1.2662517020035013,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004842576099528522,
      "loss": 5.3288,
      "mean_token_accuracy": 0.16946832239627838,
      "num_tokens": 28227963.0,
      "step": 16275
    },
    {
      "entropy": 5.523435163497925,
      "epoch": 1.2666407313752188,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004842474236093659,
      "loss": 5.2924,
      "mean_token_accuracy": 0.17364021241664887,
      "num_tokens": 28236648.0,
      "step": 16280
    },
    {
      "entropy": 5.512435960769653,
      "epoch": 1.2670297607469365,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00048423723409088306,
      "loss": 5.2871,
      "mean_token_accuracy": 0.169327412545681,
      "num_tokens": 28245221.0,
      "step": 16285
    },
    {
      "entropy": 5.537802076339721,
      "epoch": 1.267418790118654,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004842270413975582,
      "loss": 5.2682,
      "mean_token_accuracy": 0.1720076471567154,
      "num_tokens": 28254481.0,
      "step": 16290
    },
    {
      "entropy": 5.513460254669189,
      "epoch": 1.2678078194903715,
      "grad_norm": 1.6640625,
      "learning_rate": 0.000484216845529546,
      "loss": 5.2509,
      "mean_token_accuracy": 0.17465183585882188,
      "num_tokens": 28263844.0,
      "step": 16295
    },
    {
      "entropy": 5.467709922790528,
      "epoch": 1.2681968488620892,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048420664648700113,
      "loss": 5.2071,
      "mean_token_accuracy": 0.17747659385204315,
      "num_tokens": 28272500.0,
      "step": 16300
    },
    {
      "entropy": 5.558680582046509,
      "epoch": 1.2685858782338066,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004841964442700784,
      "loss": 5.3074,
      "mean_token_accuracy": 0.16695282310247422,
      "num_tokens": 28281553.0,
      "step": 16305
    },
    {
      "entropy": 5.540439224243164,
      "epoch": 1.2689749076055241,
      "grad_norm": 1.515625,
      "learning_rate": 0.00048418623887893264,
      "loss": 5.2741,
      "mean_token_accuracy": 0.1742094039916992,
      "num_tokens": 28290078.0,
      "step": 16310
    },
    {
      "entropy": 5.483782529830933,
      "epoch": 1.2693639369772418,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048417603031371867,
      "loss": 5.1407,
      "mean_token_accuracy": 0.17784553468227388,
      "num_tokens": 28298879.0,
      "step": 16315
    },
    {
      "entropy": 5.457211303710937,
      "epoch": 1.2697529663489593,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004841658185745913,
      "loss": 5.2597,
      "mean_token_accuracy": 0.18327910006046294,
      "num_tokens": 28307188.0,
      "step": 16320
    },
    {
      "entropy": 5.445938491821289,
      "epoch": 1.270141995720677,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048415560366170564,
      "loss": 5.1752,
      "mean_token_accuracy": 0.18031149506568908,
      "num_tokens": 28316012.0,
      "step": 16325
    },
    {
      "entropy": 5.530198621749878,
      "epoch": 1.2705310250923945,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004841453855752165,
      "loss": 5.3491,
      "mean_token_accuracy": 0.1687226489186287,
      "num_tokens": 28325750.0,
      "step": 16330
    },
    {
      "entropy": 5.523533058166504,
      "epoch": 1.2709200544641122,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004841351643152791,
      "loss": 5.2985,
      "mean_token_accuracy": 0.17067915201187134,
      "num_tokens": 28335217.0,
      "step": 16335
    },
    {
      "entropy": 5.577079916000367,
      "epoch": 1.2713090838358296,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004841249398820485,
      "loss": 5.2589,
      "mean_token_accuracy": 0.17277070879936218,
      "num_tokens": 28344123.0,
      "step": 16340
    },
    {
      "entropy": 5.56252384185791,
      "epoch": 1.271698113207547,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004841147122756797,
      "loss": 5.2708,
      "mean_token_accuracy": 0.17112697809934616,
      "num_tokens": 28353230.0,
      "step": 16345
    },
    {
      "entropy": 5.526676177978516,
      "epoch": 1.2720871425792648,
      "grad_norm": 1.5703125,
      "learning_rate": 0.000484104481496328,
      "loss": 5.2831,
      "mean_token_accuracy": 0.17249808758497237,
      "num_tokens": 28361636.0,
      "step": 16350
    },
    {
      "entropy": 5.504556894302368,
      "epoch": 1.2724761719509823,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004840942475441486,
      "loss": 5.2533,
      "mean_token_accuracy": 0.170597605407238,
      "num_tokens": 28370066.0,
      "step": 16355
    },
    {
      "entropy": 5.527553272247315,
      "epoch": 1.2728652013226998,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004840840104192969,
      "loss": 5.3288,
      "mean_token_accuracy": 0.16599608063697815,
      "num_tokens": 28379017.0,
      "step": 16360
    },
    {
      "entropy": 5.591423749923706,
      "epoch": 1.2732542306944175,
      "grad_norm": 2.03125,
      "learning_rate": 0.000484073770121928,
      "loss": 5.2832,
      "mean_token_accuracy": 0.17235138714313508,
      "num_tokens": 28387939.0,
      "step": 16365
    },
    {
      "entropy": 5.5084977626800535,
      "epoch": 1.273643260066135,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004840635266521975,
      "loss": 5.1952,
      "mean_token_accuracy": 0.1758476436138153,
      "num_tokens": 28396341.0,
      "step": 16370
    },
    {
      "entropy": 5.467773008346557,
      "epoch": 1.2740322894378526,
      "grad_norm": 5.78125,
      "learning_rate": 0.0004840532800102607,
      "loss": 5.2073,
      "mean_token_accuracy": 0.17458854019641876,
      "num_tokens": 28405128.0,
      "step": 16375
    },
    {
      "entropy": 5.497469186782837,
      "epoch": 1.27442131880957,
      "grad_norm": 1.796875,
      "learning_rate": 0.00048404303019627314,
      "loss": 5.1921,
      "mean_token_accuracy": 0.17867213785648345,
      "num_tokens": 28412841.0,
      "step": 16380
    },
    {
      "entropy": 5.430998992919922,
      "epoch": 1.2748103481812878,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00048403277721039036,
      "loss": 5.1505,
      "mean_token_accuracy": 0.18427162021398544,
      "num_tokens": 28421526.0,
      "step": 16385
    },
    {
      "entropy": 5.4770458221435545,
      "epoch": 1.2751993775530053,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004840225210527679,
      "loss": 5.3253,
      "mean_token_accuracy": 0.17072549164295198,
      "num_tokens": 28430070.0,
      "step": 16390
    },
    {
      "entropy": 5.460269260406494,
      "epoch": 1.2755884069247228,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004840122617235613,
      "loss": 5.2053,
      "mean_token_accuracy": 0.17117018401622772,
      "num_tokens": 28438920.0,
      "step": 16395
    },
    {
      "entropy": 5.587383127212524,
      "epoch": 1.2759774362964404,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004840019992229263,
      "loss": 5.232,
      "mean_token_accuracy": 0.17959167063236237,
      "num_tokens": 28447666.0,
      "step": 16400
    },
    {
      "entropy": 5.534450912475586,
      "epoch": 1.276366465668158,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048399173355101867,
      "loss": 5.3605,
      "mean_token_accuracy": 0.16608233749866486,
      "num_tokens": 28457341.0,
      "step": 16405
    },
    {
      "entropy": 5.470757341384887,
      "epoch": 1.2767554950398754,
      "grad_norm": 1.578125,
      "learning_rate": 0.00048398146470799417,
      "loss": 5.2504,
      "mean_token_accuracy": 0.18641512244939804,
      "num_tokens": 28466202.0,
      "step": 16410
    },
    {
      "entropy": 5.602308511734009,
      "epoch": 1.277144524411593,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00048397119269400846,
      "loss": 5.3356,
      "mean_token_accuracy": 0.17155100107192994,
      "num_tokens": 28474494.0,
      "step": 16415
    },
    {
      "entropy": 5.576149225234985,
      "epoch": 1.2775335537833106,
      "grad_norm": 1.53125,
      "learning_rate": 0.00048396091750921766,
      "loss": 5.2951,
      "mean_token_accuracy": 0.1768299549818039,
      "num_tokens": 28482931.0,
      "step": 16420
    },
    {
      "entropy": 5.48293981552124,
      "epoch": 1.2779225831550283,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004839506391537774,
      "loss": 5.2548,
      "mean_token_accuracy": 0.17274906784296035,
      "num_tokens": 28491372.0,
      "step": 16425
    },
    {
      "entropy": 5.447651052474976,
      "epoch": 1.2783116125267457,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004839403576278438,
      "loss": 5.1784,
      "mean_token_accuracy": 0.18031758815050125,
      "num_tokens": 28500201.0,
      "step": 16430
    },
    {
      "entropy": 5.464600706100464,
      "epoch": 1.2787006418984634,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004839300729315729,
      "loss": 5.2059,
      "mean_token_accuracy": 0.17613438218832017,
      "num_tokens": 28508899.0,
      "step": 16435
    },
    {
      "entropy": 5.5099687576293945,
      "epoch": 1.279089671270181,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004839197850651206,
      "loss": 5.2897,
      "mean_token_accuracy": 0.17009116113185882,
      "num_tokens": 28518280.0,
      "step": 16440
    },
    {
      "entropy": 5.590722703933716,
      "epoch": 1.2794787006418984,
      "grad_norm": 1.59375,
      "learning_rate": 0.00048390949402864317,
      "loss": 5.3411,
      "mean_token_accuracy": 0.17054921835660936,
      "num_tokens": 28527113.0,
      "step": 16445
    },
    {
      "entropy": 5.445984888076782,
      "epoch": 1.279867730013616,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00048389919982229666,
      "loss": 5.2428,
      "mean_token_accuracy": 0.1773141548037529,
      "num_tokens": 28535253.0,
      "step": 16450
    },
    {
      "entropy": 5.414700889587403,
      "epoch": 1.2802567593853336,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048388890244623723,
      "loss": 5.2753,
      "mean_token_accuracy": 0.17602664083242417,
      "num_tokens": 28543475.0,
      "step": 16455
    },
    {
      "entropy": 5.58476300239563,
      "epoch": 1.280645788757051,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004838786019006213,
      "loss": 5.3341,
      "mean_token_accuracy": 0.17287300676107406,
      "num_tokens": 28551921.0,
      "step": 16460
    },
    {
      "entropy": 5.530067205429077,
      "epoch": 1.2810348181287687,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00048386829818560493,
      "loss": 5.2924,
      "mean_token_accuracy": 0.17198163717985154,
      "num_tokens": 28560413.0,
      "step": 16465
    },
    {
      "entropy": 5.57966251373291,
      "epoch": 1.2814238475004862,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048385799130134464,
      "loss": 5.3328,
      "mean_token_accuracy": 0.17408890128135682,
      "num_tokens": 28568802.0,
      "step": 16470
    },
    {
      "entropy": 5.520438098907471,
      "epoch": 1.281812876872204,
      "grad_norm": 1.5,
      "learning_rate": 0.0004838476812479968,
      "loss": 5.4196,
      "mean_token_accuracy": 0.1665988437831402,
      "num_tokens": 28578164.0,
      "step": 16475
    },
    {
      "entropy": 5.5440600395202635,
      "epoch": 1.2822019062439214,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048383736802571774,
      "loss": 5.1965,
      "mean_token_accuracy": 0.18536286652088166,
      "num_tokens": 28586061.0,
      "step": 16480
    },
    {
      "entropy": 5.493400049209595,
      "epoch": 1.282590935615639,
      "grad_norm": 1.4375,
      "learning_rate": 0.000483827051634664,
      "loss": 5.1938,
      "mean_token_accuracy": 0.17700058668851854,
      "num_tokens": 28594185.0,
      "step": 16485
    },
    {
      "entropy": 5.458246231079102,
      "epoch": 1.2829799649873566,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00048381673207499224,
      "loss": 5.2435,
      "mean_token_accuracy": 0.1800957813858986,
      "num_tokens": 28602427.0,
      "step": 16490
    },
    {
      "entropy": 5.46396837234497,
      "epoch": 1.283368994359074,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004838064093468588,
      "loss": 5.2042,
      "mean_token_accuracy": 0.17680867612361909,
      "num_tokens": 28611608.0,
      "step": 16495
    },
    {
      "entropy": 5.487648820877075,
      "epoch": 1.2837580237307917,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004837960834504206,
      "loss": 5.204,
      "mean_token_accuracy": 0.18109429329633714,
      "num_tokens": 28620142.0,
      "step": 16500
    },
    {
      "entropy": 5.512625360488892,
      "epoch": 1.2841470531025092,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004837857543858341,
      "loss": 5.3326,
      "mean_token_accuracy": 0.17388014048337935,
      "num_tokens": 28629401.0,
      "step": 16505
    },
    {
      "entropy": 5.537943506240845,
      "epoch": 1.2845360824742267,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004837754221532561,
      "loss": 5.2438,
      "mean_token_accuracy": 0.17308045625686647,
      "num_tokens": 28637934.0,
      "step": 16510
    },
    {
      "entropy": 5.494345378875733,
      "epoch": 1.2849251118459444,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00048376508675284334,
      "loss": 5.2738,
      "mean_token_accuracy": 0.17197916358709336,
      "num_tokens": 28647145.0,
      "step": 16515
    },
    {
      "entropy": 5.511820554733276,
      "epoch": 1.2853141412176619,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00048375474818475274,
      "loss": 5.348,
      "mean_token_accuracy": 0.16744939982891083,
      "num_tokens": 28656146.0,
      "step": 16520
    },
    {
      "entropy": 5.521165418624878,
      "epoch": 1.2857031705893796,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00048374440644914104,
      "loss": 5.1696,
      "mean_token_accuracy": 0.17831232994794846,
      "num_tokens": 28664476.0,
      "step": 16525
    },
    {
      "entropy": 5.506192064285278,
      "epoch": 1.286092199961097,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048373406154616523,
      "loss": 5.1949,
      "mean_token_accuracy": 0.1829805016517639,
      "num_tokens": 28672188.0,
      "step": 16530
    },
    {
      "entropy": 5.4434877872467045,
      "epoch": 1.2864812293328147,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048372371347598226,
      "loss": 5.2932,
      "mean_token_accuracy": 0.17740696519613267,
      "num_tokens": 28680758.0,
      "step": 16535
    },
    {
      "entropy": 5.660080194473267,
      "epoch": 1.2868702587045322,
      "grad_norm": 2.515625,
      "learning_rate": 0.00048371336223874923,
      "loss": 5.4143,
      "mean_token_accuracy": 0.16645713970065118,
      "num_tokens": 28690392.0,
      "step": 16540
    },
    {
      "entropy": 5.547315073013306,
      "epoch": 1.2872592880762497,
      "grad_norm": 1.359375,
      "learning_rate": 0.000483703007834623,
      "loss": 5.2844,
      "mean_token_accuracy": 0.16991337090730668,
      "num_tokens": 28699121.0,
      "step": 16545
    },
    {
      "entropy": 5.575688982009888,
      "epoch": 1.2876483174479674,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004836926502637609,
      "loss": 5.3706,
      "mean_token_accuracy": 0.1695497527718544,
      "num_tokens": 28707692.0,
      "step": 16550
    },
    {
      "entropy": 5.533820629119873,
      "epoch": 1.2880373468196848,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048368228952632005,
      "loss": 5.3026,
      "mean_token_accuracy": 0.17484888136386872,
      "num_tokens": 28716630.0,
      "step": 16555
    },
    {
      "entropy": 5.501118946075439,
      "epoch": 1.2884263761914023,
      "grad_norm": 1.5,
      "learning_rate": 0.00048367192562245756,
      "loss": 5.1848,
      "mean_token_accuracy": 0.18588481545448304,
      "num_tokens": 28725503.0,
      "step": 16560
    },
    {
      "entropy": 5.552927112579345,
      "epoch": 1.28881540556312,
      "grad_norm": 1.5625,
      "learning_rate": 0.00048366155855233067,
      "loss": 5.3029,
      "mean_token_accuracy": 0.17311994433403016,
      "num_tokens": 28734578.0,
      "step": 16565
    },
    {
      "entropy": 5.5685888767242435,
      "epoch": 1.2892044349348377,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004836511883160968,
      "loss": 5.3426,
      "mean_token_accuracy": 0.1714356303215027,
      "num_tokens": 28743380.0,
      "step": 16570
    },
    {
      "entropy": 5.557417201995849,
      "epoch": 1.2895934643065552,
      "grad_norm": 1.5,
      "learning_rate": 0.0004836408149139133,
      "loss": 5.3036,
      "mean_token_accuracy": 0.1725929245352745,
      "num_tokens": 28752968.0,
      "step": 16575
    },
    {
      "entropy": 5.508104467391968,
      "epoch": 1.2899824936782727,
      "grad_norm": 1.625,
      "learning_rate": 0.00048363043834593743,
      "loss": 5.3067,
      "mean_token_accuracy": 0.16962089091539384,
      "num_tokens": 28761756.0,
      "step": 16580
    },
    {
      "entropy": 5.504446744918823,
      "epoch": 1.2903715230499904,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004836200586123268,
      "loss": 5.2552,
      "mean_token_accuracy": 0.17421442717313768,
      "num_tokens": 28771110.0,
      "step": 16585
    },
    {
      "entropy": 5.55663275718689,
      "epoch": 1.2907605524217078,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00048360967571323875,
      "loss": 5.3467,
      "mean_token_accuracy": 0.16930615156888962,
      "num_tokens": 28780016.0,
      "step": 16590
    },
    {
      "entropy": 5.547071695327759,
      "epoch": 1.2911495817934253,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048359928964883094,
      "loss": 5.2443,
      "mean_token_accuracy": 0.1836158812046051,
      "num_tokens": 28788150.0,
      "step": 16595
    },
    {
      "entropy": 5.515267848968506,
      "epoch": 1.291538611165143,
      "grad_norm": 1.375,
      "learning_rate": 0.000483588900419261,
      "loss": 5.2265,
      "mean_token_accuracy": 0.17374784201383592,
      "num_tokens": 28796564.0,
      "step": 16600
    },
    {
      "entropy": 5.6450110912323,
      "epoch": 1.2919276405368605,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004835785080246864,
      "loss": 5.471,
      "mean_token_accuracy": 0.16478035002946853,
      "num_tokens": 28804987.0,
      "step": 16605
    },
    {
      "entropy": 5.577270603179931,
      "epoch": 1.292316669908578,
      "grad_norm": 1.34375,
      "learning_rate": 0.000483568112465265,
      "loss": 5.2501,
      "mean_token_accuracy": 0.17944055646657944,
      "num_tokens": 28813752.0,
      "step": 16610
    },
    {
      "entropy": 5.407899522781372,
      "epoch": 1.2927056992802957,
      "grad_norm": 1.5,
      "learning_rate": 0.00048355771374115436,
      "loss": 5.1681,
      "mean_token_accuracy": 0.18100990802049638,
      "num_tokens": 28822464.0,
      "step": 16615
    },
    {
      "entropy": 5.464084005355835,
      "epoch": 1.2930947286520134,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004835473118525125,
      "loss": 5.2223,
      "mean_token_accuracy": 0.1792286142706871,
      "num_tokens": 28830997.0,
      "step": 16620
    },
    {
      "entropy": 5.505066299438477,
      "epoch": 1.2934837580237308,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004835369067994971,
      "loss": 5.1946,
      "mean_token_accuracy": 0.1861805021762848,
      "num_tokens": 28839979.0,
      "step": 16625
    },
    {
      "entropy": 5.517421722412109,
      "epoch": 1.2938727873954483,
      "grad_norm": 1.375,
      "learning_rate": 0.000483526498582266,
      "loss": 5.2183,
      "mean_token_accuracy": 0.1755661517381668,
      "num_tokens": 28848394.0,
      "step": 16630
    },
    {
      "entropy": 5.651650047302246,
      "epoch": 1.294261816767166,
      "grad_norm": 2.0625,
      "learning_rate": 0.00048351608720097715,
      "loss": 5.5204,
      "mean_token_accuracy": 0.16302997022867202,
      "num_tokens": 28857804.0,
      "step": 16635
    },
    {
      "entropy": 5.5143248558044435,
      "epoch": 1.2946508461388835,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048350567265578867,
      "loss": 5.1797,
      "mean_token_accuracy": 0.17618067264556886,
      "num_tokens": 28866405.0,
      "step": 16640
    },
    {
      "entropy": 5.540298795700073,
      "epoch": 1.295039875510601,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004834952549468584,
      "loss": 5.2881,
      "mean_token_accuracy": 0.1797433838248253,
      "num_tokens": 28874751.0,
      "step": 16645
    },
    {
      "entropy": 5.561378765106201,
      "epoch": 1.2954289048823187,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004834848340743446,
      "loss": 5.3932,
      "mean_token_accuracy": 0.16223286539316178,
      "num_tokens": 28884013.0,
      "step": 16650
    },
    {
      "entropy": 5.54998459815979,
      "epoch": 1.2958179342540361,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004834744100384052,
      "loss": 5.1959,
      "mean_token_accuracy": 0.17620742470026016,
      "num_tokens": 28892175.0,
      "step": 16655
    },
    {
      "entropy": 5.581575584411621,
      "epoch": 1.2962069636257538,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004834639828391984,
      "loss": 5.2577,
      "mean_token_accuracy": 0.1740424305200577,
      "num_tokens": 28901238.0,
      "step": 16660
    },
    {
      "entropy": 5.525385522842408,
      "epoch": 1.2965959929974713,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048345355247688256,
      "loss": 5.2794,
      "mean_token_accuracy": 0.17541230767965316,
      "num_tokens": 28910368.0,
      "step": 16665
    },
    {
      "entropy": 5.548087453842163,
      "epoch": 1.296985022369189,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004834431189516158,
      "loss": 5.3785,
      "mean_token_accuracy": 0.16417302042245865,
      "num_tokens": 28918854.0,
      "step": 16670
    },
    {
      "entropy": 5.560145330429077,
      "epoch": 1.2973740517409065,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004834326822635565,
      "loss": 5.2917,
      "mean_token_accuracy": 0.17520996034145356,
      "num_tokens": 28927103.0,
      "step": 16675
    },
    {
      "entropy": 5.516687202453613,
      "epoch": 1.297763081112624,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00048342224241286296,
      "loss": 5.2083,
      "mean_token_accuracy": 0.1745814263820648,
      "num_tokens": 28936257.0,
      "step": 16680
    },
    {
      "entropy": 5.634287881851196,
      "epoch": 1.2981521104843416,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004834117993996937,
      "loss": 5.3561,
      "mean_token_accuracy": 0.1687796711921692,
      "num_tokens": 28944991.0,
      "step": 16685
    },
    {
      "entropy": 5.563974905014038,
      "epoch": 1.2985411398560591,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004834013532242071,
      "loss": 5.3248,
      "mean_token_accuracy": 0.17341580390930175,
      "num_tokens": 28953781.0,
      "step": 16690
    },
    {
      "entropy": 5.552775430679321,
      "epoch": 1.2989301692277766,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004833909038865616,
      "loss": 5.2094,
      "mean_token_accuracy": 0.1797615483403206,
      "num_tokens": 28961916.0,
      "step": 16695
    },
    {
      "entropy": 5.501725149154663,
      "epoch": 1.2993191985994943,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004833804513869159,
      "loss": 5.2453,
      "mean_token_accuracy": 0.17446848303079604,
      "num_tokens": 28970223.0,
      "step": 16700
    },
    {
      "entropy": 5.497382974624633,
      "epoch": 1.2997082279712118,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004833699957254284,
      "loss": 5.2919,
      "mean_token_accuracy": 0.17867342829704286,
      "num_tokens": 28979063.0,
      "step": 16705
    },
    {
      "entropy": 5.640632677078247,
      "epoch": 1.3000972573429295,
      "grad_norm": 1.796875,
      "learning_rate": 0.000483359536902258,
      "loss": 5.3819,
      "mean_token_accuracy": 0.1725120261311531,
      "num_tokens": 28987229.0,
      "step": 16710
    },
    {
      "entropy": 5.578525114059448,
      "epoch": 1.300486286714647,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004833490749175632,
      "loss": 5.3407,
      "mean_token_accuracy": 0.1683332324028015,
      "num_tokens": 28995889.0,
      "step": 16715
    },
    {
      "entropy": 5.548509788513184,
      "epoch": 1.3008753160863646,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048333860977150286,
      "loss": 5.3387,
      "mean_token_accuracy": 0.16908127963542938,
      "num_tokens": 29003962.0,
      "step": 16720
    },
    {
      "entropy": 5.508148670196533,
      "epoch": 1.3012643454580821,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048332814146423566,
      "loss": 5.2254,
      "mean_token_accuracy": 0.17627947479486467,
      "num_tokens": 29012652.0,
      "step": 16725
    },
    {
      "entropy": 5.631994438171387,
      "epoch": 1.3016533748297996,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004833176699959206,
      "loss": 5.3728,
      "mean_token_accuracy": 0.16163735389709472,
      "num_tokens": 29022445.0,
      "step": 16730
    },
    {
      "entropy": 5.546034574508667,
      "epoch": 1.3020424042015173,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048330719536671633,
      "loss": 5.2529,
      "mean_token_accuracy": 0.16912207007408142,
      "num_tokens": 29030840.0,
      "step": 16735
    },
    {
      "entropy": 5.568957138061523,
      "epoch": 1.3024314335732348,
      "grad_norm": 1.421875,
      "learning_rate": 0.000483296717576782,
      "loss": 5.3078,
      "mean_token_accuracy": 0.1755591794848442,
      "num_tokens": 29039640.0,
      "step": 16740
    },
    {
      "entropy": 5.448435497283936,
      "epoch": 1.3028204629449522,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004832862366262765,
      "loss": 5.2497,
      "mean_token_accuracy": 0.17302780151367186,
      "num_tokens": 29047793.0,
      "step": 16745
    },
    {
      "entropy": 5.486004590988159,
      "epoch": 1.30320949231667,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048327575251535886,
      "loss": 5.2357,
      "mean_token_accuracy": 0.17673268914222717,
      "num_tokens": 29056880.0,
      "step": 16750
    },
    {
      "entropy": 5.54179105758667,
      "epoch": 1.3035985216883874,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004832652652441883,
      "loss": 5.2527,
      "mean_token_accuracy": 0.17667125761508942,
      "num_tokens": 29065093.0,
      "step": 16755
    },
    {
      "entropy": 5.474070596694946,
      "epoch": 1.303987551060105,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048325477481292375,
      "loss": 5.3188,
      "mean_token_accuracy": 0.17161299139261246,
      "num_tokens": 29074389.0,
      "step": 16760
    },
    {
      "entropy": 5.413042831420898,
      "epoch": 1.3043765804318226,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004832442812217244,
      "loss": 5.1535,
      "mean_token_accuracy": 0.17977993041276932,
      "num_tokens": 29082789.0,
      "step": 16765
    },
    {
      "entropy": 5.587252378463745,
      "epoch": 1.3047656098035403,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004832337844707496,
      "loss": 5.4247,
      "mean_token_accuracy": 0.16643834859132767,
      "num_tokens": 29092831.0,
      "step": 16770
    },
    {
      "entropy": 5.458986759185791,
      "epoch": 1.3051546391752578,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00048322328456015855,
      "loss": 5.1172,
      "mean_token_accuracy": 0.1841055527329445,
      "num_tokens": 29101067.0,
      "step": 16775
    },
    {
      "entropy": 5.520122098922729,
      "epoch": 1.3055436685469752,
      "grad_norm": 1.59375,
      "learning_rate": 0.00048321278149011053,
      "loss": 5.2925,
      "mean_token_accuracy": 0.1647133469581604,
      "num_tokens": 29110423.0,
      "step": 16780
    },
    {
      "entropy": 5.515471601486206,
      "epoch": 1.305932697918693,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048320227526076504,
      "loss": 5.2013,
      "mean_token_accuracy": 0.17780290842056273,
      "num_tokens": 29119508.0,
      "step": 16785
    },
    {
      "entropy": 5.467047166824341,
      "epoch": 1.3063217272904104,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004831917658722814,
      "loss": 5.2822,
      "mean_token_accuracy": 0.1781074270606041,
      "num_tokens": 29127899.0,
      "step": 16790
    },
    {
      "entropy": 5.554867172241211,
      "epoch": 1.3067107566621279,
      "grad_norm": 1.53125,
      "learning_rate": 0.00048318125332481903,
      "loss": 5.3227,
      "mean_token_accuracy": 0.17663970291614534,
      "num_tokens": 29137261.0,
      "step": 16795
    },
    {
      "entropy": 5.5556189060211185,
      "epoch": 1.3070997860338456,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048317073761853764,
      "loss": 5.3536,
      "mean_token_accuracy": 0.17130178213119507,
      "num_tokens": 29145439.0,
      "step": 16800
    },
    {
      "entropy": 5.542152452468872,
      "epoch": 1.307488815405563,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004831602187535965,
      "loss": 5.3811,
      "mean_token_accuracy": 0.16846368610858917,
      "num_tokens": 29153871.0,
      "step": 16805
    },
    {
      "entropy": 5.494944667816162,
      "epoch": 1.3078778447772808,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004831496967301554,
      "loss": 5.219,
      "mean_token_accuracy": 0.17613357603549956,
      "num_tokens": 29162343.0,
      "step": 16810
    },
    {
      "entropy": 5.4702600002288815,
      "epoch": 1.3082668741489982,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048313917154837386,
      "loss": 5.2666,
      "mean_token_accuracy": 0.1772501453757286,
      "num_tokens": 29170533.0,
      "step": 16815
    },
    {
      "entropy": 5.483978128433227,
      "epoch": 1.308655903520716,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004831286432084118,
      "loss": 5.3002,
      "mean_token_accuracy": 0.17461176365613937,
      "num_tokens": 29179353.0,
      "step": 16820
    },
    {
      "entropy": 5.505611944198608,
      "epoch": 1.3090449328924334,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004831181117104289,
      "loss": 5.2637,
      "mean_token_accuracy": 0.16662295609712602,
      "num_tokens": 29188747.0,
      "step": 16825
    },
    {
      "entropy": 5.509248971939087,
      "epoch": 1.3094339622641509,
      "grad_norm": 1.546875,
      "learning_rate": 0.00048310757705458476,
      "loss": 5.2485,
      "mean_token_accuracy": 0.17222723960876465,
      "num_tokens": 29197302.0,
      "step": 16830
    },
    {
      "entropy": 5.496852779388428,
      "epoch": 1.3098229916358686,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048309703924103944,
      "loss": 5.2638,
      "mean_token_accuracy": 0.17514885365962982,
      "num_tokens": 29205667.0,
      "step": 16835
    },
    {
      "entropy": 5.606069326400757,
      "epoch": 1.310212021007586,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048308649826995283,
      "loss": 5.331,
      "mean_token_accuracy": 0.1714055135846138,
      "num_tokens": 29214475.0,
      "step": 16840
    },
    {
      "entropy": 5.485038614273071,
      "epoch": 1.3106010503793035,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048307595414148475,
      "loss": 5.2256,
      "mean_token_accuracy": 0.1753092348575592,
      "num_tokens": 29222990.0,
      "step": 16845
    },
    {
      "entropy": 5.500304174423218,
      "epoch": 1.3109900797510212,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004830654068557952,
      "loss": 5.351,
      "mean_token_accuracy": 0.17122073471546173,
      "num_tokens": 29232351.0,
      "step": 16850
    },
    {
      "entropy": 5.591643857955932,
      "epoch": 1.3113791091227387,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004830548564130444,
      "loss": 5.2175,
      "mean_token_accuracy": 0.17079212963581086,
      "num_tokens": 29240418.0,
      "step": 16855
    },
    {
      "entropy": 5.58412914276123,
      "epoch": 1.3117681384944564,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048304430281339216,
      "loss": 5.3266,
      "mean_token_accuracy": 0.17036605775356292,
      "num_tokens": 29248975.0,
      "step": 16860
    },
    {
      "entropy": 5.466780281066894,
      "epoch": 1.3121571678661739,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004830337460569989,
      "loss": 5.299,
      "mean_token_accuracy": 0.16868091225624085,
      "num_tokens": 29257866.0,
      "step": 16865
    },
    {
      "entropy": 5.447692203521728,
      "epoch": 1.3125461972378916,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004830231861440246,
      "loss": 5.0762,
      "mean_token_accuracy": 0.19014528691768645,
      "num_tokens": 29266057.0,
      "step": 16870
    },
    {
      "entropy": 5.472238826751709,
      "epoch": 1.312935226609609,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004830126230746295,
      "loss": 5.2396,
      "mean_token_accuracy": 0.17188325077295302,
      "num_tokens": 29274752.0,
      "step": 16875
    },
    {
      "entropy": 5.591193723678589,
      "epoch": 1.3133242559813265,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048300205684897405,
      "loss": 5.2972,
      "mean_token_accuracy": 0.17175144106149673,
      "num_tokens": 29282781.0,
      "step": 16880
    },
    {
      "entropy": 5.4347821235656735,
      "epoch": 1.3137132853530442,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004829914874672184,
      "loss": 5.2579,
      "mean_token_accuracy": 0.1727503716945648,
      "num_tokens": 29291196.0,
      "step": 16885
    },
    {
      "entropy": 5.507078647613525,
      "epoch": 1.3141023147247617,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048298091492952297,
      "loss": 5.3608,
      "mean_token_accuracy": 0.17136398255825042,
      "num_tokens": 29300168.0,
      "step": 16890
    },
    {
      "entropy": 5.51510066986084,
      "epoch": 1.3144913440964792,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004829703392360482,
      "loss": 5.1742,
      "mean_token_accuracy": 0.17384043484926223,
      "num_tokens": 29308468.0,
      "step": 16895
    },
    {
      "entropy": 5.42665958404541,
      "epoch": 1.3148803734681969,
      "grad_norm": 1.625,
      "learning_rate": 0.00048295976038695455,
      "loss": 5.1956,
      "mean_token_accuracy": 0.1762784630060196,
      "num_tokens": 29317096.0,
      "step": 16900
    },
    {
      "entropy": 5.507435703277588,
      "epoch": 1.3152694028399143,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004829491783824025,
      "loss": 5.3237,
      "mean_token_accuracy": 0.16330843567848205,
      "num_tokens": 29325840.0,
      "step": 16905
    },
    {
      "entropy": 5.528168678283691,
      "epoch": 1.315658432211632,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00048293859322255276,
      "loss": 5.2574,
      "mean_token_accuracy": 0.17604086101055144,
      "num_tokens": 29333874.0,
      "step": 16910
    },
    {
      "entropy": 5.513610696792602,
      "epoch": 1.3160474615833495,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004829280049075657,
      "loss": 5.1969,
      "mean_token_accuracy": 0.17857919931411742,
      "num_tokens": 29341937.0,
      "step": 16915
    },
    {
      "entropy": 5.604503917694092,
      "epoch": 1.3164364909550672,
      "grad_norm": 1.5,
      "learning_rate": 0.00048291741343760216,
      "loss": 5.4431,
      "mean_token_accuracy": 0.16034996807575225,
      "num_tokens": 29350231.0,
      "step": 16920
    },
    {
      "entropy": 5.566038799285889,
      "epoch": 1.3168255203267847,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004829068188128229,
      "loss": 5.291,
      "mean_token_accuracy": 0.17317449748516084,
      "num_tokens": 29359036.0,
      "step": 16925
    },
    {
      "entropy": 5.547137355804443,
      "epoch": 1.3172145496985022,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004828962210333885,
      "loss": 5.2173,
      "mean_token_accuracy": 0.17305073589086534,
      "num_tokens": 29368063.0,
      "step": 16930
    },
    {
      "entropy": 5.5168146133422855,
      "epoch": 1.3176035790702199,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004828856200994598,
      "loss": 5.1926,
      "mean_token_accuracy": 0.1722488895058632,
      "num_tokens": 29376653.0,
      "step": 16935
    },
    {
      "entropy": 5.454179811477661,
      "epoch": 1.3179926084419373,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004828750160111978,
      "loss": 5.1639,
      "mean_token_accuracy": 0.1818734660744667,
      "num_tokens": 29384978.0,
      "step": 16940
    },
    {
      "entropy": 5.557752084732056,
      "epoch": 1.3183816378136548,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048286440876876325,
      "loss": 5.3509,
      "mean_token_accuracy": 0.16898144632577897,
      "num_tokens": 29393927.0,
      "step": 16945
    },
    {
      "entropy": 5.617773962020874,
      "epoch": 1.3187706671853725,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048285379837231714,
      "loss": 5.3688,
      "mean_token_accuracy": 0.17309068590402604,
      "num_tokens": 29402393.0,
      "step": 16950
    },
    {
      "entropy": 5.540349960327148,
      "epoch": 1.31915969655709,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004828431848220205,
      "loss": 5.2691,
      "mean_token_accuracy": 0.17470085322856904,
      "num_tokens": 29411011.0,
      "step": 16955
    },
    {
      "entropy": 5.511628770828247,
      "epoch": 1.3195487259288077,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004828325681180343,
      "loss": 5.3238,
      "mean_token_accuracy": 0.17451790273189544,
      "num_tokens": 29420095.0,
      "step": 16960
    },
    {
      "entropy": 5.499999713897705,
      "epoch": 1.3199377553005252,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004828219482605197,
      "loss": 5.2824,
      "mean_token_accuracy": 0.1737119346857071,
      "num_tokens": 29429274.0,
      "step": 16965
    },
    {
      "entropy": 5.573909902572632,
      "epoch": 1.3203267846722428,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00048281132524963786,
      "loss": 5.2599,
      "mean_token_accuracy": 0.1712793916463852,
      "num_tokens": 29437572.0,
      "step": 16970
    },
    {
      "entropy": 5.572024631500244,
      "epoch": 1.3207158140439603,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004828006990855499,
      "loss": 5.3103,
      "mean_token_accuracy": 0.1769681081175804,
      "num_tokens": 29446294.0,
      "step": 16975
    },
    {
      "entropy": 5.498513841629029,
      "epoch": 1.3211048434156778,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048279006976841704,
      "loss": 5.2685,
      "mean_token_accuracy": 0.17740708738565444,
      "num_tokens": 29454710.0,
      "step": 16980
    },
    {
      "entropy": 5.525652360916138,
      "epoch": 1.3214938727873955,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004827794372984007,
      "loss": 5.2778,
      "mean_token_accuracy": 0.1790226951241493,
      "num_tokens": 29463067.0,
      "step": 16985
    },
    {
      "entropy": 5.546511125564575,
      "epoch": 1.321882902159113,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000482768801675662,
      "loss": 5.2809,
      "mean_token_accuracy": 0.16328697949647902,
      "num_tokens": 29471636.0,
      "step": 16990
    },
    {
      "entropy": 5.566883897781372,
      "epoch": 1.3222719315308304,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004827581629003625,
      "loss": 5.2979,
      "mean_token_accuracy": 0.17054706811904907,
      "num_tokens": 29480772.0,
      "step": 16995
    },
    {
      "entropy": 5.467402076721191,
      "epoch": 1.3226609609025481,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048274752097266356,
      "loss": 5.2057,
      "mean_token_accuracy": 0.17657761573791503,
      "num_tokens": 29489113.0,
      "step": 17000
    },
    {
      "entropy": 5.5701793193817135,
      "epoch": 1.3230499902742658,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004827368758927266,
      "loss": 5.2834,
      "mean_token_accuracy": 0.16945007741451262,
      "num_tokens": 29497721.0,
      "step": 17005
    },
    {
      "entropy": 5.523563241958618,
      "epoch": 1.3234390196459833,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004827262276607132,
      "loss": 5.2763,
      "mean_token_accuracy": 0.17571234554052353,
      "num_tokens": 29506452.0,
      "step": 17010
    },
    {
      "entropy": 5.52751932144165,
      "epoch": 1.3238280490177008,
      "grad_norm": 1.40625,
      "learning_rate": 0.000482715576276785,
      "loss": 5.2437,
      "mean_token_accuracy": 0.17883221805095673,
      "num_tokens": 29514635.0,
      "step": 17015
    },
    {
      "entropy": 5.543888664245605,
      "epoch": 1.3242170783894185,
      "grad_norm": 1.625,
      "learning_rate": 0.0004827049217411035,
      "loss": 5.3094,
      "mean_token_accuracy": 0.17182139605283736,
      "num_tokens": 29523929.0,
      "step": 17020
    },
    {
      "entropy": 5.5267242908477785,
      "epoch": 1.324606107761136,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00048269426405383043,
      "loss": 5.2353,
      "mean_token_accuracy": 0.17125312238931656,
      "num_tokens": 29533125.0,
      "step": 17025
    },
    {
      "entropy": 5.51857590675354,
      "epoch": 1.3249951371328534,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00048268360321512745,
      "loss": 5.241,
      "mean_token_accuracy": 0.1842625692486763,
      "num_tokens": 29541604.0,
      "step": 17030
    },
    {
      "entropy": 5.5429747104644775,
      "epoch": 1.3253841665045711,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004826729392251564,
      "loss": 5.2495,
      "mean_token_accuracy": 0.17560553252696992,
      "num_tokens": 29550100.0,
      "step": 17035
    },
    {
      "entropy": 5.497852230072022,
      "epoch": 1.3257731958762886,
      "grad_norm": 1.515625,
      "learning_rate": 0.000482662272084079,
      "loss": 5.3096,
      "mean_token_accuracy": 0.17464201301336288,
      "num_tokens": 29559490.0,
      "step": 17040
    },
    {
      "entropy": 5.578866386413575,
      "epoch": 1.326162225248006,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004826516017920571,
      "loss": 5.3156,
      "mean_token_accuracy": 0.1663801148533821,
      "num_tokens": 29568734.0,
      "step": 17045
    },
    {
      "entropy": 5.558235454559326,
      "epoch": 1.3265512546197238,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004826409283492528,
      "loss": 5.3005,
      "mean_token_accuracy": 0.16661880016326905,
      "num_tokens": 29577339.0,
      "step": 17050
    },
    {
      "entropy": 5.594893550872802,
      "epoch": 1.3269402839914415,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00048263025175582783,
      "loss": 5.4069,
      "mean_token_accuracy": 0.17300795316696166,
      "num_tokens": 29585561.0,
      "step": 17055
    },
    {
      "entropy": 5.532519292831421,
      "epoch": 1.327329313363159,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004826195720119442,
      "loss": 5.3405,
      "mean_token_accuracy": 0.1671331435441971,
      "num_tokens": 29594663.0,
      "step": 17060
    },
    {
      "entropy": 5.562968492507935,
      "epoch": 1.3277183427348764,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004826088891177641,
      "loss": 5.2927,
      "mean_token_accuracy": 0.17947182208299636,
      "num_tokens": 29603537.0,
      "step": 17065
    },
    {
      "entropy": 5.561071825027466,
      "epoch": 1.3281073721065941,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048259820307344954,
      "loss": 5.3108,
      "mean_token_accuracy": 0.1721128538250923,
      "num_tokens": 29612234.0,
      "step": 17070
    },
    {
      "entropy": 5.477893447875976,
      "epoch": 1.3284964014783116,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048258751387916267,
      "loss": 5.1261,
      "mean_token_accuracy": 0.17609235048294067,
      "num_tokens": 29621435.0,
      "step": 17075
    },
    {
      "entropy": 5.65094404220581,
      "epoch": 1.328885430850029,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048257682153506564,
      "loss": 5.4109,
      "mean_token_accuracy": 0.16384723782539368,
      "num_tokens": 29630266.0,
      "step": 17080
    },
    {
      "entropy": 5.425221300125122,
      "epoch": 1.3292744602217468,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004825661260413208,
      "loss": 5.2022,
      "mean_token_accuracy": 0.17390880286693572,
      "num_tokens": 29639118.0,
      "step": 17085
    },
    {
      "entropy": 5.5009173393249515,
      "epoch": 1.3296634895934643,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00048255542739809035,
      "loss": 5.2909,
      "mean_token_accuracy": 0.1802072197198868,
      "num_tokens": 29647616.0,
      "step": 17090
    },
    {
      "entropy": 5.591170930862427,
      "epoch": 1.330052518965182,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048254472560553665,
      "loss": 5.2439,
      "mean_token_accuracy": 0.175567227602005,
      "num_tokens": 29656264.0,
      "step": 17095
    },
    {
      "entropy": 5.540565347671508,
      "epoch": 1.3304415483368994,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048253402066382207,
      "loss": 5.2545,
      "mean_token_accuracy": 0.17449957877397537,
      "num_tokens": 29664452.0,
      "step": 17100
    },
    {
      "entropy": 5.520974826812744,
      "epoch": 1.3308305777086171,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004825233125731091,
      "loss": 5.203,
      "mean_token_accuracy": 0.18167235553264618,
      "num_tokens": 29673456.0,
      "step": 17105
    },
    {
      "entropy": 5.562027359008789,
      "epoch": 1.3312196070803346,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048251260133356014,
      "loss": 5.322,
      "mean_token_accuracy": 0.17192415595054628,
      "num_tokens": 29682232.0,
      "step": 17110
    },
    {
      "entropy": 5.454007339477539,
      "epoch": 1.331608636452052,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048250188694533774,
      "loss": 5.2339,
      "mean_token_accuracy": 0.17304678857326508,
      "num_tokens": 29690421.0,
      "step": 17115
    },
    {
      "entropy": 5.4780511379241945,
      "epoch": 1.3319976658237698,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004824911694086044,
      "loss": 5.2317,
      "mean_token_accuracy": 0.17659468501806258,
      "num_tokens": 29698844.0,
      "step": 17120
    },
    {
      "entropy": 5.552738618850708,
      "epoch": 1.3323866951954872,
      "grad_norm": 1.640625,
      "learning_rate": 0.000482480448723523,
      "loss": 5.2453,
      "mean_token_accuracy": 0.17272809147834778,
      "num_tokens": 29706738.0,
      "step": 17125
    },
    {
      "entropy": 5.552623605728149,
      "epoch": 1.3327757245672047,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00048246972489025586,
      "loss": 5.3074,
      "mean_token_accuracy": 0.17191975116729735,
      "num_tokens": 29715812.0,
      "step": 17130
    },
    {
      "entropy": 5.46045331954956,
      "epoch": 1.3331647539389224,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048245899790896597,
      "loss": 5.2165,
      "mean_token_accuracy": 0.17677971422672273,
      "num_tokens": 29724488.0,
      "step": 17135
    },
    {
      "entropy": 5.487279272079467,
      "epoch": 1.33355378331064,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004824482677798159,
      "loss": 5.1985,
      "mean_token_accuracy": 0.1763208717107773,
      "num_tokens": 29733565.0,
      "step": 17140
    },
    {
      "entropy": 5.503689336776733,
      "epoch": 1.3339428126823576,
      "grad_norm": 2.046875,
      "learning_rate": 0.00048243753450296863,
      "loss": 5.2947,
      "mean_token_accuracy": 0.17290622144937515,
      "num_tokens": 29742437.0,
      "step": 17145
    },
    {
      "entropy": 5.536953592300415,
      "epoch": 1.334331842054075,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048242679807858693,
      "loss": 5.3139,
      "mean_token_accuracy": 0.1747690573334694,
      "num_tokens": 29751636.0,
      "step": 17150
    },
    {
      "entropy": 5.537477731704712,
      "epoch": 1.3347208714257928,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00048241605850683365,
      "loss": 5.2542,
      "mean_token_accuracy": 0.17375836819410323,
      "num_tokens": 29760199.0,
      "step": 17155
    },
    {
      "entropy": 5.515740203857422,
      "epoch": 1.3351099007975102,
      "grad_norm": 1.484375,
      "learning_rate": 0.00048240531578787197,
      "loss": 5.2316,
      "mean_token_accuracy": 0.17604660391807556,
      "num_tokens": 29769169.0,
      "step": 17160
    },
    {
      "entropy": 5.492473697662353,
      "epoch": 1.3354989301692277,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004823945699218647,
      "loss": 5.2275,
      "mean_token_accuracy": 0.18034135401248932,
      "num_tokens": 29777218.0,
      "step": 17165
    },
    {
      "entropy": 5.518818998336792,
      "epoch": 1.3358879595409454,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004823838209089749,
      "loss": 5.2612,
      "mean_token_accuracy": 0.17358344197273254,
      "num_tokens": 29786221.0,
      "step": 17170
    },
    {
      "entropy": 5.551379585266114,
      "epoch": 1.3362769889126629,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048237306874936565,
      "loss": 5.287,
      "mean_token_accuracy": 0.1712179586291313,
      "num_tokens": 29794659.0,
      "step": 17175
    },
    {
      "entropy": 5.61849308013916,
      "epoch": 1.3366660182843804,
      "grad_norm": 5.21875,
      "learning_rate": 0.0004823623134432001,
      "loss": 5.2747,
      "mean_token_accuracy": 0.17288178503513335,
      "num_tokens": 29804237.0,
      "step": 17180
    },
    {
      "entropy": 5.625817441940308,
      "epoch": 1.337055047656098,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048235155499064166,
      "loss": 5.3764,
      "mean_token_accuracy": 0.17292446345090867,
      "num_tokens": 29813521.0,
      "step": 17185
    },
    {
      "entropy": 5.567737722396851,
      "epoch": 1.3374440770278155,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00048234079339185335,
      "loss": 5.4118,
      "mean_token_accuracy": 0.16924850791692733,
      "num_tokens": 29822123.0,
      "step": 17190
    },
    {
      "entropy": 5.562762832641601,
      "epoch": 1.3378331063995332,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048233002864699856,
      "loss": 5.3288,
      "mean_token_accuracy": 0.17189282029867173,
      "num_tokens": 29830557.0,
      "step": 17195
    },
    {
      "entropy": 5.525648307800293,
      "epoch": 1.3382221357712507,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004823192607562405,
      "loss": 5.2644,
      "mean_token_accuracy": 0.1800930082798004,
      "num_tokens": 29839566.0,
      "step": 17200
    },
    {
      "entropy": 5.547825002670288,
      "epoch": 1.3386111651429684,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048230848971974265,
      "loss": 5.2257,
      "mean_token_accuracy": 0.18331782519817352,
      "num_tokens": 29847808.0,
      "step": 17205
    },
    {
      "entropy": 5.5210363388061525,
      "epoch": 1.3390001945146859,
      "grad_norm": 1.5,
      "learning_rate": 0.0004822977155376684,
      "loss": 5.1894,
      "mean_token_accuracy": 0.1824951261281967,
      "num_tokens": 29855974.0,
      "step": 17210
    },
    {
      "entropy": 5.562961387634277,
      "epoch": 1.3393892238864034,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004822869382101814,
      "loss": 5.3489,
      "mean_token_accuracy": 0.1708719938993454,
      "num_tokens": 29864910.0,
      "step": 17215
    },
    {
      "entropy": 5.619733810424805,
      "epoch": 1.339778253258121,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004822761577374449,
      "loss": 5.4062,
      "mean_token_accuracy": 0.17142790108919143,
      "num_tokens": 29873668.0,
      "step": 17220
    },
    {
      "entropy": 5.568241596221924,
      "epoch": 1.3401672826298385,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004822653741196227,
      "loss": 5.3239,
      "mean_token_accuracy": 0.17096898704767227,
      "num_tokens": 29882099.0,
      "step": 17225
    },
    {
      "entropy": 5.517499160766602,
      "epoch": 1.340556312001556,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004822545873568783,
      "loss": 5.2333,
      "mean_token_accuracy": 0.1769474998116493,
      "num_tokens": 29891022.0,
      "step": 17230
    },
    {
      "entropy": 5.522295522689819,
      "epoch": 1.3409453413732737,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048224379744937545,
      "loss": 5.2463,
      "mean_token_accuracy": 0.17085277736186982,
      "num_tokens": 29899270.0,
      "step": 17235
    },
    {
      "entropy": 5.584539270401001,
      "epoch": 1.3413343707449912,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048223300439727783,
      "loss": 5.3713,
      "mean_token_accuracy": 0.170196969807148,
      "num_tokens": 29908183.0,
      "step": 17240
    },
    {
      "entropy": 5.534591436386108,
      "epoch": 1.3417234001167089,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048222220820074926,
      "loss": 5.2891,
      "mean_token_accuracy": 0.1724605530500412,
      "num_tokens": 29917148.0,
      "step": 17245
    },
    {
      "entropy": 5.544017791748047,
      "epoch": 1.3421124294884264,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048221140885995346,
      "loss": 5.2134,
      "mean_token_accuracy": 0.17335040271282195,
      "num_tokens": 29926251.0,
      "step": 17250
    },
    {
      "entropy": 5.555476951599121,
      "epoch": 1.342501458860144,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004822006063750543,
      "loss": 5.3232,
      "mean_token_accuracy": 0.17271163016557695,
      "num_tokens": 29935000.0,
      "step": 17255
    },
    {
      "entropy": 5.517015266418457,
      "epoch": 1.3428904882318615,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048218980074621575,
      "loss": 5.1795,
      "mean_token_accuracy": 0.1885431557893753,
      "num_tokens": 29943046.0,
      "step": 17260
    },
    {
      "entropy": 5.550515747070312,
      "epoch": 1.343279517603579,
      "grad_norm": 1.5,
      "learning_rate": 0.00048217899197360183,
      "loss": 5.2784,
      "mean_token_accuracy": 0.17839820235967635,
      "num_tokens": 29952254.0,
      "step": 17265
    },
    {
      "entropy": 5.511128282546997,
      "epoch": 1.3436685469752967,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004821681800573764,
      "loss": 5.2408,
      "mean_token_accuracy": 0.1727580949664116,
      "num_tokens": 29961458.0,
      "step": 17270
    },
    {
      "entropy": 5.537738084793091,
      "epoch": 1.3440575763470142,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004821573649977036,
      "loss": 5.2193,
      "mean_token_accuracy": 0.17950380891561507,
      "num_tokens": 29969865.0,
      "step": 17275
    },
    {
      "entropy": 5.530964469909668,
      "epoch": 1.3444466057187316,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048214654679474753,
      "loss": 5.2639,
      "mean_token_accuracy": 0.17164423316717148,
      "num_tokens": 29978260.0,
      "step": 17280
    },
    {
      "entropy": 5.583264255523682,
      "epoch": 1.3448356350904493,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048213572544867224,
      "loss": 5.4019,
      "mean_token_accuracy": 0.1712228015065193,
      "num_tokens": 29987528.0,
      "step": 17285
    },
    {
      "entropy": 5.540467214584351,
      "epoch": 1.3452246644621668,
      "grad_norm": 1.375,
      "learning_rate": 0.00048212490095964213,
      "loss": 5.2426,
      "mean_token_accuracy": 0.17255399823188783,
      "num_tokens": 29995991.0,
      "step": 17290
    },
    {
      "entropy": 5.555293083190918,
      "epoch": 1.3456136938338845,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004821140733278212,
      "loss": 5.2137,
      "mean_token_accuracy": 0.17633496820926667,
      "num_tokens": 30004569.0,
      "step": 17295
    },
    {
      "entropy": 5.5175879955291744,
      "epoch": 1.346002723205602,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004821032425533739,
      "loss": 5.3079,
      "mean_token_accuracy": 0.17088200896978378,
      "num_tokens": 30014021.0,
      "step": 17300
    },
    {
      "entropy": 5.484115791320801,
      "epoch": 1.3463917525773197,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004820924086364646,
      "loss": 5.1275,
      "mean_token_accuracy": 0.1816444516181946,
      "num_tokens": 30021803.0,
      "step": 17305
    },
    {
      "entropy": 5.4960136890411375,
      "epoch": 1.3467807819490372,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048208157157725756,
      "loss": 5.275,
      "mean_token_accuracy": 0.17458716332912444,
      "num_tokens": 30030707.0,
      "step": 17310
    },
    {
      "entropy": 5.528691339492798,
      "epoch": 1.3471698113207546,
      "grad_norm": 1.671875,
      "learning_rate": 0.00048207073137591726,
      "loss": 5.3056,
      "mean_token_accuracy": 0.1705812394618988,
      "num_tokens": 30040235.0,
      "step": 17315
    },
    {
      "entropy": 5.610648584365845,
      "epoch": 1.3475588406924723,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048205988803260824,
      "loss": 5.3262,
      "mean_token_accuracy": 0.17061686962842942,
      "num_tokens": 30049361.0,
      "step": 17320
    },
    {
      "entropy": 5.517641830444336,
      "epoch": 1.3479478700641898,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00048204904154749496,
      "loss": 5.3024,
      "mean_token_accuracy": 0.17135556042194366,
      "num_tokens": 30058896.0,
      "step": 17325
    },
    {
      "entropy": 5.591673088073731,
      "epoch": 1.3483368994359073,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048203819192074206,
      "loss": 5.3229,
      "mean_token_accuracy": 0.16766200214624405,
      "num_tokens": 30067206.0,
      "step": 17330
    },
    {
      "entropy": 5.469931268692017,
      "epoch": 1.348725928807625,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048202733915251407,
      "loss": 5.2238,
      "mean_token_accuracy": 0.1793886184692383,
      "num_tokens": 30076286.0,
      "step": 17335
    },
    {
      "entropy": 5.378542470932007,
      "epoch": 1.3491149581793425,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00048201648324297573,
      "loss": 5.1316,
      "mean_token_accuracy": 0.17962127476930617,
      "num_tokens": 30084681.0,
      "step": 17340
    },
    {
      "entropy": 5.540548324584961,
      "epoch": 1.3495039875510602,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00048200562419229185,
      "loss": 5.2415,
      "mean_token_accuracy": 0.17583508491516114,
      "num_tokens": 30093293.0,
      "step": 17345
    },
    {
      "entropy": 5.448749971389771,
      "epoch": 1.3498930169227776,
      "grad_norm": 1.6484375,
      "learning_rate": 0.000481994762000627,
      "loss": 5.0993,
      "mean_token_accuracy": 0.18627091199159623,
      "num_tokens": 30101593.0,
      "step": 17350
    },
    {
      "entropy": 5.483889818191528,
      "epoch": 1.3502820462944953,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048198389666814616,
      "loss": 5.298,
      "mean_token_accuracy": 0.17381460219621658,
      "num_tokens": 30110328.0,
      "step": 17355
    },
    {
      "entropy": 5.494292449951172,
      "epoch": 1.3506710756662128,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048197302819501416,
      "loss": 5.2518,
      "mean_token_accuracy": 0.17856424003839494,
      "num_tokens": 30118721.0,
      "step": 17360
    },
    {
      "entropy": 5.444191646575928,
      "epoch": 1.3510601050379303,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00048196215658139586,
      "loss": 5.2113,
      "mean_token_accuracy": 0.17079658806324005,
      "num_tokens": 30127238.0,
      "step": 17365
    },
    {
      "entropy": 5.516708326339722,
      "epoch": 1.351449134409648,
      "grad_norm": 1.5,
      "learning_rate": 0.0004819512818274562,
      "loss": 5.2062,
      "mean_token_accuracy": 0.18769046068191528,
      "num_tokens": 30135049.0,
      "step": 17370
    },
    {
      "entropy": 5.38051176071167,
      "epoch": 1.3518381637813655,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004819404039333603,
      "loss": 5.2224,
      "mean_token_accuracy": 0.17592968046665192,
      "num_tokens": 30143991.0,
      "step": 17375
    },
    {
      "entropy": 5.516547632217407,
      "epoch": 1.352227193153083,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004819295228992731,
      "loss": 5.4044,
      "mean_token_accuracy": 0.16552523225545884,
      "num_tokens": 30153968.0,
      "step": 17380
    },
    {
      "entropy": 5.670028638839722,
      "epoch": 1.3526162225248006,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048191863872535984,
      "loss": 5.2668,
      "mean_token_accuracy": 0.17516425102949143,
      "num_tokens": 30162927.0,
      "step": 17385
    },
    {
      "entropy": 5.464689826965332,
      "epoch": 1.3530052518965183,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004819077514117855,
      "loss": 5.2105,
      "mean_token_accuracy": 0.18032141625881196,
      "num_tokens": 30170986.0,
      "step": 17390
    },
    {
      "entropy": 5.438656425476074,
      "epoch": 1.3533942812682358,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00048189686095871545,
      "loss": 5.2,
      "mean_token_accuracy": 0.1782979339361191,
      "num_tokens": 30179167.0,
      "step": 17395
    },
    {
      "entropy": 5.446647310256958,
      "epoch": 1.3537833106399533,
      "grad_norm": 1.5,
      "learning_rate": 0.00048188596736631475,
      "loss": 5.2096,
      "mean_token_accuracy": 0.17918047457933425,
      "num_tokens": 30187710.0,
      "step": 17400
    },
    {
      "entropy": 5.591613721847534,
      "epoch": 1.354172340011671,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048187507063474885,
      "loss": 5.3135,
      "mean_token_accuracy": 0.1697572022676468,
      "num_tokens": 30196502.0,
      "step": 17405
    },
    {
      "entropy": 5.4313451766967775,
      "epoch": 1.3545613693833884,
      "grad_norm": 1.484375,
      "learning_rate": 0.00048186417076418294,
      "loss": 5.1582,
      "mean_token_accuracy": 0.17617504000663758,
      "num_tokens": 30205070.0,
      "step": 17410
    },
    {
      "entropy": 5.485702085494995,
      "epoch": 1.354950398755106,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048185326775478256,
      "loss": 5.2269,
      "mean_token_accuracy": 0.17253556847572327,
      "num_tokens": 30213554.0,
      "step": 17415
    },
    {
      "entropy": 5.437127780914307,
      "epoch": 1.3553394281268236,
      "grad_norm": 1.484375,
      "learning_rate": 0.00048184236160671306,
      "loss": 5.1717,
      "mean_token_accuracy": 0.17828486114740372,
      "num_tokens": 30222687.0,
      "step": 17420
    },
    {
      "entropy": 5.481479597091675,
      "epoch": 1.355728457498541,
      "grad_norm": 1.625,
      "learning_rate": 0.00048183145232013995,
      "loss": 5.2522,
      "mean_token_accuracy": 0.1716803342103958,
      "num_tokens": 30231086.0,
      "step": 17425
    },
    {
      "entropy": 5.463227939605713,
      "epoch": 1.3561174868702586,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00048182053989522883,
      "loss": 5.1462,
      "mean_token_accuracy": 0.17938144952058793,
      "num_tokens": 30240355.0,
      "step": 17430
    },
    {
      "entropy": 5.56818642616272,
      "epoch": 1.3565065162419763,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048180962433214515,
      "loss": 5.3472,
      "mean_token_accuracy": 0.1726804181933403,
      "num_tokens": 30249586.0,
      "step": 17435
    },
    {
      "entropy": 5.54279408454895,
      "epoch": 1.356895545613694,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048179870563105453,
      "loss": 5.289,
      "mean_token_accuracy": 0.17133213579654694,
      "num_tokens": 30257666.0,
      "step": 17440
    },
    {
      "entropy": 5.5305335521698,
      "epoch": 1.3572845749854114,
      "grad_norm": 1.375,
      "learning_rate": 0.0004817877837921228,
      "loss": 5.4184,
      "mean_token_accuracy": 0.17413941472768785,
      "num_tokens": 30266846.0,
      "step": 17445
    },
    {
      "entropy": 5.429582405090332,
      "epoch": 1.357673604357129,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004817768588155155,
      "loss": 5.1556,
      "mean_token_accuracy": 0.18280264288187026,
      "num_tokens": 30275587.0,
      "step": 17450
    },
    {
      "entropy": 5.517132043838501,
      "epoch": 1.3580626337288466,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00048176593070139865,
      "loss": 5.251,
      "mean_token_accuracy": 0.1791014090180397,
      "num_tokens": 30283948.0,
      "step": 17455
    },
    {
      "entropy": 5.617012786865234,
      "epoch": 1.358451663100564,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00048175499944993767,
      "loss": 5.3841,
      "mean_token_accuracy": 0.17775468528270721,
      "num_tokens": 30292302.0,
      "step": 17460
    },
    {
      "entropy": 5.618616962432862,
      "epoch": 1.3588406924722816,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00048174406506129884,
      "loss": 5.3812,
      "mean_token_accuracy": 0.17075441926717758,
      "num_tokens": 30301414.0,
      "step": 17465
    },
    {
      "entropy": 5.551756906509399,
      "epoch": 1.3592297218439993,
      "grad_norm": 1.53125,
      "learning_rate": 0.00048173312753564787,
      "loss": 5.2781,
      "mean_token_accuracy": 0.18222108483314514,
      "num_tokens": 30311045.0,
      "step": 17470
    },
    {
      "entropy": 5.482461023330688,
      "epoch": 1.3596187512157167,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004817221868731506,
      "loss": 5.1715,
      "mean_token_accuracy": 0.1783289611339569,
      "num_tokens": 30318552.0,
      "step": 17475
    },
    {
      "entropy": 5.4592358589172365,
      "epoch": 1.3600077805874344,
      "grad_norm": 1.375,
      "learning_rate": 0.00048171124307397334,
      "loss": 5.2939,
      "mean_token_accuracy": 0.17215437889099122,
      "num_tokens": 30327583.0,
      "step": 17480
    },
    {
      "entropy": 5.529998731613159,
      "epoch": 1.360396809959152,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004817002961382819,
      "loss": 5.2334,
      "mean_token_accuracy": 0.17080385535955428,
      "num_tokens": 30336339.0,
      "step": 17485
    },
    {
      "entropy": 5.575060033798218,
      "epoch": 1.3607858393308696,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048168934606624255,
      "loss": 5.3262,
      "mean_token_accuracy": 0.1722218230366707,
      "num_tokens": 30344878.0,
      "step": 17490
    },
    {
      "entropy": 5.529403781890869,
      "epoch": 1.361174868702587,
      "grad_norm": 1.484375,
      "learning_rate": 0.00048167839285802116,
      "loss": 5.2777,
      "mean_token_accuracy": 0.17036345303058625,
      "num_tokens": 30353071.0,
      "step": 17495
    },
    {
      "entropy": 5.503236627578735,
      "epoch": 1.3615638980743046,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004816674365137843,
      "loss": 5.236,
      "mean_token_accuracy": 0.18010642230510712,
      "num_tokens": 30361356.0,
      "step": 17500
    },
    {
      "entropy": 5.594958639144897,
      "epoch": 1.3619529274460223,
      "grad_norm": 1.65625,
      "learning_rate": 0.000481656477033698,
      "loss": 5.2848,
      "mean_token_accuracy": 0.16840853095054625,
      "num_tokens": 30370917.0,
      "step": 17505
    },
    {
      "entropy": 5.516336965560913,
      "epoch": 1.3623419568177397,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004816455144179286,
      "loss": 5.3034,
      "mean_token_accuracy": 0.17195237725973128,
      "num_tokens": 30380304.0,
      "step": 17510
    },
    {
      "entropy": 5.569565868377685,
      "epoch": 1.3627309861894572,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004816345486666424,
      "loss": 5.2476,
      "mean_token_accuracy": 0.17904252409934998,
      "num_tokens": 30388565.0,
      "step": 17515
    },
    {
      "entropy": 5.56043963432312,
      "epoch": 1.363120015561175,
      "grad_norm": 1.375,
      "learning_rate": 0.0004816235797800058,
      "loss": 5.3612,
      "mean_token_accuracy": 0.17108625322580337,
      "num_tokens": 30397220.0,
      "step": 17520
    },
    {
      "entropy": 5.5058962345123295,
      "epoch": 1.3635090449328924,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004816126077581852,
      "loss": 5.2294,
      "mean_token_accuracy": 0.17219217568635942,
      "num_tokens": 30405235.0,
      "step": 17525
    },
    {
      "entropy": 5.501880311965943,
      "epoch": 1.36389807430461,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048160163260134724,
      "loss": 5.2692,
      "mean_token_accuracy": 0.17481008023023606,
      "num_tokens": 30414218.0,
      "step": 17530
    },
    {
      "entropy": 5.5271735191345215,
      "epoch": 1.3642871036763276,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004815906543096583,
      "loss": 5.246,
      "mean_token_accuracy": 0.1751176968216896,
      "num_tokens": 30422665.0,
      "step": 17535
    },
    {
      "entropy": 5.566660642623901,
      "epoch": 1.3646761330480452,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000481579672883285,
      "loss": 5.2742,
      "mean_token_accuracy": 0.17250326573848723,
      "num_tokens": 30431868.0,
      "step": 17540
    },
    {
      "entropy": 5.5575110912323,
      "epoch": 1.3650651624197627,
      "grad_norm": 1.375,
      "learning_rate": 0.00048156868832239396,
      "loss": 5.2896,
      "mean_token_accuracy": 0.17769983112812043,
      "num_tokens": 30440548.0,
      "step": 17545
    },
    {
      "entropy": 5.569244337081909,
      "epoch": 1.3654541917914802,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004815577006271519,
      "loss": 5.2951,
      "mean_token_accuracy": 0.16673124134540557,
      "num_tokens": 30448959.0,
      "step": 17550
    },
    {
      "entropy": 5.527101755142212,
      "epoch": 1.365843221163198,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048154670979772555,
      "loss": 5.2031,
      "mean_token_accuracy": 0.17675437778234482,
      "num_tokens": 30457193.0,
      "step": 17555
    },
    {
      "entropy": 5.593591117858887,
      "epoch": 1.3662322505349154,
      "grad_norm": 1.875,
      "learning_rate": 0.0004815357158342815,
      "loss": 5.2809,
      "mean_token_accuracy": 0.16696500182151794,
      "num_tokens": 30466253.0,
      "step": 17560
    },
    {
      "entropy": 5.470001888275147,
      "epoch": 1.3666212799066328,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004815247187369868,
      "loss": 5.1927,
      "mean_token_accuracy": 0.1841494306921959,
      "num_tokens": 30474675.0,
      "step": 17565
    },
    {
      "entropy": 5.512925243377685,
      "epoch": 1.3670103092783505,
      "grad_norm": 1.5,
      "learning_rate": 0.00048151371850600814,
      "loss": 5.2487,
      "mean_token_accuracy": 0.179536671936512,
      "num_tokens": 30483031.0,
      "step": 17570
    },
    {
      "entropy": 5.445442771911621,
      "epoch": 1.367399338650068,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00048150271514151255,
      "loss": 5.2388,
      "mean_token_accuracy": 0.17908791452646255,
      "num_tokens": 30491331.0,
      "step": 17575
    },
    {
      "entropy": 5.499799489974976,
      "epoch": 1.3677883680217857,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048149170864366693,
      "loss": 5.2732,
      "mean_token_accuracy": 0.17207299023866654,
      "num_tokens": 30499649.0,
      "step": 17580
    },
    {
      "entropy": 5.547387552261353,
      "epoch": 1.3681773973935032,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048148069901263836,
      "loss": 5.1323,
      "mean_token_accuracy": 0.18310274183750153,
      "num_tokens": 30508922.0,
      "step": 17585
    },
    {
      "entropy": 5.525827264785766,
      "epoch": 1.368566426765221,
      "grad_norm": 1.375,
      "learning_rate": 0.00048146968624859375,
      "loss": 5.2161,
      "mean_token_accuracy": 0.17420928031206132,
      "num_tokens": 30517745.0,
      "step": 17590
    },
    {
      "entropy": 5.471214771270752,
      "epoch": 1.3689554561369384,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004814586703517003,
      "loss": 5.3641,
      "mean_token_accuracy": 0.16873226016759874,
      "num_tokens": 30526597.0,
      "step": 17595
    },
    {
      "entropy": 5.624855613708496,
      "epoch": 1.3693444855086558,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004814476513221251,
      "loss": 5.2381,
      "mean_token_accuracy": 0.17717682123184203,
      "num_tokens": 30533969.0,
      "step": 17600
    },
    {
      "entropy": 5.553325605392456,
      "epoch": 1.3697335148803735,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048143662916003545,
      "loss": 5.3244,
      "mean_token_accuracy": 0.16900917142629623,
      "num_tokens": 30543277.0,
      "step": 17605
    },
    {
      "entropy": 5.552525186538697,
      "epoch": 1.370122544252091,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004814256038655985,
      "loss": 5.3019,
      "mean_token_accuracy": 0.16587134897708894,
      "num_tokens": 30551889.0,
      "step": 17610
    },
    {
      "entropy": 5.529733419418335,
      "epoch": 1.3705115736238085,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004814145754389816,
      "loss": 5.2625,
      "mean_token_accuracy": 0.17882727086544037,
      "num_tokens": 30560926.0,
      "step": 17615
    },
    {
      "entropy": 5.63389835357666,
      "epoch": 1.3709006029955262,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048140354388035204,
      "loss": 5.3334,
      "mean_token_accuracy": 0.1755084991455078,
      "num_tokens": 30570102.0,
      "step": 17620
    },
    {
      "entropy": 5.516022348403931,
      "epoch": 1.3712896323672437,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004813925091898772,
      "loss": 5.289,
      "mean_token_accuracy": 0.1758200615644455,
      "num_tokens": 30578585.0,
      "step": 17625
    },
    {
      "entropy": 5.572675895690918,
      "epoch": 1.3716786617389614,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004813814713677246,
      "loss": 5.2751,
      "mean_token_accuracy": 0.1793521448969841,
      "num_tokens": 30587594.0,
      "step": 17630
    },
    {
      "entropy": 5.639176321029663,
      "epoch": 1.3720676911106788,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004813704304140616,
      "loss": 5.2865,
      "mean_token_accuracy": 0.17619706988334655,
      "num_tokens": 30596234.0,
      "step": 17635
    },
    {
      "entropy": 5.56216402053833,
      "epoch": 1.3724567204823965,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004813593863290559,
      "loss": 5.3491,
      "mean_token_accuracy": 0.17236778140068054,
      "num_tokens": 30605822.0,
      "step": 17640
    },
    {
      "entropy": 5.5389673709869385,
      "epoch": 1.372845749854114,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004813483391128748,
      "loss": 5.2277,
      "mean_token_accuracy": 0.17901082187891007,
      "num_tokens": 30614154.0,
      "step": 17645
    },
    {
      "entropy": 5.545075607299805,
      "epoch": 1.3732347792258315,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004813372887656862,
      "loss": 5.2727,
      "mean_token_accuracy": 0.17536881566047668,
      "num_tokens": 30622393.0,
      "step": 17650
    },
    {
      "entropy": 5.468309164047241,
      "epoch": 1.3736238085975492,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004813262352876576,
      "loss": 5.314,
      "mean_token_accuracy": 0.17265358418226243,
      "num_tokens": 30632173.0,
      "step": 17655
    },
    {
      "entropy": 5.644980812072754,
      "epoch": 1.3740128379692667,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048131517867895684,
      "loss": 5.3839,
      "mean_token_accuracy": 0.16801826059818267,
      "num_tokens": 30640313.0,
      "step": 17660
    },
    {
      "entropy": 5.561770582199097,
      "epoch": 1.3744018673409841,
      "grad_norm": 1.484375,
      "learning_rate": 0.00048130411893975155,
      "loss": 5.179,
      "mean_token_accuracy": 0.17904140502214433,
      "num_tokens": 30648651.0,
      "step": 17665
    },
    {
      "entropy": 5.460551834106445,
      "epoch": 1.3747908967127018,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004812930560702097,
      "loss": 5.2722,
      "mean_token_accuracy": 0.1761232152581215,
      "num_tokens": 30657782.0,
      "step": 17670
    },
    {
      "entropy": 5.453908967971802,
      "epoch": 1.3751799260844193,
      "grad_norm": 1.375,
      "learning_rate": 0.00048128199007049907,
      "loss": 5.2695,
      "mean_token_accuracy": 0.1734609380364418,
      "num_tokens": 30666493.0,
      "step": 17675
    },
    {
      "entropy": 5.482020711898803,
      "epoch": 1.375568955456137,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048127092094078756,
      "loss": 5.1579,
      "mean_token_accuracy": 0.18722697496414184,
      "num_tokens": 30675054.0,
      "step": 17680
    },
    {
      "entropy": 5.586921834945679,
      "epoch": 1.3759579848278545,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048125984868124317,
      "loss": 5.1986,
      "mean_token_accuracy": 0.18506894409656524,
      "num_tokens": 30683927.0,
      "step": 17685
    },
    {
      "entropy": 5.448682975769043,
      "epoch": 1.3763470141995722,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004812487732920338,
      "loss": 5.1044,
      "mean_token_accuracy": 0.1847105324268341,
      "num_tokens": 30691670.0,
      "step": 17690
    },
    {
      "entropy": 5.368113422393799,
      "epoch": 1.3767360435712896,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004812376947733277,
      "loss": 5.1374,
      "mean_token_accuracy": 0.1819167971611023,
      "num_tokens": 30700312.0,
      "step": 17695
    },
    {
      "entropy": 5.484257364273072,
      "epoch": 1.3771250729430071,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00048122661312529263,
      "loss": 5.207,
      "mean_token_accuracy": 0.18260670304298401,
      "num_tokens": 30708975.0,
      "step": 17700
    },
    {
      "entropy": 5.4597815990448,
      "epoch": 1.3775141023147248,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004812155283480971,
      "loss": 5.2139,
      "mean_token_accuracy": 0.17757836878299713,
      "num_tokens": 30718314.0,
      "step": 17705
    },
    {
      "entropy": 5.502430629730225,
      "epoch": 1.3779031316864423,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004812044404419091,
      "loss": 5.2598,
      "mean_token_accuracy": 0.17979368269443513,
      "num_tokens": 30726801.0,
      "step": 17710
    },
    {
      "entropy": 5.48796010017395,
      "epoch": 1.3782921610581598,
      "grad_norm": 1.484375,
      "learning_rate": 0.000481193349406897,
      "loss": 5.1199,
      "mean_token_accuracy": 0.17985542863607407,
      "num_tokens": 30734884.0,
      "step": 17715
    },
    {
      "entropy": 5.5230245113372805,
      "epoch": 1.3786811904298775,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000481182255243229,
      "loss": 5.2694,
      "mean_token_accuracy": 0.174702250957489,
      "num_tokens": 30744986.0,
      "step": 17720
    },
    {
      "entropy": 5.448199367523193,
      "epoch": 1.379070219801595,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048117115795107335,
      "loss": 5.2396,
      "mean_token_accuracy": 0.17263717502355574,
      "num_tokens": 30753476.0,
      "step": 17725
    },
    {
      "entropy": 5.575817537307739,
      "epoch": 1.3794592491733126,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004811600575305987,
      "loss": 5.3525,
      "mean_token_accuracy": 0.16767941564321517,
      "num_tokens": 30762642.0,
      "step": 17730
    },
    {
      "entropy": 5.573722314834595,
      "epoch": 1.3798482785450301,
      "grad_norm": 1.375,
      "learning_rate": 0.0004811489539819731,
      "loss": 5.3178,
      "mean_token_accuracy": 0.17155630737543107,
      "num_tokens": 30771530.0,
      "step": 17735
    },
    {
      "entropy": 5.524826765060425,
      "epoch": 1.3802373079167478,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048113784730536544,
      "loss": 5.3021,
      "mean_token_accuracy": 0.1755194529891014,
      "num_tokens": 30780001.0,
      "step": 17740
    },
    {
      "entropy": 5.5157185077667235,
      "epoch": 1.3806263372884653,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00048112673750094396,
      "loss": 5.2185,
      "mean_token_accuracy": 0.17801757007837296,
      "num_tokens": 30788696.0,
      "step": 17745
    },
    {
      "entropy": 5.544213199615479,
      "epoch": 1.3810153666601828,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004811156245688773,
      "loss": 5.2894,
      "mean_token_accuracy": 0.17802254110574722,
      "num_tokens": 30797457.0,
      "step": 17750
    },
    {
      "entropy": 5.540354633331299,
      "epoch": 1.3814043960319005,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004811045085093342,
      "loss": 5.2623,
      "mean_token_accuracy": 0.17355971038341522,
      "num_tokens": 30807048.0,
      "step": 17755
    },
    {
      "entropy": 5.522058057785034,
      "epoch": 1.381793425403618,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004810933893224831,
      "loss": 5.2795,
      "mean_token_accuracy": 0.17771067470312119,
      "num_tokens": 30815315.0,
      "step": 17760
    },
    {
      "entropy": 5.493741416931153,
      "epoch": 1.3821824547753354,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048108226700849287,
      "loss": 5.2059,
      "mean_token_accuracy": 0.17556727230548858,
      "num_tokens": 30823221.0,
      "step": 17765
    },
    {
      "entropy": 5.653856945037842,
      "epoch": 1.382571484147053,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004810711415675323,
      "loss": 5.3886,
      "mean_token_accuracy": 0.1708286300301552,
      "num_tokens": 30832112.0,
      "step": 17770
    },
    {
      "entropy": 5.552512311935425,
      "epoch": 1.3829605135187706,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004810600129997702,
      "loss": 5.2602,
      "mean_token_accuracy": 0.18118225783109665,
      "num_tokens": 30840597.0,
      "step": 17775
    },
    {
      "entropy": 5.439036703109741,
      "epoch": 1.3833495428904883,
      "grad_norm": 1.484375,
      "learning_rate": 0.00048104888130537534,
      "loss": 5.079,
      "mean_token_accuracy": 0.18668164014816285,
      "num_tokens": 30848947.0,
      "step": 17780
    },
    {
      "entropy": 5.470128679275513,
      "epoch": 1.3837385722622058,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00048103774648451664,
      "loss": 5.2908,
      "mean_token_accuracy": 0.16934383660554886,
      "num_tokens": 30856630.0,
      "step": 17785
    },
    {
      "entropy": 5.517313623428345,
      "epoch": 1.3841276016339235,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048102660853736314,
      "loss": 5.3591,
      "mean_token_accuracy": 0.17357830703258514,
      "num_tokens": 30866032.0,
      "step": 17790
    },
    {
      "entropy": 5.559560108184814,
      "epoch": 1.384516631005641,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004810154674640837,
      "loss": 5.2005,
      "mean_token_accuracy": 0.18244669437408448,
      "num_tokens": 30874104.0,
      "step": 17795
    },
    {
      "entropy": 5.456486225128174,
      "epoch": 1.3849056603773584,
      "grad_norm": 2.265625,
      "learning_rate": 0.00048100432326484744,
      "loss": 5.1932,
      "mean_token_accuracy": 0.18531633168458939,
      "num_tokens": 30882792.0,
      "step": 17800
    },
    {
      "entropy": 5.4607233047485355,
      "epoch": 1.385294689749076,
      "grad_norm": 1.375,
      "learning_rate": 0.0004809931759398235,
      "loss": 5.1905,
      "mean_token_accuracy": 0.17663442194461823,
      "num_tokens": 30891337.0,
      "step": 17805
    },
    {
      "entropy": 5.490444087982178,
      "epoch": 1.3856837191207936,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048098202548918094,
      "loss": 5.2462,
      "mean_token_accuracy": 0.17789483666419983,
      "num_tokens": 30899589.0,
      "step": 17810
    },
    {
      "entropy": 5.548178815841675,
      "epoch": 1.386072748492511,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00048097087191308914,
      "loss": 5.248,
      "mean_token_accuracy": 0.17168817669153214,
      "num_tokens": 30907893.0,
      "step": 17815
    },
    {
      "entropy": 5.590281629562378,
      "epoch": 1.3864617778642288,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004809597152117171,
      "loss": 5.3521,
      "mean_token_accuracy": 0.17575883418321608,
      "num_tokens": 30916797.0,
      "step": 17820
    },
    {
      "entropy": 5.536968803405761,
      "epoch": 1.3868508072359464,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004809485553852342,
      "loss": 5.2271,
      "mean_token_accuracy": 0.1801610469818115,
      "num_tokens": 30925451.0,
      "step": 17825
    },
    {
      "entropy": 5.4286932945251465,
      "epoch": 1.387239836607664,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004809373924338098,
      "loss": 5.2114,
      "mean_token_accuracy": 0.1740361452102661,
      "num_tokens": 30934595.0,
      "step": 17830
    },
    {
      "entropy": 5.514843463897705,
      "epoch": 1.3876288659793814,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048092622635761316,
      "loss": 5.3527,
      "mean_token_accuracy": 0.17542300522327423,
      "num_tokens": 30943896.0,
      "step": 17835
    },
    {
      "entropy": 5.617434692382813,
      "epoch": 1.388017895351099,
      "grad_norm": 1.375,
      "learning_rate": 0.00048091505715681395,
      "loss": 5.2009,
      "mean_token_accuracy": 0.17829690277576446,
      "num_tokens": 30952255.0,
      "step": 17840
    },
    {
      "entropy": 5.536069869995117,
      "epoch": 1.3884069247228166,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004809038848315814,
      "loss": 5.2979,
      "mean_token_accuracy": 0.17594264894723893,
      "num_tokens": 30960972.0,
      "step": 17845
    },
    {
      "entropy": 5.5068258285522464,
      "epoch": 1.388795954094534,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004808927093820851,
      "loss": 5.317,
      "mean_token_accuracy": 0.1752900078892708,
      "num_tokens": 30970298.0,
      "step": 17850
    },
    {
      "entropy": 5.5917236328125,
      "epoch": 1.3891849834662517,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004808815308084947,
      "loss": 5.249,
      "mean_token_accuracy": 0.17603506594896318,
      "num_tokens": 30979355.0,
      "step": 17855
    },
    {
      "entropy": 5.469105577468872,
      "epoch": 1.3895740128379692,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004808703491109798,
      "loss": 5.2058,
      "mean_token_accuracy": 0.1828099876642227,
      "num_tokens": 30988085.0,
      "step": 17860
    },
    {
      "entropy": 5.485192489624024,
      "epoch": 1.3899630422096867,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004808591642897099,
      "loss": 5.2318,
      "mean_token_accuracy": 0.1788646697998047,
      "num_tokens": 30996039.0,
      "step": 17865
    },
    {
      "entropy": 5.622358179092407,
      "epoch": 1.3903520715814044,
      "grad_norm": 1.8125,
      "learning_rate": 0.000480847976344855,
      "loss": 5.3905,
      "mean_token_accuracy": 0.1623087152838707,
      "num_tokens": 31004664.0,
      "step": 17870
    },
    {
      "entropy": 5.525275182723999,
      "epoch": 1.390741100953122,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048083678527658465,
      "loss": 5.2237,
      "mean_token_accuracy": 0.17863224893808366,
      "num_tokens": 31012571.0,
      "step": 17875
    },
    {
      "entropy": 5.497212934494018,
      "epoch": 1.3911301303248396,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004808255910850687,
      "loss": 5.2657,
      "mean_token_accuracy": 0.17490193396806716,
      "num_tokens": 31020901.0,
      "step": 17880
    },
    {
      "entropy": 5.590429353713989,
      "epoch": 1.391519159696557,
      "grad_norm": 1.5,
      "learning_rate": 0.0004808143937704769,
      "loss": 5.2698,
      "mean_token_accuracy": 0.17845822721719742,
      "num_tokens": 31029707.0,
      "step": 17885
    },
    {
      "entropy": 5.518889379501343,
      "epoch": 1.3919081890682747,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00048080319333297937,
      "loss": 5.2573,
      "mean_token_accuracy": 0.17721348106861115,
      "num_tokens": 31037833.0,
      "step": 17890
    },
    {
      "entropy": 5.46894268989563,
      "epoch": 1.3922972184399922,
      "grad_norm": 1.609375,
      "learning_rate": 0.00048079198977274597,
      "loss": 5.2177,
      "mean_token_accuracy": 0.18497949540615083,
      "num_tokens": 31046494.0,
      "step": 17895
    },
    {
      "entropy": 5.64621934890747,
      "epoch": 1.3926862478117097,
      "grad_norm": 1.59375,
      "learning_rate": 0.00048078078308994666,
      "loss": 5.3797,
      "mean_token_accuracy": 0.16880697757005692,
      "num_tokens": 31054957.0,
      "step": 17900
    },
    {
      "entropy": 5.589640092849732,
      "epoch": 1.3930752771834274,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004807695732847515,
      "loss": 5.2143,
      "mean_token_accuracy": 0.17668064683675766,
      "num_tokens": 31063068.0,
      "step": 17905
    },
    {
      "entropy": 5.469656896591187,
      "epoch": 1.3934643065551449,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048075836035733053,
      "loss": 5.232,
      "mean_token_accuracy": 0.17443308532238005,
      "num_tokens": 31072081.0,
      "step": 17910
    },
    {
      "entropy": 5.459174633026123,
      "epoch": 1.3938533359268626,
      "grad_norm": 1.8125,
      "learning_rate": 0.000480747144307854,
      "loss": 5.1413,
      "mean_token_accuracy": 0.1869641736149788,
      "num_tokens": 31080541.0,
      "step": 17915
    },
    {
      "entropy": 5.554146337509155,
      "epoch": 1.39424236529858,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00048073592513649203,
      "loss": 5.3186,
      "mean_token_accuracy": 0.17137922942638398,
      "num_tokens": 31089035.0,
      "step": 17920
    },
    {
      "entropy": 5.503985023498535,
      "epoch": 1.3946313946702977,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004807247028434148,
      "loss": 5.3555,
      "mean_token_accuracy": 0.1599217176437378,
      "num_tokens": 31096973.0,
      "step": 17925
    },
    {
      "entropy": 5.51119384765625,
      "epoch": 1.3950204240420152,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004807134774287927,
      "loss": 5.2605,
      "mean_token_accuracy": 0.16861334443092346,
      "num_tokens": 31104526.0,
      "step": 17930
    },
    {
      "entropy": 5.5450520515441895,
      "epoch": 1.3954094534137327,
      "grad_norm": 2.953125,
      "learning_rate": 0.00048070224889279603,
      "loss": 5.1132,
      "mean_token_accuracy": 0.1863141119480133,
      "num_tokens": 31112350.0,
      "step": 17935
    },
    {
      "entropy": 5.5180140972137455,
      "epoch": 1.3957984827854504,
      "grad_norm": 1.765625,
      "learning_rate": 0.00048069101723559505,
      "loss": 5.3033,
      "mean_token_accuracy": 0.17458772361278535,
      "num_tokens": 31120333.0,
      "step": 17940
    },
    {
      "entropy": 5.510772228240967,
      "epoch": 1.3961875121571679,
      "grad_norm": 1.375,
      "learning_rate": 0.0004806797824573603,
      "loss": 5.2895,
      "mean_token_accuracy": 0.1694720506668091,
      "num_tokens": 31129652.0,
      "step": 17945
    },
    {
      "entropy": 5.636420392990113,
      "epoch": 1.3965765415288853,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004806685445582624,
      "loss": 5.2777,
      "mean_token_accuracy": 0.17210106551647186,
      "num_tokens": 31137542.0,
      "step": 17950
    },
    {
      "entropy": 5.505106592178345,
      "epoch": 1.396965570900603,
      "grad_norm": 1.5,
      "learning_rate": 0.00048065730353847143,
      "loss": 5.2651,
      "mean_token_accuracy": 0.17383188009262085,
      "num_tokens": 31145958.0,
      "step": 17955
    },
    {
      "entropy": 5.497444200515747,
      "epoch": 1.3973546002723205,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00048064605939815837,
      "loss": 5.2125,
      "mean_token_accuracy": 0.17757827937602996,
      "num_tokens": 31154267.0,
      "step": 17960
    },
    {
      "entropy": 5.5881565570831295,
      "epoch": 1.3977436296440382,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004806348121374936,
      "loss": 5.2922,
      "mean_token_accuracy": 0.1808555990457535,
      "num_tokens": 31162742.0,
      "step": 17965
    },
    {
      "entropy": 5.587315273284912,
      "epoch": 1.3981326590157557,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004806235617566479,
      "loss": 5.2793,
      "mean_token_accuracy": 0.1769614890217781,
      "num_tokens": 31171275.0,
      "step": 17970
    },
    {
      "entropy": 5.427456521987915,
      "epoch": 1.3985216883874734,
      "grad_norm": 1.46875,
      "learning_rate": 0.000480612308255792,
      "loss": 5.0029,
      "mean_token_accuracy": 0.18687910586595535,
      "num_tokens": 31179318.0,
      "step": 17975
    },
    {
      "entropy": 5.489149904251098,
      "epoch": 1.3989107177591908,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00048060105163509647,
      "loss": 5.2028,
      "mean_token_accuracy": 0.17561819553375244,
      "num_tokens": 31187129.0,
      "step": 17980
    },
    {
      "entropy": 5.546788454055786,
      "epoch": 1.3992997471309083,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004805897918947323,
      "loss": 5.3998,
      "mean_token_accuracy": 0.17156168669462205,
      "num_tokens": 31196381.0,
      "step": 17985
    },
    {
      "entropy": 5.542816305160523,
      "epoch": 1.399688776502626,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004805785290348702,
      "loss": 5.3178,
      "mean_token_accuracy": 0.17258928567171097,
      "num_tokens": 31205659.0,
      "step": 17990
    },
    {
      "entropy": 5.574976825714112,
      "epoch": 1.4000778058743435,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048056726305568124,
      "loss": 5.2832,
      "mean_token_accuracy": 0.17487747222185135,
      "num_tokens": 31214996.0,
      "step": 17995
    },
    {
      "entropy": 5.582656621932983,
      "epoch": 1.400466835246061,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048055599395733617,
      "loss": 5.2491,
      "mean_token_accuracy": 0.17277689576148986,
      "num_tokens": 31222964.0,
      "step": 18000
    },
    {
      "epoch": 1.400466835246061,
      "eval_entropy": 5.321191076364967,
      "eval_loss": 5.354572772979736,
      "eval_mean_token_accuracy": 0.1805547287212586,
      "eval_num_tokens": 31222964.0,
      "eval_runtime": 28.5359,
      "eval_samples_per_second": 1456.34,
      "eval_steps_per_second": 182.051,
      "step": 18000
    },
    {
      "entropy": 5.532103061676025,
      "epoch": 1.4008558646177787,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048054472174000615,
      "loss": 5.3504,
      "mean_token_accuracy": 0.16962839514017106,
      "num_tokens": 31231653.0,
      "step": 18005
    },
    {
      "entropy": 5.49730920791626,
      "epoch": 1.4012448939894961,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048053344640386207,
      "loss": 5.2086,
      "mean_token_accuracy": 0.1795650452375412,
      "num_tokens": 31240613.0,
      "step": 18010
    },
    {
      "entropy": 5.457658433914185,
      "epoch": 1.4016339233612138,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048052216794907505,
      "loss": 5.1542,
      "mean_token_accuracy": 0.18604094088077544,
      "num_tokens": 31249724.0,
      "step": 18015
    },
    {
      "entropy": 5.515489721298218,
      "epoch": 1.4020229527329313,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048051088637581624,
      "loss": 5.3758,
      "mean_token_accuracy": 0.17347239702939987,
      "num_tokens": 31258717.0,
      "step": 18020
    },
    {
      "entropy": 5.536655426025391,
      "epoch": 1.402411982104649,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000480499601684257,
      "loss": 5.2282,
      "mean_token_accuracy": 0.180781552195549,
      "num_tokens": 31267170.0,
      "step": 18025
    },
    {
      "entropy": 5.52217664718628,
      "epoch": 1.4028010114763665,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004804883138745682,
      "loss": 5.2772,
      "mean_token_accuracy": 0.17391119599342347,
      "num_tokens": 31277787.0,
      "step": 18030
    },
    {
      "entropy": 5.5132862567901615,
      "epoch": 1.403190040848084,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004804770229469214,
      "loss": 5.2609,
      "mean_token_accuracy": 0.17398350983858107,
      "num_tokens": 31286185.0,
      "step": 18035
    },
    {
      "entropy": 5.576523208618164,
      "epoch": 1.4035790702198017,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004804657289014878,
      "loss": 5.1974,
      "mean_token_accuracy": 0.17363112568855285,
      "num_tokens": 31295361.0,
      "step": 18040
    },
    {
      "entropy": 5.540326261520386,
      "epoch": 1.4039680995915191,
      "grad_norm": 1.484375,
      "learning_rate": 0.00048045443173843884,
      "loss": 5.3228,
      "mean_token_accuracy": 0.18048126101493836,
      "num_tokens": 31304036.0,
      "step": 18045
    },
    {
      "entropy": 5.472900772094727,
      "epoch": 1.4043571289632366,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048044313145794587,
      "loss": 5.1759,
      "mean_token_accuracy": 0.18150803595781326,
      "num_tokens": 31311919.0,
      "step": 18050
    },
    {
      "entropy": 5.553953027725219,
      "epoch": 1.4047461583349543,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048043182806018034,
      "loss": 5.2994,
      "mean_token_accuracy": 0.17473077028989792,
      "num_tokens": 31320805.0,
      "step": 18055
    },
    {
      "entropy": 5.525098085403442,
      "epoch": 1.4051351877066718,
      "grad_norm": 1.578125,
      "learning_rate": 0.00048042052154531384,
      "loss": 5.3042,
      "mean_token_accuracy": 0.17650000452995301,
      "num_tokens": 31329688.0,
      "step": 18060
    },
    {
      "entropy": 5.460553407669067,
      "epoch": 1.4055242170783895,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004804092119135179,
      "loss": 5.1551,
      "mean_token_accuracy": 0.18024409264326097,
      "num_tokens": 31338301.0,
      "step": 18065
    },
    {
      "entropy": 5.604771852493286,
      "epoch": 1.405913246450107,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004803978991649641,
      "loss": 5.3656,
      "mean_token_accuracy": 0.16851025819778442,
      "num_tokens": 31347371.0,
      "step": 18070
    },
    {
      "entropy": 5.569414234161377,
      "epoch": 1.4063022758218247,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00048038658329982404,
      "loss": 5.3098,
      "mean_token_accuracy": 0.1763094872236252,
      "num_tokens": 31355740.0,
      "step": 18075
    },
    {
      "entropy": 5.471652555465698,
      "epoch": 1.4066913051935421,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004803752643182695,
      "loss": 5.1401,
      "mean_token_accuracy": 0.18562333285808563,
      "num_tokens": 31364718.0,
      "step": 18080
    },
    {
      "entropy": 5.5960338592529295,
      "epoch": 1.4070803345652596,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004803639422204723,
      "loss": 5.2786,
      "mean_token_accuracy": 0.17359669208526612,
      "num_tokens": 31372845.0,
      "step": 18085
    },
    {
      "entropy": 5.549406337738037,
      "epoch": 1.4074693639369773,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004803526170066041,
      "loss": 5.259,
      "mean_token_accuracy": 0.18189283311367035,
      "num_tokens": 31380614.0,
      "step": 18090
    },
    {
      "entropy": 5.559660720825195,
      "epoch": 1.4078583933086948,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048034128867683674,
      "loss": 5.3279,
      "mean_token_accuracy": 0.1742562472820282,
      "num_tokens": 31389609.0,
      "step": 18095
    },
    {
      "entropy": 5.594971990585327,
      "epoch": 1.4082474226804123,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004803299572313422,
      "loss": 5.3051,
      "mean_token_accuracy": 0.17143698334693908,
      "num_tokens": 31399046.0,
      "step": 18100
    },
    {
      "entropy": 5.480393886566162,
      "epoch": 1.40863645205213,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004803186226702924,
      "loss": 5.204,
      "mean_token_accuracy": 0.18068505376577376,
      "num_tokens": 31407832.0,
      "step": 18105
    },
    {
      "entropy": 5.554833316802979,
      "epoch": 1.4090254814238474,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004803072849938592,
      "loss": 5.2712,
      "mean_token_accuracy": 0.17457137554883956,
      "num_tokens": 31417097.0,
      "step": 18110
    },
    {
      "entropy": 5.597403430938721,
      "epoch": 1.4094145107955651,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004802959442022149,
      "loss": 5.318,
      "mean_token_accuracy": 0.17648953348398208,
      "num_tokens": 31425448.0,
      "step": 18115
    },
    {
      "entropy": 5.476740503311158,
      "epoch": 1.4098035401672826,
      "grad_norm": 1.5,
      "learning_rate": 0.00048028460029553126,
      "loss": 5.1477,
      "mean_token_accuracy": 0.1920905292034149,
      "num_tokens": 31434116.0,
      "step": 18120
    },
    {
      "entropy": 5.429643869400024,
      "epoch": 1.4101925695390003,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048027325327398065,
      "loss": 5.2222,
      "mean_token_accuracy": 0.17521251887083053,
      "num_tokens": 31443456.0,
      "step": 18125
    },
    {
      "entropy": 5.503236627578735,
      "epoch": 1.4105815989107178,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004802619031377351,
      "loss": 5.2302,
      "mean_token_accuracy": 0.17537346631288528,
      "num_tokens": 31452372.0,
      "step": 18130
    },
    {
      "entropy": 5.595557832717896,
      "epoch": 1.4109706282824352,
      "grad_norm": 1.796875,
      "learning_rate": 0.00048025054988696684,
      "loss": 5.2647,
      "mean_token_accuracy": 0.1775445282459259,
      "num_tokens": 31460446.0,
      "step": 18135
    },
    {
      "entropy": 5.460465764999389,
      "epoch": 1.411359657654153,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048023919352184827,
      "loss": 5.1913,
      "mean_token_accuracy": 0.17444835752248763,
      "num_tokens": 31468814.0,
      "step": 18140
    },
    {
      "entropy": 5.440851068496704,
      "epoch": 1.4117486870258704,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048022783404255156,
      "loss": 5.1888,
      "mean_token_accuracy": 0.18189519196748732,
      "num_tokens": 31477617.0,
      "step": 18145
    },
    {
      "entropy": 5.555392789840698,
      "epoch": 1.412137716397588,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004802164714492491,
      "loss": 5.1721,
      "mean_token_accuracy": 0.18350616842508316,
      "num_tokens": 31485226.0,
      "step": 18150
    },
    {
      "entropy": 5.4960307598114015,
      "epoch": 1.4125267457693056,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048020510574211335,
      "loss": 5.213,
      "mean_token_accuracy": 0.18131631910800933,
      "num_tokens": 31493425.0,
      "step": 18155
    },
    {
      "entropy": 5.446992015838623,
      "epoch": 1.412915775141023,
      "grad_norm": 1.515625,
      "learning_rate": 0.00048019373692131674,
      "loss": 5.1858,
      "mean_token_accuracy": 0.17397785931825638,
      "num_tokens": 31502499.0,
      "step": 18160
    },
    {
      "entropy": 5.492952823638916,
      "epoch": 1.4133048045127408,
      "grad_norm": 1.6875,
      "learning_rate": 0.00048018236498703176,
      "loss": 5.2219,
      "mean_token_accuracy": 0.18217087388038636,
      "num_tokens": 31510376.0,
      "step": 18165
    },
    {
      "entropy": 5.534497308731079,
      "epoch": 1.4136938338844582,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00048017098993943097,
      "loss": 5.3843,
      "mean_token_accuracy": 0.1694500580430031,
      "num_tokens": 31519101.0,
      "step": 18170
    },
    {
      "entropy": 5.494120121002197,
      "epoch": 1.414082863256176,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000480159611778687,
      "loss": 5.2142,
      "mean_token_accuracy": 0.1815626800060272,
      "num_tokens": 31527221.0,
      "step": 18175
    },
    {
      "entropy": 5.454511451721191,
      "epoch": 1.4144718926278934,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048014823050497243,
      "loss": 5.2224,
      "mean_token_accuracy": 0.17851397693157195,
      "num_tokens": 31535711.0,
      "step": 18180
    },
    {
      "entropy": 5.554487419128418,
      "epoch": 1.4148609219996109,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048013684611846,
      "loss": 5.3671,
      "mean_token_accuracy": 0.16796732544898987,
      "num_tokens": 31545363.0,
      "step": 18185
    },
    {
      "entropy": 5.629341316223145,
      "epoch": 1.4152499513713286,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048012545861932245,
      "loss": 5.2305,
      "mean_token_accuracy": 0.1848675489425659,
      "num_tokens": 31553760.0,
      "step": 18190
    },
    {
      "entropy": 5.520274877548218,
      "epoch": 1.415638980743046,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004801140680077325,
      "loss": 5.1672,
      "mean_token_accuracy": 0.18191928416490555,
      "num_tokens": 31561625.0,
      "step": 18195
    },
    {
      "entropy": 5.431030988693237,
      "epoch": 1.4160280101147635,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004801026742838631,
      "loss": 5.2559,
      "mean_token_accuracy": 0.18150032609701156,
      "num_tokens": 31569345.0,
      "step": 18200
    },
    {
      "entropy": 5.564674043655396,
      "epoch": 1.4164170394864812,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004800912774478871,
      "loss": 5.2141,
      "mean_token_accuracy": 0.17311955988407135,
      "num_tokens": 31577990.0,
      "step": 18205
    },
    {
      "entropy": 5.449020624160767,
      "epoch": 1.416806068858199,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004800798774999773,
      "loss": 5.1736,
      "mean_token_accuracy": 0.18605172783136367,
      "num_tokens": 31586556.0,
      "step": 18210
    },
    {
      "entropy": 5.5212301254272464,
      "epoch": 1.4171950982299164,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048006847444030686,
      "loss": 5.2983,
      "mean_token_accuracy": 0.1785432368516922,
      "num_tokens": 31595697.0,
      "step": 18215
    },
    {
      "entropy": 5.457568502426147,
      "epoch": 1.4175841276016339,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004800570682690487,
      "loss": 5.1979,
      "mean_token_accuracy": 0.18486751317977906,
      "num_tokens": 31604760.0,
      "step": 18220
    },
    {
      "entropy": 5.530587768554687,
      "epoch": 1.4179731569733516,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004800456589863759,
      "loss": 5.319,
      "mean_token_accuracy": 0.1679696723818779,
      "num_tokens": 31614115.0,
      "step": 18225
    },
    {
      "entropy": 5.476338243484497,
      "epoch": 1.418362186345069,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004800342465924616,
      "loss": 5.3082,
      "mean_token_accuracy": 0.17588279098272325,
      "num_tokens": 31622492.0,
      "step": 18230
    },
    {
      "entropy": 5.564793109893799,
      "epoch": 1.4187512157167865,
      "grad_norm": 1.453125,
      "learning_rate": 0.000480022831087479,
      "loss": 5.3314,
      "mean_token_accuracy": 0.1728792354464531,
      "num_tokens": 31632269.0,
      "step": 18235
    },
    {
      "entropy": 5.582900667190552,
      "epoch": 1.4191402450885042,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004800114124716012,
      "loss": 5.3121,
      "mean_token_accuracy": 0.17432929426431656,
      "num_tokens": 31640374.0,
      "step": 18240
    },
    {
      "entropy": 5.48635573387146,
      "epoch": 1.4195292744602217,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004799999907450015,
      "loss": 5.2139,
      "mean_token_accuracy": 0.18473908305168152,
      "num_tokens": 31648406.0,
      "step": 18245
    },
    {
      "entropy": 5.486824989318848,
      "epoch": 1.4199183038319392,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004799885659078533,
      "loss": 5.1621,
      "mean_token_accuracy": 0.1814996376633644,
      "num_tokens": 31656478.0,
      "step": 18250
    },
    {
      "entropy": 5.467016363143921,
      "epoch": 1.4203073332036569,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004799771379603299,
      "loss": 5.1801,
      "mean_token_accuracy": 0.1797877460718155,
      "num_tokens": 31665106.0,
      "step": 18255
    },
    {
      "entropy": 5.577709484100342,
      "epoch": 1.4206963625753746,
      "grad_norm": 1.5,
      "learning_rate": 0.0004799657069026046,
      "loss": 5.3065,
      "mean_token_accuracy": 0.17493281960487367,
      "num_tokens": 31673557.0,
      "step": 18260
    },
    {
      "entropy": 5.481100130081177,
      "epoch": 1.421085391947092,
      "grad_norm": 1.484375,
      "learning_rate": 0.00047995427273485097,
      "loss": 5.1117,
      "mean_token_accuracy": 0.18842861652374268,
      "num_tokens": 31682123.0,
      "step": 18265
    },
    {
      "entropy": 5.520610094070435,
      "epoch": 1.4214744213188095,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00047994283545724247,
      "loss": 5.3765,
      "mean_token_accuracy": 0.16894069612026213,
      "num_tokens": 31690387.0,
      "step": 18270
    },
    {
      "entropy": 5.48298192024231,
      "epoch": 1.4218634506905272,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00047993139506995273,
      "loss": 5.2109,
      "mean_token_accuracy": 0.18396772742271422,
      "num_tokens": 31699814.0,
      "step": 18275
    },
    {
      "entropy": 5.482313394546509,
      "epoch": 1.4222524800622447,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004799199515731551,
      "loss": 5.1837,
      "mean_token_accuracy": 0.18609511256217956,
      "num_tokens": 31708437.0,
      "step": 18280
    },
    {
      "entropy": 5.479394817352295,
      "epoch": 1.4226415094339622,
      "grad_norm": 1.453125,
      "learning_rate": 0.00047990850496702346,
      "loss": 5.1567,
      "mean_token_accuracy": 0.18924205899238586,
      "num_tokens": 31716935.0,
      "step": 18285
    },
    {
      "entropy": 5.5642204761505125,
      "epoch": 1.4230305388056799,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004798970552517314,
      "loss": 5.3884,
      "mean_token_accuracy": 0.17014314979314804,
      "num_tokens": 31725997.0,
      "step": 18290
    },
    {
      "entropy": 5.4827601432800295,
      "epoch": 1.4234195681773973,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047988560242745264,
      "loss": 5.1946,
      "mean_token_accuracy": 0.17987073063850403,
      "num_tokens": 31733651.0,
      "step": 18295
    },
    {
      "entropy": 5.480494117736816,
      "epoch": 1.423808597549115,
      "grad_norm": 1.359375,
      "learning_rate": 0.000479874146494361,
      "loss": 5.195,
      "mean_token_accuracy": 0.17923596650362014,
      "num_tokens": 31742465.0,
      "step": 18300
    },
    {
      "entropy": 5.5356810092926025,
      "epoch": 1.4241976269208325,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004798626874526302,
      "loss": 5.2542,
      "mean_token_accuracy": 0.17424688786268233,
      "num_tokens": 31751547.0,
      "step": 18305
    },
    {
      "entropy": 5.50795693397522,
      "epoch": 1.4245866562925502,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00047985122530243426,
      "loss": 5.3916,
      "mean_token_accuracy": 0.16523353084921838,
      "num_tokens": 31761027.0,
      "step": 18310
    },
    {
      "entropy": 5.530346012115478,
      "epoch": 1.4249756856642677,
      "grad_norm": 1.359375,
      "learning_rate": 0.000479839760043947,
      "loss": 5.2853,
      "mean_token_accuracy": 0.17017893940210344,
      "num_tokens": 31770166.0,
      "step": 18315
    },
    {
      "entropy": 5.5170868873596195,
      "epoch": 1.4253647150359852,
      "grad_norm": 1.546875,
      "learning_rate": 0.00047982829167734245,
      "loss": 5.1459,
      "mean_token_accuracy": 0.18376674950122834,
      "num_tokens": 31778606.0,
      "step": 18320
    },
    {
      "entropy": 5.536589097976685,
      "epoch": 1.4257537444077029,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00047981682020279456,
      "loss": 5.3647,
      "mean_token_accuracy": 0.17093889117240907,
      "num_tokens": 31787637.0,
      "step": 18325
    },
    {
      "entropy": 5.438645362854004,
      "epoch": 1.4261427737794203,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004798053456204775,
      "loss": 5.1905,
      "mean_token_accuracy": 0.17995227128267288,
      "num_tokens": 31796401.0,
      "step": 18330
    },
    {
      "entropy": 5.547728776931763,
      "epoch": 1.4265318031511378,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004797938679305651,
      "loss": 5.318,
      "mean_token_accuracy": 0.17136707454919814,
      "num_tokens": 31805917.0,
      "step": 18335
    },
    {
      "entropy": 5.613900470733642,
      "epoch": 1.4269208325228555,
      "grad_norm": 1.71875,
      "learning_rate": 0.00047978238713323193,
      "loss": 5.3311,
      "mean_token_accuracy": 0.17354520857334138,
      "num_tokens": 31814890.0,
      "step": 18340
    },
    {
      "entropy": 5.4882893562316895,
      "epoch": 1.427309861894573,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047977090322865183,
      "loss": 5.186,
      "mean_token_accuracy": 0.18392392843961716,
      "num_tokens": 31822469.0,
      "step": 18345
    },
    {
      "entropy": 5.521025514602661,
      "epoch": 1.4276988912662907,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004797594162169993,
      "loss": 5.3106,
      "mean_token_accuracy": 0.1728377729654312,
      "num_tokens": 31831030.0,
      "step": 18350
    },
    {
      "entropy": 5.506049394607544,
      "epoch": 1.4280879206380082,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004797479260984485,
      "loss": 5.2157,
      "mean_token_accuracy": 0.1733946830034256,
      "num_tokens": 31840137.0,
      "step": 18355
    },
    {
      "entropy": 5.41376838684082,
      "epoch": 1.4284769500097259,
      "grad_norm": 1.4375,
      "learning_rate": 0.00047973643287317386,
      "loss": 5.1016,
      "mean_token_accuracy": 0.18575553596019745,
      "num_tokens": 31849258.0,
      "step": 18360
    },
    {
      "entropy": 5.432569313049316,
      "epoch": 1.4288659793814433,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047972493654134963,
      "loss": 5.1303,
      "mean_token_accuracy": 0.17914825528860093,
      "num_tokens": 31857506.0,
      "step": 18365
    },
    {
      "entropy": 5.534861421585083,
      "epoch": 1.4292550087531608,
      "grad_norm": 1.5,
      "learning_rate": 0.00047971343710315043,
      "loss": 5.2572,
      "mean_token_accuracy": 0.17397128641605378,
      "num_tokens": 31866685.0,
      "step": 18370
    },
    {
      "entropy": 5.589266014099121,
      "epoch": 1.4296440381248785,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004797019345587506,
      "loss": 5.3317,
      "mean_token_accuracy": 0.168351948261261,
      "num_tokens": 31874404.0,
      "step": 18375
    },
    {
      "entropy": 5.602615880966186,
      "epoch": 1.430033067496596,
      "grad_norm": 1.5,
      "learning_rate": 0.0004796904289083248,
      "loss": 5.3191,
      "mean_token_accuracy": 0.17482101172208786,
      "num_tokens": 31883075.0,
      "step": 18380
    },
    {
      "entropy": 5.518873023986816,
      "epoch": 1.4304220968683135,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004796789201520474,
      "loss": 5.2386,
      "mean_token_accuracy": 0.17895665019750595,
      "num_tokens": 31891539.0,
      "step": 18385
    },
    {
      "entropy": 5.585813522338867,
      "epoch": 1.4308111262400312,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00047966740829009333,
      "loss": 5.2729,
      "mean_token_accuracy": 0.1783260703086853,
      "num_tokens": 31900575.0,
      "step": 18390
    },
    {
      "entropy": 5.558542680740357,
      "epoch": 1.4312001556117486,
      "grad_norm": 1.5625,
      "learning_rate": 0.000479655893322637,
      "loss": 5.2645,
      "mean_token_accuracy": 0.1751030206680298,
      "num_tokens": 31909256.0,
      "step": 18395
    },
    {
      "entropy": 5.476305246353149,
      "epoch": 1.4315891849834663,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004796443752498532,
      "loss": 5.2069,
      "mean_token_accuracy": 0.17858581990003586,
      "num_tokens": 31917098.0,
      "step": 18400
    },
    {
      "entropy": 5.606909608840942,
      "epoch": 1.4319782143551838,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004796328540719169,
      "loss": 5.3353,
      "mean_token_accuracy": 0.17273962944746019,
      "num_tokens": 31925331.0,
      "step": 18405
    },
    {
      "entropy": 5.516603279113769,
      "epoch": 1.4323672437269015,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004796213297890026,
      "loss": 5.1932,
      "mean_token_accuracy": 0.17810521125793458,
      "num_tokens": 31933769.0,
      "step": 18410
    },
    {
      "entropy": 5.516629600524903,
      "epoch": 1.432756273098619,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004796098024012853,
      "loss": 5.2931,
      "mean_token_accuracy": 0.17049989551305772,
      "num_tokens": 31943089.0,
      "step": 18415
    },
    {
      "entropy": 5.542341470718384,
      "epoch": 1.4331453024703364,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00047959827190894,
      "loss": 5.2623,
      "mean_token_accuracy": 0.17716800421476364,
      "num_tokens": 31951452.0,
      "step": 18420
    },
    {
      "entropy": 5.5623345375061035,
      "epoch": 1.4335343318420541,
      "grad_norm": 1.484375,
      "learning_rate": 0.00047958673831214157,
      "loss": 5.2645,
      "mean_token_accuracy": 0.17607493400573732,
      "num_tokens": 31959625.0,
      "step": 18425
    },
    {
      "entropy": 5.45009069442749,
      "epoch": 1.4339233612137716,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00047957520161106496,
      "loss": 5.2398,
      "mean_token_accuracy": 0.17711673229932784,
      "num_tokens": 31967906.0,
      "step": 18430
    },
    {
      "entropy": 5.425977516174316,
      "epoch": 1.434312390585489,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004795636618058853,
      "loss": 5.1912,
      "mean_token_accuracy": 0.18084208518266678,
      "num_tokens": 31976183.0,
      "step": 18435
    },
    {
      "entropy": 5.4853521347045895,
      "epoch": 1.4347014199572068,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004795521188967777,
      "loss": 5.2147,
      "mean_token_accuracy": 0.1803765967488289,
      "num_tokens": 31984513.0,
      "step": 18440
    },
    {
      "entropy": 5.603140640258789,
      "epoch": 1.4350904493289243,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004795405728839172,
      "loss": 5.3823,
      "mean_token_accuracy": 0.16767289340496064,
      "num_tokens": 31992994.0,
      "step": 18445
    },
    {
      "entropy": 5.60715274810791,
      "epoch": 1.435479478700642,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004795290237674792,
      "loss": 5.2702,
      "mean_token_accuracy": 0.17277760207653045,
      "num_tokens": 32002043.0,
      "step": 18450
    },
    {
      "entropy": 5.495297861099243,
      "epoch": 1.4358685080723594,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00047951747154763866,
      "loss": 5.1444,
      "mean_token_accuracy": 0.18213044106960297,
      "num_tokens": 32009901.0,
      "step": 18455
    },
    {
      "entropy": 5.429672050476074,
      "epoch": 1.4362575374440771,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00047950591622457106,
      "loss": 5.2092,
      "mean_token_accuracy": 0.1747964233160019,
      "num_tokens": 32017947.0,
      "step": 18460
    },
    {
      "entropy": 5.491956377029419,
      "epoch": 1.4366465668157946,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047949435779845174,
      "loss": 5.2351,
      "mean_token_accuracy": 0.17535976469516754,
      "num_tokens": 32026537.0,
      "step": 18465
    },
    {
      "entropy": 5.490673208236695,
      "epoch": 1.437035596187512,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047948279626945596,
      "loss": 5.2,
      "mean_token_accuracy": 0.17987239509820938,
      "num_tokens": 32034874.0,
      "step": 18470
    },
    {
      "entropy": 5.503887081146241,
      "epoch": 1.4374246255592298,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047947123163775924,
      "loss": 5.2933,
      "mean_token_accuracy": 0.17334435433149337,
      "num_tokens": 32043829.0,
      "step": 18475
    },
    {
      "entropy": 5.574766540527344,
      "epoch": 1.4378136549309473,
      "grad_norm": 1.34375,
      "learning_rate": 0.000479459663903537,
      "loss": 5.319,
      "mean_token_accuracy": 0.17150646299123765,
      "num_tokens": 32052883.0,
      "step": 18480
    },
    {
      "entropy": 5.565496826171875,
      "epoch": 1.4382026843026647,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004794480930669649,
      "loss": 5.3384,
      "mean_token_accuracy": 0.17598079591989518,
      "num_tokens": 32061421.0,
      "step": 18485
    },
    {
      "entropy": 5.524099397659302,
      "epoch": 1.4385917136743824,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004794365191282183,
      "loss": 5.2517,
      "mean_token_accuracy": 0.1852465033531189,
      "num_tokens": 32069957.0,
      "step": 18490
    },
    {
      "entropy": 5.453353834152222,
      "epoch": 1.4389807430461,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047942494208747297,
      "loss": 5.1448,
      "mean_token_accuracy": 0.17750681787729264,
      "num_tokens": 32079375.0,
      "step": 18495
    },
    {
      "entropy": 5.513231897354126,
      "epoch": 1.4393697724178176,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004794133619449045,
      "loss": 5.1945,
      "mean_token_accuracy": 0.17859329283237457,
      "num_tokens": 32088331.0,
      "step": 18500
    },
    {
      "entropy": 5.5179328441619875,
      "epoch": 1.439758801789535,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004794017787006886,
      "loss": 5.1606,
      "mean_token_accuracy": 0.17839935719966887,
      "num_tokens": 32096897.0,
      "step": 18505
    },
    {
      "entropy": 5.462598133087158,
      "epoch": 1.4401478311612528,
      "grad_norm": 1.640625,
      "learning_rate": 0.000479390192355001,
      "loss": 5.1812,
      "mean_token_accuracy": 0.18658424019813538,
      "num_tokens": 32105942.0,
      "step": 18510
    },
    {
      "entropy": 5.582726955413818,
      "epoch": 1.4405368605329703,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004793786029080176,
      "loss": 5.3821,
      "mean_token_accuracy": 0.1678169012069702,
      "num_tokens": 32114540.0,
      "step": 18515
    },
    {
      "entropy": 5.470570611953735,
      "epoch": 1.4409258899046877,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004793670103599143,
      "loss": 5.1975,
      "mean_token_accuracy": 0.17213313430547714,
      "num_tokens": 32123214.0,
      "step": 18520
    },
    {
      "entropy": 5.520587396621704,
      "epoch": 1.4413149192764054,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047935541471086677,
      "loss": 5.2468,
      "mean_token_accuracy": 0.17993490248918534,
      "num_tokens": 32132369.0,
      "step": 18525
    },
    {
      "entropy": 5.493685579299926,
      "epoch": 1.441703948648123,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004793438159610511,
      "loss": 5.1832,
      "mean_token_accuracy": 0.17664045840501785,
      "num_tokens": 32141037.0,
      "step": 18530
    },
    {
      "entropy": 5.47927041053772,
      "epoch": 1.4420929780198404,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047933221411064334,
      "loss": 5.1904,
      "mean_token_accuracy": 0.1803370237350464,
      "num_tokens": 32149515.0,
      "step": 18535
    },
    {
      "entropy": 5.483889818191528,
      "epoch": 1.442482007391558,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004793206091598194,
      "loss": 5.2819,
      "mean_token_accuracy": 0.18418145328760147,
      "num_tokens": 32158099.0,
      "step": 18540
    },
    {
      "entropy": 5.545689821243286,
      "epoch": 1.4428710367632755,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004793090011087554,
      "loss": 5.2474,
      "mean_token_accuracy": 0.17718044966459273,
      "num_tokens": 32167059.0,
      "step": 18545
    },
    {
      "entropy": 5.628464984893799,
      "epoch": 1.4432600661349932,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047929738995762755,
      "loss": 5.4081,
      "mean_token_accuracy": 0.1738082140684128,
      "num_tokens": 32175988.0,
      "step": 18550
    },
    {
      "entropy": 5.431274652481079,
      "epoch": 1.4436490955067107,
      "grad_norm": 1.375,
      "learning_rate": 0.000479285775706612,
      "loss": 5.1472,
      "mean_token_accuracy": 0.1820313513278961,
      "num_tokens": 32185362.0,
      "step": 18555
    },
    {
      "entropy": 5.488016414642334,
      "epoch": 1.4440381248784284,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047927415835588496,
      "loss": 5.2293,
      "mean_token_accuracy": 0.1793539971113205,
      "num_tokens": 32194599.0,
      "step": 18560
    },
    {
      "entropy": 5.502857065200805,
      "epoch": 1.444427154250146,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047926253790562265,
      "loss": 5.2221,
      "mean_token_accuracy": 0.18292603492736817,
      "num_tokens": 32203106.0,
      "step": 18565
    },
    {
      "entropy": 5.462140607833862,
      "epoch": 1.4448161836218634,
      "grad_norm": 1.453125,
      "learning_rate": 0.00047925091435600147,
      "loss": 5.2762,
      "mean_token_accuracy": 0.1786574736237526,
      "num_tokens": 32211652.0,
      "step": 18570
    },
    {
      "entropy": 5.606622123718262,
      "epoch": 1.445205212993581,
      "grad_norm": 1.4375,
      "learning_rate": 0.00047923928770719776,
      "loss": 5.2767,
      "mean_token_accuracy": 0.1761241987347603,
      "num_tokens": 32220435.0,
      "step": 18575
    },
    {
      "entropy": 5.580476903915406,
      "epoch": 1.4455942423652985,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00047922765795938797,
      "loss": 5.2255,
      "mean_token_accuracy": 0.1731432482600212,
      "num_tokens": 32228368.0,
      "step": 18580
    },
    {
      "entropy": 5.499847888946533,
      "epoch": 1.445983271737016,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004792160251127485,
      "loss": 5.2905,
      "mean_token_accuracy": 0.17172825038433076,
      "num_tokens": 32237367.0,
      "step": 18585
    },
    {
      "entropy": 5.552915382385254,
      "epoch": 1.4463723011087337,
      "grad_norm": 1.796875,
      "learning_rate": 0.00047920438916745586,
      "loss": 5.281,
      "mean_token_accuracy": 0.17678776234388352,
      "num_tokens": 32245503.0,
      "step": 18590
    },
    {
      "entropy": 5.579898357391357,
      "epoch": 1.4467613304804512,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004791927501236866,
      "loss": 5.2794,
      "mean_token_accuracy": 0.17110560238361358,
      "num_tokens": 32254565.0,
      "step": 18595
    },
    {
      "entropy": 5.566863775253296,
      "epoch": 1.4471503598521689,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00047918110798161754,
      "loss": 5.3109,
      "mean_token_accuracy": 0.1716754898428917,
      "num_tokens": 32263701.0,
      "step": 18600
    },
    {
      "entropy": 5.509312582015991,
      "epoch": 1.4475393892238864,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00047916946274142503,
      "loss": 5.2271,
      "mean_token_accuracy": 0.17801002115011216,
      "num_tokens": 32272205.0,
      "step": 18605
    },
    {
      "entropy": 5.532806253433227,
      "epoch": 1.447928418595604,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00047915781440328593,
      "loss": 5.2796,
      "mean_token_accuracy": 0.1783734977245331,
      "num_tokens": 32279918.0,
      "step": 18610
    },
    {
      "entropy": 5.569362354278565,
      "epoch": 1.4483174479673215,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004791461629673769,
      "loss": 5.2782,
      "mean_token_accuracy": 0.18548174798488617,
      "num_tokens": 32288600.0,
      "step": 18615
    },
    {
      "entropy": 5.6158974170684814,
      "epoch": 1.448706477339039,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004791345084338748,
      "loss": 5.3884,
      "mean_token_accuracy": 0.1736078068614006,
      "num_tokens": 32298334.0,
      "step": 18620
    },
    {
      "entropy": 5.522739839553833,
      "epoch": 1.4490955067107567,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004791228508029565,
      "loss": 5.173,
      "mean_token_accuracy": 0.18253856599330903,
      "num_tokens": 32306785.0,
      "step": 18625
    },
    {
      "entropy": 5.447295713424682,
      "epoch": 1.4494845360824742,
      "grad_norm": 1.875,
      "learning_rate": 0.00047911119007479873,
      "loss": 5.2974,
      "mean_token_accuracy": 0.17676982432603836,
      "num_tokens": 32315419.0,
      "step": 18630
    },
    {
      "entropy": 5.498271560668945,
      "epoch": 1.4498735654541917,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047909952624957866,
      "loss": 5.1661,
      "mean_token_accuracy": 0.18285317867994308,
      "num_tokens": 32323655.0,
      "step": 18635
    },
    {
      "entropy": 5.494969701766967,
      "epoch": 1.4502625948259094,
      "grad_norm": 1.484375,
      "learning_rate": 0.000479087859327473,
      "loss": 5.1564,
      "mean_token_accuracy": 0.17849189192056655,
      "num_tokens": 32331465.0,
      "step": 18640
    },
    {
      "entropy": 5.518987464904785,
      "epoch": 1.450651624197627,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000479076189308659,
      "loss": 5.3647,
      "mean_token_accuracy": 0.16813654005527495,
      "num_tokens": 32341211.0,
      "step": 18645
    },
    {
      "entropy": 5.599263381958008,
      "epoch": 1.4510406535693445,
      "grad_norm": 1.453125,
      "learning_rate": 0.00047906451619331364,
      "loss": 5.3881,
      "mean_token_accuracy": 0.16863451153039932,
      "num_tokens": 32349344.0,
      "step": 18650
    },
    {
      "entropy": 5.597398948669434,
      "epoch": 1.451429682941062,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00047905283998161404,
      "loss": 5.2459,
      "mean_token_accuracy": 0.18101862221956252,
      "num_tokens": 32357797.0,
      "step": 18655
    },
    {
      "entropy": 5.418971300125122,
      "epoch": 1.4518187123127797,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00047904116067373743,
      "loss": 5.1741,
      "mean_token_accuracy": 0.1775391846895218,
      "num_tokens": 32366131.0,
      "step": 18660
    },
    {
      "entropy": 5.500018072128296,
      "epoch": 1.4522077416844972,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004790294782698609,
      "loss": 5.3244,
      "mean_token_accuracy": 0.17078361809253692,
      "num_tokens": 32374738.0,
      "step": 18665
    },
    {
      "entropy": 5.5818921566009525,
      "epoch": 1.4525967710562147,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004790177927701618,
      "loss": 5.2686,
      "mean_token_accuracy": 0.17400980293750762,
      "num_tokens": 32383638.0,
      "step": 18670
    },
    {
      "entropy": 5.589617919921875,
      "epoch": 1.4529858004279324,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004790061041748174,
      "loss": 5.3198,
      "mean_token_accuracy": 0.17448485642671585,
      "num_tokens": 32393269.0,
      "step": 18675
    },
    {
      "entropy": 5.578018474578857,
      "epoch": 1.4533748297996498,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004789944124840051,
      "loss": 5.2831,
      "mean_token_accuracy": 0.17404098212718963,
      "num_tokens": 32401568.0,
      "step": 18680
    },
    {
      "entropy": 5.484119415283203,
      "epoch": 1.4537638591713673,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004789827176979022,
      "loss": 5.177,
      "mean_token_accuracy": 0.18192571997642518,
      "num_tokens": 32409996.0,
      "step": 18685
    },
    {
      "entropy": 5.498085594177246,
      "epoch": 1.454152888543085,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004789710198166864,
      "loss": 5.2546,
      "mean_token_accuracy": 0.17036639899015427,
      "num_tokens": 32418886.0,
      "step": 18690
    },
    {
      "entropy": 5.4717014789581295,
      "epoch": 1.4545419179148027,
      "grad_norm": 1.4375,
      "learning_rate": 0.00047895931884053497,
      "loss": 5.088,
      "mean_token_accuracy": 0.1844627693295479,
      "num_tokens": 32427999.0,
      "step": 18695
    },
    {
      "entropy": 5.574623680114746,
      "epoch": 1.4549309472865202,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047894761476962547,
      "loss": 5.3855,
      "mean_token_accuracy": 0.1670853778719902,
      "num_tokens": 32437115.0,
      "step": 18700
    },
    {
      "entropy": 5.509613370895385,
      "epoch": 1.4553199766582376,
      "grad_norm": 1.375,
      "learning_rate": 0.00047893590760413545,
      "loss": 5.182,
      "mean_token_accuracy": 0.18272268921136856,
      "num_tokens": 32445668.0,
      "step": 18705
    },
    {
      "entropy": 5.419609689712525,
      "epoch": 1.4557090060299553,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00047892419734424276,
      "loss": 5.2173,
      "mean_token_accuracy": 0.18120238333940505,
      "num_tokens": 32454022.0,
      "step": 18710
    },
    {
      "entropy": 5.6126405715942385,
      "epoch": 1.4560980354016728,
      "grad_norm": 1.703125,
      "learning_rate": 0.00047891248399012495,
      "loss": 5.3578,
      "mean_token_accuracy": 0.16935040354728698,
      "num_tokens": 32462997.0,
      "step": 18715
    },
    {
      "entropy": 5.639453792572022,
      "epoch": 1.4564870647733903,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004789007675419597,
      "loss": 5.2013,
      "mean_token_accuracy": 0.1808380052447319,
      "num_tokens": 32471184.0,
      "step": 18720
    },
    {
      "entropy": 5.457993841171264,
      "epoch": 1.456876094145108,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047888904799992486,
      "loss": 5.2321,
      "mean_token_accuracy": 0.17736491411924363,
      "num_tokens": 32480051.0,
      "step": 18725
    },
    {
      "entropy": 5.471709728240967,
      "epoch": 1.4572651235168255,
      "grad_norm": 1.71875,
      "learning_rate": 0.00047887732536419826,
      "loss": 5.2668,
      "mean_token_accuracy": 0.17767847031354905,
      "num_tokens": 32489269.0,
      "step": 18730
    },
    {
      "entropy": 5.533928155899048,
      "epoch": 1.4576541528885432,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004788655996349577,
      "loss": 5.2032,
      "mean_token_accuracy": 0.17996295690536498,
      "num_tokens": 32497941.0,
      "step": 18735
    },
    {
      "entropy": 5.577839612960815,
      "epoch": 1.4580431822602606,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047885387081238125,
      "loss": 5.2556,
      "mean_token_accuracy": 0.17264704704284667,
      "num_tokens": 32507679.0,
      "step": 18740
    },
    {
      "entropy": 5.513876342773438,
      "epoch": 1.4584322116319783,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004788421388966467,
      "loss": 5.2543,
      "mean_token_accuracy": 0.177100370824337,
      "num_tokens": 32516964.0,
      "step": 18745
    },
    {
      "entropy": 5.489507007598877,
      "epoch": 1.4588212410036958,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047883040388793207,
      "loss": 5.0977,
      "mean_token_accuracy": 0.18875344842672348,
      "num_tokens": 32525178.0,
      "step": 18750
    },
    {
      "entropy": 5.5613339900970455,
      "epoch": 1.4592102703754133,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00047881866578641563,
      "loss": 5.2873,
      "mean_token_accuracy": 0.17423087656497954,
      "num_tokens": 32533694.0,
      "step": 18755
    },
    {
      "entropy": 5.481985378265381,
      "epoch": 1.459599299747131,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004788069245922753,
      "loss": 5.263,
      "mean_token_accuracy": 0.17914574444293976,
      "num_tokens": 32542529.0,
      "step": 18760
    },
    {
      "entropy": 5.539718818664551,
      "epoch": 1.4599883291188485,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004787951803056893,
      "loss": 5.2557,
      "mean_token_accuracy": 0.17087531983852386,
      "num_tokens": 32550987.0,
      "step": 18765
    },
    {
      "entropy": 5.461848402023316,
      "epoch": 1.460377358490566,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004787834329268358,
      "loss": 5.2196,
      "mean_token_accuracy": 0.18231728672981262,
      "num_tokens": 32559812.0,
      "step": 18770
    },
    {
      "entropy": 5.483540821075439,
      "epoch": 1.4607663878622836,
      "grad_norm": 1.546875,
      "learning_rate": 0.000478771682455893,
      "loss": 5.2575,
      "mean_token_accuracy": 0.17913838326931,
      "num_tokens": 32568477.0,
      "step": 18775
    },
    {
      "entropy": 5.595295095443726,
      "epoch": 1.461155417234001,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004787599288930393,
      "loss": 5.3042,
      "mean_token_accuracy": 0.17240114361047745,
      "num_tokens": 32577921.0,
      "step": 18780
    },
    {
      "entropy": 5.53241081237793,
      "epoch": 1.4615444466057188,
      "grad_norm": 1.546875,
      "learning_rate": 0.000478748172238453,
      "loss": 5.1865,
      "mean_token_accuracy": 0.17665198296308518,
      "num_tokens": 32587028.0,
      "step": 18785
    },
    {
      "entropy": 5.468697023391724,
      "epoch": 1.4619334759774363,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004787364124923125,
      "loss": 5.2113,
      "mean_token_accuracy": 0.17969428598880768,
      "num_tokens": 32596092.0,
      "step": 18790
    },
    {
      "entropy": 5.493045425415039,
      "epoch": 1.462322505349154,
      "grad_norm": 1.9375,
      "learning_rate": 0.00047872464965479625,
      "loss": 5.1867,
      "mean_token_accuracy": 0.17488005757331848,
      "num_tokens": 32605271.0,
      "step": 18795
    },
    {
      "entropy": 5.5779859066009525,
      "epoch": 1.4627115347208715,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004787128837260826,
      "loss": 5.3576,
      "mean_token_accuracy": 0.17157439291477203,
      "num_tokens": 32614240.0,
      "step": 18800
    },
    {
      "entropy": 5.565021753311157,
      "epoch": 1.463100564092589,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004787011147063503,
      "loss": 5.2703,
      "mean_token_accuracy": 0.1783323273062706,
      "num_tokens": 32623390.0,
      "step": 18805
    },
    {
      "entropy": 5.527627229690552,
      "epoch": 1.4634895934643066,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004786893425957777,
      "loss": 5.2488,
      "mean_token_accuracy": 0.17946368008852004,
      "num_tokens": 32631649.0,
      "step": 18810
    },
    {
      "entropy": 5.468304920196533,
      "epoch": 1.463878622836024,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004786775673945436,
      "loss": 5.1639,
      "mean_token_accuracy": 0.18047667890787125,
      "num_tokens": 32640844.0,
      "step": 18815
    },
    {
      "entropy": 5.552310752868652,
      "epoch": 1.4642676522077416,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047866578910282656,
      "loss": 5.2358,
      "mean_token_accuracy": 0.18207817822694777,
      "num_tokens": 32649458.0,
      "step": 18820
    },
    {
      "entropy": 5.4970098495483395,
      "epoch": 1.4646566815794593,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047865400772080535,
      "loss": 5.2361,
      "mean_token_accuracy": 0.18117030262947081,
      "num_tokens": 32657649.0,
      "step": 18825
    },
    {
      "entropy": 5.6064270496368405,
      "epoch": 1.4650457109511767,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00047864222324865875,
      "loss": 5.3613,
      "mean_token_accuracy": 0.17145565152168274,
      "num_tokens": 32665407.0,
      "step": 18830
    },
    {
      "entropy": 5.538425588607788,
      "epoch": 1.4654347403228944,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004786304356865655,
      "loss": 5.18,
      "mean_token_accuracy": 0.17483089864253998,
      "num_tokens": 32673760.0,
      "step": 18835
    },
    {
      "entropy": 5.4918532371521,
      "epoch": 1.465823769694612,
      "grad_norm": 2.0625,
      "learning_rate": 0.00047861864503470457,
      "loss": 5.2163,
      "mean_token_accuracy": 0.17299856692552568,
      "num_tokens": 32682118.0,
      "step": 18840
    },
    {
      "entropy": 5.533446168899536,
      "epoch": 1.4662127990663296,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004786068512932547,
      "loss": 5.2338,
      "mean_token_accuracy": 0.1782324194908142,
      "num_tokens": 32690430.0,
      "step": 18845
    },
    {
      "entropy": 5.509104585647583,
      "epoch": 1.466601828438047,
      "grad_norm": 1.578125,
      "learning_rate": 0.000478595054462395,
      "loss": 5.1005,
      "mean_token_accuracy": 0.18468573540449143,
      "num_tokens": 32698776.0,
      "step": 18850
    },
    {
      "entropy": 5.578722095489502,
      "epoch": 1.4669908578097646,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00047858325454230437,
      "loss": 5.3369,
      "mean_token_accuracy": 0.16908054798841476,
      "num_tokens": 32707299.0,
      "step": 18855
    },
    {
      "entropy": 5.461874628067017,
      "epoch": 1.4673798871814823,
      "grad_norm": 2.5,
      "learning_rate": 0.0004785714515331619,
      "loss": 5.1691,
      "mean_token_accuracy": 0.18886398226022721,
      "num_tokens": 32716041.0,
      "step": 18860
    },
    {
      "entropy": 5.558421325683594,
      "epoch": 1.4677689165531997,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00047855964543514666,
      "loss": 5.2067,
      "mean_token_accuracy": 0.18090793937444688,
      "num_tokens": 32724182.0,
      "step": 18865
    },
    {
      "entropy": 5.4812487125396725,
      "epoch": 1.4681579459249172,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047854783624843786,
      "loss": 5.1533,
      "mean_token_accuracy": 0.1838173657655716,
      "num_tokens": 32732858.0,
      "step": 18870
    },
    {
      "entropy": 5.556770753860474,
      "epoch": 1.468546975296635,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00047853602397321453,
      "loss": 5.2887,
      "mean_token_accuracy": 0.17406779378652573,
      "num_tokens": 32741422.0,
      "step": 18875
    },
    {
      "entropy": 5.587092685699463,
      "epoch": 1.4689360046683524,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00047852420860965605,
      "loss": 5.3569,
      "mean_token_accuracy": 0.17178572565317154,
      "num_tokens": 32750720.0,
      "step": 18880
    },
    {
      "entropy": 5.46769118309021,
      "epoch": 1.46932503404007,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047851239015794166,
      "loss": 5.1763,
      "mean_token_accuracy": 0.18463176041841506,
      "num_tokens": 32759418.0,
      "step": 18885
    },
    {
      "entropy": 5.475292634963989,
      "epoch": 1.4697140634117876,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00047850056861825066,
      "loss": 5.2437,
      "mean_token_accuracy": 0.17988823354244232,
      "num_tokens": 32767604.0,
      "step": 18890
    },
    {
      "entropy": 5.505810976028442,
      "epoch": 1.4701030927835053,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00047848874399076245,
      "loss": 5.3327,
      "mean_token_accuracy": 0.1643638089299202,
      "num_tokens": 32776535.0,
      "step": 18895
    },
    {
      "entropy": 5.60525918006897,
      "epoch": 1.4704921221552227,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004784769162756563,
      "loss": 5.2437,
      "mean_token_accuracy": 0.17636585980653763,
      "num_tokens": 32785633.0,
      "step": 18900
    },
    {
      "entropy": 5.555313396453857,
      "epoch": 1.4708811515269402,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000478465085473112,
      "loss": 5.1965,
      "mean_token_accuracy": 0.17897548973560334,
      "num_tokens": 32794281.0,
      "step": 18905
    },
    {
      "entropy": 5.557896709442138,
      "epoch": 1.471270180898658,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004784532515833088,
      "loss": 5.3886,
      "mean_token_accuracy": 0.17159285098314286,
      "num_tokens": 32802312.0,
      "step": 18910
    },
    {
      "entropy": 5.4439699172973635,
      "epoch": 1.4716592102703754,
      "grad_norm": 1.5625,
      "learning_rate": 0.00047844141460642636,
      "loss": 5.1483,
      "mean_token_accuracy": 0.18073673993349076,
      "num_tokens": 32810983.0,
      "step": 18915
    },
    {
      "entropy": 5.51186203956604,
      "epoch": 1.4720482396420929,
      "grad_norm": 1.5,
      "learning_rate": 0.00047842957454264425,
      "loss": 5.2088,
      "mean_token_accuracy": 0.17734425514936447,
      "num_tokens": 32820695.0,
      "step": 18920
    },
    {
      "entropy": 5.555723428726196,
      "epoch": 1.4724372690138106,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047841773139214213,
      "loss": 5.2394,
      "mean_token_accuracy": 0.17897880971431732,
      "num_tokens": 32828483.0,
      "step": 18925
    },
    {
      "entropy": 5.470044183731079,
      "epoch": 1.472826298385528,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004784058851550997,
      "loss": 5.2496,
      "mean_token_accuracy": 0.17767204642295836,
      "num_tokens": 32837827.0,
      "step": 18930
    },
    {
      "entropy": 5.560628175735474,
      "epoch": 1.4732153277572457,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004783940358316967,
      "loss": 5.338,
      "mean_token_accuracy": 0.1716095820069313,
      "num_tokens": 32846872.0,
      "step": 18935
    },
    {
      "entropy": 5.5058746337890625,
      "epoch": 1.4736043571289632,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00047838218342211296,
      "loss": 5.1407,
      "mean_token_accuracy": 0.18916736990213395,
      "num_tokens": 32855418.0,
      "step": 18940
    },
    {
      "entropy": 5.474823427200318,
      "epoch": 1.473993386500681,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047837032792652837,
      "loss": 5.2385,
      "mean_token_accuracy": 0.17264438569545745,
      "num_tokens": 32864075.0,
      "step": 18945
    },
    {
      "entropy": 5.550068044662476,
      "epoch": 1.4743824158723984,
      "grad_norm": 1.625,
      "learning_rate": 0.0004783584693451226,
      "loss": 5.3247,
      "mean_token_accuracy": 0.17061852365732194,
      "num_tokens": 32872615.0,
      "step": 18950
    },
    {
      "entropy": 5.596851968765259,
      "epoch": 1.4747714452441159,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004783466076780759,
      "loss": 5.1821,
      "mean_token_accuracy": 0.18105166256427765,
      "num_tokens": 32881313.0,
      "step": 18955
    },
    {
      "entropy": 5.535932636260986,
      "epoch": 1.4751604746158336,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004783347429255679,
      "loss": 5.2618,
      "mean_token_accuracy": 0.17378393709659576,
      "num_tokens": 32889323.0,
      "step": 18960
    },
    {
      "entropy": 5.382056617736817,
      "epoch": 1.475549503987551,
      "grad_norm": 2.5625,
      "learning_rate": 0.000478322875087779,
      "loss": 5.2122,
      "mean_token_accuracy": 0.17133685797452927,
      "num_tokens": 32898045.0,
      "step": 18965
    },
    {
      "entropy": 5.606275320053101,
      "epoch": 1.4759385333592685,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047831100416488906,
      "loss": 5.2937,
      "mean_token_accuracy": 0.16449671983718872,
      "num_tokens": 32906646.0,
      "step": 18970
    },
    {
      "entropy": 5.586366462707519,
      "epoch": 1.4763275627309862,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047829913015707816,
      "loss": 5.276,
      "mean_token_accuracy": 0.18009517788887025,
      "num_tokens": 32915714.0,
      "step": 18975
    },
    {
      "entropy": 5.513172626495361,
      "epoch": 1.4767165921027037,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047828725306452657,
      "loss": 5.3706,
      "mean_token_accuracy": 0.1678380027413368,
      "num_tokens": 32924996.0,
      "step": 18980
    },
    {
      "entropy": 5.512981271743774,
      "epoch": 1.4771056214744214,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00047827537288741453,
      "loss": 5.1739,
      "mean_token_accuracy": 0.17783922106027603,
      "num_tokens": 32933550.0,
      "step": 18985
    },
    {
      "entropy": 5.588498878479004,
      "epoch": 1.4774946508461388,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004782634896259222,
      "loss": 5.2963,
      "mean_token_accuracy": 0.17333537638187407,
      "num_tokens": 32941800.0,
      "step": 18990
    },
    {
      "entropy": 5.510610628128052,
      "epoch": 1.4778836802178565,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047825160328023,
      "loss": 5.2754,
      "mean_token_accuracy": 0.17806044071912766,
      "num_tokens": 32951674.0,
      "step": 18995
    },
    {
      "entropy": 5.515177917480469,
      "epoch": 1.478272709589574,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004782397138505181,
      "loss": 5.2085,
      "mean_token_accuracy": 0.1780326247215271,
      "num_tokens": 32959880.0,
      "step": 19000
    },
    {
      "entropy": 5.46801233291626,
      "epoch": 1.4786617389612915,
      "grad_norm": 1.53125,
      "learning_rate": 0.00047822782133696715,
      "loss": 5.2089,
      "mean_token_accuracy": 0.1779392957687378,
      "num_tokens": 32968557.0,
      "step": 19005
    },
    {
      "entropy": 5.643283653259277,
      "epoch": 1.4790507683330092,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004782159257397574,
      "loss": 5.3013,
      "mean_token_accuracy": 0.1736760750412941,
      "num_tokens": 32977519.0,
      "step": 19010
    },
    {
      "entropy": 5.605164241790772,
      "epoch": 1.4794397977047267,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00047820402705906953,
      "loss": 5.2983,
      "mean_token_accuracy": 0.1751009404659271,
      "num_tokens": 32985917.0,
      "step": 19015
    },
    {
      "entropy": 5.442546033859253,
      "epoch": 1.4798288270764441,
      "grad_norm": 1.375,
      "learning_rate": 0.00047819212529508394,
      "loss": 5.2219,
      "mean_token_accuracy": 0.17615674436092377,
      "num_tokens": 32994612.0,
      "step": 19020
    },
    {
      "entropy": 5.510252857208252,
      "epoch": 1.4802178564481618,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004781802204479812,
      "loss": 5.3181,
      "mean_token_accuracy": 0.17805822640657426,
      "num_tokens": 33003528.0,
      "step": 19025
    },
    {
      "entropy": 5.508291482925415,
      "epoch": 1.4806068858198795,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004781683125179421,
      "loss": 5.1985,
      "mean_token_accuracy": 0.18396976590156555,
      "num_tokens": 33011685.0,
      "step": 19030
    },
    {
      "entropy": 5.5322174549102785,
      "epoch": 1.480995915191597,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004781564015051472,
      "loss": 5.241,
      "mean_token_accuracy": 0.18288617730140685,
      "num_tokens": 33020313.0,
      "step": 19035
    },
    {
      "entropy": 5.47178168296814,
      "epoch": 1.4813849445633145,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004781444874097772,
      "loss": 5.2781,
      "mean_token_accuracy": 0.17470365464687349,
      "num_tokens": 33029143.0,
      "step": 19040
    },
    {
      "entropy": 5.494315099716187,
      "epoch": 1.4817739739350322,
      "grad_norm": 2.09375,
      "learning_rate": 0.00047813257023201307,
      "loss": 5.1964,
      "mean_token_accuracy": 0.1762500986456871,
      "num_tokens": 33038359.0,
      "step": 19045
    },
    {
      "entropy": 5.467061376571655,
      "epoch": 1.4821630033067497,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004781206499720354,
      "loss": 5.2529,
      "mean_token_accuracy": 0.17411217987537383,
      "num_tokens": 33046914.0,
      "step": 19050
    },
    {
      "entropy": 5.591938924789429,
      "epoch": 1.4825520326784671,
      "grad_norm": 1.5,
      "learning_rate": 0.00047810872663002523,
      "loss": 5.3421,
      "mean_token_accuracy": 0.17366454303264617,
      "num_tokens": 33054490.0,
      "step": 19055
    },
    {
      "entropy": 5.562242221832276,
      "epoch": 1.4829410620501848,
      "grad_norm": 1.734375,
      "learning_rate": 0.00047809680020616333,
      "loss": 5.391,
      "mean_token_accuracy": 0.17157071232795715,
      "num_tokens": 33064624.0,
      "step": 19060
    },
    {
      "entropy": 5.484333562850952,
      "epoch": 1.4833300914219023,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047808487070063083,
      "loss": 5.1165,
      "mean_token_accuracy": 0.17980176955461502,
      "num_tokens": 33072581.0,
      "step": 19065
    },
    {
      "entropy": 5.520661783218384,
      "epoch": 1.4837191207936198,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004780729381136086,
      "loss": 5.2819,
      "mean_token_accuracy": 0.178500497341156,
      "num_tokens": 33081696.0,
      "step": 19070
    },
    {
      "entropy": 5.533953046798706,
      "epoch": 1.4841081501653375,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004780610024452778,
      "loss": 5.2499,
      "mean_token_accuracy": 0.17403241097927094,
      "num_tokens": 33090470.0,
      "step": 19075
    },
    {
      "entropy": 5.5420433521270756,
      "epoch": 1.4844971795370552,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00047804906369581954,
      "loss": 5.3068,
      "mean_token_accuracy": 0.1782301813364029,
      "num_tokens": 33099006.0,
      "step": 19080
    },
    {
      "entropy": 5.4641313552856445,
      "epoch": 1.4848862089087727,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004780371218654148,
      "loss": 5.1575,
      "mean_token_accuracy": 0.18589748591184616,
      "num_tokens": 33107715.0,
      "step": 19085
    },
    {
      "entropy": 5.525591039657593,
      "epoch": 1.4852752382804901,
      "grad_norm": 1.625,
      "learning_rate": 0.00047802517695424496,
      "loss": 5.1965,
      "mean_token_accuracy": 0.18338321894407272,
      "num_tokens": 33116895.0,
      "step": 19090
    },
    {
      "entropy": 5.5436680793762205,
      "epoch": 1.4856642676522078,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004780132289624913,
      "loss": 5.2482,
      "mean_token_accuracy": 0.17710049748420714,
      "num_tokens": 33125676.0,
      "step": 19095
    },
    {
      "entropy": 5.504342174530029,
      "epoch": 1.4860532970239253,
      "grad_norm": 1.875,
      "learning_rate": 0.0004780012778903349,
      "loss": 5.2414,
      "mean_token_accuracy": 0.1806422621011734,
      "num_tokens": 33134870.0,
      "step": 19100
    },
    {
      "entropy": 5.506136846542359,
      "epoch": 1.4864423263956428,
      "grad_norm": 1.546875,
      "learning_rate": 0.00047798932373795724,
      "loss": 5.3259,
      "mean_token_accuracy": 0.17531076222658157,
      "num_tokens": 33144258.0,
      "step": 19105
    },
    {
      "entropy": 5.574120378494262,
      "epoch": 1.4868313557673605,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00047797736650553977,
      "loss": 5.2662,
      "mean_token_accuracy": 0.17567249685525893,
      "num_tokens": 33153703.0,
      "step": 19110
    },
    {
      "entropy": 5.570472526550293,
      "epoch": 1.487220385139078,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00047796540619326366,
      "loss": 5.247,
      "mean_token_accuracy": 0.1701607197523117,
      "num_tokens": 33163092.0,
      "step": 19115
    },
    {
      "entropy": 5.511020660400391,
      "epoch": 1.4876094145107956,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004779534428013107,
      "loss": 5.2518,
      "mean_token_accuracy": 0.17861293256282806,
      "num_tokens": 33172133.0,
      "step": 19120
    },
    {
      "entropy": 5.5301741600036625,
      "epoch": 1.4879984438825131,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047794147632986223,
      "loss": 5.2577,
      "mean_token_accuracy": 0.17505211383104324,
      "num_tokens": 33180629.0,
      "step": 19125
    },
    {
      "entropy": 5.515226984024048,
      "epoch": 1.4883874732542308,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004779295067790999,
      "loss": 5.209,
      "mean_token_accuracy": 0.17765021622180938,
      "num_tokens": 33189602.0,
      "step": 19130
    },
    {
      "entropy": 5.523668098449707,
      "epoch": 1.4887765026259483,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004779175341492053,
      "loss": 5.2393,
      "mean_token_accuracy": 0.17363129705190658,
      "num_tokens": 33198403.0,
      "step": 19135
    },
    {
      "entropy": 5.474996280670166,
      "epoch": 1.4891655319976658,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004779055584403601,
      "loss": 5.1822,
      "mean_token_accuracy": 0.17516040056943893,
      "num_tokens": 33207131.0,
      "step": 19140
    },
    {
      "entropy": 5.5032196044921875,
      "epoch": 1.4895545613693835,
      "grad_norm": 1.40625,
      "learning_rate": 0.000477893579652746,
      "loss": 5.2319,
      "mean_token_accuracy": 0.1749684691429138,
      "num_tokens": 33216143.0,
      "step": 19145
    },
    {
      "entropy": 5.491715717315674,
      "epoch": 1.489943590741101,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047788159778654475,
      "loss": 5.227,
      "mean_token_accuracy": 0.18794695436954498,
      "num_tokens": 33224558.0,
      "step": 19150
    },
    {
      "entropy": 5.418473196029663,
      "epoch": 1.4903326201128184,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047786961284193813,
      "loss": 5.2249,
      "mean_token_accuracy": 0.1761377602815628,
      "num_tokens": 33232737.0,
      "step": 19155
    },
    {
      "entropy": 5.546581697463989,
      "epoch": 1.4907216494845361,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004778576248191081,
      "loss": 5.2419,
      "mean_token_accuracy": 0.17629723995923996,
      "num_tokens": 33240845.0,
      "step": 19160
    },
    {
      "entropy": 5.50310640335083,
      "epoch": 1.4911106788562536,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004778456337182365,
      "loss": 5.1486,
      "mean_token_accuracy": 0.18558304905891418,
      "num_tokens": 33249101.0,
      "step": 19165
    },
    {
      "entropy": 5.570547819137573,
      "epoch": 1.4914997082279713,
      "grad_norm": 1.625,
      "learning_rate": 0.0004778336395395052,
      "loss": 5.2546,
      "mean_token_accuracy": 0.17909433245658873,
      "num_tokens": 33258403.0,
      "step": 19170
    },
    {
      "entropy": 5.3771405696868895,
      "epoch": 1.4918887375996888,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047782164228309636,
      "loss": 5.1617,
      "mean_token_accuracy": 0.18375119864940642,
      "num_tokens": 33266679.0,
      "step": 19175
    },
    {
      "entropy": 5.477028846740723,
      "epoch": 1.4922777669714065,
      "grad_norm": 1.59375,
      "learning_rate": 0.00047780964194919193,
      "loss": 5.1756,
      "mean_token_accuracy": 0.1825355038046837,
      "num_tokens": 33274996.0,
      "step": 19180
    },
    {
      "entropy": 5.6039050102233885,
      "epoch": 1.492666796343124,
      "grad_norm": 1.578125,
      "learning_rate": 0.00047779763853797387,
      "loss": 5.297,
      "mean_token_accuracy": 0.17526201605796815,
      "num_tokens": 33282558.0,
      "step": 19185
    },
    {
      "entropy": 5.521491050720215,
      "epoch": 1.4930558257148414,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004777856320496245,
      "loss": 5.2857,
      "mean_token_accuracy": 0.17085435837507248,
      "num_tokens": 33291233.0,
      "step": 19190
    },
    {
      "entropy": 5.516939544677735,
      "epoch": 1.493444855086559,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000477773622484326,
      "loss": 5.1519,
      "mean_token_accuracy": 0.18402533680200578,
      "num_tokens": 33299821.0,
      "step": 19195
    },
    {
      "entropy": 5.497485160827637,
      "epoch": 1.4938338844582766,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004777616098422604,
      "loss": 5.243,
      "mean_token_accuracy": 0.1784694239497185,
      "num_tokens": 33309624.0,
      "step": 19200
    },
    {
      "entropy": 5.548103713989258,
      "epoch": 1.494222913829994,
      "grad_norm": 1.671875,
      "learning_rate": 0.00047774959412361014,
      "loss": 5.2534,
      "mean_token_accuracy": 0.17316729128360747,
      "num_tokens": 33318727.0,
      "step": 19205
    },
    {
      "entropy": 5.556207752227783,
      "epoch": 1.4946119432017118,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004777375753285575,
      "loss": 5.2643,
      "mean_token_accuracy": 0.17649584859609604,
      "num_tokens": 33327219.0,
      "step": 19210
    },
    {
      "entropy": 5.49752836227417,
      "epoch": 1.4950009725734292,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00047772555345728486,
      "loss": 5.2436,
      "mean_token_accuracy": 0.17766701579093933,
      "num_tokens": 33336466.0,
      "step": 19215
    },
    {
      "entropy": 5.639565515518188,
      "epoch": 1.495390001945147,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004777135285099746,
      "loss": 5.3138,
      "mean_token_accuracy": 0.1677102193236351,
      "num_tokens": 33346040.0,
      "step": 19220
    },
    {
      "entropy": 5.583371782302857,
      "epoch": 1.4957790313168644,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004777015004868092,
      "loss": 5.2554,
      "mean_token_accuracy": 0.17835576236248016,
      "num_tokens": 33355476.0,
      "step": 19225
    },
    {
      "entropy": 5.427486371994019,
      "epoch": 1.496168060688582,
      "grad_norm": 1.5,
      "learning_rate": 0.0004776894693879712,
      "loss": 5.2101,
      "mean_token_accuracy": 0.1823199763894081,
      "num_tokens": 33364455.0,
      "step": 19230
    },
    {
      "entropy": 5.449832153320313,
      "epoch": 1.4965570900602996,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004776774352136431,
      "loss": 5.1963,
      "mean_token_accuracy": 0.17985452711582184,
      "num_tokens": 33373129.0,
      "step": 19235
    },
    {
      "entropy": 5.468187570571899,
      "epoch": 1.496946119432017,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00047766539796400756,
      "loss": 5.2227,
      "mean_token_accuracy": 0.17842733412981032,
      "num_tokens": 33382775.0,
      "step": 19240
    },
    {
      "entropy": 5.522367668151856,
      "epoch": 1.4973351488037348,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004776533576392471,
      "loss": 5.2179,
      "mean_token_accuracy": 0.17783847004175185,
      "num_tokens": 33391370.0,
      "step": 19245
    },
    {
      "entropy": 5.577925491333008,
      "epoch": 1.4977241781754522,
      "grad_norm": 1.25,
      "learning_rate": 0.0004776413142395445,
      "loss": 5.2231,
      "mean_token_accuracy": 0.18265827596187592,
      "num_tokens": 33399832.0,
      "step": 19250
    },
    {
      "entropy": 5.447705411911011,
      "epoch": 1.4981132075471697,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047762926776508244,
      "loss": 5.2702,
      "mean_token_accuracy": 0.17773692309856415,
      "num_tokens": 33408709.0,
      "step": 19255
    },
    {
      "entropy": 5.469012689590454,
      "epoch": 1.4985022369188874,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00047761721821604387,
      "loss": 5.2526,
      "mean_token_accuracy": 0.17171256840229035,
      "num_tokens": 33416983.0,
      "step": 19260
    },
    {
      "entropy": 5.547253656387329,
      "epoch": 1.4988912662906049,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004776051655926115,
      "loss": 5.2792,
      "mean_token_accuracy": 0.17764000296592714,
      "num_tokens": 33426692.0,
      "step": 19265
    },
    {
      "entropy": 5.525631237030029,
      "epoch": 1.4992802956623226,
      "grad_norm": 1.75,
      "learning_rate": 0.00047759310989496813,
      "loss": 5.2771,
      "mean_token_accuracy": 0.1786951094865799,
      "num_tokens": 33434895.0,
      "step": 19270
    },
    {
      "entropy": 5.48480863571167,
      "epoch": 1.49966932503404,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004775810511232969,
      "loss": 5.1799,
      "mean_token_accuracy": 0.17740183025598527,
      "num_tokens": 33443378.0,
      "step": 19275
    },
    {
      "entropy": 5.6220931053161625,
      "epoch": 1.5000583544057577,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047756898927778056,
      "loss": 5.3216,
      "mean_token_accuracy": 0.17830412834882736,
      "num_tokens": 33452753.0,
      "step": 19280
    },
    {
      "entropy": 5.612939500808716,
      "epoch": 1.5004473837774752,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047755692435860227,
      "loss": 5.2736,
      "mean_token_accuracy": 0.17088256925344467,
      "num_tokens": 33461625.0,
      "step": 19285
    },
    {
      "entropy": 5.475265979766846,
      "epoch": 1.5008364131491927,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000477544856365945,
      "loss": 5.2059,
      "mean_token_accuracy": 0.179485522210598,
      "num_tokens": 33471228.0,
      "step": 19290
    },
    {
      "entropy": 5.471999597549439,
      "epoch": 1.5012254425209104,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047753278529999205,
      "loss": 5.2291,
      "mean_token_accuracy": 0.17918296307325363,
      "num_tokens": 33480137.0,
      "step": 19295
    },
    {
      "entropy": 5.535181331634521,
      "epoch": 1.5016144718926279,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047752071116092637,
      "loss": 5.2295,
      "mean_token_accuracy": 0.18213903903961182,
      "num_tokens": 33488860.0,
      "step": 19300
    },
    {
      "entropy": 5.5006327629089355,
      "epoch": 1.5020035012643453,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004775086339489312,
      "loss": 5.1929,
      "mean_token_accuracy": 0.17780565321445466,
      "num_tokens": 33497625.0,
      "step": 19305
    },
    {
      "entropy": 5.493137454986572,
      "epoch": 1.502392530636063,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004774965536641899,
      "loss": 5.3801,
      "mean_token_accuracy": 0.16805230230093002,
      "num_tokens": 33506422.0,
      "step": 19310
    },
    {
      "entropy": 5.495547437667847,
      "epoch": 1.5027815600077807,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047748447030688575,
      "loss": 5.1149,
      "mean_token_accuracy": 0.18306494504213333,
      "num_tokens": 33515638.0,
      "step": 19315
    },
    {
      "entropy": 5.507063627243042,
      "epoch": 1.503170589379498,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047747238387720194,
      "loss": 5.1579,
      "mean_token_accuracy": 0.17949606329202653,
      "num_tokens": 33524287.0,
      "step": 19320
    },
    {
      "entropy": 5.469194650650024,
      "epoch": 1.5035596187512157,
      "grad_norm": 1.328125,
      "learning_rate": 0.000477460294375322,
      "loss": 5.169,
      "mean_token_accuracy": 0.17187461405992507,
      "num_tokens": 33532579.0,
      "step": 19325
    },
    {
      "entropy": 5.462903022766113,
      "epoch": 1.5039486481229334,
      "grad_norm": 1.65625,
      "learning_rate": 0.00047744820180142935,
      "loss": 5.2507,
      "mean_token_accuracy": 0.17614063024520873,
      "num_tokens": 33540816.0,
      "step": 19330
    },
    {
      "entropy": 5.43666410446167,
      "epoch": 1.5043376774946509,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00047743610615570746,
      "loss": 5.1207,
      "mean_token_accuracy": 0.18905248045921325,
      "num_tokens": 33549383.0,
      "step": 19335
    },
    {
      "entropy": 5.44313154220581,
      "epoch": 1.5047267068663683,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047742400743833993,
      "loss": 5.1711,
      "mean_token_accuracy": 0.1888299450278282,
      "num_tokens": 33557457.0,
      "step": 19340
    },
    {
      "entropy": 5.536280107498169,
      "epoch": 1.505115736238086,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004774119056495102,
      "loss": 5.2773,
      "mean_token_accuracy": 0.1775692045688629,
      "num_tokens": 33565479.0,
      "step": 19345
    },
    {
      "entropy": 5.490532875061035,
      "epoch": 1.5055047656098035,
      "grad_norm": 1.375,
      "learning_rate": 0.00047739980078940206,
      "loss": 5.2402,
      "mean_token_accuracy": 0.1738107830286026,
      "num_tokens": 33574691.0,
      "step": 19350
    },
    {
      "entropy": 5.49760069847107,
      "epoch": 1.505893794981521,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047738769285819894,
      "loss": 5.0974,
      "mean_token_accuracy": 0.19047680944204332,
      "num_tokens": 33582360.0,
      "step": 19355
    },
    {
      "entropy": 5.417484474182129,
      "epoch": 1.5062828243532387,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004773755818560849,
      "loss": 5.2145,
      "mean_token_accuracy": 0.18156716525554656,
      "num_tokens": 33590977.0,
      "step": 19360
    },
    {
      "entropy": 5.421353960037232,
      "epoch": 1.5066718537249564,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004773634677832434,
      "loss": 5.1858,
      "mean_token_accuracy": 0.17915512025356292,
      "num_tokens": 33600227.0,
      "step": 19365
    },
    {
      "entropy": 5.555703830718994,
      "epoch": 1.5070608830966739,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004773513506398584,
      "loss": 5.2418,
      "mean_token_accuracy": 0.1850794032216072,
      "num_tokens": 33608408.0,
      "step": 19370
    },
    {
      "entropy": 5.5497194766998295,
      "epoch": 1.5074499124683913,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004773392304261137,
      "loss": 5.163,
      "mean_token_accuracy": 0.17952251434326172,
      "num_tokens": 33616892.0,
      "step": 19375
    },
    {
      "entropy": 5.471491575241089,
      "epoch": 1.507838941840109,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004773271071421933,
      "loss": 5.2719,
      "mean_token_accuracy": 0.17584420144557952,
      "num_tokens": 33626090.0,
      "step": 19380
    },
    {
      "entropy": 5.371228408813477,
      "epoch": 1.5082279712118265,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00047731498078828105,
      "loss": 5.113,
      "mean_token_accuracy": 0.18904853612184525,
      "num_tokens": 33634855.0,
      "step": 19385
    },
    {
      "entropy": 5.534246635437012,
      "epoch": 1.508617000583544,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000477302851364561,
      "loss": 5.2704,
      "mean_token_accuracy": 0.17462550103664398,
      "num_tokens": 33643425.0,
      "step": 19390
    },
    {
      "entropy": 5.6518120765686035,
      "epoch": 1.5090060299552617,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047729071887121717,
      "loss": 5.3694,
      "mean_token_accuracy": 0.16821225732564926,
      "num_tokens": 33652926.0,
      "step": 19395
    },
    {
      "entropy": 5.577523469924927,
      "epoch": 1.5093950593269791,
      "grad_norm": 1.625,
      "learning_rate": 0.0004772785833084337,
      "loss": 5.2849,
      "mean_token_accuracy": 0.16904430240392684,
      "num_tokens": 33661279.0,
      "step": 19400
    },
    {
      "entropy": 5.459644746780396,
      "epoch": 1.5097840886986966,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004772664446763946,
      "loss": 5.1542,
      "mean_token_accuracy": 0.17910317182540894,
      "num_tokens": 33669717.0,
      "step": 19405
    },
    {
      "entropy": 5.519716501235962,
      "epoch": 1.5101731180704143,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004772543029752842,
      "loss": 5.2944,
      "mean_token_accuracy": 0.17271620333194732,
      "num_tokens": 33679141.0,
      "step": 19410
    },
    {
      "entropy": 5.503009366989136,
      "epoch": 1.510562147442132,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047724215820528666,
      "loss": 5.1043,
      "mean_token_accuracy": 0.1840835064649582,
      "num_tokens": 33686867.0,
      "step": 19415
    },
    {
      "entropy": 5.514965438842774,
      "epoch": 1.5109511768138495,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004772300103665863,
      "loss": 5.2052,
      "mean_token_accuracy": 0.18849213123321534,
      "num_tokens": 33695073.0,
      "step": 19420
    },
    {
      "entropy": 5.492791175842285,
      "epoch": 1.511340206185567,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047721785945936733,
      "loss": 5.2865,
      "mean_token_accuracy": 0.17294924110174179,
      "num_tokens": 33703212.0,
      "step": 19425
    },
    {
      "entropy": 5.543099737167358,
      "epoch": 1.5117292355572847,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004772057054838143,
      "loss": 5.2325,
      "mean_token_accuracy": 0.178933984041214,
      "num_tokens": 33711487.0,
      "step": 19430
    },
    {
      "entropy": 5.539937543869018,
      "epoch": 1.5121182649290021,
      "grad_norm": 1.640625,
      "learning_rate": 0.00047719354844011146,
      "loss": 5.3027,
      "mean_token_accuracy": 0.17686383575201034,
      "num_tokens": 33719642.0,
      "step": 19435
    },
    {
      "entropy": 5.452017545700073,
      "epoch": 1.5125072943007196,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004771813883284434,
      "loss": 5.2132,
      "mean_token_accuracy": 0.1760488346219063,
      "num_tokens": 33728167.0,
      "step": 19440
    },
    {
      "entropy": 5.487132215499878,
      "epoch": 1.5128963236724373,
      "grad_norm": 1.546875,
      "learning_rate": 0.00047716922514899455,
      "loss": 5.2183,
      "mean_token_accuracy": 0.18252888768911363,
      "num_tokens": 33736094.0,
      "step": 19445
    },
    {
      "entropy": 5.490319633483887,
      "epoch": 1.5132853530441548,
      "grad_norm": 1.6875,
      "learning_rate": 0.00047715705890194953,
      "loss": 5.1025,
      "mean_token_accuracy": 0.18678970783948898,
      "num_tokens": 33744416.0,
      "step": 19450
    },
    {
      "entropy": 5.535049533843994,
      "epoch": 1.5136743824158723,
      "grad_norm": 2.140625,
      "learning_rate": 0.00047714488958749285,
      "loss": 5.2294,
      "mean_token_accuracy": 0.18198212683200837,
      "num_tokens": 33752873.0,
      "step": 19455
    },
    {
      "entropy": 5.4473429203033445,
      "epoch": 1.51406341178759,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00047713271720580924,
      "loss": 5.2055,
      "mean_token_accuracy": 0.17665968537330629,
      "num_tokens": 33761604.0,
      "step": 19460
    },
    {
      "entropy": 5.523878526687622,
      "epoch": 1.5144524411593077,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004771205417570834,
      "loss": 5.2378,
      "mean_token_accuracy": 0.18237500339746476,
      "num_tokens": 33770562.0,
      "step": 19465
    },
    {
      "entropy": 5.526638126373291,
      "epoch": 1.5148414705310251,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047710836324150004,
      "loss": 5.1784,
      "mean_token_accuracy": 0.18329360336065292,
      "num_tokens": 33779037.0,
      "step": 19470
    },
    {
      "entropy": 5.4901021957397464,
      "epoch": 1.5152304999027426,
      "grad_norm": 1.546875,
      "learning_rate": 0.000477096181659244,
      "loss": 5.2452,
      "mean_token_accuracy": 0.18164824694395065,
      "num_tokens": 33787949.0,
      "step": 19475
    },
    {
      "entropy": 5.500197124481201,
      "epoch": 1.5156195292744603,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004770839970105,
      "loss": 5.161,
      "mean_token_accuracy": 0.18387614339590072,
      "num_tokens": 33796783.0,
      "step": 19480
    },
    {
      "entropy": 5.528604984283447,
      "epoch": 1.5160085586461778,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047707180929545295,
      "loss": 5.2021,
      "mean_token_accuracy": 0.17828374952077866,
      "num_tokens": 33804975.0,
      "step": 19485
    },
    {
      "entropy": 5.4207807064056395,
      "epoch": 1.5163975880178953,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047705961851428786,
      "loss": 5.217,
      "mean_token_accuracy": 0.17787167727947234,
      "num_tokens": 33813355.0,
      "step": 19490
    },
    {
      "entropy": 5.450259447097778,
      "epoch": 1.516786617389613,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047704742466718973,
      "loss": 5.1799,
      "mean_token_accuracy": 0.18868510276079178,
      "num_tokens": 33821710.0,
      "step": 19495
    },
    {
      "entropy": 5.585661935806274,
      "epoch": 1.5171756467613304,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047703522775434354,
      "loss": 5.3466,
      "mean_token_accuracy": 0.17114655673503876,
      "num_tokens": 33830674.0,
      "step": 19500
    },
    {
      "entropy": 5.433118104934692,
      "epoch": 1.517564676133048,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047702302777593425,
      "loss": 5.1585,
      "mean_token_accuracy": 0.1804536372423172,
      "num_tokens": 33838963.0,
      "step": 19505
    },
    {
      "entropy": 5.537739515304565,
      "epoch": 1.5179537055047656,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047701082473214715,
      "loss": 5.2981,
      "mean_token_accuracy": 0.17711785733699797,
      "num_tokens": 33847728.0,
      "step": 19510
    },
    {
      "entropy": 5.5240199089050295,
      "epoch": 1.5183427348764833,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047699861862316725,
      "loss": 5.2805,
      "mean_token_accuracy": 0.18331485688686372,
      "num_tokens": 33856170.0,
      "step": 19515
    },
    {
      "entropy": 5.534817743301391,
      "epoch": 1.5187317642482008,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004769864094491798,
      "loss": 5.2693,
      "mean_token_accuracy": 0.17890494912862778,
      "num_tokens": 33865531.0,
      "step": 19520
    },
    {
      "entropy": 5.564637756347656,
      "epoch": 1.5191207936199183,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004769741972103702,
      "loss": 5.1741,
      "mean_token_accuracy": 0.1755641967058182,
      "num_tokens": 33873976.0,
      "step": 19525
    },
    {
      "entropy": 5.477769804000855,
      "epoch": 1.519509822991636,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047696198190692353,
      "loss": 5.0558,
      "mean_token_accuracy": 0.19165010005235672,
      "num_tokens": 33881884.0,
      "step": 19530
    },
    {
      "entropy": 5.481703615188598,
      "epoch": 1.5198988523633534,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004769497635390253,
      "loss": 5.2998,
      "mean_token_accuracy": 0.17802365124225616,
      "num_tokens": 33890932.0,
      "step": 19535
    },
    {
      "entropy": 5.5093635559082035,
      "epoch": 1.520287881735071,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004769375421068608,
      "loss": 5.2395,
      "mean_token_accuracy": 0.17986425906419753,
      "num_tokens": 33899271.0,
      "step": 19540
    },
    {
      "entropy": 5.48005690574646,
      "epoch": 1.5206769111067886,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00047692531761061555,
      "loss": 5.2722,
      "mean_token_accuracy": 0.1734055608510971,
      "num_tokens": 33908776.0,
      "step": 19545
    },
    {
      "entropy": 5.477955961227417,
      "epoch": 1.521065940478506,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000476913090050475,
      "loss": 5.1584,
      "mean_token_accuracy": 0.18549179583787917,
      "num_tokens": 33917424.0,
      "step": 19550
    },
    {
      "entropy": 5.561444520950317,
      "epoch": 1.5214549698502235,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047690085942662464,
      "loss": 5.2409,
      "mean_token_accuracy": 0.1757921427488327,
      "num_tokens": 33925948.0,
      "step": 19555
    },
    {
      "entropy": 5.540966939926148,
      "epoch": 1.5218439992219412,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047688862573925015,
      "loss": 5.2657,
      "mean_token_accuracy": 0.1819721519947052,
      "num_tokens": 33934623.0,
      "step": 19560
    },
    {
      "entropy": 5.499468040466309,
      "epoch": 1.522233028593659,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047687638898853707,
      "loss": 5.2651,
      "mean_token_accuracy": 0.17442897260189055,
      "num_tokens": 33943871.0,
      "step": 19565
    },
    {
      "entropy": 5.559149694442749,
      "epoch": 1.5226220579653764,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004768641491746711,
      "loss": 5.2328,
      "mean_token_accuracy": 0.17475976943969726,
      "num_tokens": 33953032.0,
      "step": 19570
    },
    {
      "entropy": 5.515017557144165,
      "epoch": 1.523011087337094,
      "grad_norm": 1.4375,
      "learning_rate": 0.000476851906297838,
      "loss": 5.2703,
      "mean_token_accuracy": 0.17934996038675308,
      "num_tokens": 33961702.0,
      "step": 19575
    },
    {
      "entropy": 5.515548610687256,
      "epoch": 1.5234001167088116,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047683966035822346,
      "loss": 5.2907,
      "mean_token_accuracy": 0.17442075312137603,
      "num_tokens": 33970996.0,
      "step": 19580
    },
    {
      "entropy": 5.560163354873657,
      "epoch": 1.523789146080529,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047682741135601336,
      "loss": 5.2671,
      "mean_token_accuracy": 0.17682661414146422,
      "num_tokens": 33980445.0,
      "step": 19585
    },
    {
      "entropy": 5.546480560302735,
      "epoch": 1.5241781754522465,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00047681515929139356,
      "loss": 5.1924,
      "mean_token_accuracy": 0.17978233397006987,
      "num_tokens": 33989042.0,
      "step": 19590
    },
    {
      "entropy": 5.400068187713623,
      "epoch": 1.5245672048239642,
      "grad_norm": 1.53125,
      "learning_rate": 0.00047680290416454995,
      "loss": 5.087,
      "mean_token_accuracy": 0.1935490220785141,
      "num_tokens": 33998162.0,
      "step": 19595
    },
    {
      "entropy": 5.455098581314087,
      "epoch": 1.524956234195682,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004767906459756684,
      "loss": 5.2001,
      "mean_token_accuracy": 0.18574739545583724,
      "num_tokens": 34007480.0,
      "step": 19600
    },
    {
      "entropy": 5.476698589324951,
      "epoch": 1.5253452635673992,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00047677838472493504,
      "loss": 5.2068,
      "mean_token_accuracy": 0.17758456617593765,
      "num_tokens": 34015605.0,
      "step": 19605
    },
    {
      "entropy": 5.539534568786621,
      "epoch": 1.5257342929391169,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004767661204125358,
      "loss": 5.3378,
      "mean_token_accuracy": 0.172260582447052,
      "num_tokens": 34024847.0,
      "step": 19610
    },
    {
      "entropy": 5.507574892044067,
      "epoch": 1.5261233223108346,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004767538530386569,
      "loss": 5.2114,
      "mean_token_accuracy": 0.17498655766248702,
      "num_tokens": 34034205.0,
      "step": 19615
    },
    {
      "entropy": 5.542208528518676,
      "epoch": 1.526512351682552,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00047674158260348437,
      "loss": 5.1349,
      "mean_token_accuracy": 0.18341722786426545,
      "num_tokens": 34042578.0,
      "step": 19620
    },
    {
      "entropy": 5.490110778808594,
      "epoch": 1.5269013810542695,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047672930910720436,
      "loss": 5.2068,
      "mean_token_accuracy": 0.18112893104553224,
      "num_tokens": 34051699.0,
      "step": 19625
    },
    {
      "entropy": 5.412960624694824,
      "epoch": 1.5272904104259872,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047671703255000326,
      "loss": 5.2156,
      "mean_token_accuracy": 0.17743537127971648,
      "num_tokens": 34060872.0,
      "step": 19630
    },
    {
      "entropy": 5.495071268081665,
      "epoch": 1.5276794397977047,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004767047529320673,
      "loss": 5.2585,
      "mean_token_accuracy": 0.17518007159233093,
      "num_tokens": 34069501.0,
      "step": 19635
    },
    {
      "entropy": 5.5749565124511715,
      "epoch": 1.5280684691694222,
      "grad_norm": 1.25,
      "learning_rate": 0.00047669247025358257,
      "loss": 5.3057,
      "mean_token_accuracy": 0.1733539342880249,
      "num_tokens": 34078573.0,
      "step": 19640
    },
    {
      "entropy": 5.506174993515015,
      "epoch": 1.5284574985411399,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047668018451473584,
      "loss": 5.2502,
      "mean_token_accuracy": 0.17640733271837233,
      "num_tokens": 34087270.0,
      "step": 19645
    },
    {
      "entropy": 5.487158155441284,
      "epoch": 1.5288465279128576,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004766678957157132,
      "loss": 5.279,
      "mean_token_accuracy": 0.17368052005767823,
      "num_tokens": 34096205.0,
      "step": 19650
    },
    {
      "entropy": 5.477304410934448,
      "epoch": 1.5292355572845748,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004766556038567013,
      "loss": 5.1091,
      "mean_token_accuracy": 0.1881142407655716,
      "num_tokens": 34104736.0,
      "step": 19655
    },
    {
      "entropy": 5.521236848831177,
      "epoch": 1.5296245866562925,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004766433089378865,
      "loss": 5.2285,
      "mean_token_accuracy": 0.18070223331451415,
      "num_tokens": 34113420.0,
      "step": 19660
    },
    {
      "entropy": 5.556590986251831,
      "epoch": 1.5300136160280102,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047663101095945544,
      "loss": 5.3078,
      "mean_token_accuracy": 0.16629046946763992,
      "num_tokens": 34121702.0,
      "step": 19665
    },
    {
      "entropy": 5.5381275653839115,
      "epoch": 1.5304026453997277,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047661870992159476,
      "loss": 5.1991,
      "mean_token_accuracy": 0.17533429712057114,
      "num_tokens": 34129657.0,
      "step": 19670
    },
    {
      "entropy": 5.510898780822754,
      "epoch": 1.5307916747714452,
      "grad_norm": 1.75,
      "learning_rate": 0.00047660640582449106,
      "loss": 5.3392,
      "mean_token_accuracy": 0.1726617157459259,
      "num_tokens": 34138078.0,
      "step": 19675
    },
    {
      "entropy": 5.532526254653931,
      "epoch": 1.5311807041431629,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047659409866833104,
      "loss": 5.208,
      "mean_token_accuracy": 0.17718277871608734,
      "num_tokens": 34146402.0,
      "step": 19680
    },
    {
      "entropy": 5.593402957916259,
      "epoch": 1.5315697335148803,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004765817884533014,
      "loss": 5.3406,
      "mean_token_accuracy": 0.17491628229618073,
      "num_tokens": 34155909.0,
      "step": 19685
    },
    {
      "entropy": 5.443002796173095,
      "epoch": 1.5319587628865978,
      "grad_norm": 2.859375,
      "learning_rate": 0.000476569475179589,
      "loss": 5.1955,
      "mean_token_accuracy": 0.17626997977495193,
      "num_tokens": 34164598.0,
      "step": 19690
    },
    {
      "entropy": 5.490495204925537,
      "epoch": 1.5323477922583155,
      "grad_norm": 1.578125,
      "learning_rate": 0.00047655715884738074,
      "loss": 5.2364,
      "mean_token_accuracy": 0.17988378256559373,
      "num_tokens": 34173476.0,
      "step": 19695
    },
    {
      "entropy": 5.51881685256958,
      "epoch": 1.5327368216300332,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004765448394568632,
      "loss": 5.2607,
      "mean_token_accuracy": 0.17252431511878968,
      "num_tokens": 34181986.0,
      "step": 19700
    },
    {
      "entropy": 5.490952825546264,
      "epoch": 1.5331258510017505,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004765325170082237,
      "loss": 5.2343,
      "mean_token_accuracy": 0.17885182052850723,
      "num_tokens": 34190477.0,
      "step": 19705
    },
    {
      "entropy": 5.420037317276001,
      "epoch": 1.5335148803734682,
      "grad_norm": 1.484375,
      "learning_rate": 0.000476520191501649,
      "loss": 5.1337,
      "mean_token_accuracy": 0.18653440922498704,
      "num_tokens": 34198955.0,
      "step": 19710
    },
    {
      "entropy": 5.56107873916626,
      "epoch": 1.5339039097451859,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047650786293732607,
      "loss": 5.3278,
      "mean_token_accuracy": 0.1705892026424408,
      "num_tokens": 34207744.0,
      "step": 19715
    },
    {
      "entropy": 5.500405502319336,
      "epoch": 1.5342929391169033,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004764955313154421,
      "loss": 5.2267,
      "mean_token_accuracy": 0.1827854871749878,
      "num_tokens": 34215935.0,
      "step": 19720
    },
    {
      "entropy": 5.500259351730347,
      "epoch": 1.5346819684886208,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047648319663618415,
      "loss": 5.2572,
      "mean_token_accuracy": 0.17511722147464753,
      "num_tokens": 34224990.0,
      "step": 19725
    },
    {
      "entropy": 5.490764236450195,
      "epoch": 1.5350709978603385,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047647085889973936,
      "loss": 5.1639,
      "mean_token_accuracy": 0.17535528540611267,
      "num_tokens": 34232928.0,
      "step": 19730
    },
    {
      "entropy": 5.4304577827453615,
      "epoch": 1.535460027232056,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047645851810629503,
      "loss": 5.1248,
      "mean_token_accuracy": 0.18983405232429504,
      "num_tokens": 34241434.0,
      "step": 19735
    },
    {
      "entropy": 5.448640394210815,
      "epoch": 1.5358490566037735,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047644617425603825,
      "loss": 5.1878,
      "mean_token_accuracy": 0.18371172100305558,
      "num_tokens": 34250067.0,
      "step": 19740
    },
    {
      "entropy": 5.5160462856292725,
      "epoch": 1.5362380859754912,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004764338273491565,
      "loss": 5.2585,
      "mean_token_accuracy": 0.17849910706281663,
      "num_tokens": 34258656.0,
      "step": 19745
    },
    {
      "entropy": 5.497633218765259,
      "epoch": 1.5366271153472089,
      "grad_norm": 1.375,
      "learning_rate": 0.00047642147738583695,
      "loss": 5.3363,
      "mean_token_accuracy": 0.1677684485912323,
      "num_tokens": 34267417.0,
      "step": 19750
    },
    {
      "entropy": 5.501345634460449,
      "epoch": 1.5370161447189261,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004764091243662671,
      "loss": 5.1136,
      "mean_token_accuracy": 0.18635567128658295,
      "num_tokens": 34275334.0,
      "step": 19755
    },
    {
      "entropy": 5.474584245681763,
      "epoch": 1.5374051740906438,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047639676829063437,
      "loss": 5.267,
      "mean_token_accuracy": 0.17711216062307358,
      "num_tokens": 34284016.0,
      "step": 19760
    },
    {
      "entropy": 5.422528028488159,
      "epoch": 1.5377942034623615,
      "grad_norm": 1.640625,
      "learning_rate": 0.00047638440915912623,
      "loss": 5.1803,
      "mean_token_accuracy": 0.18633621633052827,
      "num_tokens": 34292747.0,
      "step": 19765
    },
    {
      "entropy": 5.4854660987854,
      "epoch": 1.538183232834079,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004763720469719303,
      "loss": 5.1284,
      "mean_token_accuracy": 0.18211973756551741,
      "num_tokens": 34301764.0,
      "step": 19770
    },
    {
      "entropy": 5.45488920211792,
      "epoch": 1.5385722622057965,
      "grad_norm": 1.5703125,
      "learning_rate": 0.000476359681729234,
      "loss": 5.2554,
      "mean_token_accuracy": 0.180397367477417,
      "num_tokens": 34310522.0,
      "step": 19775
    },
    {
      "entropy": 5.5550501346588135,
      "epoch": 1.5389612915775142,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000476347313431225,
      "loss": 5.2454,
      "mean_token_accuracy": 0.1829739987850189,
      "num_tokens": 34319342.0,
      "step": 19780
    },
    {
      "entropy": 5.654476547241211,
      "epoch": 1.5393503209492316,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00047633494207809096,
      "loss": 5.349,
      "mean_token_accuracy": 0.17210122644901277,
      "num_tokens": 34327468.0,
      "step": 19785
    },
    {
      "entropy": 5.552998638153076,
      "epoch": 1.539739350320949,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047632256767001977,
      "loss": 5.3085,
      "mean_token_accuracy": 0.17831750959157944,
      "num_tokens": 34336484.0,
      "step": 19790
    },
    {
      "entropy": 5.433817720413208,
      "epoch": 1.5401283796926668,
      "grad_norm": 1.390625,
      "learning_rate": 0.000476310190207199,
      "loss": 5.121,
      "mean_token_accuracy": 0.18866233229637147,
      "num_tokens": 34344708.0,
      "step": 19795
    },
    {
      "entropy": 5.497208499908448,
      "epoch": 1.5405174090643845,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00047629780968981653,
      "loss": 5.1789,
      "mean_token_accuracy": 0.17989312559366227,
      "num_tokens": 34352757.0,
      "step": 19800
    },
    {
      "entropy": 5.498340320587158,
      "epoch": 1.540906438436102,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00047628542611806007,
      "loss": 5.1837,
      "mean_token_accuracy": 0.18371548503637314,
      "num_tokens": 34362070.0,
      "step": 19805
    },
    {
      "entropy": 5.472554445266724,
      "epoch": 1.5412954678078195,
      "grad_norm": 1.84375,
      "learning_rate": 0.00047627303949211773,
      "loss": 5.2068,
      "mean_token_accuracy": 0.17690205425024033,
      "num_tokens": 34370136.0,
      "step": 19810
    },
    {
      "entropy": 5.441941928863526,
      "epoch": 1.5416844971795371,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004762606498121774,
      "loss": 5.1586,
      "mean_token_accuracy": 0.18312578350305558,
      "num_tokens": 34378534.0,
      "step": 19815
    },
    {
      "entropy": 5.4551536560058596,
      "epoch": 1.5420735265512546,
      "grad_norm": 1.390625,
      "learning_rate": 0.000476248257078427,
      "loss": 5.155,
      "mean_token_accuracy": 0.18011748045682907,
      "num_tokens": 34386313.0,
      "step": 19820
    },
    {
      "entropy": 5.466526460647583,
      "epoch": 1.542462555922972,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004762358612910547,
      "loss": 5.2162,
      "mean_token_accuracy": 0.18401468992233277,
      "num_tokens": 34394566.0,
      "step": 19825
    },
    {
      "entropy": 5.451198577880859,
      "epoch": 1.5428515852946898,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004762234624502484,
      "loss": 5.1038,
      "mean_token_accuracy": 0.18378186076879502,
      "num_tokens": 34403177.0,
      "step": 19830
    },
    {
      "entropy": 5.5477441310882565,
      "epoch": 1.5432406146664073,
      "grad_norm": 1.75,
      "learning_rate": 0.00047621106055619644,
      "loss": 5.2504,
      "mean_token_accuracy": 0.18197994977235793,
      "num_tokens": 34411577.0,
      "step": 19835
    },
    {
      "entropy": 5.44796199798584,
      "epoch": 1.5436296440381247,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047619865560908687,
      "loss": 5.2403,
      "mean_token_accuracy": 0.17166987657546998,
      "num_tokens": 34420649.0,
      "step": 19840
    },
    {
      "entropy": 5.490853643417358,
      "epoch": 1.5440186734098424,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004761862476091079,
      "loss": 5.1491,
      "mean_token_accuracy": 0.1853147879242897,
      "num_tokens": 34429778.0,
      "step": 19845
    },
    {
      "entropy": 5.536733770370484,
      "epoch": 1.5444077027815601,
      "grad_norm": 2.0,
      "learning_rate": 0.00047617383655644785,
      "loss": 5.2474,
      "mean_token_accuracy": 0.17991104274988173,
      "num_tokens": 34438412.0,
      "step": 19850
    },
    {
      "entropy": 5.500594711303711,
      "epoch": 1.5447967321532776,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004761614224512951,
      "loss": 5.2341,
      "mean_token_accuracy": 0.17307254076004028,
      "num_tokens": 34447022.0,
      "step": 19855
    },
    {
      "entropy": 5.505466651916504,
      "epoch": 1.545185761524995,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004761490052938379,
      "loss": 5.2752,
      "mean_token_accuracy": 0.17882970869541168,
      "num_tokens": 34455502.0,
      "step": 19860
    },
    {
      "entropy": 5.46995792388916,
      "epoch": 1.5455747908967128,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004761365850842646,
      "loss": 5.2091,
      "mean_token_accuracy": 0.17881093323230743,
      "num_tokens": 34463682.0,
      "step": 19865
    },
    {
      "entropy": 5.4828651428222654,
      "epoch": 1.5459638202684303,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004761241618227639,
      "loss": 5.1368,
      "mean_token_accuracy": 0.18842589110136032,
      "num_tokens": 34471678.0,
      "step": 19870
    },
    {
      "entropy": 5.447758388519287,
      "epoch": 1.5463528496401477,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00047611173550952414,
      "loss": 5.1516,
      "mean_token_accuracy": 0.18642189353704453,
      "num_tokens": 34479830.0,
      "step": 19875
    },
    {
      "entropy": 5.490526628494263,
      "epoch": 1.5467418790118654,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047609930614473387,
      "loss": 5.2329,
      "mean_token_accuracy": 0.1749920517206192,
      "num_tokens": 34487445.0,
      "step": 19880
    },
    {
      "entropy": 5.439973926544189,
      "epoch": 1.547130908383583,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047608687372858175,
      "loss": 5.1097,
      "mean_token_accuracy": 0.18481873273849486,
      "num_tokens": 34496306.0,
      "step": 19885
    },
    {
      "entropy": 5.4423408031463625,
      "epoch": 1.5475199377553004,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047607443826125633,
      "loss": 5.223,
      "mean_token_accuracy": 0.18217843919992446,
      "num_tokens": 34505067.0,
      "step": 19890
    },
    {
      "entropy": 5.577670431137085,
      "epoch": 1.547908967127018,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00047606199974294637,
      "loss": 5.3308,
      "mean_token_accuracy": 0.17201985120773317,
      "num_tokens": 34513947.0,
      "step": 19895
    },
    {
      "entropy": 5.595619869232178,
      "epoch": 1.5482979964987358,
      "grad_norm": 1.75,
      "learning_rate": 0.0004760495581738406,
      "loss": 5.3579,
      "mean_token_accuracy": 0.1698410466313362,
      "num_tokens": 34523383.0,
      "step": 19900
    },
    {
      "entropy": 5.517604017257691,
      "epoch": 1.5486870258704533,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004760371135541278,
      "loss": 5.1609,
      "mean_token_accuracy": 0.1815445527434349,
      "num_tokens": 34532376.0,
      "step": 19905
    },
    {
      "entropy": 5.633794975280762,
      "epoch": 1.5490760552421707,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004760246658839967,
      "loss": 5.3244,
      "mean_token_accuracy": 0.1725177586078644,
      "num_tokens": 34541075.0,
      "step": 19910
    },
    {
      "entropy": 5.488174247741699,
      "epoch": 1.5494650846138884,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004760122151636364,
      "loss": 5.1303,
      "mean_token_accuracy": 0.18467805832624434,
      "num_tokens": 34550296.0,
      "step": 19915
    },
    {
      "entropy": 5.496546077728271,
      "epoch": 1.549854113985606,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004759997613932356,
      "loss": 5.2673,
      "mean_token_accuracy": 0.17745690494775773,
      "num_tokens": 34559185.0,
      "step": 19920
    },
    {
      "entropy": 5.591800928115845,
      "epoch": 1.5502431433573234,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047598730457298335,
      "loss": 5.2671,
      "mean_token_accuracy": 0.17782671451568605,
      "num_tokens": 34568018.0,
      "step": 19925
    },
    {
      "entropy": 5.539928197860718,
      "epoch": 1.550632172729041,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047597484470306866,
      "loss": 5.2757,
      "mean_token_accuracy": 0.18031223565340043,
      "num_tokens": 34576835.0,
      "step": 19930
    },
    {
      "entropy": 5.514178514480591,
      "epoch": 1.5510212021007586,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004759623817836806,
      "loss": 5.2019,
      "mean_token_accuracy": 0.18242556899785994,
      "num_tokens": 34585426.0,
      "step": 19935
    },
    {
      "entropy": 5.47841248512268,
      "epoch": 1.551410231472476,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004759499158150082,
      "loss": 5.2561,
      "mean_token_accuracy": 0.17484237104654313,
      "num_tokens": 34594428.0,
      "step": 19940
    },
    {
      "entropy": 5.550037002563476,
      "epoch": 1.5517992608441937,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00047593744679724076,
      "loss": 5.2841,
      "mean_token_accuracy": 0.1782570242881775,
      "num_tokens": 34603736.0,
      "step": 19945
    },
    {
      "entropy": 5.572537994384765,
      "epoch": 1.5521882902159114,
      "grad_norm": 1.484375,
      "learning_rate": 0.00047592497473056733,
      "loss": 5.2563,
      "mean_token_accuracy": 0.17629345506429672,
      "num_tokens": 34613295.0,
      "step": 19950
    },
    {
      "entropy": 5.442608213424682,
      "epoch": 1.552577319587629,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047591249961517724,
      "loss": 5.2058,
      "mean_token_accuracy": 0.1787840574979782,
      "num_tokens": 34622955.0,
      "step": 19955
    },
    {
      "entropy": 5.43725905418396,
      "epoch": 1.5529663489593464,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004759000214512598,
      "loss": 5.209,
      "mean_token_accuracy": 0.1802055209875107,
      "num_tokens": 34631866.0,
      "step": 19960
    },
    {
      "entropy": 5.546178340911865,
      "epoch": 1.553355378331064,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004758875402390042,
      "loss": 5.2882,
      "mean_token_accuracy": 0.17883485406637192,
      "num_tokens": 34640281.0,
      "step": 19965
    },
    {
      "entropy": 5.563056135177613,
      "epoch": 1.5537444077027815,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00047587505597859996,
      "loss": 5.2577,
      "mean_token_accuracy": 0.1743539720773697,
      "num_tokens": 34649080.0,
      "step": 19970
    },
    {
      "entropy": 5.471347713470459,
      "epoch": 1.554133437074499,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00047586256867023646,
      "loss": 5.2033,
      "mean_token_accuracy": 0.182217738032341,
      "num_tokens": 34657969.0,
      "step": 19975
    },
    {
      "entropy": 5.536583137512207,
      "epoch": 1.5545224664462167,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004758500783141032,
      "loss": 5.3468,
      "mean_token_accuracy": 0.17523048669099808,
      "num_tokens": 34667187.0,
      "step": 19980
    },
    {
      "entropy": 5.51764669418335,
      "epoch": 1.5549114958179344,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004758375849103897,
      "loss": 5.1836,
      "mean_token_accuracy": 0.17960923463106154,
      "num_tokens": 34676292.0,
      "step": 19985
    },
    {
      "entropy": 5.548939990997314,
      "epoch": 1.5553005251896517,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004758250884592855,
      "loss": 5.2907,
      "mean_token_accuracy": 0.17781132757663726,
      "num_tokens": 34685198.0,
      "step": 19990
    },
    {
      "entropy": 5.479723691940308,
      "epoch": 1.5556895545613694,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047581258896098024,
      "loss": 5.109,
      "mean_token_accuracy": 0.19083650261163712,
      "num_tokens": 34694019.0,
      "step": 19995
    },
    {
      "entropy": 5.526100778579712,
      "epoch": 1.556078583933087,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004758000864156636,
      "loss": 5.2558,
      "mean_token_accuracy": 0.17508168518543243,
      "num_tokens": 34703183.0,
      "step": 20000
    },
    {
      "entropy": 5.500713443756103,
      "epoch": 1.5564676133048045,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047578758082352527,
      "loss": 5.2231,
      "mean_token_accuracy": 0.175339911878109,
      "num_tokens": 34711049.0,
      "step": 20005
    },
    {
      "entropy": 5.505970239639282,
      "epoch": 1.556856642676522,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047577507218475487,
      "loss": 5.3026,
      "mean_token_accuracy": 0.1678979679942131,
      "num_tokens": 34720130.0,
      "step": 20010
    },
    {
      "entropy": 5.581413412094117,
      "epoch": 1.5572456720482397,
      "grad_norm": 1.375,
      "learning_rate": 0.00047576256049954236,
      "loss": 5.2633,
      "mean_token_accuracy": 0.18482479006052016,
      "num_tokens": 34728791.0,
      "step": 20015
    },
    {
      "entropy": 5.416379833221436,
      "epoch": 1.5576347014199572,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004757500457680776,
      "loss": 5.047,
      "mean_token_accuracy": 0.18813843578100203,
      "num_tokens": 34737470.0,
      "step": 20020
    },
    {
      "entropy": 5.446556568145752,
      "epoch": 1.5580237307916747,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004757375279905504,
      "loss": 5.2406,
      "mean_token_accuracy": 0.17581817060708999,
      "num_tokens": 34745774.0,
      "step": 20025
    },
    {
      "entropy": 5.398502254486084,
      "epoch": 1.5584127601633924,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047572500716715075,
      "loss": 5.1243,
      "mean_token_accuracy": 0.1879748821258545,
      "num_tokens": 34753649.0,
      "step": 20030
    },
    {
      "entropy": 5.539699745178223,
      "epoch": 1.55880178953511,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047571248329806855,
      "loss": 5.2084,
      "mean_token_accuracy": 0.18214959949254989,
      "num_tokens": 34761833.0,
      "step": 20035
    },
    {
      "entropy": 5.467098426818848,
      "epoch": 1.5591908189068273,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047569995638349383,
      "loss": 5.1837,
      "mean_token_accuracy": 0.17986234724521638,
      "num_tokens": 34770969.0,
      "step": 20040
    },
    {
      "entropy": 5.504928064346314,
      "epoch": 1.559579848278545,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004756874264236168,
      "loss": 5.2555,
      "mean_token_accuracy": 0.1732724726200104,
      "num_tokens": 34779860.0,
      "step": 20045
    },
    {
      "entropy": 5.512750959396362,
      "epoch": 1.5599688776502627,
      "grad_norm": 1.53125,
      "learning_rate": 0.00047567489341862753,
      "loss": 5.1722,
      "mean_token_accuracy": 0.18275004476308823,
      "num_tokens": 34788816.0,
      "step": 20050
    },
    {
      "entropy": 5.417970657348633,
      "epoch": 1.5603579070219802,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047566235736871607,
      "loss": 5.2677,
      "mean_token_accuracy": 0.17720874398946762,
      "num_tokens": 34797784.0,
      "step": 20055
    },
    {
      "entropy": 5.46078052520752,
      "epoch": 1.5607469363936977,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00047564981827407277,
      "loss": 5.1076,
      "mean_token_accuracy": 0.19123952388763427,
      "num_tokens": 34806553.0,
      "step": 20060
    },
    {
      "entropy": 5.473185634613037,
      "epoch": 1.5611359657654154,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047563727613488785,
      "loss": 5.1359,
      "mean_token_accuracy": 0.1809887170791626,
      "num_tokens": 34814854.0,
      "step": 20065
    },
    {
      "entropy": 5.4507176876068115,
      "epoch": 1.5615249951371328,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047562473095135154,
      "loss": 5.317,
      "mean_token_accuracy": 0.1838538318872452,
      "num_tokens": 34823147.0,
      "step": 20070
    },
    {
      "entropy": 5.517060995101929,
      "epoch": 1.5619140245088503,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004756121827236544,
      "loss": 5.1878,
      "mean_token_accuracy": 0.188933427631855,
      "num_tokens": 34831335.0,
      "step": 20075
    },
    {
      "entropy": 5.562550687789917,
      "epoch": 1.562303053880568,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004755996314519866,
      "loss": 5.2583,
      "mean_token_accuracy": 0.17847694605588912,
      "num_tokens": 34840118.0,
      "step": 20080
    },
    {
      "entropy": 5.479040861129761,
      "epoch": 1.5626920832522857,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004755870771365387,
      "loss": 5.2501,
      "mean_token_accuracy": 0.17598040252923966,
      "num_tokens": 34848851.0,
      "step": 20085
    },
    {
      "entropy": 5.474196910858154,
      "epoch": 1.563081112624003,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047557451977750113,
      "loss": 5.1292,
      "mean_token_accuracy": 0.18570237308740617,
      "num_tokens": 34857660.0,
      "step": 20090
    },
    {
      "entropy": 5.477176094055176,
      "epoch": 1.5634701419957207,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004755619593750645,
      "loss": 5.1808,
      "mean_token_accuracy": 0.18108201622962952,
      "num_tokens": 34866828.0,
      "step": 20095
    },
    {
      "entropy": 5.542474794387817,
      "epoch": 1.5638591713674383,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004755493959294194,
      "loss": 5.3232,
      "mean_token_accuracy": 0.1698934778571129,
      "num_tokens": 34875461.0,
      "step": 20100
    },
    {
      "entropy": 5.429262256622314,
      "epoch": 1.5642482007391558,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004755368294407564,
      "loss": 5.1674,
      "mean_token_accuracy": 0.17874481230974198,
      "num_tokens": 34883914.0,
      "step": 20105
    },
    {
      "entropy": 5.504143095016479,
      "epoch": 1.5646372301108733,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004755242599092662,
      "loss": 5.3103,
      "mean_token_accuracy": 0.1787399411201477,
      "num_tokens": 34892191.0,
      "step": 20110
    },
    {
      "entropy": 5.572123861312866,
      "epoch": 1.565026259482591,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047551168733513956,
      "loss": 5.2645,
      "mean_token_accuracy": 0.17563781291246414,
      "num_tokens": 34900450.0,
      "step": 20115
    },
    {
      "entropy": 5.533158302307129,
      "epoch": 1.5654152888543085,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047549911171856717,
      "loss": 5.1451,
      "mean_token_accuracy": 0.18031956851482392,
      "num_tokens": 34908706.0,
      "step": 20120
    },
    {
      "entropy": 5.489985609054566,
      "epoch": 1.565804318226026,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004754865330597398,
      "loss": 5.1831,
      "mean_token_accuracy": 0.18891116380691528,
      "num_tokens": 34916598.0,
      "step": 20125
    },
    {
      "entropy": 5.485782814025879,
      "epoch": 1.5661933475977436,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00047547395135884854,
      "loss": 5.2131,
      "mean_token_accuracy": 0.18263823240995408,
      "num_tokens": 34925850.0,
      "step": 20130
    },
    {
      "entropy": 5.5575401306152346,
      "epoch": 1.5665823769694613,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004754613666160841,
      "loss": 5.2962,
      "mean_token_accuracy": 0.17317381054162978,
      "num_tokens": 34934919.0,
      "step": 20135
    },
    {
      "entropy": 5.6056372165679935,
      "epoch": 1.5669714063411786,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047544877883163753,
      "loss": 5.3957,
      "mean_token_accuracy": 0.16409746706485748,
      "num_tokens": 34943875.0,
      "step": 20140
    },
    {
      "entropy": 5.497797155380249,
      "epoch": 1.5673604357128963,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047543618800569975,
      "loss": 5.1688,
      "mean_token_accuracy": 0.18242404907941817,
      "num_tokens": 34952458.0,
      "step": 20145
    },
    {
      "entropy": 5.5054436206817625,
      "epoch": 1.567749465084614,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047542359413846184,
      "loss": 5.2815,
      "mean_token_accuracy": 0.17552659511566163,
      "num_tokens": 34961550.0,
      "step": 20150
    },
    {
      "entropy": 5.490035438537598,
      "epoch": 1.5681384944563315,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004754109972301149,
      "loss": 5.1875,
      "mean_token_accuracy": 0.17088261991739273,
      "num_tokens": 34969561.0,
      "step": 20155
    },
    {
      "entropy": 5.533228731155395,
      "epoch": 1.568527523828049,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047539839728085007,
      "loss": 5.2692,
      "mean_token_accuracy": 0.1801791563630104,
      "num_tokens": 34977904.0,
      "step": 20160
    },
    {
      "entropy": 5.47243299484253,
      "epoch": 1.5689165531997666,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004753857942908585,
      "loss": 5.2963,
      "mean_token_accuracy": 0.17523210495710373,
      "num_tokens": 34986646.0,
      "step": 20165
    },
    {
      "entropy": 5.574240875244141,
      "epoch": 1.5693055825714841,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004753731882603315,
      "loss": 5.3001,
      "mean_token_accuracy": 0.17411949932575227,
      "num_tokens": 34995387.0,
      "step": 20170
    },
    {
      "entropy": 5.571469640731811,
      "epoch": 1.5696946119432016,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047536057918946026,
      "loss": 5.2341,
      "mean_token_accuracy": 0.18052912801504134,
      "num_tokens": 35004668.0,
      "step": 20175
    },
    {
      "entropy": 5.4023552417755125,
      "epoch": 1.5700836413149193,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004753479670784361,
      "loss": 5.1006,
      "mean_token_accuracy": 0.19072446525096892,
      "num_tokens": 35012881.0,
      "step": 20180
    },
    {
      "entropy": 5.545693445205688,
      "epoch": 1.570472670686637,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004753353519274505,
      "loss": 5.2295,
      "mean_token_accuracy": 0.17299910485744477,
      "num_tokens": 35022127.0,
      "step": 20185
    },
    {
      "entropy": 5.501242589950562,
      "epoch": 1.5708617000583545,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004753227337366948,
      "loss": 5.1233,
      "mean_token_accuracy": 0.18250587880611419,
      "num_tokens": 35030066.0,
      "step": 20190
    },
    {
      "entropy": 5.393864011764526,
      "epoch": 1.571250729430072,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047531011250636046,
      "loss": 5.1584,
      "mean_token_accuracy": 0.18586154878139496,
      "num_tokens": 35038560.0,
      "step": 20195
    },
    {
      "entropy": 5.496399593353272,
      "epoch": 1.5716397588017896,
      "grad_norm": 1.515625,
      "learning_rate": 0.00047529748823663896,
      "loss": 5.3378,
      "mean_token_accuracy": 0.1725176230072975,
      "num_tokens": 35048011.0,
      "step": 20200
    },
    {
      "entropy": 5.610437440872192,
      "epoch": 1.572028788173507,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004752848609277219,
      "loss": 5.2989,
      "mean_token_accuracy": 0.17597989439964296,
      "num_tokens": 35056247.0,
      "step": 20205
    },
    {
      "entropy": 5.532671070098877,
      "epoch": 1.5724178175452246,
      "grad_norm": 1.453125,
      "learning_rate": 0.00047527223057980086,
      "loss": 5.2149,
      "mean_token_accuracy": 0.17512428164482116,
      "num_tokens": 35064885.0,
      "step": 20210
    },
    {
      "entropy": 5.4634042263031,
      "epoch": 1.5728068469169423,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047525959719306736,
      "loss": 5.1533,
      "mean_token_accuracy": 0.1735348418354988,
      "num_tokens": 35073452.0,
      "step": 20215
    },
    {
      "entropy": 5.50248498916626,
      "epoch": 1.5731958762886598,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004752469607677134,
      "loss": 5.3041,
      "mean_token_accuracy": 0.16941092163324356,
      "num_tokens": 35082459.0,
      "step": 20220
    },
    {
      "entropy": 5.544084072113037,
      "epoch": 1.5735849056603772,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004752343213039305,
      "loss": 5.256,
      "mean_token_accuracy": 0.17132338881492615,
      "num_tokens": 35091336.0,
      "step": 20225
    },
    {
      "entropy": 5.502958154678344,
      "epoch": 1.573973935032095,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004752216788019104,
      "loss": 5.2031,
      "mean_token_accuracy": 0.1825808823108673,
      "num_tokens": 35099802.0,
      "step": 20230
    },
    {
      "entropy": 5.484231853485108,
      "epoch": 1.5743629644038126,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004752090332618451,
      "loss": 5.2221,
      "mean_token_accuracy": 0.17584270983934402,
      "num_tokens": 35108520.0,
      "step": 20235
    },
    {
      "entropy": 5.576858997344971,
      "epoch": 1.57475199377553,
      "grad_norm": 1.375,
      "learning_rate": 0.0004751963846839263,
      "loss": 5.2748,
      "mean_token_accuracy": 0.1817975491285324,
      "num_tokens": 35117481.0,
      "step": 20240
    },
    {
      "entropy": 5.492777347564697,
      "epoch": 1.5751410231472476,
      "grad_norm": 1.375,
      "learning_rate": 0.00047518373306834605,
      "loss": 5.2381,
      "mean_token_accuracy": 0.17642291337251664,
      "num_tokens": 35126605.0,
      "step": 20245
    },
    {
      "entropy": 5.503046464920044,
      "epoch": 1.5755300525189653,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047517107841529626,
      "loss": 5.1378,
      "mean_token_accuracy": 0.189200721681118,
      "num_tokens": 35135312.0,
      "step": 20250
    },
    {
      "entropy": 5.534327745437622,
      "epoch": 1.5759190818906827,
      "grad_norm": 1.640625,
      "learning_rate": 0.000475158420724969,
      "loss": 5.2003,
      "mean_token_accuracy": 0.17962053567171096,
      "num_tokens": 35144759.0,
      "step": 20255
    },
    {
      "entropy": 5.524717998504639,
      "epoch": 1.5763081112624002,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047514575999755627,
      "loss": 5.2953,
      "mean_token_accuracy": 0.17909365892410278,
      "num_tokens": 35153701.0,
      "step": 20260
    },
    {
      "entropy": 5.538240051269531,
      "epoch": 1.576697140634118,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047513309623325024,
      "loss": 5.3798,
      "mean_token_accuracy": 0.1722281128168106,
      "num_tokens": 35162503.0,
      "step": 20265
    },
    {
      "entropy": 5.459548711776733,
      "epoch": 1.5770861700058354,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004751204294322429,
      "loss": 5.2024,
      "mean_token_accuracy": 0.18295868188142778,
      "num_tokens": 35171209.0,
      "step": 20270
    },
    {
      "entropy": 5.464225101470947,
      "epoch": 1.5774751993775529,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047510775959472675,
      "loss": 5.203,
      "mean_token_accuracy": 0.17535534501075745,
      "num_tokens": 35179623.0,
      "step": 20275
    },
    {
      "entropy": 5.508400106430054,
      "epoch": 1.5778642287492706,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004750950867208937,
      "loss": 5.189,
      "mean_token_accuracy": 0.18121856600046157,
      "num_tokens": 35188679.0,
      "step": 20280
    },
    {
      "entropy": 5.464699602127075,
      "epoch": 1.5782532581209883,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004750824108109362,
      "loss": 5.1719,
      "mean_token_accuracy": 0.18520854711532592,
      "num_tokens": 35197325.0,
      "step": 20285
    },
    {
      "entropy": 5.620121860504151,
      "epoch": 1.5786422874927057,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004750697318650466,
      "loss": 5.3222,
      "mean_token_accuracy": 0.17438203543424607,
      "num_tokens": 35206741.0,
      "step": 20290
    },
    {
      "entropy": 5.571952676773071,
      "epoch": 1.5790313168644232,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004750570498834173,
      "loss": 5.2589,
      "mean_token_accuracy": 0.1712891712784767,
      "num_tokens": 35215346.0,
      "step": 20295
    },
    {
      "entropy": 5.484773921966553,
      "epoch": 1.579420346236141,
      "grad_norm": 1.375,
      "learning_rate": 0.0004750443648662407,
      "loss": 5.1765,
      "mean_token_accuracy": 0.1844707801938057,
      "num_tokens": 35223811.0,
      "step": 20300
    },
    {
      "entropy": 5.521778964996338,
      "epoch": 1.5798093756078584,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047503167681370924,
      "loss": 5.2587,
      "mean_token_accuracy": 0.17331053912639618,
      "num_tokens": 35232438.0,
      "step": 20305
    },
    {
      "entropy": 5.507816410064697,
      "epoch": 1.5801984049795759,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004750189857260154,
      "loss": 5.2342,
      "mean_token_accuracy": 0.16795391738414764,
      "num_tokens": 35242529.0,
      "step": 20310
    },
    {
      "entropy": 5.394672775268555,
      "epoch": 1.5805874343512936,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004750062916033519,
      "loss": 5.0468,
      "mean_token_accuracy": 0.18975515067577362,
      "num_tokens": 35250695.0,
      "step": 20315
    },
    {
      "entropy": 5.486641788482666,
      "epoch": 1.580976463723011,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004749935944459113,
      "loss": 5.1744,
      "mean_token_accuracy": 0.18354397267103195,
      "num_tokens": 35259212.0,
      "step": 20320
    },
    {
      "entropy": 5.459771680831909,
      "epoch": 1.5813654930947285,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004749808942538862,
      "loss": 5.1817,
      "mean_token_accuracy": 0.1863773837685585,
      "num_tokens": 35267667.0,
      "step": 20325
    },
    {
      "entropy": 5.490675592422486,
      "epoch": 1.5817545224664462,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004749681910274693,
      "loss": 5.2145,
      "mean_token_accuracy": 0.18251511007547377,
      "num_tokens": 35275850.0,
      "step": 20330
    },
    {
      "entropy": 5.604182004928589,
      "epoch": 1.582143551838164,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00047495548476685334,
      "loss": 5.2147,
      "mean_token_accuracy": 0.17405499815940856,
      "num_tokens": 35284635.0,
      "step": 20335
    },
    {
      "entropy": 5.5078987121582035,
      "epoch": 1.5825325812098814,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047494277547223125,
      "loss": 5.2323,
      "mean_token_accuracy": 0.18257189840078353,
      "num_tokens": 35292335.0,
      "step": 20340
    },
    {
      "entropy": 5.429032039642334,
      "epoch": 1.5829216105815989,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047493006314379573,
      "loss": 5.2113,
      "mean_token_accuracy": 0.18066998422145844,
      "num_tokens": 35300709.0,
      "step": 20345
    },
    {
      "entropy": 5.4936658382415775,
      "epoch": 1.5833106399533166,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047491734778173976,
      "loss": 5.1801,
      "mean_token_accuracy": 0.18381191343069075,
      "num_tokens": 35308956.0,
      "step": 20350
    },
    {
      "entropy": 5.424949693679809,
      "epoch": 1.583699669325034,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047490462938625617,
      "loss": 5.198,
      "mean_token_accuracy": 0.17705115526914597,
      "num_tokens": 35317663.0,
      "step": 20355
    },
    {
      "entropy": 5.470248889923096,
      "epoch": 1.5840886986967515,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047489190795753806,
      "loss": 5.2847,
      "mean_token_accuracy": 0.17230038046836854,
      "num_tokens": 35327412.0,
      "step": 20360
    },
    {
      "entropy": 5.6052069664001465,
      "epoch": 1.5844777280684692,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004748791834957784,
      "loss": 5.2948,
      "mean_token_accuracy": 0.17563155442476272,
      "num_tokens": 35335888.0,
      "step": 20365
    },
    {
      "entropy": 5.557424688339234,
      "epoch": 1.5848667574401867,
      "grad_norm": 2.515625,
      "learning_rate": 0.00047486645600117037,
      "loss": 5.2407,
      "mean_token_accuracy": 0.1836341843008995,
      "num_tokens": 35345122.0,
      "step": 20370
    },
    {
      "entropy": 5.537162637710571,
      "epoch": 1.5852557868119042,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004748537254739068,
      "loss": 5.3098,
      "mean_token_accuracy": 0.18009259998798371,
      "num_tokens": 35353751.0,
      "step": 20375
    },
    {
      "entropy": 5.5538407325744625,
      "epoch": 1.5856448161836219,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004748409919141812,
      "loss": 5.246,
      "mean_token_accuracy": 0.1785751089453697,
      "num_tokens": 35363083.0,
      "step": 20380
    },
    {
      "entropy": 5.448006725311279,
      "epoch": 1.5860338455553395,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004748282553221865,
      "loss": 5.1688,
      "mean_token_accuracy": 0.187142077088356,
      "num_tokens": 35371198.0,
      "step": 20385
    },
    {
      "entropy": 5.47127137184143,
      "epoch": 1.586422874927057,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004748155156981162,
      "loss": 5.2586,
      "mean_token_accuracy": 0.18322091847658156,
      "num_tokens": 35379633.0,
      "step": 20390
    },
    {
      "entropy": 5.5214379787445065,
      "epoch": 1.5868119042987745,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047480277304216346,
      "loss": 5.2137,
      "mean_token_accuracy": 0.1787571743130684,
      "num_tokens": 35388172.0,
      "step": 20395
    },
    {
      "entropy": 5.466855525970459,
      "epoch": 1.5872009336704922,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047479002735452164,
      "loss": 5.1879,
      "mean_token_accuracy": 0.17919579297304153,
      "num_tokens": 35397126.0,
      "step": 20400
    },
    {
      "entropy": 5.496103429794312,
      "epoch": 1.5875899630422097,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047477727863538415,
      "loss": 5.1711,
      "mean_token_accuracy": 0.18472271859645845,
      "num_tokens": 35405364.0,
      "step": 20405
    },
    {
      "entropy": 5.476044321060181,
      "epoch": 1.5879789924139271,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047476452688494444,
      "loss": 5.1923,
      "mean_token_accuracy": 0.18309639245271683,
      "num_tokens": 35414187.0,
      "step": 20410
    },
    {
      "entropy": 5.501715040206909,
      "epoch": 1.5883680217856448,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000474751772103396,
      "loss": 5.1853,
      "mean_token_accuracy": 0.18391240686178206,
      "num_tokens": 35422694.0,
      "step": 20415
    },
    {
      "entropy": 5.502434825897216,
      "epoch": 1.5887570511573625,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004747390142909323,
      "loss": 5.3234,
      "mean_token_accuracy": 0.1774757221341133,
      "num_tokens": 35431176.0,
      "step": 20420
    },
    {
      "entropy": 5.539558696746826,
      "epoch": 1.5891460805290798,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047472625344774713,
      "loss": 5.2502,
      "mean_token_accuracy": 0.17063300162553788,
      "num_tokens": 35439938.0,
      "step": 20425
    },
    {
      "entropy": 5.549200963973999,
      "epoch": 1.5895351099007975,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047471348957403374,
      "loss": 5.1734,
      "mean_token_accuracy": 0.1832476183772087,
      "num_tokens": 35447864.0,
      "step": 20430
    },
    {
      "entropy": 5.535675382614135,
      "epoch": 1.5899241392725152,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004747007226699862,
      "loss": 5.2277,
      "mean_token_accuracy": 0.18092171102762222,
      "num_tokens": 35456047.0,
      "step": 20435
    },
    {
      "entropy": 5.414860343933105,
      "epoch": 1.5903131686442327,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047468795273579803,
      "loss": 5.1655,
      "mean_token_accuracy": 0.185432069003582,
      "num_tokens": 35464510.0,
      "step": 20440
    },
    {
      "entropy": 5.539222860336304,
      "epoch": 1.5907021980159501,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004746751797716629,
      "loss": 5.287,
      "mean_token_accuracy": 0.1778273656964302,
      "num_tokens": 35472715.0,
      "step": 20445
    },
    {
      "entropy": 5.464086818695068,
      "epoch": 1.5910912273876678,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004746624037777748,
      "loss": 5.1588,
      "mean_token_accuracy": 0.1783091187477112,
      "num_tokens": 35482094.0,
      "step": 20450
    },
    {
      "entropy": 5.495008754730224,
      "epoch": 1.5914802567593853,
      "grad_norm": 1.6875,
      "learning_rate": 0.00047464962475432754,
      "loss": 5.2007,
      "mean_token_accuracy": 0.17638317197561265,
      "num_tokens": 35489864.0,
      "step": 20455
    },
    {
      "entropy": 5.524823331832886,
      "epoch": 1.5918692861311028,
      "grad_norm": 1.515625,
      "learning_rate": 0.000474636842701515,
      "loss": 5.2853,
      "mean_token_accuracy": 0.17576328068971633,
      "num_tokens": 35498840.0,
      "step": 20460
    },
    {
      "entropy": 5.476958084106445,
      "epoch": 1.5922583155028205,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004746240576195311,
      "loss": 5.1186,
      "mean_token_accuracy": 0.17764958292245864,
      "num_tokens": 35506954.0,
      "step": 20465
    },
    {
      "entropy": 5.490156459808349,
      "epoch": 1.5926473448745382,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047461126950856987,
      "loss": 5.2748,
      "mean_token_accuracy": 0.17365874946117402,
      "num_tokens": 35515796.0,
      "step": 20470
    },
    {
      "entropy": 5.501256084442138,
      "epoch": 1.5930363742462554,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004745984783688253,
      "loss": 5.1867,
      "mean_token_accuracy": 0.18192726820707322,
      "num_tokens": 35523713.0,
      "step": 20475
    },
    {
      "entropy": 5.509800577163697,
      "epoch": 1.5934254036179731,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047458568420049153,
      "loss": 5.3246,
      "mean_token_accuracy": 0.1736530378460884,
      "num_tokens": 35532800.0,
      "step": 20480
    },
    {
      "entropy": 5.5810600280761715,
      "epoch": 1.5938144329896908,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047457288700376274,
      "loss": 5.2308,
      "mean_token_accuracy": 0.18621969372034072,
      "num_tokens": 35540865.0,
      "step": 20485
    },
    {
      "entropy": 5.528258323669434,
      "epoch": 1.5942034623614083,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047456008677883304,
      "loss": 5.3214,
      "mean_token_accuracy": 0.17472659200429916,
      "num_tokens": 35549498.0,
      "step": 20490
    },
    {
      "entropy": 5.52451343536377,
      "epoch": 1.5945924917331258,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004745472835258966,
      "loss": 5.2187,
      "mean_token_accuracy": 0.18188000321388245,
      "num_tokens": 35557756.0,
      "step": 20495
    },
    {
      "entropy": 5.586997318267822,
      "epoch": 1.5949815211048435,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047453447724514773,
      "loss": 5.3117,
      "mean_token_accuracy": 0.17629093527793885,
      "num_tokens": 35566062.0,
      "step": 20500
    },
    {
      "entropy": 5.530290269851685,
      "epoch": 1.595370550476561,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004745216679367808,
      "loss": 5.1428,
      "mean_token_accuracy": 0.18013672083616256,
      "num_tokens": 35574377.0,
      "step": 20505
    },
    {
      "entropy": 5.52440710067749,
      "epoch": 1.5957595798482784,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004745088556009901,
      "loss": 5.2307,
      "mean_token_accuracy": 0.1787106916308403,
      "num_tokens": 35582254.0,
      "step": 20510
    },
    {
      "entropy": 5.534545516967773,
      "epoch": 1.5961486092199961,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004744960402379701,
      "loss": 5.2749,
      "mean_token_accuracy": 0.17506421804428102,
      "num_tokens": 35590804.0,
      "step": 20515
    },
    {
      "entropy": 5.520808601379395,
      "epoch": 1.5965376385917138,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00047448322184791525,
      "loss": 5.2238,
      "mean_token_accuracy": 0.1801039084792137,
      "num_tokens": 35599547.0,
      "step": 20520
    },
    {
      "entropy": 5.598384952545166,
      "epoch": 1.596926667963431,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00047447040043101994,
      "loss": 5.3329,
      "mean_token_accuracy": 0.17442476749420166,
      "num_tokens": 35608948.0,
      "step": 20525
    },
    {
      "entropy": 5.602685832977295,
      "epoch": 1.5973156973351488,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00047445757598747886,
      "loss": 5.4273,
      "mean_token_accuracy": 0.16351246759295462,
      "num_tokens": 35618243.0,
      "step": 20530
    },
    {
      "entropy": 5.503303718566895,
      "epoch": 1.5977047267068665,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004744447485174864,
      "loss": 5.2108,
      "mean_token_accuracy": 0.1850836083292961,
      "num_tokens": 35626938.0,
      "step": 20535
    },
    {
      "entropy": 5.566260528564453,
      "epoch": 1.598093756078584,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00047443191802123746,
      "loss": 5.2608,
      "mean_token_accuracy": 0.18745750486850737,
      "num_tokens": 35635612.0,
      "step": 20540
    },
    {
      "entropy": 5.497407865524292,
      "epoch": 1.5984827854503014,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047441908449892664,
      "loss": 5.2542,
      "mean_token_accuracy": 0.1845002830028534,
      "num_tokens": 35644320.0,
      "step": 20545
    },
    {
      "entropy": 5.451295042037964,
      "epoch": 1.5988718148220191,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047440624795074855,
      "loss": 5.1568,
      "mean_token_accuracy": 0.18606287091970444,
      "num_tokens": 35652606.0,
      "step": 20550
    },
    {
      "entropy": 5.475770568847656,
      "epoch": 1.5992608441937366,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047439340837689804,
      "loss": 5.1453,
      "mean_token_accuracy": 0.18382026851177216,
      "num_tokens": 35661082.0,
      "step": 20555
    },
    {
      "entropy": 5.48795371055603,
      "epoch": 1.599649873565454,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047438056577756986,
      "loss": 5.2585,
      "mean_token_accuracy": 0.17349821031093599,
      "num_tokens": 35669926.0,
      "step": 20560
    },
    {
      "entropy": 5.532926225662232,
      "epoch": 1.6000389029371718,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047436772015295903,
      "loss": 5.2502,
      "mean_token_accuracy": 0.17873613089323043,
      "num_tokens": 35678964.0,
      "step": 20565
    },
    {
      "entropy": 5.524367427825927,
      "epoch": 1.6004279323088895,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047435487150326036,
      "loss": 5.1992,
      "mean_token_accuracy": 0.1857219621539116,
      "num_tokens": 35687943.0,
      "step": 20570
    },
    {
      "entropy": 5.4822197437286375,
      "epoch": 1.6008169616806067,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004743420198286688,
      "loss": 5.2474,
      "mean_token_accuracy": 0.17616415917873382,
      "num_tokens": 35696684.0,
      "step": 20575
    },
    {
      "entropy": 5.466225242614746,
      "epoch": 1.6012059910523244,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047432916512937936,
      "loss": 5.2522,
      "mean_token_accuracy": 0.17492427974939345,
      "num_tokens": 35705038.0,
      "step": 20580
    },
    {
      "entropy": 5.586339712142944,
      "epoch": 1.6015950204240421,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004743163074055871,
      "loss": 5.2317,
      "mean_token_accuracy": 0.1801270604133606,
      "num_tokens": 35713638.0,
      "step": 20585
    },
    {
      "entropy": 5.591613149642944,
      "epoch": 1.6019840497957596,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004743034466574871,
      "loss": 5.2268,
      "mean_token_accuracy": 0.17935534566640854,
      "num_tokens": 35721935.0,
      "step": 20590
    },
    {
      "entropy": 5.51762900352478,
      "epoch": 1.602373079167477,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004742905828852746,
      "loss": 5.2819,
      "mean_token_accuracy": 0.17517272382974625,
      "num_tokens": 35731610.0,
      "step": 20595
    },
    {
      "entropy": 5.554374933242798,
      "epoch": 1.6027621085391948,
      "grad_norm": 1.375,
      "learning_rate": 0.0004742777160891447,
      "loss": 5.2825,
      "mean_token_accuracy": 0.17923903167247773,
      "num_tokens": 35740956.0,
      "step": 20600
    },
    {
      "entropy": 5.472058820724487,
      "epoch": 1.6031511379109122,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004742648462692926,
      "loss": 5.1309,
      "mean_token_accuracy": 0.18912305533885956,
      "num_tokens": 35749316.0,
      "step": 20605
    },
    {
      "entropy": 5.5658728122711185,
      "epoch": 1.6035401672826297,
      "grad_norm": 1.59375,
      "learning_rate": 0.00047425197342591363,
      "loss": 5.3211,
      "mean_token_accuracy": 0.18015774935483933,
      "num_tokens": 35758875.0,
      "step": 20610
    },
    {
      "entropy": 5.455133056640625,
      "epoch": 1.6039291966543474,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004742390975592031,
      "loss": 5.2631,
      "mean_token_accuracy": 0.1832687199115753,
      "num_tokens": 35767851.0,
      "step": 20615
    },
    {
      "entropy": 5.5684990882873535,
      "epoch": 1.604318226026065,
      "grad_norm": 1.5,
      "learning_rate": 0.00047422621866935645,
      "loss": 5.1664,
      "mean_token_accuracy": 0.18225348889827728,
      "num_tokens": 35776393.0,
      "step": 20620
    },
    {
      "entropy": 5.482385444641113,
      "epoch": 1.6047072553977826,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000474213336756569,
      "loss": 5.139,
      "mean_token_accuracy": 0.18780395090579988,
      "num_tokens": 35784868.0,
      "step": 20625
    },
    {
      "entropy": 5.474336242675781,
      "epoch": 1.6050962847695,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004742004518210362,
      "loss": 5.2337,
      "mean_token_accuracy": 0.1708284929394722,
      "num_tokens": 35793812.0,
      "step": 20630
    },
    {
      "entropy": 5.509834098815918,
      "epoch": 1.6054853141412178,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004741875638629536,
      "loss": 5.2303,
      "mean_token_accuracy": 0.17225415557622908,
      "num_tokens": 35802206.0,
      "step": 20635
    },
    {
      "entropy": 5.551030969619751,
      "epoch": 1.6058743435129352,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004741746728825168,
      "loss": 5.2035,
      "mean_token_accuracy": 0.17514467090368271,
      "num_tokens": 35811095.0,
      "step": 20640
    },
    {
      "entropy": 5.6419744968414305,
      "epoch": 1.6062633728846527,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004741617788799213,
      "loss": 5.3242,
      "mean_token_accuracy": 0.1752131998538971,
      "num_tokens": 35819799.0,
      "step": 20645
    },
    {
      "entropy": 5.452331590652466,
      "epoch": 1.6066524022563704,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00047414888185536285,
      "loss": 5.2251,
      "mean_token_accuracy": 0.17969018518924712,
      "num_tokens": 35829130.0,
      "step": 20650
    },
    {
      "entropy": 5.4639064311981205,
      "epoch": 1.6070414316280879,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004741359818090371,
      "loss": 5.1843,
      "mean_token_accuracy": 0.18253854364156724,
      "num_tokens": 35838020.0,
      "step": 20655
    },
    {
      "entropy": 5.4403825283050535,
      "epoch": 1.6074304609998054,
      "grad_norm": 2.34375,
      "learning_rate": 0.00047412307874113976,
      "loss": 5.1156,
      "mean_token_accuracy": 0.18635677099227904,
      "num_tokens": 35845843.0,
      "step": 20660
    },
    {
      "entropy": 5.446789121627807,
      "epoch": 1.607819490371523,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004741101726518667,
      "loss": 5.1243,
      "mean_token_accuracy": 0.18147239238023757,
      "num_tokens": 35854505.0,
      "step": 20665
    },
    {
      "entropy": 5.465287542343139,
      "epoch": 1.6082085197432407,
      "grad_norm": 1.375,
      "learning_rate": 0.0004740972635414136,
      "loss": 5.2151,
      "mean_token_accuracy": 0.18238385915756225,
      "num_tokens": 35862847.0,
      "step": 20670
    },
    {
      "entropy": 5.450656318664551,
      "epoch": 1.6085975491149582,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004740843514099765,
      "loss": 5.183,
      "mean_token_accuracy": 0.18003190457820892,
      "num_tokens": 35871471.0,
      "step": 20675
    },
    {
      "entropy": 5.546759843826294,
      "epoch": 1.6089865784866757,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004740714362577512,
      "loss": 5.2022,
      "mean_token_accuracy": 0.17458469718694686,
      "num_tokens": 35880266.0,
      "step": 20680
    },
    {
      "entropy": 5.507287645339966,
      "epoch": 1.6093756078583934,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047405851808493364,
      "loss": 5.1285,
      "mean_token_accuracy": 0.1825320303440094,
      "num_tokens": 35888727.0,
      "step": 20685
    },
    {
      "entropy": 5.449972677230835,
      "epoch": 1.6097646372301109,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047404559689172006,
      "loss": 5.215,
      "mean_token_accuracy": 0.18035564869642257,
      "num_tokens": 35897750.0,
      "step": 20690
    },
    {
      "entropy": 5.462679386138916,
      "epoch": 1.6101536666018283,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047403267267830617,
      "loss": 5.1597,
      "mean_token_accuracy": 0.18867727369070053,
      "num_tokens": 35906254.0,
      "step": 20695
    },
    {
      "entropy": 5.484301376342773,
      "epoch": 1.610542695973546,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047401974544488844,
      "loss": 5.2796,
      "mean_token_accuracy": 0.16822059601545333,
      "num_tokens": 35915180.0,
      "step": 20700
    },
    {
      "entropy": 5.592352342605591,
      "epoch": 1.6109317253452635,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004740068151916628,
      "loss": 5.2399,
      "mean_token_accuracy": 0.1708812490105629,
      "num_tokens": 35924336.0,
      "step": 20705
    },
    {
      "entropy": 5.547119665145874,
      "epoch": 1.611320754716981,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004739938819188255,
      "loss": 5.2514,
      "mean_token_accuracy": 0.17776671648025513,
      "num_tokens": 35933031.0,
      "step": 20710
    },
    {
      "entropy": 5.507070016860962,
      "epoch": 1.6117097840886987,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004739809456265727,
      "loss": 5.0965,
      "mean_token_accuracy": 0.19073860198259354,
      "num_tokens": 35941794.0,
      "step": 20715
    },
    {
      "entropy": 5.5489341735839846,
      "epoch": 1.6120988134604164,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004739680063151008,
      "loss": 5.3039,
      "mean_token_accuracy": 0.1768066868185997,
      "num_tokens": 35949636.0,
      "step": 20720
    },
    {
      "entropy": 5.509795761108398,
      "epoch": 1.6124878428321339,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004739550639846061,
      "loss": 5.3201,
      "mean_token_accuracy": 0.17850505858659743,
      "num_tokens": 35958842.0,
      "step": 20725
    },
    {
      "entropy": 5.488194561004638,
      "epoch": 1.6128768722038513,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00047394211863528496,
      "loss": 5.1651,
      "mean_token_accuracy": 0.17458401918411254,
      "num_tokens": 35967897.0,
      "step": 20730
    },
    {
      "entropy": 5.435737419128418,
      "epoch": 1.613265901575569,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004739291702673338,
      "loss": 5.0806,
      "mean_token_accuracy": 0.18676541596651078,
      "num_tokens": 35976309.0,
      "step": 20735
    },
    {
      "entropy": 5.58004412651062,
      "epoch": 1.6136549309472865,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004739162188809492,
      "loss": 5.3278,
      "mean_token_accuracy": 0.17659748047590257,
      "num_tokens": 35984401.0,
      "step": 20740
    },
    {
      "entropy": 5.485971975326538,
      "epoch": 1.614043960319004,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047390326447632747,
      "loss": 5.2191,
      "mean_token_accuracy": 0.1749339923262596,
      "num_tokens": 35993068.0,
      "step": 20745
    },
    {
      "entropy": 5.597433567047119,
      "epoch": 1.6144329896907217,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004738903070536654,
      "loss": 5.3715,
      "mean_token_accuracy": 0.17616495341062546,
      "num_tokens": 36001559.0,
      "step": 20750
    },
    {
      "entropy": 5.546025609970092,
      "epoch": 1.6148220190624392,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004738773466131594,
      "loss": 5.2857,
      "mean_token_accuracy": 0.1728214368224144,
      "num_tokens": 36010675.0,
      "step": 20755
    },
    {
      "entropy": 5.4294198513031,
      "epoch": 1.6152110484341566,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004738643831550063,
      "loss": 5.0305,
      "mean_token_accuracy": 0.19377739876508712,
      "num_tokens": 36019498.0,
      "step": 20760
    },
    {
      "entropy": 5.397676038742065,
      "epoch": 1.6156000778058743,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004738514166794026,
      "loss": 5.1076,
      "mean_token_accuracy": 0.19262148737907409,
      "num_tokens": 36028206.0,
      "step": 20765
    },
    {
      "entropy": 5.544890689849853,
      "epoch": 1.615989107177592,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047383844718654525,
      "loss": 5.3169,
      "mean_token_accuracy": 0.1798240751028061,
      "num_tokens": 36037341.0,
      "step": 20770
    },
    {
      "entropy": 5.5295360565185545,
      "epoch": 1.6163781365493095,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004738254746766309,
      "loss": 5.193,
      "mean_token_accuracy": 0.1816869631409645,
      "num_tokens": 36046596.0,
      "step": 20775
    },
    {
      "entropy": 5.507260847091675,
      "epoch": 1.616767165921027,
      "grad_norm": 1.375,
      "learning_rate": 0.0004738124991498565,
      "loss": 5.1667,
      "mean_token_accuracy": 0.18574288934469224,
      "num_tokens": 36055031.0,
      "step": 20780
    },
    {
      "entropy": 5.561546039581299,
      "epoch": 1.6171561952927447,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00047379952060641866,
      "loss": 5.3078,
      "mean_token_accuracy": 0.18006632775068282,
      "num_tokens": 36064049.0,
      "step": 20785
    },
    {
      "entropy": 5.5834431648254395,
      "epoch": 1.6175452246644622,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047378653904651475,
      "loss": 5.2795,
      "mean_token_accuracy": 0.17170560359954834,
      "num_tokens": 36072886.0,
      "step": 20790
    },
    {
      "entropy": 5.545154571533203,
      "epoch": 1.6179342540361796,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00047377355447034136,
      "loss": 5.28,
      "mean_token_accuracy": 0.18235398083925247,
      "num_tokens": 36082133.0,
      "step": 20795
    },
    {
      "entropy": 5.5143474578857425,
      "epoch": 1.6183232834078973,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004737605668780958,
      "loss": 5.1922,
      "mean_token_accuracy": 0.182487191259861,
      "num_tokens": 36090980.0,
      "step": 20800
    },
    {
      "entropy": 5.5199206352233885,
      "epoch": 1.618712312779615,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047374757626997494,
      "loss": 5.2586,
      "mean_token_accuracy": 0.17965995520353317,
      "num_tokens": 36099584.0,
      "step": 20805
    },
    {
      "entropy": 5.484288454055786,
      "epoch": 1.6191013421513323,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004737345826461759,
      "loss": 5.2013,
      "mean_token_accuracy": 0.17327310889959335,
      "num_tokens": 36108052.0,
      "step": 20810
    },
    {
      "entropy": 5.492011308670044,
      "epoch": 1.61949037152305,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004737215860068959,
      "loss": 5.0859,
      "mean_token_accuracy": 0.1897791400551796,
      "num_tokens": 36116623.0,
      "step": 20815
    },
    {
      "entropy": 5.523939800262451,
      "epoch": 1.6198794008947677,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047370858635233223,
      "loss": 5.2208,
      "mean_token_accuracy": 0.1824929416179657,
      "num_tokens": 36124953.0,
      "step": 20820
    },
    {
      "entropy": 5.4983409404754635,
      "epoch": 1.6202684302664851,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047369558368268194,
      "loss": 5.2047,
      "mean_token_accuracy": 0.17592939585447312,
      "num_tokens": 36133500.0,
      "step": 20825
    },
    {
      "entropy": 5.473618841171264,
      "epoch": 1.6206574596382026,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004736825779981424,
      "loss": 5.175,
      "mean_token_accuracy": 0.18104309290647508,
      "num_tokens": 36141934.0,
      "step": 20830
    },
    {
      "entropy": 5.456574249267578,
      "epoch": 1.6210464890099203,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004736695692989111,
      "loss": 5.1853,
      "mean_token_accuracy": 0.1817682757973671,
      "num_tokens": 36150592.0,
      "step": 20835
    },
    {
      "entropy": 5.473036098480224,
      "epoch": 1.6214355183816378,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004736565575851852,
      "loss": 5.0635,
      "mean_token_accuracy": 0.1875779777765274,
      "num_tokens": 36159377.0,
      "step": 20840
    },
    {
      "entropy": 5.500918817520142,
      "epoch": 1.6218245477533553,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004736435428571622,
      "loss": 5.1734,
      "mean_token_accuracy": 0.1775798186659813,
      "num_tokens": 36168159.0,
      "step": 20845
    },
    {
      "entropy": 5.38011531829834,
      "epoch": 1.622213577125073,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004736305251150397,
      "loss": 5.1387,
      "mean_token_accuracy": 0.1821829557418823,
      "num_tokens": 36176108.0,
      "step": 20850
    },
    {
      "entropy": 5.485075855255127,
      "epoch": 1.6226026064967907,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004736175043590151,
      "loss": 5.3189,
      "mean_token_accuracy": 0.17256989181041718,
      "num_tokens": 36184930.0,
      "step": 20855
    },
    {
      "entropy": 5.582487964630127,
      "epoch": 1.622991635868508,
      "grad_norm": 1.46875,
      "learning_rate": 0.000473604480589286,
      "loss": 5.241,
      "mean_token_accuracy": 0.1799950748682022,
      "num_tokens": 36193308.0,
      "step": 20860
    },
    {
      "entropy": 5.502960252761841,
      "epoch": 1.6233806652402256,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047359145380605007,
      "loss": 5.1849,
      "mean_token_accuracy": 0.18289902210235595,
      "num_tokens": 36202016.0,
      "step": 20865
    },
    {
      "entropy": 5.553669023513794,
      "epoch": 1.6237696946119433,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004735784240095049,
      "loss": 5.1828,
      "mean_token_accuracy": 0.1815961256623268,
      "num_tokens": 36210757.0,
      "step": 20870
    },
    {
      "entropy": 5.4530699253082275,
      "epoch": 1.6241587239836608,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047356539119984813,
      "loss": 5.1234,
      "mean_token_accuracy": 0.18398251086473466,
      "num_tokens": 36219052.0,
      "step": 20875
    },
    {
      "entropy": 5.406506204605103,
      "epoch": 1.6245477533553783,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004735523553772777,
      "loss": 5.1114,
      "mean_token_accuracy": 0.18620857000350952,
      "num_tokens": 36227890.0,
      "step": 20880
    },
    {
      "entropy": 5.4753025531768795,
      "epoch": 1.624936782727096,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004735393165419912,
      "loss": 5.2433,
      "mean_token_accuracy": 0.17671396881341933,
      "num_tokens": 36236725.0,
      "step": 20885
    },
    {
      "entropy": 5.585158157348633,
      "epoch": 1.6253258120988134,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047352627469418666,
      "loss": 5.2346,
      "mean_token_accuracy": 0.17834849357604982,
      "num_tokens": 36246220.0,
      "step": 20890
    },
    {
      "entropy": 5.558758211135864,
      "epoch": 1.625714841470531,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004735132298340619,
      "loss": 5.2747,
      "mean_token_accuracy": 0.17417685985565184,
      "num_tokens": 36254662.0,
      "step": 20895
    },
    {
      "entropy": 5.476184272766114,
      "epoch": 1.6261038708422486,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004735001819618148,
      "loss": 5.2274,
      "mean_token_accuracy": 0.1811300739645958,
      "num_tokens": 36262446.0,
      "step": 20900
    },
    {
      "entropy": 5.5535718441009525,
      "epoch": 1.6264929002139663,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004734871310776434,
      "loss": 5.222,
      "mean_token_accuracy": 0.1802380919456482,
      "num_tokens": 36271284.0,
      "step": 20905
    },
    {
      "entropy": 5.469676160812378,
      "epoch": 1.6268819295856836,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004734740771817457,
      "loss": 5.0896,
      "mean_token_accuracy": 0.19211526364088058,
      "num_tokens": 36279180.0,
      "step": 20910
    },
    {
      "entropy": 5.489728689193726,
      "epoch": 1.6272709589574013,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004734610202743198,
      "loss": 5.2406,
      "mean_token_accuracy": 0.1792478233575821,
      "num_tokens": 36287714.0,
      "step": 20915
    },
    {
      "entropy": 5.504299211502075,
      "epoch": 1.627659988329119,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004734479603555638,
      "loss": 5.2513,
      "mean_token_accuracy": 0.17651249468326569,
      "num_tokens": 36297120.0,
      "step": 20920
    },
    {
      "entropy": 5.534188652038575,
      "epoch": 1.6280490177008364,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047343489742567593,
      "loss": 5.2398,
      "mean_token_accuracy": 0.17953075468540192,
      "num_tokens": 36305232.0,
      "step": 20925
    },
    {
      "entropy": 5.5675560474395756,
      "epoch": 1.628438047072554,
      "grad_norm": 1.375,
      "learning_rate": 0.00047342183148485424,
      "loss": 5.273,
      "mean_token_accuracy": 0.18312925547361375,
      "num_tokens": 36313587.0,
      "step": 20930
    },
    {
      "entropy": 5.484789562225342,
      "epoch": 1.6288270764442716,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047340876253329706,
      "loss": 5.2047,
      "mean_token_accuracy": 0.18920795172452926,
      "num_tokens": 36322054.0,
      "step": 20935
    },
    {
      "entropy": 5.509126329421997,
      "epoch": 1.629216105815989,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047339569057120275,
      "loss": 5.127,
      "mean_token_accuracy": 0.18547769635915756,
      "num_tokens": 36330226.0,
      "step": 20940
    },
    {
      "entropy": 5.44800295829773,
      "epoch": 1.6296051351877066,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00047338261559876966,
      "loss": 5.1991,
      "mean_token_accuracy": 0.173978354036808,
      "num_tokens": 36338436.0,
      "step": 20945
    },
    {
      "entropy": 5.402836322784424,
      "epoch": 1.6299941645594243,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047336953761619604,
      "loss": 5.169,
      "mean_token_accuracy": 0.17679326236248016,
      "num_tokens": 36347204.0,
      "step": 20950
    },
    {
      "entropy": 5.414460039138794,
      "epoch": 1.630383193931142,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047335645662368046,
      "loss": 5.1761,
      "mean_token_accuracy": 0.1849618673324585,
      "num_tokens": 36356225.0,
      "step": 20955
    },
    {
      "entropy": 5.484632778167724,
      "epoch": 1.6307722233028592,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004733433726214214,
      "loss": 5.2419,
      "mean_token_accuracy": 0.17961994856595992,
      "num_tokens": 36364526.0,
      "step": 20960
    },
    {
      "entropy": 5.586671209335327,
      "epoch": 1.631161252674577,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047333028560961733,
      "loss": 5.3042,
      "mean_token_accuracy": 0.17767173647880555,
      "num_tokens": 36373162.0,
      "step": 20965
    },
    {
      "entropy": 5.55222373008728,
      "epoch": 1.6315502820462946,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047331719558846687,
      "loss": 5.2133,
      "mean_token_accuracy": 0.17999052703380586,
      "num_tokens": 36381545.0,
      "step": 20970
    },
    {
      "entropy": 5.478394937515259,
      "epoch": 1.631939311418012,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047330410255816863,
      "loss": 5.179,
      "mean_token_accuracy": 0.18163208067417144,
      "num_tokens": 36390267.0,
      "step": 20975
    },
    {
      "entropy": 5.512081956863403,
      "epoch": 1.6323283407897295,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047329100651892123,
      "loss": 5.2118,
      "mean_token_accuracy": 0.18402910530567168,
      "num_tokens": 36398910.0,
      "step": 20980
    },
    {
      "entropy": 5.502491140365601,
      "epoch": 1.6327173701614472,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004732779074709234,
      "loss": 5.1919,
      "mean_token_accuracy": 0.18053594082593918,
      "num_tokens": 36407382.0,
      "step": 20985
    },
    {
      "entropy": 5.452455234527588,
      "epoch": 1.6331063995331647,
      "grad_norm": 1.34375,
      "learning_rate": 0.000473264805414374,
      "loss": 5.1769,
      "mean_token_accuracy": 0.18799318224191666,
      "num_tokens": 36415734.0,
      "step": 20990
    },
    {
      "entropy": 5.504100704193116,
      "epoch": 1.6334954289048822,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00047325170034947167,
      "loss": 5.2637,
      "mean_token_accuracy": 0.17064776122570038,
      "num_tokens": 36424537.0,
      "step": 20995
    },
    {
      "entropy": 5.513808393478394,
      "epoch": 1.6338844582766,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047323859227641537,
      "loss": 5.1551,
      "mean_token_accuracy": 0.18044808208942414,
      "num_tokens": 36433276.0,
      "step": 21000
    },
    {
      "epoch": 1.6338844582766,
      "eval_entropy": 5.372300026299749,
      "eval_loss": 5.31075382232666,
      "eval_mean_token_accuracy": 0.1839048819603888,
      "eval_num_tokens": 36433276.0,
      "eval_runtime": 28.7266,
      "eval_samples_per_second": 1446.672,
      "eval_steps_per_second": 180.843,
      "step": 21000
    },
    {
      "entropy": 5.501618051528931,
      "epoch": 1.6342734876483176,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00047322548119540397,
      "loss": 5.2173,
      "mean_token_accuracy": 0.17700836956501007,
      "num_tokens": 36442154.0,
      "step": 21005
    },
    {
      "entropy": 5.398017358779907,
      "epoch": 1.634662517020035,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047321236710663636,
      "loss": 5.064,
      "mean_token_accuracy": 0.18820375800132752,
      "num_tokens": 36450625.0,
      "step": 21010
    },
    {
      "entropy": 5.560705900192261,
      "epoch": 1.6350515463917525,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047319925001031165,
      "loss": 5.2152,
      "mean_token_accuracy": 0.17989173531532288,
      "num_tokens": 36460082.0,
      "step": 21015
    },
    {
      "entropy": 5.546498870849609,
      "epoch": 1.6354405757634702,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004731861299066287,
      "loss": 5.2649,
      "mean_token_accuracy": 0.17282200008630752,
      "num_tokens": 36469372.0,
      "step": 21020
    },
    {
      "entropy": 5.420505666732788,
      "epoch": 1.6358296051351877,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004731730067957867,
      "loss": 5.0368,
      "mean_token_accuracy": 0.1866752788424492,
      "num_tokens": 36478190.0,
      "step": 21025
    },
    {
      "entropy": 5.387681293487549,
      "epoch": 1.6362186345069052,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047315988067798476,
      "loss": 5.1613,
      "mean_token_accuracy": 0.18450152724981309,
      "num_tokens": 36486245.0,
      "step": 21030
    },
    {
      "entropy": 5.4909279346466064,
      "epoch": 1.6366076638786229,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000473146751553422,
      "loss": 5.2498,
      "mean_token_accuracy": 0.17672774493694304,
      "num_tokens": 36495858.0,
      "step": 21035
    },
    {
      "entropy": 5.599689865112305,
      "epoch": 1.6369966932503404,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004731336194222978,
      "loss": 5.2539,
      "mean_token_accuracy": 0.17409946024417877,
      "num_tokens": 36504564.0,
      "step": 21040
    },
    {
      "entropy": 5.353814888000488,
      "epoch": 1.6373857226220578,
      "grad_norm": 1.53125,
      "learning_rate": 0.00047312048428481114,
      "loss": 5.1251,
      "mean_token_accuracy": 0.1855647921562195,
      "num_tokens": 36512758.0,
      "step": 21045
    },
    {
      "entropy": 5.3884512424469,
      "epoch": 1.6377747519937755,
      "grad_norm": 1.546875,
      "learning_rate": 0.00047310734614116153,
      "loss": 5.1124,
      "mean_token_accuracy": 0.18222106844186783,
      "num_tokens": 36521429.0,
      "step": 21050
    },
    {
      "entropy": 5.514833450317383,
      "epoch": 1.6381637813654932,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004730942049915483,
      "loss": 5.2004,
      "mean_token_accuracy": 0.18652649223804474,
      "num_tokens": 36530161.0,
      "step": 21055
    },
    {
      "entropy": 5.496166181564331,
      "epoch": 1.6385528107372107,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004730810608361707,
      "loss": 5.1813,
      "mean_token_accuracy": 0.18224625438451766,
      "num_tokens": 36538184.0,
      "step": 21060
    },
    {
      "entropy": 5.4722819328308105,
      "epoch": 1.6389418401089282,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047306791367522833,
      "loss": 5.1678,
      "mean_token_accuracy": 0.1848045125603676,
      "num_tokens": 36547418.0,
      "step": 21065
    },
    {
      "entropy": 5.614156723022461,
      "epoch": 1.6393308694806459,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004730547635089206,
      "loss": 5.3604,
      "mean_token_accuracy": 0.1707831457257271,
      "num_tokens": 36556542.0,
      "step": 21070
    },
    {
      "entropy": 5.5106339931488035,
      "epoch": 1.6397198988523634,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047304161033744714,
      "loss": 5.1586,
      "mean_token_accuracy": 0.17730581611394883,
      "num_tokens": 36565903.0,
      "step": 21075
    },
    {
      "entropy": 5.429777765274048,
      "epoch": 1.6401089282240808,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004730284541610075,
      "loss": 5.1755,
      "mean_token_accuracy": 0.17911712378263472,
      "num_tokens": 36574010.0,
      "step": 21080
    },
    {
      "entropy": 5.45843014717102,
      "epoch": 1.6404979575957985,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004730152949798012,
      "loss": 5.1999,
      "mean_token_accuracy": 0.1825070008635521,
      "num_tokens": 36582142.0,
      "step": 21085
    },
    {
      "entropy": 5.455931711196899,
      "epoch": 1.640886986967516,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00047300213279402787,
      "loss": 5.1971,
      "mean_token_accuracy": 0.18478820770978927,
      "num_tokens": 36590676.0,
      "step": 21090
    },
    {
      "entropy": 5.477128267288208,
      "epoch": 1.6412760163392335,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047298896760388745,
      "loss": 5.0925,
      "mean_token_accuracy": 0.19207348823547363,
      "num_tokens": 36599489.0,
      "step": 21095
    },
    {
      "entropy": 5.491489315032959,
      "epoch": 1.6416650457109512,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004729757994095796,
      "loss": 5.1659,
      "mean_token_accuracy": 0.17730745822191238,
      "num_tokens": 36608231.0,
      "step": 21100
    },
    {
      "entropy": 5.419163703918457,
      "epoch": 1.6420540750826689,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047296262821130405,
      "loss": 5.2344,
      "mean_token_accuracy": 0.17491605430841445,
      "num_tokens": 36617335.0,
      "step": 21105
    },
    {
      "entropy": 5.54743390083313,
      "epoch": 1.6424431044543863,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047294945400926065,
      "loss": 5.2425,
      "mean_token_accuracy": 0.17947182357311248,
      "num_tokens": 36626249.0,
      "step": 21110
    },
    {
      "entropy": 5.563682126998901,
      "epoch": 1.6428321338261038,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004729362768036494,
      "loss": 5.3014,
      "mean_token_accuracy": 0.1732996955513954,
      "num_tokens": 36634602.0,
      "step": 21115
    },
    {
      "entropy": 5.485794401168823,
      "epoch": 1.6432211631978215,
      "grad_norm": 1.53125,
      "learning_rate": 0.00047292309659467024,
      "loss": 5.2352,
      "mean_token_accuracy": 0.1770779922604561,
      "num_tokens": 36643091.0,
      "step": 21120
    },
    {
      "entropy": 5.541834020614624,
      "epoch": 1.643610192569539,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004729099133825231,
      "loss": 5.2513,
      "mean_token_accuracy": 0.18132439106702805,
      "num_tokens": 36652530.0,
      "step": 21125
    },
    {
      "entropy": 5.5994336128234865,
      "epoch": 1.6439992219412565,
      "grad_norm": 1.546875,
      "learning_rate": 0.000472896727167408,
      "loss": 5.2581,
      "mean_token_accuracy": 0.17381585389375687,
      "num_tokens": 36660858.0,
      "step": 21130
    },
    {
      "entropy": 5.61146092414856,
      "epoch": 1.6443882513129742,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00047288353794952505,
      "loss": 5.3237,
      "mean_token_accuracy": 0.16663160622119905,
      "num_tokens": 36670211.0,
      "step": 21135
    },
    {
      "entropy": 5.502262973785401,
      "epoch": 1.6447772806846916,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004728703457290744,
      "loss": 5.2337,
      "mean_token_accuracy": 0.1756765738129616,
      "num_tokens": 36679604.0,
      "step": 21140
    },
    {
      "entropy": 5.548543882369995,
      "epoch": 1.6451663100564091,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004728571505062562,
      "loss": 5.2994,
      "mean_token_accuracy": 0.17042364180088043,
      "num_tokens": 36687908.0,
      "step": 21145
    },
    {
      "entropy": 5.509022760391235,
      "epoch": 1.6455553394281268,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004728439522812707,
      "loss": 5.1156,
      "mean_token_accuracy": 0.1840012028813362,
      "num_tokens": 36696675.0,
      "step": 21150
    },
    {
      "entropy": 5.464862585067749,
      "epoch": 1.6459443687998445,
      "grad_norm": 1.4375,
      "learning_rate": 0.00047283075105431806,
      "loss": 5.2001,
      "mean_token_accuracy": 0.17636069655418396,
      "num_tokens": 36705594.0,
      "step": 21155
    },
    {
      "entropy": 5.543157863616943,
      "epoch": 1.646333398171562,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004728175468255987,
      "loss": 5.2583,
      "mean_token_accuracy": 0.17793198972940444,
      "num_tokens": 36714008.0,
      "step": 21160
    },
    {
      "entropy": 5.492327642440796,
      "epoch": 1.6467224275432795,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00047280433959531287,
      "loss": 5.1043,
      "mean_token_accuracy": 0.19281417727470399,
      "num_tokens": 36723093.0,
      "step": 21165
    },
    {
      "entropy": 5.449915790557862,
      "epoch": 1.6471114569149972,
      "grad_norm": 1.328125,
      "learning_rate": 0.000472791129363661,
      "loss": 5.1076,
      "mean_token_accuracy": 0.18590016067028045,
      "num_tokens": 36732119.0,
      "step": 21170
    },
    {
      "entropy": 5.393493413925171,
      "epoch": 1.6475004862867146,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004727779161308436,
      "loss": 5.1792,
      "mean_token_accuracy": 0.17904208302497865,
      "num_tokens": 36740593.0,
      "step": 21175
    },
    {
      "entropy": 5.561703491210937,
      "epoch": 1.6478895156584321,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004727646998970612,
      "loss": 5.3688,
      "mean_token_accuracy": 0.17116793543100356,
      "num_tokens": 36749083.0,
      "step": 21180
    },
    {
      "entropy": 5.580496835708618,
      "epoch": 1.6482785450301498,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047275148066251417,
      "loss": 5.1801,
      "mean_token_accuracy": 0.1789710432291031,
      "num_tokens": 36757469.0,
      "step": 21185
    },
    {
      "entropy": 5.593105030059815,
      "epoch": 1.6486675744018673,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004727382584274032,
      "loss": 5.3309,
      "mean_token_accuracy": 0.16570038050413133,
      "num_tokens": 36766197.0,
      "step": 21190
    },
    {
      "entropy": 5.44945387840271,
      "epoch": 1.6490566037735848,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047272503319192887,
      "loss": 5.1772,
      "mean_token_accuracy": 0.18537746071815492,
      "num_tokens": 36774172.0,
      "step": 21195
    },
    {
      "entropy": 5.498950624465943,
      "epoch": 1.6494456331453025,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004727118049562919,
      "loss": 5.2195,
      "mean_token_accuracy": 0.17729924619197845,
      "num_tokens": 36783549.0,
      "step": 21200
    },
    {
      "entropy": 5.603929471969605,
      "epoch": 1.6498346625170202,
      "grad_norm": 1.6171875,
      "learning_rate": 0.000472698573720693,
      "loss": 5.2722,
      "mean_token_accuracy": 0.17381686121225357,
      "num_tokens": 36791755.0,
      "step": 21205
    },
    {
      "entropy": 5.5498090267181395,
      "epoch": 1.6502236918887376,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004726853394853329,
      "loss": 5.3341,
      "mean_token_accuracy": 0.17136198431253433,
      "num_tokens": 36801253.0,
      "step": 21210
    },
    {
      "entropy": 5.473941421508789,
      "epoch": 1.650612721260455,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047267210225041234,
      "loss": 5.1686,
      "mean_token_accuracy": 0.1872959852218628,
      "num_tokens": 36810497.0,
      "step": 21215
    },
    {
      "entropy": 5.479462480545044,
      "epoch": 1.6510017506321728,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00047265886201613234,
      "loss": 5.1576,
      "mean_token_accuracy": 0.18567696511745452,
      "num_tokens": 36819823.0,
      "step": 21220
    },
    {
      "entropy": 5.476222896575928,
      "epoch": 1.6513907800038903,
      "grad_norm": 1.5625,
      "learning_rate": 0.00047264561878269365,
      "loss": 5.1423,
      "mean_token_accuracy": 0.18462543934583664,
      "num_tokens": 36828935.0,
      "step": 21225
    },
    {
      "entropy": 5.468758678436279,
      "epoch": 1.6517798093756078,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004726323725502973,
      "loss": 5.0973,
      "mean_token_accuracy": 0.18429700881242753,
      "num_tokens": 36837906.0,
      "step": 21230
    },
    {
      "entropy": 5.498550271987915,
      "epoch": 1.6521688387473255,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004726191233191443,
      "loss": 5.1622,
      "mean_token_accuracy": 0.1857444688677788,
      "num_tokens": 36846041.0,
      "step": 21235
    },
    {
      "entropy": 5.542250347137451,
      "epoch": 1.6525578681190431,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004726058710894357,
      "loss": 5.2919,
      "mean_token_accuracy": 0.1728598564863205,
      "num_tokens": 36854493.0,
      "step": 21240
    },
    {
      "entropy": 5.445422792434693,
      "epoch": 1.6529468974907604,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004725926158613724,
      "loss": 5.0903,
      "mean_token_accuracy": 0.1866839811205864,
      "num_tokens": 36863042.0,
      "step": 21245
    },
    {
      "entropy": 5.528048992156982,
      "epoch": 1.653335926862478,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004725793576351557,
      "loss": 5.2602,
      "mean_token_accuracy": 0.17213737666606904,
      "num_tokens": 36872513.0,
      "step": 21250
    },
    {
      "entropy": 5.522940444946289,
      "epoch": 1.6537249562341958,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004725660964109867,
      "loss": 5.2161,
      "mean_token_accuracy": 0.1788783550262451,
      "num_tokens": 36881149.0,
      "step": 21255
    },
    {
      "entropy": 5.488233470916748,
      "epoch": 1.6541139856059133,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047255283218906673,
      "loss": 5.2395,
      "mean_token_accuracy": 0.18180759847164155,
      "num_tokens": 36889372.0,
      "step": 21260
    },
    {
      "entropy": 5.48690447807312,
      "epoch": 1.6545030149776307,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000472539564969597,
      "loss": 5.1756,
      "mean_token_accuracy": 0.18047224432229997,
      "num_tokens": 36897727.0,
      "step": 21265
    },
    {
      "entropy": 5.498178625106812,
      "epoch": 1.6548920443493484,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047252629475277863,
      "loss": 5.188,
      "mean_token_accuracy": 0.18057161420583726,
      "num_tokens": 36906288.0,
      "step": 21270
    },
    {
      "entropy": 5.528655481338501,
      "epoch": 1.655281073721066,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004725130215388132,
      "loss": 5.248,
      "mean_token_accuracy": 0.1735369861125946,
      "num_tokens": 36915614.0,
      "step": 21275
    },
    {
      "entropy": 5.472884035110473,
      "epoch": 1.6556701030927834,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004724997453279021,
      "loss": 5.1117,
      "mean_token_accuracy": 0.18641575276851655,
      "num_tokens": 36923631.0,
      "step": 21280
    },
    {
      "entropy": 5.512561893463134,
      "epoch": 1.656059132464501,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004724864661202467,
      "loss": 5.2837,
      "mean_token_accuracy": 0.1790953740477562,
      "num_tokens": 36932296.0,
      "step": 21285
    },
    {
      "entropy": 5.4006248950958256,
      "epoch": 1.6564481618362188,
      "grad_norm": 1.671875,
      "learning_rate": 0.00047247318391604846,
      "loss": 5.0735,
      "mean_token_accuracy": 0.19231092929840088,
      "num_tokens": 36940785.0,
      "step": 21290
    },
    {
      "entropy": 5.509478616714477,
      "epoch": 1.656837191207936,
      "grad_norm": 1.4375,
      "learning_rate": 0.00047245989871550897,
      "loss": 5.2028,
      "mean_token_accuracy": 0.1878322422504425,
      "num_tokens": 36948797.0,
      "step": 21295
    },
    {
      "entropy": 5.554609918594361,
      "epoch": 1.6572262205796537,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004724466105188299,
      "loss": 5.3133,
      "mean_token_accuracy": 0.175143563747406,
      "num_tokens": 36957808.0,
      "step": 21300
    },
    {
      "entropy": 5.529737281799316,
      "epoch": 1.6576152499513714,
      "grad_norm": 1.453125,
      "learning_rate": 0.00047243331932621263,
      "loss": 5.1916,
      "mean_token_accuracy": 0.18748618215322493,
      "num_tokens": 36966260.0,
      "step": 21305
    },
    {
      "entropy": 5.544512367248535,
      "epoch": 1.658004279323089,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004724200251378591,
      "loss": 5.2393,
      "mean_token_accuracy": 0.1755203902721405,
      "num_tokens": 36975436.0,
      "step": 21310
    },
    {
      "entropy": 5.607953500747681,
      "epoch": 1.6583933086948064,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004724067279539709,
      "loss": 5.3472,
      "mean_token_accuracy": 0.17324303537607194,
      "num_tokens": 36983485.0,
      "step": 21315
    },
    {
      "entropy": 5.649571800231934,
      "epoch": 1.658782338066524,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047239342777474975,
      "loss": 5.3442,
      "mean_token_accuracy": 0.17102190107107162,
      "num_tokens": 36992632.0,
      "step": 21320
    },
    {
      "entropy": 5.498779916763306,
      "epoch": 1.6591713674382416,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047238012460039765,
      "loss": 5.1282,
      "mean_token_accuracy": 0.184706611931324,
      "num_tokens": 37000694.0,
      "step": 21325
    },
    {
      "entropy": 5.43941969871521,
      "epoch": 1.659560396809959,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004723668184311162,
      "loss": 5.195,
      "mean_token_accuracy": 0.18460538238286972,
      "num_tokens": 37009514.0,
      "step": 21330
    },
    {
      "entropy": 5.535276889801025,
      "epoch": 1.6599494261816767,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004723535092671074,
      "loss": 5.3332,
      "mean_token_accuracy": 0.1774955838918686,
      "num_tokens": 37019434.0,
      "step": 21335
    },
    {
      "entropy": 5.482357740402222,
      "epoch": 1.6603384555533944,
      "grad_norm": 1.453125,
      "learning_rate": 0.00047234019710857335,
      "loss": 5.1406,
      "mean_token_accuracy": 0.17885607331991196,
      "num_tokens": 37028673.0,
      "step": 21340
    },
    {
      "entropy": 5.458214187622071,
      "epoch": 1.6607274849251117,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004723268819557158,
      "loss": 5.1822,
      "mean_token_accuracy": 0.18142179548740386,
      "num_tokens": 37038204.0,
      "step": 21345
    },
    {
      "entropy": 5.411493015289307,
      "epoch": 1.6611165142968294,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00047231356380873687,
      "loss": 5.1029,
      "mean_token_accuracy": 0.18283302634954451,
      "num_tokens": 37047301.0,
      "step": 21350
    },
    {
      "entropy": 5.473569631576538,
      "epoch": 1.661505543668547,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004723002426678388,
      "loss": 5.2148,
      "mean_token_accuracy": 0.1792337715625763,
      "num_tokens": 37054892.0,
      "step": 21355
    },
    {
      "entropy": 5.460709905624389,
      "epoch": 1.6618945730402646,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004722869185332235,
      "loss": 5.1015,
      "mean_token_accuracy": 0.18274426758289336,
      "num_tokens": 37063139.0,
      "step": 21360
    },
    {
      "entropy": 5.507890367507935,
      "epoch": 1.662283602411982,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047227359140509324,
      "loss": 5.2773,
      "mean_token_accuracy": 0.17320998758077621,
      "num_tokens": 37072019.0,
      "step": 21365
    },
    {
      "entropy": 5.508670902252197,
      "epoch": 1.6626726317836997,
      "grad_norm": 1.5,
      "learning_rate": 0.0004722602612836501,
      "loss": 5.3563,
      "mean_token_accuracy": 0.17287321388721466,
      "num_tokens": 37081209.0,
      "step": 21370
    },
    {
      "entropy": 5.416639614105224,
      "epoch": 1.6630616611554172,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047224692816909665,
      "loss": 5.0897,
      "mean_token_accuracy": 0.18702242225408555,
      "num_tokens": 37089464.0,
      "step": 21375
    },
    {
      "entropy": 5.436346960067749,
      "epoch": 1.6634506905271347,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004722335920616349,
      "loss": 5.0992,
      "mean_token_accuracy": 0.18826077431440352,
      "num_tokens": 37098224.0,
      "step": 21380
    },
    {
      "entropy": 5.464160394668579,
      "epoch": 1.6638397198988524,
      "grad_norm": 1.4375,
      "learning_rate": 0.00047222025296146745,
      "loss": 5.2496,
      "mean_token_accuracy": 0.18096317499876022,
      "num_tokens": 37107039.0,
      "step": 21385
    },
    {
      "entropy": 5.578289270401001,
      "epoch": 1.66422874927057,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047220691086879643,
      "loss": 5.264,
      "mean_token_accuracy": 0.17396355122327806,
      "num_tokens": 37116029.0,
      "step": 21390
    },
    {
      "entropy": 5.490733337402344,
      "epoch": 1.6646177786422873,
      "grad_norm": 1.375,
      "learning_rate": 0.0004721935657838245,
      "loss": 5.0709,
      "mean_token_accuracy": 0.19254263937473298,
      "num_tokens": 37124553.0,
      "step": 21395
    },
    {
      "entropy": 5.4725157737731935,
      "epoch": 1.665006808014005,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000472180217706754,
      "loss": 5.2293,
      "mean_token_accuracy": 0.18405308574438095,
      "num_tokens": 37132670.0,
      "step": 21400
    },
    {
      "entropy": 5.497640800476074,
      "epoch": 1.6653958373857227,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004721668666377876,
      "loss": 5.3252,
      "mean_token_accuracy": 0.17594234198331832,
      "num_tokens": 37141534.0,
      "step": 21405
    },
    {
      "entropy": 5.513368940353393,
      "epoch": 1.6657848667574402,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047215351257712783,
      "loss": 5.2368,
      "mean_token_accuracy": 0.17468259036540984,
      "num_tokens": 37150895.0,
      "step": 21410
    },
    {
      "entropy": 5.502500820159912,
      "epoch": 1.6661738961291577,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047214015552497735,
      "loss": 5.1871,
      "mean_token_accuracy": 0.18133655339479446,
      "num_tokens": 37159244.0,
      "step": 21415
    },
    {
      "entropy": 5.4912059783935545,
      "epoch": 1.6665629255008754,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047212679548153867,
      "loss": 5.2469,
      "mean_token_accuracy": 0.18683581799268723,
      "num_tokens": 37167762.0,
      "step": 21420
    },
    {
      "entropy": 5.4224341869354244,
      "epoch": 1.6669519548725928,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047211343244701475,
      "loss": 5.1685,
      "mean_token_accuracy": 0.18159301280975343,
      "num_tokens": 37176470.0,
      "step": 21425
    },
    {
      "entropy": 5.531132411956787,
      "epoch": 1.6673409842443103,
      "grad_norm": 1.5,
      "learning_rate": 0.0004721000664216081,
      "loss": 5.2758,
      "mean_token_accuracy": 0.17480931729078292,
      "num_tokens": 37185923.0,
      "step": 21430
    },
    {
      "entropy": 5.496185970306397,
      "epoch": 1.667730013616028,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004720866974055219,
      "loss": 5.2788,
      "mean_token_accuracy": 0.17899755835533143,
      "num_tokens": 37195400.0,
      "step": 21435
    },
    {
      "entropy": 5.557925462722778,
      "epoch": 1.6681190429877457,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004720733253989585,
      "loss": 5.3076,
      "mean_token_accuracy": 0.17580303698778152,
      "num_tokens": 37203862.0,
      "step": 21440
    },
    {
      "entropy": 5.477754020690918,
      "epoch": 1.6685080723594632,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047205995040212114,
      "loss": 5.0736,
      "mean_token_accuracy": 0.18728139102458954,
      "num_tokens": 37212518.0,
      "step": 21445
    },
    {
      "entropy": 5.564401006698608,
      "epoch": 1.6688971017311807,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004720465724152127,
      "loss": 5.3096,
      "mean_token_accuracy": 0.17177732437849044,
      "num_tokens": 37221342.0,
      "step": 21450
    },
    {
      "entropy": 5.5528092861175535,
      "epoch": 1.6692861311028984,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00047203319143843613,
      "loss": 5.2324,
      "mean_token_accuracy": 0.1832715541124344,
      "num_tokens": 37230477.0,
      "step": 21455
    },
    {
      "entropy": 5.557088422775268,
      "epoch": 1.6696751604746158,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004720198074719946,
      "loss": 5.1887,
      "mean_token_accuracy": 0.1888517916202545,
      "num_tokens": 37238590.0,
      "step": 21460
    },
    {
      "entropy": 5.467580127716064,
      "epoch": 1.6700641898463333,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047200642051609094,
      "loss": 5.2181,
      "mean_token_accuracy": 0.18081460446119307,
      "num_tokens": 37248065.0,
      "step": 21465
    },
    {
      "entropy": 5.492550086975098,
      "epoch": 1.670453219218051,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004719930305709285,
      "loss": 5.2125,
      "mean_token_accuracy": 0.18174617886543273,
      "num_tokens": 37256467.0,
      "step": 21470
    },
    {
      "entropy": 5.553583717346191,
      "epoch": 1.6708422485897685,
      "grad_norm": 1.65625,
      "learning_rate": 0.00047197963763671033,
      "loss": 5.1984,
      "mean_token_accuracy": 0.1813953474164009,
      "num_tokens": 37264255.0,
      "step": 21475
    },
    {
      "entropy": 5.460461521148682,
      "epoch": 1.671231277961486,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004719662417136397,
      "loss": 5.1129,
      "mean_token_accuracy": 0.18463306725025178,
      "num_tokens": 37272671.0,
      "step": 21480
    },
    {
      "entropy": 5.456419610977173,
      "epoch": 1.6716203073332037,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00047195284280191987,
      "loss": 5.2427,
      "mean_token_accuracy": 0.18076366037130356,
      "num_tokens": 37281895.0,
      "step": 21485
    },
    {
      "entropy": 5.401412487030029,
      "epoch": 1.6720093367049214,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004719394409017541,
      "loss": 5.125,
      "mean_token_accuracy": 0.18857089728116988,
      "num_tokens": 37290098.0,
      "step": 21490
    },
    {
      "entropy": 5.506783533096313,
      "epoch": 1.6723983660766388,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00047192603601334585,
      "loss": 5.1737,
      "mean_token_accuracy": 0.18363878577947618,
      "num_tokens": 37299494.0,
      "step": 21495
    },
    {
      "entropy": 5.502944612503052,
      "epoch": 1.6727873954483563,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00047191262813689835,
      "loss": 5.2464,
      "mean_token_accuracy": 0.17842065989971162,
      "num_tokens": 37308366.0,
      "step": 21500
    },
    {
      "entropy": 5.476567268371582,
      "epoch": 1.673176424820074,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004718992172726152,
      "loss": 5.219,
      "mean_token_accuracy": 0.17957063466310502,
      "num_tokens": 37316774.0,
      "step": 21505
    },
    {
      "entropy": 5.520447397232056,
      "epoch": 1.6735654541917915,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00047188580342069983,
      "loss": 5.2503,
      "mean_token_accuracy": 0.17251145243644714,
      "num_tokens": 37325388.0,
      "step": 21510
    },
    {
      "entropy": 5.562305021286011,
      "epoch": 1.673954483563509,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004718723865813557,
      "loss": 5.2768,
      "mean_token_accuracy": 0.17611507922410966,
      "num_tokens": 37333476.0,
      "step": 21515
    },
    {
      "entropy": 5.5756597995758055,
      "epoch": 1.6743435129352267,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004718589667547865,
      "loss": 5.3421,
      "mean_token_accuracy": 0.17612942755222322,
      "num_tokens": 37342556.0,
      "step": 21520
    },
    {
      "entropy": 5.53990421295166,
      "epoch": 1.6747325423069441,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004718455439411958,
      "loss": 5.2205,
      "mean_token_accuracy": 0.18342919945716857,
      "num_tokens": 37351082.0,
      "step": 21525
    },
    {
      "entropy": 5.537401437759399,
      "epoch": 1.6751215716786616,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004718321181407873,
      "loss": 5.1775,
      "mean_token_accuracy": 0.1762392118573189,
      "num_tokens": 37359083.0,
      "step": 21530
    },
    {
      "entropy": 5.466626596450806,
      "epoch": 1.6755106010503793,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004718186893537647,
      "loss": 5.1831,
      "mean_token_accuracy": 0.18693919479846954,
      "num_tokens": 37367982.0,
      "step": 21535
    },
    {
      "entropy": 5.570475101470947,
      "epoch": 1.675899630422097,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004718052575803318,
      "loss": 5.2665,
      "mean_token_accuracy": 0.17587852030992507,
      "num_tokens": 37376932.0,
      "step": 21540
    },
    {
      "entropy": 5.53325834274292,
      "epoch": 1.6762886597938145,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004717918228206923,
      "loss": 5.2771,
      "mean_token_accuracy": 0.1744777664542198,
      "num_tokens": 37386248.0,
      "step": 21545
    },
    {
      "entropy": 5.5484436511993405,
      "epoch": 1.676677689165532,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004717783850750501,
      "loss": 5.1991,
      "mean_token_accuracy": 0.1772058054804802,
      "num_tokens": 37395412.0,
      "step": 21550
    },
    {
      "entropy": 5.542940521240235,
      "epoch": 1.6770667185372496,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004717649443436091,
      "loss": 5.161,
      "mean_token_accuracy": 0.18279172480106354,
      "num_tokens": 37404332.0,
      "step": 21555
    },
    {
      "entropy": 5.538176393508911,
      "epoch": 1.6774557479089671,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047175150062657336,
      "loss": 5.2719,
      "mean_token_accuracy": 0.1777028486132622,
      "num_tokens": 37413365.0,
      "step": 21560
    },
    {
      "entropy": 5.5362122535705565,
      "epoch": 1.6778447772806846,
      "grad_norm": 1.484375,
      "learning_rate": 0.00047173805392414664,
      "loss": 5.1549,
      "mean_token_accuracy": 0.1803798645734787,
      "num_tokens": 37422164.0,
      "step": 21565
    },
    {
      "entropy": 5.529918813705445,
      "epoch": 1.6782338066524023,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004717246042365332,
      "loss": 5.2371,
      "mean_token_accuracy": 0.18031506687402726,
      "num_tokens": 37431092.0,
      "step": 21570
    },
    {
      "entropy": 5.522467136383057,
      "epoch": 1.6786228360241198,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047171115156393695,
      "loss": 5.3072,
      "mean_token_accuracy": 0.1718997463583946,
      "num_tokens": 37440832.0,
      "step": 21575
    },
    {
      "entropy": 5.519169569015503,
      "epoch": 1.6790118653958372,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047169769590656207,
      "loss": 5.2358,
      "mean_token_accuracy": 0.1741737648844719,
      "num_tokens": 37449317.0,
      "step": 21580
    },
    {
      "entropy": 5.547975540161133,
      "epoch": 1.679400894767555,
      "grad_norm": 1.640625,
      "learning_rate": 0.00047168423726461273,
      "loss": 5.2718,
      "mean_token_accuracy": 0.17741824686527252,
      "num_tokens": 37457820.0,
      "step": 21585
    },
    {
      "entropy": 5.516720199584961,
      "epoch": 1.6797899241392726,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004716707756382931,
      "loss": 5.1137,
      "mean_token_accuracy": 0.19077614545822144,
      "num_tokens": 37466378.0,
      "step": 21590
    },
    {
      "entropy": 5.543765926361084,
      "epoch": 1.6801789535109901,
      "grad_norm": 1.4375,
      "learning_rate": 0.00047165731102780757,
      "loss": 5.2484,
      "mean_token_accuracy": 0.17531055808067322,
      "num_tokens": 37475114.0,
      "step": 21595
    },
    {
      "entropy": 5.5077352046966555,
      "epoch": 1.6805679828827076,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047164384343336037,
      "loss": 5.2652,
      "mean_token_accuracy": 0.168734173476696,
      "num_tokens": 37483898.0,
      "step": 21600
    },
    {
      "entropy": 5.498862171173096,
      "epoch": 1.6809570122544253,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047163037285515583,
      "loss": 5.2499,
      "mean_token_accuracy": 0.18168826997280121,
      "num_tokens": 37492420.0,
      "step": 21605
    },
    {
      "entropy": 5.501647472381592,
      "epoch": 1.6813460416261428,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004716168992933982,
      "loss": 5.1504,
      "mean_token_accuracy": 0.18668870478868485,
      "num_tokens": 37501090.0,
      "step": 21610
    },
    {
      "entropy": 5.513693380355835,
      "epoch": 1.6817350709978602,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047160342274829234,
      "loss": 5.3013,
      "mean_token_accuracy": 0.1726093515753746,
      "num_tokens": 37510465.0,
      "step": 21615
    },
    {
      "entropy": 5.48835506439209,
      "epoch": 1.682124100369578,
      "grad_norm": 1.453125,
      "learning_rate": 0.00047158994322004223,
      "loss": 5.2147,
      "mean_token_accuracy": 0.18208342641592026,
      "num_tokens": 37519155.0,
      "step": 21620
    },
    {
      "entropy": 5.484408950805664,
      "epoch": 1.6825131297412956,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004715764607088527,
      "loss": 5.1904,
      "mean_token_accuracy": 0.18758898377418518,
      "num_tokens": 37527134.0,
      "step": 21625
    },
    {
      "entropy": 5.511429738998413,
      "epoch": 1.6829021591130129,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004715629752149283,
      "loss": 5.2217,
      "mean_token_accuracy": 0.18314019441604615,
      "num_tokens": 37536087.0,
      "step": 21630
    },
    {
      "entropy": 5.458495378494263,
      "epoch": 1.6832911884847306,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047154948673847356,
      "loss": 5.1611,
      "mean_token_accuracy": 0.1872522711753845,
      "num_tokens": 37543896.0,
      "step": 21635
    },
    {
      "entropy": 5.465446090698242,
      "epoch": 1.6836802178564483,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004715359952796933,
      "loss": 5.2192,
      "mean_token_accuracy": 0.18115221709012985,
      "num_tokens": 37552571.0,
      "step": 21640
    },
    {
      "entropy": 5.5718724727630615,
      "epoch": 1.6840692472281658,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047152250083879203,
      "loss": 5.2636,
      "mean_token_accuracy": 0.17640819847583772,
      "num_tokens": 37560823.0,
      "step": 21645
    },
    {
      "entropy": 5.463348865509033,
      "epoch": 1.6844582765998832,
      "grad_norm": 1.515625,
      "learning_rate": 0.00047150900341597464,
      "loss": 5.1546,
      "mean_token_accuracy": 0.1767340987920761,
      "num_tokens": 37568312.0,
      "step": 21650
    },
    {
      "entropy": 5.449061489105224,
      "epoch": 1.684847305971601,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004714955030114459,
      "loss": 5.2171,
      "mean_token_accuracy": 0.18659872114658355,
      "num_tokens": 37577008.0,
      "step": 21655
    },
    {
      "entropy": 5.521494674682617,
      "epoch": 1.6852363353433184,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047148199962541065,
      "loss": 5.2181,
      "mean_token_accuracy": 0.17695734649896622,
      "num_tokens": 37586252.0,
      "step": 21660
    },
    {
      "entropy": 5.523296928405761,
      "epoch": 1.6856253647150359,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004714684932580738,
      "loss": 5.2095,
      "mean_token_accuracy": 0.18079616725444794,
      "num_tokens": 37594699.0,
      "step": 21665
    },
    {
      "entropy": 5.479553937911987,
      "epoch": 1.6860143940867536,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004714549839096403,
      "loss": 5.1891,
      "mean_token_accuracy": 0.17756162583827972,
      "num_tokens": 37603891.0,
      "step": 21670
    },
    {
      "entropy": 5.500479507446289,
      "epoch": 1.6864034234584713,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047144147158031507,
      "loss": 5.2086,
      "mean_token_accuracy": 0.17659277319908143,
      "num_tokens": 37612809.0,
      "step": 21675
    },
    {
      "entropy": 5.52520112991333,
      "epoch": 1.6867924528301885,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004714279562703032,
      "loss": 5.1665,
      "mean_token_accuracy": 0.17749557346105577,
      "num_tokens": 37621787.0,
      "step": 21680
    },
    {
      "entropy": 5.433181810379028,
      "epoch": 1.6871814822019062,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004714144379798098,
      "loss": 5.1155,
      "mean_token_accuracy": 0.19358740895986556,
      "num_tokens": 37629796.0,
      "step": 21685
    },
    {
      "entropy": 5.502446794509888,
      "epoch": 1.687570511573624,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00047140091670903986,
      "loss": 5.1658,
      "mean_token_accuracy": 0.18205602169036866,
      "num_tokens": 37638467.0,
      "step": 21690
    },
    {
      "entropy": 5.47499532699585,
      "epoch": 1.6879595409453414,
      "grad_norm": 1.546875,
      "learning_rate": 0.00047138739245819857,
      "loss": 5.2202,
      "mean_token_accuracy": 0.1819450080394745,
      "num_tokens": 37647159.0,
      "step": 21695
    },
    {
      "entropy": 5.50620813369751,
      "epoch": 1.6883485703170589,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00047137386522749124,
      "loss": 5.2061,
      "mean_token_accuracy": 0.17646756619215012,
      "num_tokens": 37655986.0,
      "step": 21700
    },
    {
      "entropy": 5.458985996246338,
      "epoch": 1.6887375996887766,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004713603350171231,
      "loss": 5.101,
      "mean_token_accuracy": 0.19006167352199554,
      "num_tokens": 37663932.0,
      "step": 21705
    },
    {
      "entropy": 5.545205211639404,
      "epoch": 1.689126629060494,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047134680182729926,
      "loss": 5.2627,
      "mean_token_accuracy": 0.171118625998497,
      "num_tokens": 37672142.0,
      "step": 21710
    },
    {
      "entropy": 5.491796255111694,
      "epoch": 1.6895156584322115,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004713332656582254,
      "loss": 5.1516,
      "mean_token_accuracy": 0.1812073454260826,
      "num_tokens": 37680569.0,
      "step": 21715
    },
    {
      "entropy": 5.4419197082519535,
      "epoch": 1.6899046878039292,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004713197265101066,
      "loss": 5.2284,
      "mean_token_accuracy": 0.1740078255534172,
      "num_tokens": 37688852.0,
      "step": 21720
    },
    {
      "entropy": 5.538496255874634,
      "epoch": 1.690293717175647,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00047130618438314856,
      "loss": 5.2009,
      "mean_token_accuracy": 0.18574654161930085,
      "num_tokens": 37696900.0,
      "step": 21725
    },
    {
      "entropy": 5.569355010986328,
      "epoch": 1.6906827465473642,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004712926392775565,
      "loss": 5.3547,
      "mean_token_accuracy": 0.17537433207035064,
      "num_tokens": 37706292.0,
      "step": 21730
    },
    {
      "entropy": 5.422090911865235,
      "epoch": 1.6910717759190819,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004712790911935362,
      "loss": 5.1559,
      "mean_token_accuracy": 0.18529009073972702,
      "num_tokens": 37715025.0,
      "step": 21735
    },
    {
      "entropy": 5.5033763408660885,
      "epoch": 1.6914608052907996,
      "grad_norm": 1.453125,
      "learning_rate": 0.000471265540131293,
      "loss": 5.2751,
      "mean_token_accuracy": 0.17744905650615692,
      "num_tokens": 37724124.0,
      "step": 21740
    },
    {
      "entropy": 5.535478305816651,
      "epoch": 1.691849834662517,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047125198609103267,
      "loss": 5.2153,
      "mean_token_accuracy": 0.1779232829809189,
      "num_tokens": 37732689.0,
      "step": 21745
    },
    {
      "entropy": 5.5286273002624515,
      "epoch": 1.6922388640342345,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004712384290729607,
      "loss": 5.179,
      "mean_token_accuracy": 0.1785541445016861,
      "num_tokens": 37741206.0,
      "step": 21750
    },
    {
      "entropy": 5.514972352981568,
      "epoch": 1.6926278934059522,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000471224869077283,
      "loss": 5.1799,
      "mean_token_accuracy": 0.1855989173054695,
      "num_tokens": 37749549.0,
      "step": 21755
    },
    {
      "entropy": 5.397480154037476,
      "epoch": 1.6930169227776697,
      "grad_norm": 1.578125,
      "learning_rate": 0.00047121130610420527,
      "loss": 4.9959,
      "mean_token_accuracy": 0.19770121127367019,
      "num_tokens": 37758332.0,
      "step": 21760
    },
    {
      "entropy": 5.4677262783050535,
      "epoch": 1.6934059521493872,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047119774015393323,
      "loss": 5.1968,
      "mean_token_accuracy": 0.17989788800477982,
      "num_tokens": 37767368.0,
      "step": 21765
    },
    {
      "entropy": 5.492316675186157,
      "epoch": 1.6937949815211049,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047118417122667285,
      "loss": 5.1695,
      "mean_token_accuracy": 0.18027150332927705,
      "num_tokens": 37776426.0,
      "step": 21770
    },
    {
      "entropy": 5.59453330039978,
      "epoch": 1.6941840108928226,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004711705993226298,
      "loss": 5.3459,
      "mean_token_accuracy": 0.17371474504470824,
      "num_tokens": 37784464.0,
      "step": 21775
    },
    {
      "entropy": 5.468116283416748,
      "epoch": 1.6945730402645398,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004711570244420102,
      "loss": 5.1767,
      "mean_token_accuracy": 0.17527460753917695,
      "num_tokens": 37794307.0,
      "step": 21780
    },
    {
      "entropy": 5.538769292831421,
      "epoch": 1.6949620696362575,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004711434465850199,
      "loss": 5.251,
      "mean_token_accuracy": 0.1727452203631401,
      "num_tokens": 37803151.0,
      "step": 21785
    },
    {
      "entropy": 5.4954376220703125,
      "epoch": 1.6953510990079752,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004711298657518651,
      "loss": 5.2201,
      "mean_token_accuracy": 0.18278917968273162,
      "num_tokens": 37812433.0,
      "step": 21790
    },
    {
      "entropy": 5.4799943447113035,
      "epoch": 1.6957401283796927,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004711162819427518,
      "loss": 5.1538,
      "mean_token_accuracy": 0.17833563685417175,
      "num_tokens": 37820627.0,
      "step": 21795
    },
    {
      "entropy": 5.515842437744141,
      "epoch": 1.6961291577514102,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000471102695157886,
      "loss": 5.209,
      "mean_token_accuracy": 0.179173843562603,
      "num_tokens": 37829015.0,
      "step": 21800
    },
    {
      "entropy": 5.544214344024658,
      "epoch": 1.6965181871231279,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004710891053974739,
      "loss": 5.2767,
      "mean_token_accuracy": 0.18105421662330629,
      "num_tokens": 37837097.0,
      "step": 21805
    },
    {
      "entropy": 5.483367586135865,
      "epoch": 1.6969072164948453,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004710755126617217,
      "loss": 5.2337,
      "mean_token_accuracy": 0.1754517912864685,
      "num_tokens": 37845414.0,
      "step": 21810
    },
    {
      "entropy": 5.5808460235595705,
      "epoch": 1.6972962458665628,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004710619169508358,
      "loss": 5.2418,
      "mean_token_accuracy": 0.17700733840465546,
      "num_tokens": 37855006.0,
      "step": 21815
    },
    {
      "entropy": 5.608074283599853,
      "epoch": 1.6976852752382805,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004710483182650223,
      "loss": 5.3714,
      "mean_token_accuracy": 0.17176454812288283,
      "num_tokens": 37863611.0,
      "step": 21820
    },
    {
      "entropy": 5.488765859603882,
      "epoch": 1.6980743046099982,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00047103471660448767,
      "loss": 5.2412,
      "mean_token_accuracy": 0.1852096512913704,
      "num_tokens": 37871935.0,
      "step": 21825
    },
    {
      "entropy": 5.453651762008667,
      "epoch": 1.6984633339817157,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047102111196943813,
      "loss": 5.194,
      "mean_token_accuracy": 0.17935559153556824,
      "num_tokens": 37880760.0,
      "step": 21830
    },
    {
      "entropy": 5.534674739837646,
      "epoch": 1.6988523633534331,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047100750436008035,
      "loss": 5.1333,
      "mean_token_accuracy": 0.18111956119537354,
      "num_tokens": 37889457.0,
      "step": 21835
    },
    {
      "entropy": 5.509087371826172,
      "epoch": 1.6992413927251508,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047099389377662054,
      "loss": 5.2159,
      "mean_token_accuracy": 0.17657059729099273,
      "num_tokens": 37897805.0,
      "step": 21840
    },
    {
      "entropy": 5.471603918075561,
      "epoch": 1.6996304220968683,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004709802802192654,
      "loss": 5.2611,
      "mean_token_accuracy": 0.17692361027002335,
      "num_tokens": 37906292.0,
      "step": 21845
    },
    {
      "entropy": 5.475686740875244,
      "epoch": 1.7000194514685858,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047096666368822153,
      "loss": 5.1862,
      "mean_token_accuracy": 0.17862415462732315,
      "num_tokens": 37914513.0,
      "step": 21850
    },
    {
      "entropy": 5.549863481521607,
      "epoch": 1.7004084808403035,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047095304418369536,
      "loss": 5.1783,
      "mean_token_accuracy": 0.18084910213947297,
      "num_tokens": 37923097.0,
      "step": 21855
    },
    {
      "entropy": 5.511747026443482,
      "epoch": 1.700797510212021,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004709394217058936,
      "loss": 5.2284,
      "mean_token_accuracy": 0.18504218459129335,
      "num_tokens": 37931617.0,
      "step": 21860
    },
    {
      "entropy": 5.544911241531372,
      "epoch": 1.7011865395837384,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004709257962550231,
      "loss": 5.1756,
      "mean_token_accuracy": 0.1816698357462883,
      "num_tokens": 37939777.0,
      "step": 21865
    },
    {
      "entropy": 5.488839864730835,
      "epoch": 1.7015755689554561,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00047091216783129035,
      "loss": 5.2202,
      "mean_token_accuracy": 0.17924584299325944,
      "num_tokens": 37949375.0,
      "step": 21870
    },
    {
      "entropy": 5.465655088424683,
      "epoch": 1.7019645983271738,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004708985364349024,
      "loss": 5.1785,
      "mean_token_accuracy": 0.18088839203119278,
      "num_tokens": 37958724.0,
      "step": 21875
    },
    {
      "entropy": 5.5307880401611325,
      "epoch": 1.7023536276988913,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047088490206606586,
      "loss": 5.159,
      "mean_token_accuracy": 0.18678024411201477,
      "num_tokens": 37967646.0,
      "step": 21880
    },
    {
      "entropy": 5.494955682754517,
      "epoch": 1.7027426570706088,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00047087126472498785,
      "loss": 5.2442,
      "mean_token_accuracy": 0.17920563966035843,
      "num_tokens": 37976058.0,
      "step": 21885
    },
    {
      "entropy": 5.3977477073669435,
      "epoch": 1.7031316864423265,
      "grad_norm": 1.4375,
      "learning_rate": 0.000470857624411875,
      "loss": 5.1493,
      "mean_token_accuracy": 0.18142630904912949,
      "num_tokens": 37984895.0,
      "step": 21890
    },
    {
      "entropy": 5.495544576644898,
      "epoch": 1.703520715814044,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047084398112693456,
      "loss": 5.1429,
      "mean_token_accuracy": 0.18670977503061295,
      "num_tokens": 37993514.0,
      "step": 21895
    },
    {
      "entropy": 5.509128093719482,
      "epoch": 1.7039097451857614,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004708303348703734,
      "loss": 5.1887,
      "mean_token_accuracy": 0.17902112603187562,
      "num_tokens": 38002274.0,
      "step": 21900
    },
    {
      "entropy": 5.456932926177979,
      "epoch": 1.7042987745574791,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004708166856423986,
      "loss": 5.0952,
      "mean_token_accuracy": 0.18529791682958602,
      "num_tokens": 38010336.0,
      "step": 21905
    },
    {
      "entropy": 5.461038494110108,
      "epoch": 1.7046878039291966,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047080303344321727,
      "loss": 5.1723,
      "mean_token_accuracy": 0.18365278393030166,
      "num_tokens": 38019166.0,
      "step": 21910
    },
    {
      "entropy": 5.4248772144317625,
      "epoch": 1.705076833300914,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047078937827303653,
      "loss": 5.2018,
      "mean_token_accuracy": 0.18425847440958024,
      "num_tokens": 38027634.0,
      "step": 21915
    },
    {
      "entropy": 5.502353763580322,
      "epoch": 1.7054658626726318,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004707757201320636,
      "loss": 5.197,
      "mean_token_accuracy": 0.17829769402742385,
      "num_tokens": 38036611.0,
      "step": 21920
    },
    {
      "entropy": 5.413452196121216,
      "epoch": 1.7058548920443495,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047076205902050577,
      "loss": 5.1171,
      "mean_token_accuracy": 0.18649646937847136,
      "num_tokens": 38045815.0,
      "step": 21925
    },
    {
      "entropy": 5.568564081192017,
      "epoch": 1.706243921416067,
      "grad_norm": 1.25,
      "learning_rate": 0.00047074839493857024,
      "loss": 5.2784,
      "mean_token_accuracy": 0.17126722782850265,
      "num_tokens": 38054973.0,
      "step": 21930
    },
    {
      "entropy": 5.538995170593262,
      "epoch": 1.7066329507877844,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004707347278864644,
      "loss": 5.1426,
      "mean_token_accuracy": 0.18554636240005493,
      "num_tokens": 38064084.0,
      "step": 21935
    },
    {
      "entropy": 5.42212495803833,
      "epoch": 1.7070219801595021,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00047072105786439563,
      "loss": 5.1768,
      "mean_token_accuracy": 0.18434233367443084,
      "num_tokens": 38073838.0,
      "step": 21940
    },
    {
      "entropy": 5.566514158248902,
      "epoch": 1.7074110095312196,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00047070738487257137,
      "loss": 5.283,
      "mean_token_accuracy": 0.17126548290252686,
      "num_tokens": 38082531.0,
      "step": 21945
    },
    {
      "entropy": 5.5509467124938965,
      "epoch": 1.707800038902937,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047069370891119895,
      "loss": 5.276,
      "mean_token_accuracy": 0.1744527518749237,
      "num_tokens": 38091499.0,
      "step": 21950
    },
    {
      "entropy": 5.473871040344238,
      "epoch": 1.7081890682746548,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000470680029980486,
      "loss": 5.1044,
      "mean_token_accuracy": 0.1944224163889885,
      "num_tokens": 38099488.0,
      "step": 21955
    },
    {
      "entropy": 5.53945574760437,
      "epoch": 1.7085780976463723,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047066634808064006,
      "loss": 5.2293,
      "mean_token_accuracy": 0.18228229135274887,
      "num_tokens": 38107535.0,
      "step": 21960
    },
    {
      "entropy": 5.525403070449829,
      "epoch": 1.7089671270180897,
      "grad_norm": 1.6875,
      "learning_rate": 0.00047065266321186873,
      "loss": 5.1981,
      "mean_token_accuracy": 0.17936547249555587,
      "num_tokens": 38116457.0,
      "step": 21965
    },
    {
      "entropy": 5.5263231754302975,
      "epoch": 1.7093561563898074,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004706389753743797,
      "loss": 5.1201,
      "mean_token_accuracy": 0.18372064232826232,
      "num_tokens": 38124571.0,
      "step": 21970
    },
    {
      "entropy": 5.4254436016082765,
      "epoch": 1.7097451857615251,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004706252845683805,
      "loss": 5.1668,
      "mean_token_accuracy": 0.17873062938451767,
      "num_tokens": 38132789.0,
      "step": 21975
    },
    {
      "entropy": 5.346435642242431,
      "epoch": 1.7101342151332426,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047061159079407903,
      "loss": 5.0742,
      "mean_token_accuracy": 0.18696712106466293,
      "num_tokens": 38141413.0,
      "step": 21980
    },
    {
      "entropy": 5.444990396499634,
      "epoch": 1.71052324450496,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000470597894051683,
      "loss": 5.0933,
      "mean_token_accuracy": 0.18352185934782028,
      "num_tokens": 38149869.0,
      "step": 21985
    },
    {
      "entropy": 5.424140930175781,
      "epoch": 1.7109122738766778,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004705841943414003,
      "loss": 5.0313,
      "mean_token_accuracy": 0.19739166647195816,
      "num_tokens": 38157356.0,
      "step": 21990
    },
    {
      "entropy": 5.463647413253784,
      "epoch": 1.7113013032483952,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00047057049166343876,
      "loss": 5.2161,
      "mean_token_accuracy": 0.1833688959479332,
      "num_tokens": 38166125.0,
      "step": 21995
    },
    {
      "entropy": 5.442975950241089,
      "epoch": 1.7116903326201127,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047055678601800623,
      "loss": 5.1479,
      "mean_token_accuracy": 0.1867290273308754,
      "num_tokens": 38174171.0,
      "step": 22000
    },
    {
      "entropy": 5.558338594436646,
      "epoch": 1.7120793619918304,
      "grad_norm": 1.375,
      "learning_rate": 0.00047054307740531076,
      "loss": 5.2564,
      "mean_token_accuracy": 0.1801733061671257,
      "num_tokens": 38182755.0,
      "step": 22005
    },
    {
      "entropy": 5.509387636184693,
      "epoch": 1.712468391363548,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047052936582556035,
      "loss": 5.191,
      "mean_token_accuracy": 0.18543592542409898,
      "num_tokens": 38191289.0,
      "step": 22010
    },
    {
      "entropy": 5.545600175857544,
      "epoch": 1.7128574207352654,
      "grad_norm": 1.625,
      "learning_rate": 0.00047051565127896307,
      "loss": 5.3311,
      "mean_token_accuracy": 0.17620862424373626,
      "num_tokens": 38200624.0,
      "step": 22015
    },
    {
      "entropy": 5.543854522705078,
      "epoch": 1.713246450106983,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047050193376572686,
      "loss": 5.1212,
      "mean_token_accuracy": 0.1847473829984665,
      "num_tokens": 38209272.0,
      "step": 22020
    },
    {
      "entropy": 5.560673522949219,
      "epoch": 1.7136354794787008,
      "grad_norm": 1.5,
      "learning_rate": 0.00047048821328606,
      "loss": 5.3079,
      "mean_token_accuracy": 0.17380361557006835,
      "num_tokens": 38218851.0,
      "step": 22025
    },
    {
      "entropy": 5.499703502655029,
      "epoch": 1.7140245088504182,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004704744898401708,
      "loss": 5.1976,
      "mean_token_accuracy": 0.18027321845293046,
      "num_tokens": 38227367.0,
      "step": 22030
    },
    {
      "entropy": 5.4966553211212155,
      "epoch": 1.7144135382221357,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004704607634282672,
      "loss": 5.1966,
      "mean_token_accuracy": 0.18162908107042314,
      "num_tokens": 38235721.0,
      "step": 22035
    },
    {
      "entropy": 5.536008262634278,
      "epoch": 1.7148025675938534,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047044703405055765,
      "loss": 5.2701,
      "mean_token_accuracy": 0.17266861498355865,
      "num_tokens": 38244686.0,
      "step": 22040
    },
    {
      "entropy": 5.525236415863037,
      "epoch": 1.7151915969655709,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00047043330170725046,
      "loss": 5.2671,
      "mean_token_accuracy": 0.17743858098983764,
      "num_tokens": 38253609.0,
      "step": 22045
    },
    {
      "entropy": 5.480854845046997,
      "epoch": 1.7155806263372884,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047041956639855406,
      "loss": 5.1883,
      "mean_token_accuracy": 0.17956473380327226,
      "num_tokens": 38261746.0,
      "step": 22050
    },
    {
      "entropy": 5.504291582107544,
      "epoch": 1.715969655709006,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004704058281246766,
      "loss": 5.1998,
      "mean_token_accuracy": 0.18395068347454072,
      "num_tokens": 38270573.0,
      "step": 22055
    },
    {
      "entropy": 5.509825658798218,
      "epoch": 1.7163586850807238,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004703920868858268,
      "loss": 5.209,
      "mean_token_accuracy": 0.18277066498994826,
      "num_tokens": 38279031.0,
      "step": 22060
    },
    {
      "entropy": 5.498215675354004,
      "epoch": 1.716747714452441,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047037834268221315,
      "loss": 5.2314,
      "mean_token_accuracy": 0.17874146848917008,
      "num_tokens": 38287060.0,
      "step": 22065
    },
    {
      "entropy": 5.5415667533874515,
      "epoch": 1.7171367438241587,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004703645955140441,
      "loss": 5.2182,
      "mean_token_accuracy": 0.1846153825521469,
      "num_tokens": 38295192.0,
      "step": 22070
    },
    {
      "entropy": 5.474501466751098,
      "epoch": 1.7175257731958764,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00047035084538152815,
      "loss": 5.1256,
      "mean_token_accuracy": 0.18454911559820175,
      "num_tokens": 38303932.0,
      "step": 22075
    },
    {
      "entropy": 5.502408647537232,
      "epoch": 1.7179148025675939,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004703370922848742,
      "loss": 5.2879,
      "mean_token_accuracy": 0.17059511840343475,
      "num_tokens": 38312906.0,
      "step": 22080
    },
    {
      "entropy": 5.51400785446167,
      "epoch": 1.7183038319393114,
      "grad_norm": 1.484375,
      "learning_rate": 0.00047032333622429074,
      "loss": 5.2646,
      "mean_token_accuracy": 0.17891405075788497,
      "num_tokens": 38322359.0,
      "step": 22085
    },
    {
      "entropy": 5.609464263916015,
      "epoch": 1.718692861311029,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047030957719998656,
      "loss": 5.2912,
      "mean_token_accuracy": 0.17101875692605972,
      "num_tokens": 38331880.0,
      "step": 22090
    },
    {
      "entropy": 5.5549389839172365,
      "epoch": 1.7190818906827465,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004702958152121704,
      "loss": 5.3634,
      "mean_token_accuracy": 0.1668516829609871,
      "num_tokens": 38340763.0,
      "step": 22095
    },
    {
      "entropy": 5.556585216522217,
      "epoch": 1.719470920054464,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004702820502610511,
      "loss": 5.1855,
      "mean_token_accuracy": 0.17876700460910797,
      "num_tokens": 38349427.0,
      "step": 22100
    },
    {
      "entropy": 5.499334669113159,
      "epoch": 1.7198599494261817,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004702682823468375,
      "loss": 5.1542,
      "mean_token_accuracy": 0.18651724606752396,
      "num_tokens": 38358025.0,
      "step": 22105
    },
    {
      "entropy": 5.530140399932861,
      "epoch": 1.7202489787978994,
      "grad_norm": 1.375,
      "learning_rate": 0.00047025451146973844,
      "loss": 5.1656,
      "mean_token_accuracy": 0.17682957351207734,
      "num_tokens": 38366923.0,
      "step": 22110
    },
    {
      "entropy": 5.516496276855468,
      "epoch": 1.7206380081696167,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047024073762996305,
      "loss": 5.2308,
      "mean_token_accuracy": 0.18244261890649796,
      "num_tokens": 38375397.0,
      "step": 22115
    },
    {
      "entropy": 5.4580292224884035,
      "epoch": 1.7210270375413343,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004702269608277202,
      "loss": 5.1358,
      "mean_token_accuracy": 0.18412314355373383,
      "num_tokens": 38384279.0,
      "step": 22120
    },
    {
      "entropy": 5.5016316890716555,
      "epoch": 1.721416066913052,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004702131810632189,
      "loss": 5.1912,
      "mean_token_accuracy": 0.18388418555259706,
      "num_tokens": 38392087.0,
      "step": 22125
    },
    {
      "entropy": 5.500098419189453,
      "epoch": 1.7218050962847695,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047019939833666837,
      "loss": 5.2948,
      "mean_token_accuracy": 0.17404642701148987,
      "num_tokens": 38400616.0,
      "step": 22130
    },
    {
      "entropy": 5.513371467590332,
      "epoch": 1.722194125656487,
      "grad_norm": 1.375,
      "learning_rate": 0.0004701856126482776,
      "loss": 5.184,
      "mean_token_accuracy": 0.1760253816843033,
      "num_tokens": 38408975.0,
      "step": 22135
    },
    {
      "entropy": 5.535154390335083,
      "epoch": 1.7225831550282047,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004701718239982559,
      "loss": 5.304,
      "mean_token_accuracy": 0.1801455423235893,
      "num_tokens": 38417161.0,
      "step": 22140
    },
    {
      "entropy": 5.525936412811279,
      "epoch": 1.7229721843999222,
      "grad_norm": 1.5625,
      "learning_rate": 0.00047015803238681234,
      "loss": 5.2696,
      "mean_token_accuracy": 0.179266294836998,
      "num_tokens": 38425078.0,
      "step": 22145
    },
    {
      "entropy": 5.494637393951416,
      "epoch": 1.7233612137716396,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004701442378141563,
      "loss": 5.1512,
      "mean_token_accuracy": 0.1798102930188179,
      "num_tokens": 38433347.0,
      "step": 22150
    },
    {
      "entropy": 5.558014678955078,
      "epoch": 1.7237502431433573,
      "grad_norm": 1.234375,
      "learning_rate": 0.000470130440280497,
      "loss": 5.2635,
      "mean_token_accuracy": 0.17664334326982498,
      "num_tokens": 38442373.0,
      "step": 22155
    },
    {
      "entropy": 5.548906898498535,
      "epoch": 1.724139272515075,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004701166397860439,
      "loss": 5.3186,
      "mean_token_accuracy": 0.16979156732559203,
      "num_tokens": 38452325.0,
      "step": 22160
    },
    {
      "entropy": 5.578975248336792,
      "epoch": 1.7245283018867923,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004701028363310064,
      "loss": 5.3118,
      "mean_token_accuracy": 0.17591005116701125,
      "num_tokens": 38462317.0,
      "step": 22165
    },
    {
      "entropy": 5.5366353511810305,
      "epoch": 1.72491733125851,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047008902991559385,
      "loss": 5.1478,
      "mean_token_accuracy": 0.18252284079790115,
      "num_tokens": 38471149.0,
      "step": 22170
    },
    {
      "entropy": 5.564240980148315,
      "epoch": 1.7253063606302277,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004700752205400158,
      "loss": 5.2652,
      "mean_token_accuracy": 0.17625696063041688,
      "num_tokens": 38480427.0,
      "step": 22175
    },
    {
      "entropy": 5.452456521987915,
      "epoch": 1.7256953900019452,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047006140820448174,
      "loss": 5.1653,
      "mean_token_accuracy": 0.18430205136537553,
      "num_tokens": 38489283.0,
      "step": 22180
    },
    {
      "entropy": 5.53076663017273,
      "epoch": 1.7260844193736626,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047004759290920136,
      "loss": 5.2096,
      "mean_token_accuracy": 0.1815559223294258,
      "num_tokens": 38497557.0,
      "step": 22185
    },
    {
      "entropy": 5.482972574234009,
      "epoch": 1.7264734487453803,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004700337746543841,
      "loss": 5.2468,
      "mean_token_accuracy": 0.18089222013950348,
      "num_tokens": 38506589.0,
      "step": 22190
    },
    {
      "entropy": 5.491979026794434,
      "epoch": 1.7268624781170978,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004700199534402398,
      "loss": 5.2318,
      "mean_token_accuracy": 0.18370293825864792,
      "num_tokens": 38515016.0,
      "step": 22195
    },
    {
      "entropy": 5.597206020355225,
      "epoch": 1.7272515074888153,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004700061292669781,
      "loss": 5.2663,
      "mean_token_accuracy": 0.1809206947684288,
      "num_tokens": 38523792.0,
      "step": 22200
    },
    {
      "entropy": 5.434280061721802,
      "epoch": 1.727640536860533,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004699923021348088,
      "loss": 5.0711,
      "mean_token_accuracy": 0.18863318711519242,
      "num_tokens": 38532007.0,
      "step": 22205
    },
    {
      "entropy": 5.447651624679565,
      "epoch": 1.7280295662322507,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00046997847204394166,
      "loss": 5.2029,
      "mean_token_accuracy": 0.17935859113931657,
      "num_tokens": 38541044.0,
      "step": 22210
    },
    {
      "entropy": 5.547362852096557,
      "epoch": 1.728418595603968,
      "grad_norm": 1.46875,
      "learning_rate": 0.00046996463899458653,
      "loss": 5.137,
      "mean_token_accuracy": 0.19051269590854644,
      "num_tokens": 38549365.0,
      "step": 22215
    },
    {
      "entropy": 5.478987360000611,
      "epoch": 1.7288076249756856,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004699508029869533,
      "loss": 5.1462,
      "mean_token_accuracy": 0.19089205861091613,
      "num_tokens": 38558289.0,
      "step": 22220
    },
    {
      "entropy": 5.458957481384277,
      "epoch": 1.7291966543474033,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00046993696402125205,
      "loss": 5.2771,
      "mean_token_accuracy": 0.17165924608707428,
      "num_tokens": 38568422.0,
      "step": 22225
    },
    {
      "entropy": 5.53229169845581,
      "epoch": 1.7295856837191208,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004699231220976925,
      "loss": 5.2043,
      "mean_token_accuracy": 0.18020358085632324,
      "num_tokens": 38577009.0,
      "step": 22230
    },
    {
      "entropy": 5.477518129348755,
      "epoch": 1.7299747130908383,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004699092772164849,
      "loss": 5.0347,
      "mean_token_accuracy": 0.19384513199329376,
      "num_tokens": 38584920.0,
      "step": 22235
    },
    {
      "entropy": 5.3449609756469725,
      "epoch": 1.730363742462556,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004698954293778392,
      "loss": 5.1218,
      "mean_token_accuracy": 0.19012910425662993,
      "num_tokens": 38593533.0,
      "step": 22240
    },
    {
      "entropy": 5.405400085449219,
      "epoch": 1.7307527718342735,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00046988157858196555,
      "loss": 5.1222,
      "mean_token_accuracy": 0.18851638436317444,
      "num_tokens": 38602300.0,
      "step": 22245
    },
    {
      "entropy": 5.507129287719726,
      "epoch": 1.731141801205991,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046986772482907426,
      "loss": 5.2196,
      "mean_token_accuracy": 0.18343135416507722,
      "num_tokens": 38610729.0,
      "step": 22250
    },
    {
      "entropy": 5.5146712303161625,
      "epoch": 1.7315308305777086,
      "grad_norm": 3.84375,
      "learning_rate": 0.0004698538681193754,
      "loss": 5.0851,
      "mean_token_accuracy": 0.18869762420654296,
      "num_tokens": 38618192.0,
      "step": 22255
    },
    {
      "entropy": 5.419455528259277,
      "epoch": 1.7319198599494263,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046984000845307907,
      "loss": 5.1844,
      "mean_token_accuracy": 0.18631936907768248,
      "num_tokens": 38627030.0,
      "step": 22260
    },
    {
      "entropy": 5.482566833496094,
      "epoch": 1.7323088893211438,
      "grad_norm": 1.40625,
      "learning_rate": 0.00046982614583039584,
      "loss": 5.2838,
      "mean_token_accuracy": 0.17830957919359208,
      "num_tokens": 38635761.0,
      "step": 22265
    },
    {
      "entropy": 5.4713677883148195,
      "epoch": 1.7326979186928613,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004698122802515359,
      "loss": 5.1944,
      "mean_token_accuracy": 0.1820732206106186,
      "num_tokens": 38643952.0,
      "step": 22270
    },
    {
      "entropy": 5.499280500411987,
      "epoch": 1.733086948064579,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004697984117167097,
      "loss": 5.1417,
      "mean_token_accuracy": 0.1906971201300621,
      "num_tokens": 38651888.0,
      "step": 22275
    },
    {
      "entropy": 5.505435848236084,
      "epoch": 1.7334759774362964,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046978454022612767,
      "loss": 5.2712,
      "mean_token_accuracy": 0.1786338835954666,
      "num_tokens": 38660466.0,
      "step": 22280
    },
    {
      "entropy": 5.571185445785522,
      "epoch": 1.733865006808014,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00046977066578000025,
      "loss": 5.2213,
      "mean_token_accuracy": 0.18112041056156158,
      "num_tokens": 38668429.0,
      "step": 22285
    },
    {
      "entropy": 5.504121255874634,
      "epoch": 1.7342540361797316,
      "grad_norm": 2.578125,
      "learning_rate": 0.00046975678837853806,
      "loss": 5.2393,
      "mean_token_accuracy": 0.18523801118135452,
      "num_tokens": 38676391.0,
      "step": 22290
    },
    {
      "entropy": 5.515945100784302,
      "epoch": 1.734643065551449,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046974290802195156,
      "loss": 5.2463,
      "mean_token_accuracy": 0.184119313955307,
      "num_tokens": 38684662.0,
      "step": 22295
    },
    {
      "entropy": 5.484704446792603,
      "epoch": 1.7350320949231666,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004697290247104513,
      "loss": 5.1657,
      "mean_token_accuracy": 0.18281205892562866,
      "num_tokens": 38693497.0,
      "step": 22300
    },
    {
      "entropy": 5.408651876449585,
      "epoch": 1.7354211242948843,
      "grad_norm": 1.984375,
      "learning_rate": 0.00046971513844424814,
      "loss": 5.0972,
      "mean_token_accuracy": 0.18519675880670547,
      "num_tokens": 38701898.0,
      "step": 22305
    },
    {
      "entropy": 5.551790380477906,
      "epoch": 1.735810153666602,
      "grad_norm": 1.40625,
      "learning_rate": 0.00046970124922355265,
      "loss": 5.2179,
      "mean_token_accuracy": 0.17562125772237777,
      "num_tokens": 38710910.0,
      "step": 22310
    },
    {
      "entropy": 5.5886858940124515,
      "epoch": 1.7361991830383194,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004696873570485756,
      "loss": 5.2399,
      "mean_token_accuracy": 0.1784719482064247,
      "num_tokens": 38719612.0,
      "step": 22315
    },
    {
      "entropy": 5.356118202209473,
      "epoch": 1.736588212410037,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004696734619195278,
      "loss": 5.2285,
      "mean_token_accuracy": 0.17571593225002288,
      "num_tokens": 38728332.0,
      "step": 22320
    },
    {
      "entropy": 5.470266437530517,
      "epoch": 1.7369772417817546,
      "grad_norm": 1.25,
      "learning_rate": 0.00046965956383662,
      "loss": 5.229,
      "mean_token_accuracy": 0.17644769251346587,
      "num_tokens": 38736830.0,
      "step": 22325
    },
    {
      "entropy": 5.559103775024414,
      "epoch": 1.737366271153472,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004696456628000632,
      "loss": 5.1997,
      "mean_token_accuracy": 0.18392471969127655,
      "num_tokens": 38745972.0,
      "step": 22330
    },
    {
      "entropy": 5.447118139266967,
      "epoch": 1.7377553005251896,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046963175881006825,
      "loss": 5.0933,
      "mean_token_accuracy": 0.1814733162522316,
      "num_tokens": 38754287.0,
      "step": 22335
    },
    {
      "entropy": 5.391161632537842,
      "epoch": 1.7381443298969073,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004696178518668462,
      "loss": 5.1891,
      "mean_token_accuracy": 0.18238743394613266,
      "num_tokens": 38763212.0,
      "step": 22340
    },
    {
      "entropy": 5.408348035812378,
      "epoch": 1.7385333592686247,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046960394197060804,
      "loss": 5.0209,
      "mean_token_accuracy": 0.1893278107047081,
      "num_tokens": 38771668.0,
      "step": 22345
    },
    {
      "entropy": 5.5382458686828615,
      "epoch": 1.7389223886403422,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046959002912156473,
      "loss": 5.2641,
      "mean_token_accuracy": 0.1779286101460457,
      "num_tokens": 38781184.0,
      "step": 22350
    },
    {
      "entropy": 5.463222551345825,
      "epoch": 1.73931141801206,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004695761133199275,
      "loss": 5.0973,
      "mean_token_accuracy": 0.19137011766433715,
      "num_tokens": 38789622.0,
      "step": 22355
    },
    {
      "entropy": 5.506410932540893,
      "epoch": 1.7397004473837776,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004695621945659074,
      "loss": 5.1774,
      "mean_token_accuracy": 0.18483394682407378,
      "num_tokens": 38798250.0,
      "step": 22360
    },
    {
      "entropy": 5.504396533966064,
      "epoch": 1.740089476755495,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004695482728597157,
      "loss": 5.199,
      "mean_token_accuracy": 0.18315800428390502,
      "num_tokens": 38807724.0,
      "step": 22365
    },
    {
      "entropy": 5.503889656066894,
      "epoch": 1.7404785061272126,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046953434820156363,
      "loss": 5.237,
      "mean_token_accuracy": 0.18558563888072968,
      "num_tokens": 38816017.0,
      "step": 22370
    },
    {
      "entropy": 5.538943910598755,
      "epoch": 1.7408675354989303,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004695204205916624,
      "loss": 5.1541,
      "mean_token_accuracy": 0.19018099904060365,
      "num_tokens": 38824976.0,
      "step": 22375
    },
    {
      "entropy": 5.439003801345825,
      "epoch": 1.7412565648706477,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004695064900302235,
      "loss": 5.1473,
      "mean_token_accuracy": 0.18844833970069885,
      "num_tokens": 38834358.0,
      "step": 22380
    },
    {
      "entropy": 5.478014039993286,
      "epoch": 1.7416455942423652,
      "grad_norm": 1.375,
      "learning_rate": 0.0004694925565174581,
      "loss": 5.2348,
      "mean_token_accuracy": 0.1731978103518486,
      "num_tokens": 38842801.0,
      "step": 22385
    },
    {
      "entropy": 5.52531156539917,
      "epoch": 1.742034623614083,
      "grad_norm": 1.25,
      "learning_rate": 0.00046947862005357777,
      "loss": 5.2576,
      "mean_token_accuracy": 0.17530435025691987,
      "num_tokens": 38851335.0,
      "step": 22390
    },
    {
      "entropy": 5.575278425216675,
      "epoch": 1.7424236529858004,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004694646806387939,
      "loss": 5.2511,
      "mean_token_accuracy": 0.17375997006893157,
      "num_tokens": 38860074.0,
      "step": 22395
    },
    {
      "entropy": 5.494519662857056,
      "epoch": 1.7428126823575179,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004694507382733181,
      "loss": 5.1679,
      "mean_token_accuracy": 0.18731395602226258,
      "num_tokens": 38868319.0,
      "step": 22400
    },
    {
      "entropy": 5.441346216201782,
      "epoch": 1.7432017117292355,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004694367929573618,
      "loss": 5.2222,
      "mean_token_accuracy": 0.18582238852977753,
      "num_tokens": 38876971.0,
      "step": 22405
    },
    {
      "entropy": 5.547492027282715,
      "epoch": 1.7435907411009532,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004694228446911367,
      "loss": 5.3315,
      "mean_token_accuracy": 0.16878755688667296,
      "num_tokens": 38886032.0,
      "step": 22410
    },
    {
      "entropy": 5.583523797988891,
      "epoch": 1.7439797704726707,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004694088934748542,
      "loss": 5.197,
      "mean_token_accuracy": 0.18434173315763475,
      "num_tokens": 38894505.0,
      "step": 22415
    },
    {
      "entropy": 5.393583488464356,
      "epoch": 1.7443687998443882,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004693949393087263,
      "loss": 5.0982,
      "mean_token_accuracy": 0.18977867513895036,
      "num_tokens": 38903450.0,
      "step": 22420
    },
    {
      "entropy": 5.474939775466919,
      "epoch": 1.744757829216106,
      "grad_norm": 1.5,
      "learning_rate": 0.0004693809821929647,
      "loss": 5.1824,
      "mean_token_accuracy": 0.18529875129461287,
      "num_tokens": 38911326.0,
      "step": 22425
    },
    {
      "entropy": 5.499625587463379,
      "epoch": 1.7451468585878234,
      "grad_norm": 1.484375,
      "learning_rate": 0.000469367022127781,
      "loss": 5.1559,
      "mean_token_accuracy": 0.18469337821006776,
      "num_tokens": 38920035.0,
      "step": 22430
    },
    {
      "entropy": 5.5433837890625,
      "epoch": 1.7455358879595408,
      "grad_norm": 1.40625,
      "learning_rate": 0.00046935305911338703,
      "loss": 5.2099,
      "mean_token_accuracy": 0.1855094775557518,
      "num_tokens": 38928542.0,
      "step": 22435
    },
    {
      "entropy": 5.521333980560303,
      "epoch": 1.7459249173312585,
      "grad_norm": 1.375,
      "learning_rate": 0.0004693390931499948,
      "loss": 5.1954,
      "mean_token_accuracy": 0.18716973960399627,
      "num_tokens": 38937486.0,
      "step": 22440
    },
    {
      "entropy": 5.436490297317505,
      "epoch": 1.7463139467029762,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004693251242378162,
      "loss": 5.1272,
      "mean_token_accuracy": 0.1873105689883232,
      "num_tokens": 38946171.0,
      "step": 22445
    },
    {
      "entropy": 5.436745452880859,
      "epoch": 1.7467029760746935,
      "grad_norm": 1.453125,
      "learning_rate": 0.00046931115237706304,
      "loss": 5.1178,
      "mean_token_accuracy": 0.18590092658996582,
      "num_tokens": 38954214.0,
      "step": 22450
    },
    {
      "entropy": 5.598425674438476,
      "epoch": 1.7470920054464112,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004692971775679475,
      "loss": 5.2671,
      "mean_token_accuracy": 0.1758560135960579,
      "num_tokens": 38962102.0,
      "step": 22455
    },
    {
      "entropy": 5.5146181106567385,
      "epoch": 1.7474810348181289,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00046928319981068154,
      "loss": 5.1608,
      "mean_token_accuracy": 0.18913196325302123,
      "num_tokens": 38970472.0,
      "step": 22460
    },
    {
      "entropy": 5.461085557937622,
      "epoch": 1.7478700641898464,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004692692191054773,
      "loss": 5.1618,
      "mean_token_accuracy": 0.18234552890062333,
      "num_tokens": 38979151.0,
      "step": 22465
    },
    {
      "entropy": 5.5228314876556395,
      "epoch": 1.7482590935615638,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004692552354525468,
      "loss": 5.1692,
      "mean_token_accuracy": 0.1873565971851349,
      "num_tokens": 38987044.0,
      "step": 22470
    },
    {
      "entropy": 5.495915079116822,
      "epoch": 1.7486481229332815,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004692412488521023,
      "loss": 5.2679,
      "mean_token_accuracy": 0.1838843435049057,
      "num_tokens": 38995882.0,
      "step": 22475
    },
    {
      "entropy": 5.477252435684204,
      "epoch": 1.749037152304999,
      "grad_norm": 1.796875,
      "learning_rate": 0.000469227259304356,
      "loss": 5.2418,
      "mean_token_accuracy": 0.17672751396894454,
      "num_tokens": 39004456.0,
      "step": 22480
    },
    {
      "entropy": 5.532679891586303,
      "epoch": 1.7494261816767165,
      "grad_norm": 1.53125,
      "learning_rate": 0.00046921326680952023,
      "loss": 5.2262,
      "mean_token_accuracy": 0.18299031853675843,
      "num_tokens": 39013363.0,
      "step": 22485
    },
    {
      "entropy": 5.520345544815063,
      "epoch": 1.7498152110484342,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004691992713678072,
      "loss": 5.1744,
      "mean_token_accuracy": 0.18481789976358415,
      "num_tokens": 39022116.0,
      "step": 22490
    },
    {
      "entropy": 5.517253398895264,
      "epoch": 1.7502042404201519,
      "grad_norm": 1.375,
      "learning_rate": 0.0004691852729794293,
      "loss": 5.2093,
      "mean_token_accuracy": 0.18476366698741914,
      "num_tokens": 39030737.0,
      "step": 22495
    },
    {
      "entropy": 5.454028797149658,
      "epoch": 1.7505932697918691,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004691712716445991,
      "loss": 5.1222,
      "mean_token_accuracy": 0.18870659172534943,
      "num_tokens": 39039511.0,
      "step": 22500
    },
    {
      "entropy": 5.481350231170654,
      "epoch": 1.7509822991635868,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004691572673635288,
      "loss": 5.1309,
      "mean_token_accuracy": 0.18534973114728928,
      "num_tokens": 39048745.0,
      "step": 22505
    },
    {
      "entropy": 5.49375638961792,
      "epoch": 1.7513713285353045,
      "grad_norm": 1.484375,
      "learning_rate": 0.000469143260136431,
      "loss": 5.2153,
      "mean_token_accuracy": 0.1792880430817604,
      "num_tokens": 39057311.0,
      "step": 22510
    },
    {
      "entropy": 5.527177715301514,
      "epoch": 1.751760357907022,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004691292499635183,
      "loss": 5.2139,
      "mean_token_accuracy": 0.18225941956043243,
      "num_tokens": 39065630.0,
      "step": 22515
    },
    {
      "entropy": 5.495059776306152,
      "epoch": 1.7521493872787395,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004691152368450032,
      "loss": 5.177,
      "mean_token_accuracy": 0.1807619497179985,
      "num_tokens": 39074284.0,
      "step": 22520
    },
    {
      "entropy": 5.492371320724487,
      "epoch": 1.7525384166504572,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046910122078109835,
      "loss": 5.2081,
      "mean_token_accuracy": 0.18475724905729293,
      "num_tokens": 39083284.0,
      "step": 22525
    },
    {
      "entropy": 5.487291622161865,
      "epoch": 1.7529274460221747,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004690872017720163,
      "loss": 5.1628,
      "mean_token_accuracy": 0.18781275004148484,
      "num_tokens": 39091106.0,
      "step": 22530
    },
    {
      "entropy": 5.407898187637329,
      "epoch": 1.7533164753938921,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046907317981797006,
      "loss": 5.1078,
      "mean_token_accuracy": 0.19382071644067764,
      "num_tokens": 39100499.0,
      "step": 22535
    },
    {
      "entropy": 5.4645740509033205,
      "epoch": 1.7537055047656098,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00046905915491917213,
      "loss": 5.2208,
      "mean_token_accuracy": 0.190023709833622,
      "num_tokens": 39108895.0,
      "step": 22540
    },
    {
      "entropy": 5.429339456558227,
      "epoch": 1.7540945341373275,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004690451270758354,
      "loss": 5.0473,
      "mean_token_accuracy": 0.19826958626508712,
      "num_tokens": 39117053.0,
      "step": 22545
    },
    {
      "entropy": 5.490063762664795,
      "epoch": 1.7544835635090448,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00046903109628817276,
      "loss": 5.1995,
      "mean_token_accuracy": 0.18657177239656447,
      "num_tokens": 39125750.0,
      "step": 22550
    },
    {
      "entropy": 5.435941076278686,
      "epoch": 1.7548725928807625,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046901706255639703,
      "loss": 5.0856,
      "mean_token_accuracy": 0.18898576200008393,
      "num_tokens": 39134105.0,
      "step": 22555
    },
    {
      "entropy": 5.6019457340240475,
      "epoch": 1.7552616222524802,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004690030258807212,
      "loss": 5.3492,
      "mean_token_accuracy": 0.17937949150800706,
      "num_tokens": 39143543.0,
      "step": 22560
    },
    {
      "entropy": 5.595874404907226,
      "epoch": 1.7556506516241976,
      "grad_norm": 3.203125,
      "learning_rate": 0.00046898898626135827,
      "loss": 5.2413,
      "mean_token_accuracy": 0.18138222843408586,
      "num_tokens": 39152826.0,
      "step": 22565
    },
    {
      "entropy": 5.5266258239746096,
      "epoch": 1.7560396809959151,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004689749436985211,
      "loss": 5.1481,
      "mean_token_accuracy": 0.18144762068986892,
      "num_tokens": 39160999.0,
      "step": 22570
    },
    {
      "entropy": 5.496015787124634,
      "epoch": 1.7564287103676328,
      "grad_norm": 1.453125,
      "learning_rate": 0.00046896089819242304,
      "loss": 5.1942,
      "mean_token_accuracy": 0.18176806569099427,
      "num_tokens": 39169134.0,
      "step": 22575
    },
    {
      "entropy": 5.460223579406739,
      "epoch": 1.7568177397393503,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004689468497432771,
      "loss": 5.1316,
      "mean_token_accuracy": 0.18542095720767976,
      "num_tokens": 39177545.0,
      "step": 22580
    },
    {
      "entropy": 5.441378879547119,
      "epoch": 1.7572067691110678,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004689327983512963,
      "loss": 5.0744,
      "mean_token_accuracy": 0.18980902731418609,
      "num_tokens": 39185671.0,
      "step": 22585
    },
    {
      "entropy": 5.54808406829834,
      "epoch": 1.7575957984827855,
      "grad_norm": 1.59375,
      "learning_rate": 0.00046891874401669404,
      "loss": 5.2269,
      "mean_token_accuracy": 0.17572949230670928,
      "num_tokens": 39194420.0,
      "step": 22590
    },
    {
      "entropy": 5.470951938629151,
      "epoch": 1.7579848278545032,
      "grad_norm": 1.375,
      "learning_rate": 0.0004689046867396834,
      "loss": 5.1883,
      "mean_token_accuracy": 0.1830612137913704,
      "num_tokens": 39203295.0,
      "step": 22595
    },
    {
      "entropy": 5.464349031448364,
      "epoch": 1.7583738572262204,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004688906265204779,
      "loss": 5.1586,
      "mean_token_accuracy": 0.18421381413936616,
      "num_tokens": 39211838.0,
      "step": 22600
    },
    {
      "entropy": 5.514088821411133,
      "epoch": 1.7587628865979381,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004688765633592907,
      "loss": 5.1161,
      "mean_token_accuracy": 0.1813815116882324,
      "num_tokens": 39219842.0,
      "step": 22605
    },
    {
      "entropy": 5.412389659881592,
      "epoch": 1.7591519159696558,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004688624972563352,
      "loss": 5.1958,
      "mean_token_accuracy": 0.18454648107290267,
      "num_tokens": 39228167.0,
      "step": 22610
    },
    {
      "entropy": 5.497512006759644,
      "epoch": 1.7595409453413733,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004688484282118249,
      "loss": 5.2265,
      "mean_token_accuracy": 0.1761507660150528,
      "num_tokens": 39236546.0,
      "step": 22615
    },
    {
      "entropy": 5.5328000545501705,
      "epoch": 1.7599299747130908,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004688343562259732,
      "loss": 5.1581,
      "mean_token_accuracy": 0.18949703574180604,
      "num_tokens": 39244212.0,
      "step": 22620
    },
    {
      "entropy": 5.424105262756347,
      "epoch": 1.7603190040848085,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004688202812989937,
      "loss": 5.1137,
      "mean_token_accuracy": 0.18745281398296357,
      "num_tokens": 39252892.0,
      "step": 22625
    },
    {
      "entropy": 5.412605905532837,
      "epoch": 1.760708033456526,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004688062034311,
      "loss": 5.1206,
      "mean_token_accuracy": 0.18772437125444413,
      "num_tokens": 39262261.0,
      "step": 22630
    },
    {
      "entropy": 5.517810869216919,
      "epoch": 1.7610970628282434,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00046879212262250546,
      "loss": 5.1548,
      "mean_token_accuracy": 0.18764956146478654,
      "num_tokens": 39270790.0,
      "step": 22635
    },
    {
      "entropy": 5.462061357498169,
      "epoch": 1.761486092199961,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00046877803887342406,
      "loss": 5.1739,
      "mean_token_accuracy": 0.1800505116581917,
      "num_tokens": 39278825.0,
      "step": 22640
    },
    {
      "entropy": 5.526913261413574,
      "epoch": 1.7618751215716788,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004687639521840693,
      "loss": 5.1968,
      "mean_token_accuracy": 0.18272940814495087,
      "num_tokens": 39288343.0,
      "step": 22645
    },
    {
      "entropy": 5.580000877380371,
      "epoch": 1.7622641509433963,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046874986255465495,
      "loss": 5.2227,
      "mean_token_accuracy": 0.17613643109798433,
      "num_tokens": 39296562.0,
      "step": 22650
    },
    {
      "entropy": 5.541043996810913,
      "epoch": 1.7626531803151138,
      "grad_norm": 1.390625,
      "learning_rate": 0.00046873576998539485,
      "loss": 5.1769,
      "mean_token_accuracy": 0.1873554453253746,
      "num_tokens": 39305433.0,
      "step": 22655
    },
    {
      "entropy": 5.422834300994873,
      "epoch": 1.7630422096868315,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004687216744765028,
      "loss": 5.2317,
      "mean_token_accuracy": 0.1813822418451309,
      "num_tokens": 39313244.0,
      "step": 22660
    },
    {
      "entropy": 5.529487085342407,
      "epoch": 1.763431239058549,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004687075760281927,
      "loss": 5.269,
      "mean_token_accuracy": 0.1812152922153473,
      "num_tokens": 39321933.0,
      "step": 22665
    },
    {
      "entropy": 5.576750469207764,
      "epoch": 1.7638202684302664,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004686934746406784,
      "loss": 5.2517,
      "mean_token_accuracy": 0.18140339553356172,
      "num_tokens": 39331194.0,
      "step": 22670
    },
    {
      "entropy": 5.5439536571502686,
      "epoch": 1.764209297801984,
      "grad_norm": 1.421875,
      "learning_rate": 0.00046867937031417397,
      "loss": 5.1582,
      "mean_token_accuracy": 0.1755712017416954,
      "num_tokens": 39339762.0,
      "step": 22675
    },
    {
      "entropy": 5.508156824111938,
      "epoch": 1.7645983271737016,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004686652630488933,
      "loss": 5.2376,
      "mean_token_accuracy": 0.1777756154537201,
      "num_tokens": 39349020.0,
      "step": 22680
    },
    {
      "entropy": 5.558917760848999,
      "epoch": 1.764987356545419,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046865115284505063,
      "loss": 5.3292,
      "mean_token_accuracy": 0.1744164377450943,
      "num_tokens": 39357889.0,
      "step": 22685
    },
    {
      "entropy": 5.5419670104980465,
      "epoch": 1.7653763859171367,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00046863703970285986,
      "loss": 5.1779,
      "mean_token_accuracy": 0.18267302364110946,
      "num_tokens": 39366120.0,
      "step": 22690
    },
    {
      "entropy": 5.510168504714966,
      "epoch": 1.7657654152888544,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004686229236225352,
      "loss": 5.0463,
      "mean_token_accuracy": 0.1917452484369278,
      "num_tokens": 39374056.0,
      "step": 22695
    },
    {
      "entropy": 5.514323616027832,
      "epoch": 1.766154444660572,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004686088046042909,
      "loss": 5.243,
      "mean_token_accuracy": 0.18376550674438477,
      "num_tokens": 39383170.0,
      "step": 22700
    },
    {
      "entropy": 5.514267206192017,
      "epoch": 1.7665434740322894,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046859468264834114,
      "loss": 5.2028,
      "mean_token_accuracy": 0.17862859666347503,
      "num_tokens": 39391599.0,
      "step": 22705
    },
    {
      "entropy": 5.510101461410523,
      "epoch": 1.766932503404007,
      "grad_norm": 1.359375,
      "learning_rate": 0.00046858055775490017,
      "loss": 5.2653,
      "mean_token_accuracy": 0.17287039011716843,
      "num_tokens": 39400217.0,
      "step": 22710
    },
    {
      "entropy": 5.522445344924927,
      "epoch": 1.7673215327757246,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004685664299241825,
      "loss": 5.2213,
      "mean_token_accuracy": 0.1826951175928116,
      "num_tokens": 39408943.0,
      "step": 22715
    },
    {
      "entropy": 5.5075657844543455,
      "epoch": 1.767710562147442,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004685522991564022,
      "loss": 5.1799,
      "mean_token_accuracy": 0.17735762000083924,
      "num_tokens": 39417361.0,
      "step": 22720
    },
    {
      "entropy": 5.349432849884034,
      "epoch": 1.7680995915191597,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004685381654517738,
      "loss": 4.9576,
      "mean_token_accuracy": 0.19321755021810533,
      "num_tokens": 39425896.0,
      "step": 22725
    },
    {
      "entropy": 5.423156452178955,
      "epoch": 1.7684886208908772,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004685240288105119,
      "loss": 5.1905,
      "mean_token_accuracy": 0.18617706149816513,
      "num_tokens": 39434719.0,
      "step": 22730
    },
    {
      "entropy": 5.592346811294556,
      "epoch": 1.7688776502625947,
      "grad_norm": 2.046875,
      "learning_rate": 0.00046850988923283085,
      "loss": 5.1784,
      "mean_token_accuracy": 0.1781913235783577,
      "num_tokens": 39443731.0,
      "step": 22735
    },
    {
      "entropy": 5.548964405059815,
      "epoch": 1.7692666796343124,
      "grad_norm": 1.46875,
      "learning_rate": 0.00046849574671894523,
      "loss": 5.2615,
      "mean_token_accuracy": 0.1811185136437416,
      "num_tokens": 39452430.0,
      "step": 22740
    },
    {
      "entropy": 5.468866395950317,
      "epoch": 1.76965570900603,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00046848160126906956,
      "loss": 5.0788,
      "mean_token_accuracy": 0.191403166949749,
      "num_tokens": 39460560.0,
      "step": 22745
    },
    {
      "entropy": 5.48011884689331,
      "epoch": 1.7700447383777476,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004684674528834186,
      "loss": 5.2173,
      "mean_token_accuracy": 0.17509957253932953,
      "num_tokens": 39469575.0,
      "step": 22750
    },
    {
      "entropy": 5.486629152297974,
      "epoch": 1.770433767749465,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00046845330156220703,
      "loss": 5.1754,
      "mean_token_accuracy": 0.18481565713882447,
      "num_tokens": 39478297.0,
      "step": 22755
    },
    {
      "entropy": 5.554228639602661,
      "epoch": 1.7708227971211827,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004684391473056495,
      "loss": 5.1999,
      "mean_token_accuracy": 0.17505336850881575,
      "num_tokens": 39486947.0,
      "step": 22760
    },
    {
      "entropy": 5.554534959793091,
      "epoch": 1.7712118264929002,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00046842499011396076,
      "loss": 5.193,
      "mean_token_accuracy": 0.18447613716125488,
      "num_tokens": 39495649.0,
      "step": 22765
    },
    {
      "entropy": 5.419743442535401,
      "epoch": 1.7716008558646177,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00046841082998735575,
      "loss": 5.1744,
      "mean_token_accuracy": 0.18257228285074234,
      "num_tokens": 39504733.0,
      "step": 22770
    },
    {
      "entropy": 5.422462463378906,
      "epoch": 1.7719898852363354,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00046839666692604916,
      "loss": 5.1816,
      "mean_token_accuracy": 0.1864844396710396,
      "num_tokens": 39513007.0,
      "step": 22775
    },
    {
      "entropy": 5.541841173171997,
      "epoch": 1.7723789146080529,
      "grad_norm": 1.359375,
      "learning_rate": 0.000468382500930256,
      "loss": 5.2592,
      "mean_token_accuracy": 0.18428375869989394,
      "num_tokens": 39521734.0,
      "step": 22780
    },
    {
      "entropy": 5.574861764907837,
      "epoch": 1.7727679439797703,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046836833200019116,
      "loss": 5.2422,
      "mean_token_accuracy": 0.17813752144575118,
      "num_tokens": 39530525.0,
      "step": 22785
    },
    {
      "entropy": 5.566452264785767,
      "epoch": 1.773156973351488,
      "grad_norm": 1.25,
      "learning_rate": 0.0004683541601360697,
      "loss": 5.1746,
      "mean_token_accuracy": 0.19332297891378403,
      "num_tokens": 39538938.0,
      "step": 22790
    },
    {
      "entropy": 5.428329515457153,
      "epoch": 1.7735460027232057,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046833998533810653,
      "loss": 5.0483,
      "mean_token_accuracy": 0.19870121479034425,
      "num_tokens": 39547342.0,
      "step": 22795
    },
    {
      "entropy": 5.4057197093963625,
      "epoch": 1.7739350320949232,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004683258076065169,
      "loss": 5.1484,
      "mean_token_accuracy": 0.17728945463895798,
      "num_tokens": 39555107.0,
      "step": 22800
    },
    {
      "entropy": 5.552070045471192,
      "epoch": 1.7743240614666407,
      "grad_norm": 1.359375,
      "learning_rate": 0.00046831162694151586,
      "loss": 5.1907,
      "mean_token_accuracy": 0.17948800474405288,
      "num_tokens": 39563226.0,
      "step": 22805
    },
    {
      "entropy": 5.5458996295928955,
      "epoch": 1.7747130908383584,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00046829744334331855,
      "loss": 5.2526,
      "mean_token_accuracy": 0.17849595546722413,
      "num_tokens": 39571689.0,
      "step": 22810
    },
    {
      "entropy": 5.43891487121582,
      "epoch": 1.7751021202100759,
      "grad_norm": 1.40625,
      "learning_rate": 0.00046828325681214013,
      "loss": 5.1281,
      "mean_token_accuracy": 0.1900151178240776,
      "num_tokens": 39579985.0,
      "step": 22815
    },
    {
      "entropy": 5.496083498001099,
      "epoch": 1.7754911495817933,
      "grad_norm": 1.4375,
      "learning_rate": 0.000468269067348196,
      "loss": 5.214,
      "mean_token_accuracy": 0.1786200374364853,
      "num_tokens": 39587892.0,
      "step": 22820
    },
    {
      "entropy": 5.525401544570923,
      "epoch": 1.775880178953511,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004682548749517014,
      "loss": 5.2747,
      "mean_token_accuracy": 0.17936124205589293,
      "num_tokens": 39597516.0,
      "step": 22825
    },
    {
      "entropy": 5.560754442214966,
      "epoch": 1.7762692083252287,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046824067962287163,
      "loss": 5.1933,
      "mean_token_accuracy": 0.18506172597408294,
      "num_tokens": 39605860.0,
      "step": 22830
    },
    {
      "entropy": 5.515400123596192,
      "epoch": 1.776658237696946,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004682264813619221,
      "loss": 5.2599,
      "mean_token_accuracy": 0.1773095592856407,
      "num_tokens": 39614915.0,
      "step": 22835
    },
    {
      "entropy": 5.579865598678589,
      "epoch": 1.7770472670686637,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00046821228016906836,
      "loss": 5.2239,
      "mean_token_accuracy": 0.17707508504390718,
      "num_tokens": 39623588.0,
      "step": 22840
    },
    {
      "entropy": 5.543531656265259,
      "epoch": 1.7774362964403814,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004681980760445257,
      "loss": 5.2679,
      "mean_token_accuracy": 0.17618009597063064,
      "num_tokens": 39632351.0,
      "step": 22845
    },
    {
      "entropy": 5.481593418121338,
      "epoch": 1.7778253258120988,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004681838689885098,
      "loss": 5.2033,
      "mean_token_accuracy": 0.1831686705350876,
      "num_tokens": 39641053.0,
      "step": 22850
    },
    {
      "entropy": 5.413815546035766,
      "epoch": 1.7782143551838163,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004681696590012362,
      "loss": 5.0452,
      "mean_token_accuracy": 0.18759573251008987,
      "num_tokens": 39649838.0,
      "step": 22855
    },
    {
      "entropy": 5.44483118057251,
      "epoch": 1.778603384555534,
      "grad_norm": 1.25,
      "learning_rate": 0.00046815544608292043,
      "loss": 5.1042,
      "mean_token_accuracy": 0.18544670045375825,
      "num_tokens": 39658886.0,
      "step": 22860
    },
    {
      "entropy": 5.5596668243408205,
      "epoch": 1.7789924139272515,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004681412302337782,
      "loss": 5.251,
      "mean_token_accuracy": 0.1861257806420326,
      "num_tokens": 39667279.0,
      "step": 22865
    },
    {
      "entropy": 5.47145447731018,
      "epoch": 1.779381443298969,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004681270114540252,
      "loss": 5.1637,
      "mean_token_accuracy": 0.17995944023132324,
      "num_tokens": 39676213.0,
      "step": 22870
    },
    {
      "entropy": 5.480303144454956,
      "epoch": 1.7797704726706867,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004681127897438772,
      "loss": 5.1823,
      "mean_token_accuracy": 0.18513032644987107,
      "num_tokens": 39685859.0,
      "step": 22875
    },
    {
      "entropy": 5.470049285888672,
      "epoch": 1.7801595020424044,
      "grad_norm": 1.515625,
      "learning_rate": 0.00046809856510355007,
      "loss": 5.1297,
      "mean_token_accuracy": 0.1814592659473419,
      "num_tokens": 39695240.0,
      "step": 22880
    },
    {
      "entropy": 5.456104183197022,
      "epoch": 1.7805485314141216,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004680843375332595,
      "loss": 5.1702,
      "mean_token_accuracy": 0.18085871934890746,
      "num_tokens": 39704441.0,
      "step": 22885
    },
    {
      "entropy": 5.558865356445312,
      "epoch": 1.7809375607858393,
      "grad_norm": 1.25,
      "learning_rate": 0.00046807010703322143,
      "loss": 5.2474,
      "mean_token_accuracy": 0.1789546474814415,
      "num_tokens": 39714074.0,
      "step": 22890
    },
    {
      "entropy": 5.577106094360351,
      "epoch": 1.781326590157557,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004680558736036518,
      "loss": 5.2467,
      "mean_token_accuracy": 0.1729050636291504,
      "num_tokens": 39722527.0,
      "step": 22895
    },
    {
      "entropy": 5.4906635761260985,
      "epoch": 1.7817156195292745,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004680416372447666,
      "loss": 5.2061,
      "mean_token_accuracy": 0.18229880183935165,
      "num_tokens": 39730983.0,
      "step": 22900
    },
    {
      "entropy": 5.54629373550415,
      "epoch": 1.782104648900992,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046802739795678177,
      "loss": 5.268,
      "mean_token_accuracy": 0.17313648909330367,
      "num_tokens": 39741114.0,
      "step": 22905
    },
    {
      "entropy": 5.573707532882691,
      "epoch": 1.7824936782727097,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046801315573991346,
      "loss": 5.1811,
      "mean_token_accuracy": 0.17849989533424376,
      "num_tokens": 39749281.0,
      "step": 22910
    },
    {
      "entropy": 5.503108787536621,
      "epoch": 1.7828827076444271,
      "grad_norm": 1.359375,
      "learning_rate": 0.00046799891059437764,
      "loss": 5.0259,
      "mean_token_accuracy": 0.1950564280152321,
      "num_tokens": 39758136.0,
      "step": 22915
    },
    {
      "entropy": 5.43076810836792,
      "epoch": 1.7832717370161446,
      "grad_norm": 4.0,
      "learning_rate": 0.00046798466252039056,
      "loss": 5.2394,
      "mean_token_accuracy": 0.18145583122968673,
      "num_tokens": 39767104.0,
      "step": 22920
    },
    {
      "entropy": 5.493445348739624,
      "epoch": 1.7836607663878623,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046797041151816845,
      "loss": 5.1866,
      "mean_token_accuracy": 0.1837488204240799,
      "num_tokens": 39775979.0,
      "step": 22925
    },
    {
      "entropy": 5.4416172981262205,
      "epoch": 1.78404979575958,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00046795615758792747,
      "loss": 5.0824,
      "mean_token_accuracy": 0.18693752437829972,
      "num_tokens": 39783740.0,
      "step": 22930
    },
    {
      "entropy": 5.483373308181763,
      "epoch": 1.7844388251312973,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004679419007298839,
      "loss": 5.2096,
      "mean_token_accuracy": 0.18720052540302276,
      "num_tokens": 39791836.0,
      "step": 22935
    },
    {
      "entropy": 5.499473667144775,
      "epoch": 1.784827854503015,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004679276409442541,
      "loss": 5.1021,
      "mean_token_accuracy": 0.18431679755449296,
      "num_tokens": 39800694.0,
      "step": 22940
    },
    {
      "entropy": 5.445596694946289,
      "epoch": 1.7852168838747327,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004679133782312544,
      "loss": 5.2307,
      "mean_token_accuracy": 0.18080616295337676,
      "num_tokens": 39809108.0,
      "step": 22945
    },
    {
      "entropy": 5.505818843841553,
      "epoch": 1.7856059132464501,
      "grad_norm": 1.375,
      "learning_rate": 0.0004678991125911013,
      "loss": 5.1656,
      "mean_token_accuracy": 0.18493040949106215,
      "num_tokens": 39817608.0,
      "step": 22950
    },
    {
      "entropy": 5.4679711818695065,
      "epoch": 1.7859949426181676,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004678848440240111,
      "loss": 5.1446,
      "mean_token_accuracy": 0.18441020399332048,
      "num_tokens": 39826596.0,
      "step": 22955
    },
    {
      "entropy": 5.478511381149292,
      "epoch": 1.7863839719898853,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004678705725302005,
      "loss": 5.1663,
      "mean_token_accuracy": 0.18216990828514099,
      "num_tokens": 39834440.0,
      "step": 22960
    },
    {
      "entropy": 5.480071926116944,
      "epoch": 1.7867730013616028,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004678562981098859,
      "loss": 5.1997,
      "mean_token_accuracy": 0.17654727697372435,
      "num_tokens": 39842617.0,
      "step": 22965
    },
    {
      "entropy": 5.419719266891479,
      "epoch": 1.7871620307333203,
      "grad_norm": 1.25,
      "learning_rate": 0.00046784202076328394,
      "loss": 5.0483,
      "mean_token_accuracy": 0.1962967559695244,
      "num_tokens": 39850688.0,
      "step": 22970
    },
    {
      "entropy": 5.475227403640747,
      "epoch": 1.787551060105038,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00046782774049061124,
      "loss": 5.1795,
      "mean_token_accuracy": 0.18305582106113433,
      "num_tokens": 39859347.0,
      "step": 22975
    },
    {
      "entropy": 5.495944023132324,
      "epoch": 1.7879400894767556,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004678134572920845,
      "loss": 5.1794,
      "mean_token_accuracy": 0.1887215331196785,
      "num_tokens": 39867090.0,
      "step": 22980
    },
    {
      "entropy": 5.474986362457275,
      "epoch": 1.788329118848473,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004677991711679204,
      "loss": 5.1155,
      "mean_token_accuracy": 0.18230823427438736,
      "num_tokens": 39875653.0,
      "step": 22985
    },
    {
      "entropy": 5.487527179718017,
      "epoch": 1.7887181482201906,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046778488211833585,
      "loss": 5.2208,
      "mean_token_accuracy": 0.17306694239377976,
      "num_tokens": 39884781.0,
      "step": 22990
    },
    {
      "entropy": 5.518486738204956,
      "epoch": 1.7891071775919083,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004677705901435475,
      "loss": 5.1292,
      "mean_token_accuracy": 0.1844154417514801,
      "num_tokens": 39892619.0,
      "step": 22995
    },
    {
      "entropy": 5.510538101196289,
      "epoch": 1.7894962069636258,
      "grad_norm": 1.375,
      "learning_rate": 0.0004677562952437723,
      "loss": 5.183,
      "mean_token_accuracy": 0.1811816766858101,
      "num_tokens": 39901518.0,
      "step": 23000
    },
    {
      "entropy": 5.439651870727539,
      "epoch": 1.7898852363353432,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046774199741922705,
      "loss": 5.1437,
      "mean_token_accuracy": 0.1903405472636223,
      "num_tokens": 39909550.0,
      "step": 23005
    },
    {
      "entropy": 5.521391296386719,
      "epoch": 1.790274265707061,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046772769667012884,
      "loss": 5.2126,
      "mean_token_accuracy": 0.1858076810836792,
      "num_tokens": 39918177.0,
      "step": 23010
    },
    {
      "entropy": 5.446392250061035,
      "epoch": 1.7906632950787784,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004677133929966946,
      "loss": 5.179,
      "mean_token_accuracy": 0.1791824772953987,
      "num_tokens": 39926812.0,
      "step": 23015
    },
    {
      "entropy": 5.492099046707153,
      "epoch": 1.791052324450496,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046769908639914134,
      "loss": 5.1825,
      "mean_token_accuracy": 0.18112296760082244,
      "num_tokens": 39934506.0,
      "step": 23020
    },
    {
      "entropy": 5.450998115539551,
      "epoch": 1.7914413538222136,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004676847768776861,
      "loss": 5.1649,
      "mean_token_accuracy": 0.18170108199119567,
      "num_tokens": 39943252.0,
      "step": 23025
    },
    {
      "entropy": 5.458129501342773,
      "epoch": 1.7918303831939313,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004676704644325462,
      "loss": 5.2178,
      "mean_token_accuracy": 0.17964989691972733,
      "num_tokens": 39952585.0,
      "step": 23030
    },
    {
      "entropy": 5.532080841064453,
      "epoch": 1.7922194125656488,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004676561490639385,
      "loss": 5.1815,
      "mean_token_accuracy": 0.18414133191108703,
      "num_tokens": 39961805.0,
      "step": 23035
    },
    {
      "entropy": 5.525699806213379,
      "epoch": 1.7926084419373662,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004676418307720805,
      "loss": 5.1918,
      "mean_token_accuracy": 0.18118867725133897,
      "num_tokens": 39970045.0,
      "step": 23040
    },
    {
      "entropy": 5.492172050476074,
      "epoch": 1.792997471309084,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004676275095571892,
      "loss": 5.1346,
      "mean_token_accuracy": 0.17861605882644654,
      "num_tokens": 39979073.0,
      "step": 23045
    },
    {
      "entropy": 5.449691534042358,
      "epoch": 1.7933865006808014,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046761318541948215,
      "loss": 5.1832,
      "mean_token_accuracy": 0.18431015759706498,
      "num_tokens": 39987936.0,
      "step": 23050
    },
    {
      "entropy": 5.4313208103179935,
      "epoch": 1.7937755300525189,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004675988583591766,
      "loss": 5.2141,
      "mean_token_accuracy": 0.1806098312139511,
      "num_tokens": 39996547.0,
      "step": 23055
    },
    {
      "entropy": 5.563972997665405,
      "epoch": 1.7941645594242366,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046758452837648997,
      "loss": 5.2608,
      "mean_token_accuracy": 0.17358036935329438,
      "num_tokens": 40005717.0,
      "step": 23060
    },
    {
      "entropy": 5.447782325744629,
      "epoch": 1.794553588795954,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004675701954716395,
      "loss": 5.0644,
      "mean_token_accuracy": 0.18892951160669327,
      "num_tokens": 40014799.0,
      "step": 23065
    },
    {
      "entropy": 5.446631574630738,
      "epoch": 1.7949426181676715,
      "grad_norm": 1.4375,
      "learning_rate": 0.00046755585964484286,
      "loss": 5.1922,
      "mean_token_accuracy": 0.18078854829072952,
      "num_tokens": 40023899.0,
      "step": 23070
    },
    {
      "entropy": 5.485397863388061,
      "epoch": 1.7953316475393892,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004675415208963177,
      "loss": 5.1668,
      "mean_token_accuracy": 0.183061483502388,
      "num_tokens": 40032629.0,
      "step": 23075
    },
    {
      "entropy": 5.493171834945679,
      "epoch": 1.795720676911107,
      "grad_norm": 1.25,
      "learning_rate": 0.00046752717922628125,
      "loss": 5.217,
      "mean_token_accuracy": 0.18633842021226882,
      "num_tokens": 40040575.0,
      "step": 23080
    },
    {
      "entropy": 5.468233203887939,
      "epoch": 1.7961097062828244,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004675128346349514,
      "loss": 5.2118,
      "mean_token_accuracy": 0.17855575531721116,
      "num_tokens": 40049221.0,
      "step": 23085
    },
    {
      "entropy": 5.594485759735107,
      "epoch": 1.7964987356545419,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046749848712254554,
      "loss": 5.2256,
      "mean_token_accuracy": 0.17665435820817948,
      "num_tokens": 40057737.0,
      "step": 23090
    },
    {
      "entropy": 5.4544459819793705,
      "epoch": 1.7968877650262596,
      "grad_norm": 1.359375,
      "learning_rate": 0.00046748413668928164,
      "loss": 5.1747,
      "mean_token_accuracy": 0.19249273985624313,
      "num_tokens": 40066447.0,
      "step": 23095
    },
    {
      "entropy": 5.480179500579834,
      "epoch": 1.797276794397977,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004674697833353774,
      "loss": 5.2121,
      "mean_token_accuracy": 0.18554846495389937,
      "num_tokens": 40075825.0,
      "step": 23100
    },
    {
      "entropy": 5.538825845718383,
      "epoch": 1.7976658237696945,
      "grad_norm": 1.25,
      "learning_rate": 0.00046745542706105045,
      "loss": 5.2243,
      "mean_token_accuracy": 0.1842022031545639,
      "num_tokens": 40084342.0,
      "step": 23105
    },
    {
      "entropy": 5.613399791717529,
      "epoch": 1.7980548531414122,
      "grad_norm": 1.25,
      "learning_rate": 0.00046744106786651875,
      "loss": 5.2522,
      "mean_token_accuracy": 0.17147384732961654,
      "num_tokens": 40092650.0,
      "step": 23110
    },
    {
      "entropy": 5.4534307479858395,
      "epoch": 1.7984438825131297,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004674267057520001,
      "loss": 5.1474,
      "mean_token_accuracy": 0.17812027782201767,
      "num_tokens": 40100738.0,
      "step": 23115
    },
    {
      "entropy": 5.401161527633667,
      "epoch": 1.7988329118848472,
      "grad_norm": 1.25,
      "learning_rate": 0.0004674123407177125,
      "loss": 5.0574,
      "mean_token_accuracy": 0.1911110833287239,
      "num_tokens": 40109213.0,
      "step": 23120
    },
    {
      "entropy": 5.427142906188965,
      "epoch": 1.7992219412565649,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046739797276387394,
      "loss": 5.0204,
      "mean_token_accuracy": 0.19041806310415268,
      "num_tokens": 40117622.0,
      "step": 23125
    },
    {
      "entropy": 5.468235206604004,
      "epoch": 1.7996109706282826,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046738360189070235,
      "loss": 5.1772,
      "mean_token_accuracy": 0.18247490525245666,
      "num_tokens": 40126347.0,
      "step": 23130
    },
    {
      "entropy": 5.604040431976318,
      "epoch": 1.8,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004673692280984157,
      "loss": 5.2536,
      "mean_token_accuracy": 0.17334724515676497,
      "num_tokens": 40134245.0,
      "step": 23135
    },
    {
      "entropy": 5.507966136932373,
      "epoch": 1.8003890293717175,
      "grad_norm": 1.25,
      "learning_rate": 0.0004673548513872324,
      "loss": 5.1277,
      "mean_token_accuracy": 0.17861067950725557,
      "num_tokens": 40142304.0,
      "step": 23140
    },
    {
      "entropy": 5.4621185779571535,
      "epoch": 1.8007780587434352,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046734047175737026,
      "loss": 5.1258,
      "mean_token_accuracy": 0.1822003170847893,
      "num_tokens": 40150999.0,
      "step": 23145
    },
    {
      "entropy": 5.515125894546509,
      "epoch": 1.8011670881151527,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046732608920904754,
      "loss": 5.159,
      "mean_token_accuracy": 0.1836692363023758,
      "num_tokens": 40158659.0,
      "step": 23150
    },
    {
      "entropy": 5.425184202194214,
      "epoch": 1.8015561174868702,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004673117037424827,
      "loss": 5.0997,
      "mean_token_accuracy": 0.1865812435746193,
      "num_tokens": 40167343.0,
      "step": 23155
    },
    {
      "entropy": 5.451471042633057,
      "epoch": 1.8019451468585879,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046729731535789375,
      "loss": 5.1448,
      "mean_token_accuracy": 0.18296872079372406,
      "num_tokens": 40176433.0,
      "step": 23160
    },
    {
      "entropy": 5.554238319396973,
      "epoch": 1.8023341762303053,
      "grad_norm": 1.578125,
      "learning_rate": 0.00046728292405549913,
      "loss": 5.2296,
      "mean_token_accuracy": 0.17968525886535644,
      "num_tokens": 40185826.0,
      "step": 23165
    },
    {
      "entropy": 5.465466070175171,
      "epoch": 1.8027232056020228,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004672685298355172,
      "loss": 5.23,
      "mean_token_accuracy": 0.18256641030311585,
      "num_tokens": 40194359.0,
      "step": 23170
    },
    {
      "entropy": 5.422066640853882,
      "epoch": 1.8031122349737405,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004672541326981664,
      "loss": 5.0435,
      "mean_token_accuracy": 0.18694728761911392,
      "num_tokens": 40202812.0,
      "step": 23175
    },
    {
      "entropy": 5.45535945892334,
      "epoch": 1.8035012643454582,
      "grad_norm": 1.234375,
      "learning_rate": 0.000467239732643665,
      "loss": 5.2081,
      "mean_token_accuracy": 0.17527190446853638,
      "num_tokens": 40212344.0,
      "step": 23180
    },
    {
      "entropy": 5.534019470214844,
      "epoch": 1.8038902937171757,
      "grad_norm": 1.421875,
      "learning_rate": 0.00046722532967223183,
      "loss": 5.2556,
      "mean_token_accuracy": 0.18014490604400635,
      "num_tokens": 40220974.0,
      "step": 23185
    },
    {
      "entropy": 5.518744993209839,
      "epoch": 1.8042793230888932,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004672109237840851,
      "loss": 5.2045,
      "mean_token_accuracy": 0.17542807757854462,
      "num_tokens": 40229616.0,
      "step": 23190
    },
    {
      "entropy": 5.488847780227661,
      "epoch": 1.8046683524606109,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046719651497944355,
      "loss": 5.1973,
      "mean_token_accuracy": 0.1816132053732872,
      "num_tokens": 40238505.0,
      "step": 23195
    },
    {
      "entropy": 5.612068223953247,
      "epoch": 1.8050573818323283,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004671821032585259,
      "loss": 5.3215,
      "mean_token_accuracy": 0.17440221458673477,
      "num_tokens": 40246716.0,
      "step": 23200
    },
    {
      "entropy": 5.584719133377075,
      "epoch": 1.8054464112040458,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004671676886215506,
      "loss": 5.2903,
      "mean_token_accuracy": 0.16984851956367492,
      "num_tokens": 40256146.0,
      "step": 23205
    },
    {
      "entropy": 5.568271446228027,
      "epoch": 1.8058354405757635,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004671532710687365,
      "loss": 5.26,
      "mean_token_accuracy": 0.18023647218942643,
      "num_tokens": 40265304.0,
      "step": 23210
    },
    {
      "entropy": 5.412523889541626,
      "epoch": 1.806224469947481,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004671388506003024,
      "loss": 5.039,
      "mean_token_accuracy": 0.188178214430809,
      "num_tokens": 40274055.0,
      "step": 23215
    },
    {
      "entropy": 5.4580771923065186,
      "epoch": 1.8066134993191985,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004671244272164671,
      "loss": 5.1532,
      "mean_token_accuracy": 0.18401886969804765,
      "num_tokens": 40283184.0,
      "step": 23220
    },
    {
      "entropy": 5.604478549957276,
      "epoch": 1.8070025286909162,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00046711000091744935,
      "loss": 5.235,
      "mean_token_accuracy": 0.18231750279664993,
      "num_tokens": 40292604.0,
      "step": 23225
    },
    {
      "entropy": 5.415930461883545,
      "epoch": 1.8073915580626339,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004670955717034681,
      "loss": 4.9732,
      "mean_token_accuracy": 0.19660312831401824,
      "num_tokens": 40301096.0,
      "step": 23230
    },
    {
      "entropy": 5.508109188079834,
      "epoch": 1.8077805874343513,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00046708113957474233,
      "loss": 5.3716,
      "mean_token_accuracy": 0.17006833404302596,
      "num_tokens": 40309976.0,
      "step": 23235
    },
    {
      "entropy": 5.606958389282227,
      "epoch": 1.8081696168060688,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000467066704531491,
      "loss": 5.3082,
      "mean_token_accuracy": 0.18004206269979478,
      "num_tokens": 40318594.0,
      "step": 23240
    },
    {
      "entropy": 5.504400062561035,
      "epoch": 1.8085586461777865,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004670522665739332,
      "loss": 5.1413,
      "mean_token_accuracy": 0.1838740348815918,
      "num_tokens": 40326480.0,
      "step": 23245
    },
    {
      "entropy": 5.428885221481323,
      "epoch": 1.808947675549504,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004670378257022878,
      "loss": 5.2448,
      "mean_token_accuracy": 0.18750713765621185,
      "num_tokens": 40334646.0,
      "step": 23250
    },
    {
      "entropy": 5.518796825408936,
      "epoch": 1.8093367049212215,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046702338191677414,
      "loss": 5.1961,
      "mean_token_accuracy": 0.1821474015712738,
      "num_tokens": 40344007.0,
      "step": 23255
    },
    {
      "entropy": 5.577638959884643,
      "epoch": 1.8097257342929391,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004670089352176113,
      "loss": 5.2341,
      "mean_token_accuracy": 0.1759342595934868,
      "num_tokens": 40352075.0,
      "step": 23260
    },
    {
      "entropy": 5.442049741744995,
      "epoch": 1.8101147636646568,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046699448560501847,
      "loss": 5.1686,
      "mean_token_accuracy": 0.1791912466287613,
      "num_tokens": 40360725.0,
      "step": 23265
    },
    {
      "entropy": 5.484342193603515,
      "epoch": 1.810503793036374,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004669800330792149,
      "loss": 5.1646,
      "mean_token_accuracy": 0.18470221310853957,
      "num_tokens": 40369022.0,
      "step": 23270
    },
    {
      "entropy": 5.494751214981079,
      "epoch": 1.8108928224080918,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046696557764041994,
      "loss": 5.1522,
      "mean_token_accuracy": 0.18368327766656875,
      "num_tokens": 40377624.0,
      "step": 23275
    },
    {
      "entropy": 5.4696708679199215,
      "epoch": 1.8112818517798095,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004669511192888528,
      "loss": 5.1885,
      "mean_token_accuracy": 0.18576640635728836,
      "num_tokens": 40385678.0,
      "step": 23280
    },
    {
      "entropy": 5.480497980117798,
      "epoch": 1.811670881151527,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046693665802473294,
      "loss": 5.2107,
      "mean_token_accuracy": 0.1820526510477066,
      "num_tokens": 40394565.0,
      "step": 23285
    },
    {
      "entropy": 5.542156457901001,
      "epoch": 1.8120599105232444,
      "grad_norm": 1.375,
      "learning_rate": 0.00046692219384827986,
      "loss": 5.2067,
      "mean_token_accuracy": 0.18153039067983628,
      "num_tokens": 40402910.0,
      "step": 23290
    },
    {
      "entropy": 5.460059452056885,
      "epoch": 1.8124489398949621,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004669077267597129,
      "loss": 5.0896,
      "mean_token_accuracy": 0.18731612861156463,
      "num_tokens": 40411352.0,
      "step": 23295
    },
    {
      "entropy": 5.453590774536133,
      "epoch": 1.8128379692666796,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046689325675925174,
      "loss": 5.2308,
      "mean_token_accuracy": 0.18438348025083542,
      "num_tokens": 40419925.0,
      "step": 23300
    },
    {
      "entropy": 5.50316424369812,
      "epoch": 1.813226998638397,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046687878384711574,
      "loss": 5.1742,
      "mean_token_accuracy": 0.17969903200864792,
      "num_tokens": 40428948.0,
      "step": 23305
    },
    {
      "entropy": 5.5396623611450195,
      "epoch": 1.8136160280101148,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046686430802352476,
      "loss": 5.2201,
      "mean_token_accuracy": 0.17791296541690826,
      "num_tokens": 40437625.0,
      "step": 23310
    },
    {
      "entropy": 5.446634721755982,
      "epoch": 1.8140050573818325,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004668498292886982,
      "loss": 5.1431,
      "mean_token_accuracy": 0.18832213878631593,
      "num_tokens": 40446435.0,
      "step": 23315
    },
    {
      "entropy": 5.554263544082642,
      "epoch": 1.8143940867535497,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046683534764285577,
      "loss": 5.1869,
      "mean_token_accuracy": 0.1816474676132202,
      "num_tokens": 40455368.0,
      "step": 23320
    },
    {
      "entropy": 5.484384536743164,
      "epoch": 1.8147831161252674,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046682086308621725,
      "loss": 5.1837,
      "mean_token_accuracy": 0.1831982284784317,
      "num_tokens": 40464461.0,
      "step": 23325
    },
    {
      "entropy": 5.529201889038086,
      "epoch": 1.8151721454969851,
      "grad_norm": 1.375,
      "learning_rate": 0.0004668063756190026,
      "loss": 5.2968,
      "mean_token_accuracy": 0.17010256499052048,
      "num_tokens": 40473143.0,
      "step": 23330
    },
    {
      "entropy": 5.487561178207398,
      "epoch": 1.8155611748687026,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046679188524143136,
      "loss": 5.1471,
      "mean_token_accuracy": 0.1879039317369461,
      "num_tokens": 40481592.0,
      "step": 23335
    },
    {
      "entropy": 5.383489322662354,
      "epoch": 1.81595020424042,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004667773919537235,
      "loss": 5.0204,
      "mean_token_accuracy": 0.1969506859779358,
      "num_tokens": 40489445.0,
      "step": 23340
    },
    {
      "entropy": 5.417173719406128,
      "epoch": 1.8163392336121378,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046676289575609914,
      "loss": 5.1207,
      "mean_token_accuracy": 0.18748218864202498,
      "num_tokens": 40498279.0,
      "step": 23345
    },
    {
      "entropy": 5.534570360183716,
      "epoch": 1.8167282629838553,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046674839664877796,
      "loss": 5.1053,
      "mean_token_accuracy": 0.18652520179748536,
      "num_tokens": 40507051.0,
      "step": 23350
    },
    {
      "entropy": 5.456001138687133,
      "epoch": 1.8171172923555727,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004667338946319801,
      "loss": 5.0735,
      "mean_token_accuracy": 0.18819492310285568,
      "num_tokens": 40515213.0,
      "step": 23355
    },
    {
      "entropy": 5.46458888053894,
      "epoch": 1.8175063217272904,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004667193897059255,
      "loss": 5.2267,
      "mean_token_accuracy": 0.17738055437803268,
      "num_tokens": 40523859.0,
      "step": 23360
    },
    {
      "entropy": 5.568717908859253,
      "epoch": 1.8178953510990081,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00046670488187083436,
      "loss": 5.2959,
      "mean_token_accuracy": 0.17637215554714203,
      "num_tokens": 40533236.0,
      "step": 23365
    },
    {
      "entropy": 5.513077259063721,
      "epoch": 1.8182843804707254,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004666903711269267,
      "loss": 5.1055,
      "mean_token_accuracy": 0.1881256178021431,
      "num_tokens": 40541670.0,
      "step": 23370
    },
    {
      "entropy": 5.4451593399047855,
      "epoch": 1.818673409842443,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004666758574744228,
      "loss": 5.2193,
      "mean_token_accuracy": 0.18748000264167786,
      "num_tokens": 40551411.0,
      "step": 23375
    },
    {
      "entropy": 5.531672859191895,
      "epoch": 1.8190624392141608,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004666613409135429,
      "loss": 5.1716,
      "mean_token_accuracy": 0.18129484057426454,
      "num_tokens": 40560351.0,
      "step": 23380
    },
    {
      "entropy": 5.447461080551148,
      "epoch": 1.8194514685858783,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004666468214445072,
      "loss": 5.1098,
      "mean_token_accuracy": 0.18999021053314208,
      "num_tokens": 40569363.0,
      "step": 23385
    },
    {
      "entropy": 5.505829620361328,
      "epoch": 1.8198404979575957,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046663229906753595,
      "loss": 5.1667,
      "mean_token_accuracy": 0.1801112785935402,
      "num_tokens": 40577425.0,
      "step": 23390
    },
    {
      "entropy": 5.499072456359864,
      "epoch": 1.8202295273293134,
      "grad_norm": 1.765625,
      "learning_rate": 0.00046661777378284965,
      "loss": 5.1896,
      "mean_token_accuracy": 0.179530830681324,
      "num_tokens": 40585718.0,
      "step": 23395
    },
    {
      "entropy": 5.581491231918335,
      "epoch": 1.820618556701031,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004666032455906685,
      "loss": 5.2666,
      "mean_token_accuracy": 0.17839665561914445,
      "num_tokens": 40594886.0,
      "step": 23400
    },
    {
      "entropy": 5.5808652400970455,
      "epoch": 1.8210075860727484,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046658871449121325,
      "loss": 5.2988,
      "mean_token_accuracy": 0.17446724623441695,
      "num_tokens": 40603737.0,
      "step": 23405
    },
    {
      "entropy": 5.467912340164185,
      "epoch": 1.821396615444466,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004665741804847041,
      "loss": 5.1805,
      "mean_token_accuracy": 0.1730229914188385,
      "num_tokens": 40612473.0,
      "step": 23410
    },
    {
      "entropy": 5.496291303634644,
      "epoch": 1.8217856448161838,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046655964357136164,
      "loss": 5.1235,
      "mean_token_accuracy": 0.18517027497291566,
      "num_tokens": 40620517.0,
      "step": 23415
    },
    {
      "entropy": 5.549793004989624,
      "epoch": 1.822174674187901,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046654510375140657,
      "loss": 5.2365,
      "mean_token_accuracy": 0.17439647316932677,
      "num_tokens": 40629530.0,
      "step": 23420
    },
    {
      "entropy": 5.491628360748291,
      "epoch": 1.8225637035596187,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004665305610250594,
      "loss": 5.1486,
      "mean_token_accuracy": 0.18616321086883544,
      "num_tokens": 40638616.0,
      "step": 23425
    },
    {
      "entropy": 5.578245544433594,
      "epoch": 1.8229527329313364,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004665160153925408,
      "loss": 5.1989,
      "mean_token_accuracy": 0.17650567293167113,
      "num_tokens": 40646945.0,
      "step": 23430
    },
    {
      "entropy": 5.467570972442627,
      "epoch": 1.823341762303054,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046650146685407154,
      "loss": 5.1703,
      "mean_token_accuracy": 0.18351942002773286,
      "num_tokens": 40654845.0,
      "step": 23435
    },
    {
      "entropy": 5.3574388980865475,
      "epoch": 1.8237307916747714,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046648691540987226,
      "loss": 5.0334,
      "mean_token_accuracy": 0.19450154155492783,
      "num_tokens": 40663776.0,
      "step": 23440
    },
    {
      "entropy": 5.47280969619751,
      "epoch": 1.824119821046489,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046647236106016387,
      "loss": 5.0844,
      "mean_token_accuracy": 0.19155684858560562,
      "num_tokens": 40672071.0,
      "step": 23445
    },
    {
      "entropy": 5.498898601531982,
      "epoch": 1.8245088504182065,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004664578038051672,
      "loss": 5.2523,
      "mean_token_accuracy": 0.1806219130754471,
      "num_tokens": 40681264.0,
      "step": 23450
    },
    {
      "entropy": 5.429758167266845,
      "epoch": 1.824897879789924,
      "grad_norm": 1.4375,
      "learning_rate": 0.000466443243645103,
      "loss": 5.1497,
      "mean_token_accuracy": 0.18900523632764815,
      "num_tokens": 40690091.0,
      "step": 23455
    },
    {
      "entropy": 5.576818513870239,
      "epoch": 1.8252869091616417,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004664286805801923,
      "loss": 5.2983,
      "mean_token_accuracy": 0.17648597657680512,
      "num_tokens": 40699623.0,
      "step": 23460
    },
    {
      "entropy": 5.499347591400147,
      "epoch": 1.8256759385333594,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004664141146106562,
      "loss": 5.1282,
      "mean_token_accuracy": 0.1847424954175949,
      "num_tokens": 40708288.0,
      "step": 23465
    },
    {
      "entropy": 5.556072473526001,
      "epoch": 1.8260649679050769,
      "grad_norm": 1.4375,
      "learning_rate": 0.00046639954573671547,
      "loss": 5.2437,
      "mean_token_accuracy": 0.18135821372270583,
      "num_tokens": 40716357.0,
      "step": 23470
    },
    {
      "entropy": 5.553737926483154,
      "epoch": 1.8264539972767944,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046638497395859127,
      "loss": 5.2386,
      "mean_token_accuracy": 0.18301109224557877,
      "num_tokens": 40724775.0,
      "step": 23475
    },
    {
      "entropy": 5.457752656936646,
      "epoch": 1.826843026648512,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004663703992765047,
      "loss": 5.1913,
      "mean_token_accuracy": 0.18251526802778245,
      "num_tokens": 40733642.0,
      "step": 23480
    },
    {
      "entropy": 5.4011266231536865,
      "epoch": 1.8272320560202295,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00046635582169067693,
      "loss": 5.122,
      "mean_token_accuracy": 0.18980038911104202,
      "num_tokens": 40742201.0,
      "step": 23485
    },
    {
      "entropy": 5.438978242874145,
      "epoch": 1.827621085391947,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046634124120132915,
      "loss": 5.0967,
      "mean_token_accuracy": 0.1914320170879364,
      "num_tokens": 40750379.0,
      "step": 23490
    },
    {
      "entropy": 5.480813598632812,
      "epoch": 1.8280101147636647,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004663266578086826,
      "loss": 5.2383,
      "mean_token_accuracy": 0.17355093359947205,
      "num_tokens": 40759232.0,
      "step": 23495
    },
    {
      "entropy": 5.474311828613281,
      "epoch": 1.8283991441353822,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004663120715129585,
      "loss": 5.1978,
      "mean_token_accuracy": 0.18082138895988464,
      "num_tokens": 40768247.0,
      "step": 23500
    },
    {
      "entropy": 5.5469437599182125,
      "epoch": 1.8287881735070997,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004662974823143783,
      "loss": 5.2511,
      "mean_token_accuracy": 0.1832208201289177,
      "num_tokens": 40777085.0,
      "step": 23505
    },
    {
      "entropy": 5.502522993087768,
      "epoch": 1.8291772028788174,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004662828902131632,
      "loss": 5.258,
      "mean_token_accuracy": 0.1753576397895813,
      "num_tokens": 40785466.0,
      "step": 23510
    },
    {
      "entropy": 5.51013560295105,
      "epoch": 1.829566232250535,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004662682952095348,
      "loss": 5.1652,
      "mean_token_accuracy": 0.18816448152065277,
      "num_tokens": 40794165.0,
      "step": 23515
    },
    {
      "entropy": 5.461502552032471,
      "epoch": 1.8299552616222525,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046625369730371436,
      "loss": 5.1328,
      "mean_token_accuracy": 0.18191335797309877,
      "num_tokens": 40802497.0,
      "step": 23520
    },
    {
      "entropy": 5.473326206207275,
      "epoch": 1.83034429099397,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004662390964959235,
      "loss": 5.1166,
      "mean_token_accuracy": 0.19084784835577012,
      "num_tokens": 40811022.0,
      "step": 23525
    },
    {
      "entropy": 5.530590963363648,
      "epoch": 1.8307333203656877,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046622449278638375,
      "loss": 5.2097,
      "mean_token_accuracy": 0.1784398466348648,
      "num_tokens": 40820430.0,
      "step": 23530
    },
    {
      "entropy": 5.477851629257202,
      "epoch": 1.8311223497374052,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004662098861753167,
      "loss": 5.2232,
      "mean_token_accuracy": 0.18283409476280213,
      "num_tokens": 40829245.0,
      "step": 23535
    },
    {
      "entropy": 5.439356994628906,
      "epoch": 1.8315113791091227,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046619527666294394,
      "loss": 5.1304,
      "mean_token_accuracy": 0.1887688234448433,
      "num_tokens": 40838576.0,
      "step": 23540
    },
    {
      "entropy": 5.550945854187011,
      "epoch": 1.8319004084808403,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004661806642494872,
      "loss": 5.2592,
      "mean_token_accuracy": 0.1754223644733429,
      "num_tokens": 40847097.0,
      "step": 23545
    },
    {
      "entropy": 5.542883348464966,
      "epoch": 1.8322894378525578,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004661660489351681,
      "loss": 5.1328,
      "mean_token_accuracy": 0.1895716205239296,
      "num_tokens": 40855135.0,
      "step": 23550
    },
    {
      "entropy": 5.482326221466065,
      "epoch": 1.8326784672242753,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004661514307202086,
      "loss": 5.2422,
      "mean_token_accuracy": 0.1790841907262802,
      "num_tokens": 40864218.0,
      "step": 23555
    },
    {
      "entropy": 5.475128602981568,
      "epoch": 1.833067496595993,
      "grad_norm": 1.390625,
      "learning_rate": 0.00046613680960483036,
      "loss": 5.1861,
      "mean_token_accuracy": 0.18098906576633453,
      "num_tokens": 40873147.0,
      "step": 23560
    },
    {
      "entropy": 5.533008718490601,
      "epoch": 1.8334565259677107,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004661221855892552,
      "loss": 5.2578,
      "mean_token_accuracy": 0.18332785964012147,
      "num_tokens": 40881340.0,
      "step": 23565
    },
    {
      "entropy": 5.475386428833008,
      "epoch": 1.8338455553394282,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046610755867370506,
      "loss": 5.1069,
      "mean_token_accuracy": 0.18876246809959413,
      "num_tokens": 40889246.0,
      "step": 23570
    },
    {
      "entropy": 5.476675271987915,
      "epoch": 1.8342345847111456,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000466092928858402,
      "loss": 5.1995,
      "mean_token_accuracy": 0.18428771644830705,
      "num_tokens": 40898285.0,
      "step": 23575
    },
    {
      "entropy": 5.466903162002564,
      "epoch": 1.8346236140828633,
      "grad_norm": 1.375,
      "learning_rate": 0.00046607829614356787,
      "loss": 5.2434,
      "mean_token_accuracy": 0.17963754683732985,
      "num_tokens": 40907012.0,
      "step": 23580
    },
    {
      "entropy": 5.555818128585815,
      "epoch": 1.8350126434545808,
      "grad_norm": 1.25,
      "learning_rate": 0.0004660636605294247,
      "loss": 5.1084,
      "mean_token_accuracy": 0.18875508159399032,
      "num_tokens": 40915857.0,
      "step": 23585
    },
    {
      "entropy": 5.551228189468384,
      "epoch": 1.8354016728262983,
      "grad_norm": 1.25,
      "learning_rate": 0.0004660490220161946,
      "loss": 5.1505,
      "mean_token_accuracy": 0.18369343429803847,
      "num_tokens": 40924364.0,
      "step": 23590
    },
    {
      "entropy": 5.403105735778809,
      "epoch": 1.835790702198016,
      "grad_norm": 1.34375,
      "learning_rate": 0.00046603438060409966,
      "loss": 5.1389,
      "mean_token_accuracy": 0.18578112721443177,
      "num_tokens": 40934092.0,
      "step": 23595
    },
    {
      "entropy": 5.441193532943726,
      "epoch": 1.8361797315697335,
      "grad_norm": 1.3125,
      "learning_rate": 0.000466019736293362,
      "loss": 5.0242,
      "mean_token_accuracy": 0.19129632115364076,
      "num_tokens": 40942066.0,
      "step": 23600
    },
    {
      "entropy": 5.503836393356323,
      "epoch": 1.836568760941451,
      "grad_norm": 1.34375,
      "learning_rate": 0.00046600508908420396,
      "loss": 5.1888,
      "mean_token_accuracy": 0.18000709414482116,
      "num_tokens": 40951445.0,
      "step": 23605
    },
    {
      "entropy": 5.47071852684021,
      "epoch": 1.8369577903131686,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046599043897684766,
      "loss": 5.1631,
      "mean_token_accuracy": 0.19170099049806594,
      "num_tokens": 40960111.0,
      "step": 23610
    },
    {
      "entropy": 5.474275779724121,
      "epoch": 1.8373468196848863,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004659757859715155,
      "loss": 5.12,
      "mean_token_accuracy": 0.19222436398267745,
      "num_tokens": 40968036.0,
      "step": 23615
    },
    {
      "entropy": 5.421841764450074,
      "epoch": 1.8377358490566038,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00046596113006842975,
      "loss": 5.1786,
      "mean_token_accuracy": 0.18294933438301086,
      "num_tokens": 40977032.0,
      "step": 23620
    },
    {
      "entropy": 5.470087480545044,
      "epoch": 1.8381248784283213,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004659464712678128,
      "loss": 5.1641,
      "mean_token_accuracy": 0.18142516314983367,
      "num_tokens": 40985523.0,
      "step": 23625
    },
    {
      "entropy": 5.432314777374268,
      "epoch": 1.838513907800039,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004659318095698871,
      "loss": 5.0356,
      "mean_token_accuracy": 0.19299081712961197,
      "num_tokens": 40993617.0,
      "step": 23630
    },
    {
      "entropy": 5.368931865692138,
      "epoch": 1.8389029371717565,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000465917144974875,
      "loss": 5.1335,
      "mean_token_accuracy": 0.18121035397052765,
      "num_tokens": 41002424.0,
      "step": 23635
    },
    {
      "entropy": 5.491849613189697,
      "epoch": 1.839291966543474,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004659024774829992,
      "loss": 5.2486,
      "mean_token_accuracy": 0.17130196839571,
      "num_tokens": 41011150.0,
      "step": 23640
    },
    {
      "entropy": 5.552239274978637,
      "epoch": 1.8396809959151916,
      "grad_norm": 2.25,
      "learning_rate": 0.000465887807094482,
      "loss": 5.1302,
      "mean_token_accuracy": 0.1828191176056862,
      "num_tokens": 41020112.0,
      "step": 23645
    },
    {
      "entropy": 5.504865074157715,
      "epoch": 1.8400700252869093,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046587313380954635,
      "loss": 5.189,
      "mean_token_accuracy": 0.17841228544712068,
      "num_tokens": 41029470.0,
      "step": 23650
    },
    {
      "entropy": 5.476781511306763,
      "epoch": 1.8404590546586266,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046585845762841453,
      "loss": 5.1989,
      "mean_token_accuracy": 0.1734434798359871,
      "num_tokens": 41038622.0,
      "step": 23655
    },
    {
      "entropy": 5.577941226959228,
      "epoch": 1.8408480840303443,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004658437785513095,
      "loss": 5.3194,
      "mean_token_accuracy": 0.17199227809906006,
      "num_tokens": 41047573.0,
      "step": 23660
    },
    {
      "entropy": 5.522489166259765,
      "epoch": 1.841237113402062,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004658290965784539,
      "loss": 5.1362,
      "mean_token_accuracy": 0.18965699076652526,
      "num_tokens": 41057084.0,
      "step": 23665
    },
    {
      "entropy": 5.4084742069244385,
      "epoch": 1.8416261427737795,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004658144117100704,
      "loss": 5.0528,
      "mean_token_accuracy": 0.1881210520863533,
      "num_tokens": 41065297.0,
      "step": 23670
    },
    {
      "entropy": 5.42309536933899,
      "epoch": 1.842015172145497,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046579972394638204,
      "loss": 5.1551,
      "mean_token_accuracy": 0.18398280888795854,
      "num_tokens": 41073748.0,
      "step": 23675
    },
    {
      "entropy": 5.509632349014282,
      "epoch": 1.8424042015172146,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046578503328761146,
      "loss": 5.1021,
      "mean_token_accuracy": 0.18898945152759553,
      "num_tokens": 41082304.0,
      "step": 23680
    },
    {
      "entropy": 5.518460369110107,
      "epoch": 1.842793230888932,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046577033973398173,
      "loss": 5.2645,
      "mean_token_accuracy": 0.17780764997005463,
      "num_tokens": 41091130.0,
      "step": 23685
    },
    {
      "entropy": 5.4526628971099855,
      "epoch": 1.8431822602606496,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004657556432857157,
      "loss": 5.1702,
      "mean_token_accuracy": 0.18572618812322617,
      "num_tokens": 41100413.0,
      "step": 23690
    },
    {
      "entropy": 5.356188869476318,
      "epoch": 1.8435712896323673,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004657409439430364,
      "loss": 5.0225,
      "mean_token_accuracy": 0.19837235361337663,
      "num_tokens": 41108582.0,
      "step": 23695
    },
    {
      "entropy": 5.4476786136627195,
      "epoch": 1.843960319004085,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004657262417061669,
      "loss": 5.1055,
      "mean_token_accuracy": 0.18668131977319719,
      "num_tokens": 41117067.0,
      "step": 23700
    },
    {
      "entropy": 5.4262354373931885,
      "epoch": 1.8443493483758022,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004657115365753302,
      "loss": 5.1173,
      "mean_token_accuracy": 0.18363260775804519,
      "num_tokens": 41125559.0,
      "step": 23705
    },
    {
      "entropy": 5.448144626617432,
      "epoch": 1.84473837774752,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046569682855074953,
      "loss": 5.1526,
      "mean_token_accuracy": 0.18344168215990067,
      "num_tokens": 41134503.0,
      "step": 23710
    },
    {
      "entropy": 5.575972414016723,
      "epoch": 1.8451274071192376,
      "grad_norm": 1.671875,
      "learning_rate": 0.00046568211763264796,
      "loss": 5.2782,
      "mean_token_accuracy": 0.17267971634864807,
      "num_tokens": 41143903.0,
      "step": 23715
    },
    {
      "entropy": 5.487414407730102,
      "epoch": 1.845516436490955,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004656674038212488,
      "loss": 5.1694,
      "mean_token_accuracy": 0.17904836535453797,
      "num_tokens": 41153082.0,
      "step": 23720
    },
    {
      "entropy": 5.542892169952393,
      "epoch": 1.8459054658626726,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046565268711677525,
      "loss": 5.1426,
      "mean_token_accuracy": 0.1869928389787674,
      "num_tokens": 41161749.0,
      "step": 23725
    },
    {
      "entropy": 5.43172550201416,
      "epoch": 1.8462944952343903,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046563796751945065,
      "loss": 5.1835,
      "mean_token_accuracy": 0.1804962933063507,
      "num_tokens": 41170969.0,
      "step": 23730
    },
    {
      "entropy": 5.377286386489868,
      "epoch": 1.8466835246061077,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00046562324502949834,
      "loss": 5.0312,
      "mean_token_accuracy": 0.19420293420553209,
      "num_tokens": 41179512.0,
      "step": 23735
    },
    {
      "entropy": 5.438078737258911,
      "epoch": 1.8470725539778252,
      "grad_norm": 1.25,
      "learning_rate": 0.0004656085196471416,
      "loss": 5.0873,
      "mean_token_accuracy": 0.18688165992498398,
      "num_tokens": 41188109.0,
      "step": 23740
    },
    {
      "entropy": 5.480934524536133,
      "epoch": 1.847461583349543,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046559379137260404,
      "loss": 5.2016,
      "mean_token_accuracy": 0.1750790774822235,
      "num_tokens": 41196708.0,
      "step": 23745
    },
    {
      "entropy": 5.556172180175781,
      "epoch": 1.8478506127212606,
      "grad_norm": 1.359375,
      "learning_rate": 0.000465579060206109,
      "loss": 5.3081,
      "mean_token_accuracy": 0.1780659958720207,
      "num_tokens": 41205013.0,
      "step": 23750
    },
    {
      "entropy": 5.448770952224732,
      "epoch": 1.8482396420929779,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00046556432614788015,
      "loss": 5.027,
      "mean_token_accuracy": 0.19285576045513153,
      "num_tokens": 41213701.0,
      "step": 23755
    },
    {
      "entropy": 5.386996078491211,
      "epoch": 1.8486286714646956,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004655495891981409,
      "loss": 5.0756,
      "mean_token_accuracy": 0.18846024125814437,
      "num_tokens": 41222450.0,
      "step": 23760
    },
    {
      "entropy": 5.467958879470825,
      "epoch": 1.8490177008364133,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000465534849357115,
      "loss": 5.1881,
      "mean_token_accuracy": 0.17802358865737916,
      "num_tokens": 41230259.0,
      "step": 23765
    },
    {
      "entropy": 5.458728790283203,
      "epoch": 1.8494067302081307,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046552010662502595,
      "loss": 5.1285,
      "mean_token_accuracy": 0.1926506504416466,
      "num_tokens": 41238863.0,
      "step": 23770
    },
    {
      "entropy": 5.428018379211426,
      "epoch": 1.8497957595798482,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004655053610020976,
      "loss": 5.1631,
      "mean_token_accuracy": 0.18430890291929244,
      "num_tokens": 41247260.0,
      "step": 23775
    },
    {
      "entropy": 5.4210710525512695,
      "epoch": 1.850184788951566,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004654906124885535,
      "loss": 5.0658,
      "mean_token_accuracy": 0.19114517271518708,
      "num_tokens": 41255475.0,
      "step": 23780
    },
    {
      "entropy": 5.3939659118652346,
      "epoch": 1.8505738183232834,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004654758610846176,
      "loss": 4.9588,
      "mean_token_accuracy": 0.20290385335683822,
      "num_tokens": 41263858.0,
      "step": 23785
    },
    {
      "entropy": 5.421311855316162,
      "epoch": 1.8509628476950009,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004654611067905137,
      "loss": 5.2203,
      "mean_token_accuracy": 0.18146197348833085,
      "num_tokens": 41272527.0,
      "step": 23790
    },
    {
      "entropy": 5.504159450531006,
      "epoch": 1.8513518770667186,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046544634960646563,
      "loss": 5.1659,
      "mean_token_accuracy": 0.1872281089425087,
      "num_tokens": 41280647.0,
      "step": 23795
    },
    {
      "entropy": 5.54730076789856,
      "epoch": 1.8517409064384363,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004654315895326974,
      "loss": 5.1859,
      "mean_token_accuracy": 0.18824778646230697,
      "num_tokens": 41289462.0,
      "step": 23800
    },
    {
      "entropy": 5.539541530609131,
      "epoch": 1.8521299358101535,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004654168265694328,
      "loss": 5.1159,
      "mean_token_accuracy": 0.18879576474428178,
      "num_tokens": 41297178.0,
      "step": 23805
    },
    {
      "entropy": 5.448823928833008,
      "epoch": 1.8525189651818712,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000465402060716896,
      "loss": 5.0943,
      "mean_token_accuracy": 0.18625584095716477,
      "num_tokens": 41305405.0,
      "step": 23810
    },
    {
      "entropy": 5.440502166748047,
      "epoch": 1.852907994553589,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000465387291975311,
      "loss": 5.106,
      "mean_token_accuracy": 0.18837257474660873,
      "num_tokens": 41313330.0,
      "step": 23815
    },
    {
      "entropy": 5.4377179622650145,
      "epoch": 1.8532970239253064,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046537252034490185,
      "loss": 5.1436,
      "mean_token_accuracy": 0.19087498188018798,
      "num_tokens": 41321245.0,
      "step": 23820
    },
    {
      "entropy": 5.548234605789185,
      "epoch": 1.8536860532970239,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046535774582589275,
      "loss": 5.2058,
      "mean_token_accuracy": 0.17530859708786012,
      "num_tokens": 41330013.0,
      "step": 23825
    },
    {
      "entropy": 5.541493844985962,
      "epoch": 1.8540750826687415,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046534296841850776,
      "loss": 5.1687,
      "mean_token_accuracy": 0.1860146403312683,
      "num_tokens": 41338134.0,
      "step": 23830
    },
    {
      "entropy": 5.41396951675415,
      "epoch": 1.854464112040459,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046532818812297124,
      "loss": 5.0782,
      "mean_token_accuracy": 0.18359216600656508,
      "num_tokens": 41346693.0,
      "step": 23835
    },
    {
      "entropy": 5.4653685092926025,
      "epoch": 1.8548531414121765,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004653134049395074,
      "loss": 5.2347,
      "mean_token_accuracy": 0.17438702881336213,
      "num_tokens": 41354987.0,
      "step": 23840
    },
    {
      "entropy": 5.508549308776855,
      "epoch": 1.8552421707838942,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004652986188683406,
      "loss": 5.2194,
      "mean_token_accuracy": 0.17582044005393982,
      "num_tokens": 41363703.0,
      "step": 23845
    },
    {
      "entropy": 5.500749349594116,
      "epoch": 1.855631200155612,
      "grad_norm": 1.34375,
      "learning_rate": 0.00046528382990969504,
      "loss": 5.165,
      "mean_token_accuracy": 0.1863449066877365,
      "num_tokens": 41371983.0,
      "step": 23850
    },
    {
      "entropy": 5.507573080062866,
      "epoch": 1.8560202295273294,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004652690380637953,
      "loss": 5.1995,
      "mean_token_accuracy": 0.18139823228120805,
      "num_tokens": 41380215.0,
      "step": 23855
    },
    {
      "entropy": 5.484664440155029,
      "epoch": 1.8564092588990468,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004652542433308657,
      "loss": 5.128,
      "mean_token_accuracy": 0.1867167368531227,
      "num_tokens": 41388927.0,
      "step": 23860
    },
    {
      "entropy": 5.5138843059539795,
      "epoch": 1.8567982882707645,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004652394457111309,
      "loss": 5.2225,
      "mean_token_accuracy": 0.17612198293209075,
      "num_tokens": 41397823.0,
      "step": 23865
    },
    {
      "entropy": 5.510759353637695,
      "epoch": 1.857187317642482,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046522464520481517,
      "loss": 5.1895,
      "mean_token_accuracy": 0.1852448746562004,
      "num_tokens": 41406322.0,
      "step": 23870
    },
    {
      "entropy": 5.4510509967803955,
      "epoch": 1.8575763470141995,
      "grad_norm": 1.359375,
      "learning_rate": 0.00046520984181214333,
      "loss": 5.1502,
      "mean_token_accuracy": 0.18764548748731613,
      "num_tokens": 41415118.0,
      "step": 23875
    },
    {
      "entropy": 5.473931169509887,
      "epoch": 1.8579653763859172,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004651950355333398,
      "loss": 5.2019,
      "mean_token_accuracy": 0.1841976210474968,
      "num_tokens": 41424139.0,
      "step": 23880
    },
    {
      "entropy": 5.504588270187378,
      "epoch": 1.8583544057576347,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004651802263686294,
      "loss": 5.2343,
      "mean_token_accuracy": 0.17744267880916595,
      "num_tokens": 41432017.0,
      "step": 23885
    },
    {
      "entropy": 5.512833881378174,
      "epoch": 1.8587434351293521,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004651654143182367,
      "loss": 5.1336,
      "mean_token_accuracy": 0.18234180510044098,
      "num_tokens": 41440279.0,
      "step": 23890
    },
    {
      "entropy": 5.433212518692017,
      "epoch": 1.8591324645010698,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004651505993823866,
      "loss": 5.0622,
      "mean_token_accuracy": 0.186395999789238,
      "num_tokens": 41449325.0,
      "step": 23895
    },
    {
      "entropy": 5.4236030101776125,
      "epoch": 1.8595214938727875,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046513578156130383,
      "loss": 5.0586,
      "mean_token_accuracy": 0.1806375950574875,
      "num_tokens": 41457898.0,
      "step": 23900
    },
    {
      "entropy": 5.427620458602905,
      "epoch": 1.859910523244505,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004651209608552131,
      "loss": 5.0865,
      "mean_token_accuracy": 0.19528905749320985,
      "num_tokens": 41466410.0,
      "step": 23905
    },
    {
      "entropy": 5.555057525634766,
      "epoch": 1.8602995526162225,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046510613726433945,
      "loss": 5.208,
      "mean_token_accuracy": 0.18541549742221833,
      "num_tokens": 41475061.0,
      "step": 23910
    },
    {
      "entropy": 5.376064300537109,
      "epoch": 1.8606885819879402,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004650913107889078,
      "loss": 5.1645,
      "mean_token_accuracy": 0.18871715813875198,
      "num_tokens": 41483569.0,
      "step": 23915
    },
    {
      "entropy": 5.420612573623657,
      "epoch": 1.8610776113596577,
      "grad_norm": 1.40625,
      "learning_rate": 0.000465076481429143,
      "loss": 5.188,
      "mean_token_accuracy": 0.18206114321947098,
      "num_tokens": 41492113.0,
      "step": 23920
    },
    {
      "entropy": 5.473980522155761,
      "epoch": 1.8614666407313751,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004650616491852701,
      "loss": 5.2377,
      "mean_token_accuracy": 0.17541804611682893,
      "num_tokens": 41500722.0,
      "step": 23925
    },
    {
      "entropy": 5.442581462860107,
      "epoch": 1.8618556701030928,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046504681405751426,
      "loss": 5.0891,
      "mean_token_accuracy": 0.18984262645244598,
      "num_tokens": 41509272.0,
      "step": 23930
    },
    {
      "entropy": 5.497873210906983,
      "epoch": 1.8622446994748103,
      "grad_norm": 1.390625,
      "learning_rate": 0.00046503197604610047,
      "loss": 5.2517,
      "mean_token_accuracy": 0.18093563467264176,
      "num_tokens": 41517379.0,
      "step": 23935
    },
    {
      "entropy": 5.4139515399932865,
      "epoch": 1.8626337288465278,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00046501713515125393,
      "loss": 5.1129,
      "mean_token_accuracy": 0.18834685683250427,
      "num_tokens": 41525737.0,
      "step": 23940
    },
    {
      "entropy": 5.4388192653656,
      "epoch": 1.8630227582182455,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004650022913731998,
      "loss": 5.1585,
      "mean_token_accuracy": 0.18253383487462999,
      "num_tokens": 41535016.0,
      "step": 23945
    },
    {
      "entropy": 5.518453454971313,
      "epoch": 1.8634117875899632,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046498744471216335,
      "loss": 5.2418,
      "mean_token_accuracy": 0.18166171461343766,
      "num_tokens": 41544253.0,
      "step": 23950
    },
    {
      "entropy": 5.529545450210572,
      "epoch": 1.8638008169616807,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046497259516836974,
      "loss": 5.2009,
      "mean_token_accuracy": 0.17735646516084672,
      "num_tokens": 41551907.0,
      "step": 23955
    },
    {
      "entropy": 5.558815240859985,
      "epoch": 1.8641898463333981,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046495774274204446,
      "loss": 5.2102,
      "mean_token_accuracy": 0.18527324199676515,
      "num_tokens": 41560675.0,
      "step": 23960
    },
    {
      "entropy": 5.519184160232544,
      "epoch": 1.8645788757051158,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004649428874334127,
      "loss": 5.2233,
      "mean_token_accuracy": 0.177040958404541,
      "num_tokens": 41570256.0,
      "step": 23965
    },
    {
      "entropy": 5.51069393157959,
      "epoch": 1.8649679050768333,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004649280292427,
      "loss": 5.1806,
      "mean_token_accuracy": 0.18845358341932297,
      "num_tokens": 41578990.0,
      "step": 23970
    },
    {
      "entropy": 5.440972852706909,
      "epoch": 1.8653569344485508,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004649131681701318,
      "loss": 5.1814,
      "mean_token_accuracy": 0.18484075665473937,
      "num_tokens": 41587896.0,
      "step": 23975
    },
    {
      "entropy": 5.451095104217529,
      "epoch": 1.8657459638202685,
      "grad_norm": 1.25,
      "learning_rate": 0.00046489830421593347,
      "loss": 5.1587,
      "mean_token_accuracy": 0.18506552428007125,
      "num_tokens": 41596160.0,
      "step": 23980
    },
    {
      "entropy": 5.490046596527099,
      "epoch": 1.866134993191986,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004648834373803307,
      "loss": 5.1716,
      "mean_token_accuracy": 0.17922349125146866,
      "num_tokens": 41605025.0,
      "step": 23985
    },
    {
      "entropy": 5.488406658172607,
      "epoch": 1.8665240225637034,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046486856766354893,
      "loss": 5.1424,
      "mean_token_accuracy": 0.18891462832689285,
      "num_tokens": 41612768.0,
      "step": 23990
    },
    {
      "entropy": 5.4166882038116455,
      "epoch": 1.8669130519354211,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046485369506581387,
      "loss": 5.1814,
      "mean_token_accuracy": 0.182580429315567,
      "num_tokens": 41620892.0,
      "step": 23995
    },
    {
      "entropy": 5.530186462402344,
      "epoch": 1.8673020813071388,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046483881958735124,
      "loss": 5.2636,
      "mean_token_accuracy": 0.18245457261800765,
      "num_tokens": 41629708.0,
      "step": 24000
    },
    {
      "epoch": 1.8673020813071388,
      "eval_entropy": 5.377515880962882,
      "eval_loss": 5.254362106323242,
      "eval_mean_token_accuracy": 0.1882645305780348,
      "eval_num_tokens": 41629708.0,
      "eval_runtime": 28.6314,
      "eval_samples_per_second": 1451.483,
      "eval_steps_per_second": 181.444,
      "step": 24000
    },
    {
      "entropy": 5.564175701141357,
      "epoch": 1.8676911106788563,
      "grad_norm": 1.25,
      "learning_rate": 0.00046482394122838663,
      "loss": 5.0646,
      "mean_token_accuracy": 0.19183979481458663,
      "num_tokens": 41637966.0,
      "step": 24005
    },
    {
      "entropy": 5.396935224533081,
      "epoch": 1.8680801400505738,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046480905998914587,
      "loss": 5.1292,
      "mean_token_accuracy": 0.18848343342542648,
      "num_tokens": 41646726.0,
      "step": 24010
    },
    {
      "entropy": 5.402428436279297,
      "epoch": 1.8684691694222915,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004647941758698547,
      "loss": 5.1355,
      "mean_token_accuracy": 0.1816083699464798,
      "num_tokens": 41655450.0,
      "step": 24015
    },
    {
      "entropy": 5.58498477935791,
      "epoch": 1.868858198794009,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000464779288870739,
      "loss": 5.2626,
      "mean_token_accuracy": 0.18196014761924745,
      "num_tokens": 41664295.0,
      "step": 24020
    },
    {
      "entropy": 5.379287385940552,
      "epoch": 1.8692472281657264,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046476439899202464,
      "loss": 4.9557,
      "mean_token_accuracy": 0.1983610764145851,
      "num_tokens": 41672025.0,
      "step": 24025
    },
    {
      "entropy": 5.375784301757813,
      "epoch": 1.8696362575374441,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00046474950623393756,
      "loss": 5.1824,
      "mean_token_accuracy": 0.1824754446744919,
      "num_tokens": 41680360.0,
      "step": 24030
    },
    {
      "entropy": 5.512145566940307,
      "epoch": 1.8700252869091616,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004647346105967038,
      "loss": 5.1572,
      "mean_token_accuracy": 0.18014441430568695,
      "num_tokens": 41689865.0,
      "step": 24035
    },
    {
      "entropy": 5.519852638244629,
      "epoch": 1.870414316280879,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004647197120805493,
      "loss": 5.3154,
      "mean_token_accuracy": 0.17120766043663024,
      "num_tokens": 41698583.0,
      "step": 24040
    },
    {
      "entropy": 5.48077802658081,
      "epoch": 1.8708033456525968,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046470481068570013,
      "loss": 5.3424,
      "mean_token_accuracy": 0.17267362624406815,
      "num_tokens": 41708107.0,
      "step": 24045
    },
    {
      "entropy": 5.5476541996002195,
      "epoch": 1.8711923750243145,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004646899064123824,
      "loss": 5.2229,
      "mean_token_accuracy": 0.17752458453178405,
      "num_tokens": 41716446.0,
      "step": 24050
    },
    {
      "entropy": 5.483295059204101,
      "epoch": 1.871581404396032,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004646749992608223,
      "loss": 5.1782,
      "mean_token_accuracy": 0.18513074070215224,
      "num_tokens": 41726142.0,
      "step": 24055
    },
    {
      "entropy": 5.475309705734253,
      "epoch": 1.8719704337677494,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004646600892312459,
      "loss": 5.1525,
      "mean_token_accuracy": 0.18742657452821732,
      "num_tokens": 41734488.0,
      "step": 24060
    },
    {
      "entropy": 5.4400660514831545,
      "epoch": 1.872359463139467,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004646451763238796,
      "loss": 5.1278,
      "mean_token_accuracy": 0.18037073761224748,
      "num_tokens": 41743164.0,
      "step": 24065
    },
    {
      "entropy": 5.506045007705689,
      "epoch": 1.8727484925111846,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004646302605389496,
      "loss": 5.2149,
      "mean_token_accuracy": 0.1824511393904686,
      "num_tokens": 41751994.0,
      "step": 24070
    },
    {
      "entropy": 5.550379228591919,
      "epoch": 1.873137521882902,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004646153418766822,
      "loss": 5.266,
      "mean_token_accuracy": 0.18220161646604538,
      "num_tokens": 41761801.0,
      "step": 24075
    },
    {
      "entropy": 5.586080265045166,
      "epoch": 1.8735265512546198,
      "grad_norm": 1.25,
      "learning_rate": 0.00046460042033730377,
      "loss": 5.2756,
      "mean_token_accuracy": 0.182007272541523,
      "num_tokens": 41770771.0,
      "step": 24080
    },
    {
      "entropy": 5.506754589080811,
      "epoch": 1.8739155806263375,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004645854959210408,
      "loss": 5.2068,
      "mean_token_accuracy": 0.18073942214250566,
      "num_tokens": 41779813.0,
      "step": 24085
    },
    {
      "entropy": 5.52459306716919,
      "epoch": 1.8743046099980547,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046457056862811956,
      "loss": 5.2526,
      "mean_token_accuracy": 0.18616203218698502,
      "num_tokens": 41790076.0,
      "step": 24090
    },
    {
      "entropy": 5.523430728912354,
      "epoch": 1.8746936393697724,
      "grad_norm": 1.25,
      "learning_rate": 0.0004645556384587668,
      "loss": 5.1363,
      "mean_token_accuracy": 0.18820858895778655,
      "num_tokens": 41799448.0,
      "step": 24095
    },
    {
      "entropy": 5.537123966217041,
      "epoch": 1.87508266874149,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004645407054132089,
      "loss": 5.2244,
      "mean_token_accuracy": 0.17518313080072404,
      "num_tokens": 41808117.0,
      "step": 24100
    },
    {
      "entropy": 5.434284687042236,
      "epoch": 1.8754716981132076,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004645257694916725,
      "loss": 5.0941,
      "mean_token_accuracy": 0.19359175860881805,
      "num_tokens": 41816176.0,
      "step": 24105
    },
    {
      "entropy": 5.4685076713562015,
      "epoch": 1.875860727484925,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004645108306943842,
      "loss": 5.1868,
      "mean_token_accuracy": 0.18438277244567872,
      "num_tokens": 41825093.0,
      "step": 24110
    },
    {
      "entropy": 5.467846632003784,
      "epoch": 1.8762497568566427,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004644958890215707,
      "loss": 5.2758,
      "mean_token_accuracy": 0.17675289511680603,
      "num_tokens": 41834524.0,
      "step": 24115
    },
    {
      "entropy": 5.4337804317474365,
      "epoch": 1.8766387862283602,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004644809444734586,
      "loss": 5.0686,
      "mean_token_accuracy": 0.18909573405981064,
      "num_tokens": 41843188.0,
      "step": 24120
    },
    {
      "entropy": 5.484651756286621,
      "epoch": 1.8770278156000777,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046446599705027487,
      "loss": 5.1909,
      "mean_token_accuracy": 0.18062212616205214,
      "num_tokens": 41852296.0,
      "step": 24125
    },
    {
      "entropy": 5.566081094741821,
      "epoch": 1.8774168449717954,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046445104675224607,
      "loss": 5.2102,
      "mean_token_accuracy": 0.1837824121117592,
      "num_tokens": 41861641.0,
      "step": 24130
    },
    {
      "entropy": 5.422137689590454,
      "epoch": 1.877805874343513,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004644360935795993,
      "loss": 5.1286,
      "mean_token_accuracy": 0.17997875958681106,
      "num_tokens": 41870463.0,
      "step": 24135
    },
    {
      "entropy": 5.457820034027099,
      "epoch": 1.8781949037152303,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046442113753256126,
      "loss": 5.2089,
      "mean_token_accuracy": 0.1828642323613167,
      "num_tokens": 41878739.0,
      "step": 24140
    },
    {
      "entropy": 5.518969440460205,
      "epoch": 1.878583933086948,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046440617861135905,
      "loss": 5.1974,
      "mean_token_accuracy": 0.18267695009708404,
      "num_tokens": 41887211.0,
      "step": 24145
    },
    {
      "entropy": 5.512998580932617,
      "epoch": 1.8789729624586657,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004643912168162194,
      "loss": 5.1056,
      "mean_token_accuracy": 0.18735189139842987,
      "num_tokens": 41895742.0,
      "step": 24150
    },
    {
      "entropy": 5.4816575050354,
      "epoch": 1.8793619918303832,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004643762521473696,
      "loss": 5.165,
      "mean_token_accuracy": 0.18422200977802278,
      "num_tokens": 41904866.0,
      "step": 24155
    },
    {
      "entropy": 5.523362159729004,
      "epoch": 1.8797510212021007,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004643612846050366,
      "loss": 5.2842,
      "mean_token_accuracy": 0.17461914867162703,
      "num_tokens": 41914237.0,
      "step": 24160
    },
    {
      "entropy": 5.398854398727417,
      "epoch": 1.8801400505738184,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046434631418944744,
      "loss": 5.1434,
      "mean_token_accuracy": 0.18566177040338516,
      "num_tokens": 41923070.0,
      "step": 24165
    },
    {
      "entropy": 5.49472861289978,
      "epoch": 1.8805290799455359,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004643313409008294,
      "loss": 5.1399,
      "mean_token_accuracy": 0.18025409281253815,
      "num_tokens": 41932076.0,
      "step": 24170
    },
    {
      "entropy": 5.472627592086792,
      "epoch": 1.8809181093172533,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046431636473940956,
      "loss": 5.1894,
      "mean_token_accuracy": 0.1793360412120819,
      "num_tokens": 41941169.0,
      "step": 24175
    },
    {
      "entropy": 5.41436357498169,
      "epoch": 1.881307138688971,
      "grad_norm": 1.25,
      "learning_rate": 0.0004643013857054152,
      "loss": 5.0684,
      "mean_token_accuracy": 0.18709639757871627,
      "num_tokens": 41949499.0,
      "step": 24180
    },
    {
      "entropy": 5.552463579177856,
      "epoch": 1.8816961680606887,
      "grad_norm": 1.765625,
      "learning_rate": 0.00046428640379907367,
      "loss": 5.2551,
      "mean_token_accuracy": 0.1800643637776375,
      "num_tokens": 41957274.0,
      "step": 24185
    },
    {
      "entropy": 5.42193717956543,
      "epoch": 1.882085197432406,
      "grad_norm": 1.53125,
      "learning_rate": 0.00046427141902061225,
      "loss": 5.089,
      "mean_token_accuracy": 0.1853225514292717,
      "num_tokens": 41965996.0,
      "step": 24190
    },
    {
      "entropy": 5.471542453765869,
      "epoch": 1.8824742268041237,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004642564313702582,
      "loss": 5.1442,
      "mean_token_accuracy": 0.18504322618246077,
      "num_tokens": 41974652.0,
      "step": 24195
    },
    {
      "entropy": 5.5182465553283695,
      "epoch": 1.8828632561758414,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046424144084823916,
      "loss": 5.165,
      "mean_token_accuracy": 0.17763276398181915,
      "num_tokens": 41982699.0,
      "step": 24200
    },
    {
      "entropy": 5.501750469207764,
      "epoch": 1.8832522855475589,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004642264474547824,
      "loss": 5.1158,
      "mean_token_accuracy": 0.18840053528547288,
      "num_tokens": 41991441.0,
      "step": 24205
    },
    {
      "entropy": 5.409266805648803,
      "epoch": 1.8836413149192763,
      "grad_norm": 1.5,
      "learning_rate": 0.00046421145119011556,
      "loss": 5.1069,
      "mean_token_accuracy": 0.1937762826681137,
      "num_tokens": 42000683.0,
      "step": 24210
    },
    {
      "entropy": 5.4898370742797855,
      "epoch": 1.884030344290994,
      "grad_norm": 1.3125,
      "learning_rate": 0.000464196452054466,
      "loss": 5.2135,
      "mean_token_accuracy": 0.18022435009479523,
      "num_tokens": 42009895.0,
      "step": 24215
    },
    {
      "entropy": 5.475457000732422,
      "epoch": 1.8844193736627115,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004641814500480615,
      "loss": 5.0522,
      "mean_token_accuracy": 0.18679832071065902,
      "num_tokens": 42018086.0,
      "step": 24220
    },
    {
      "entropy": 5.497677850723266,
      "epoch": 1.884808403034429,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004641664451711296,
      "loss": 5.2198,
      "mean_token_accuracy": 0.1766365259885788,
      "num_tokens": 42026959.0,
      "step": 24225
    },
    {
      "entropy": 5.376885557174683,
      "epoch": 1.8851974324061467,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046415143742389793,
      "loss": 5.1518,
      "mean_token_accuracy": 0.18459801077842714,
      "num_tokens": 42035641.0,
      "step": 24230
    },
    {
      "entropy": 5.471140909194946,
      "epoch": 1.8855864617778644,
      "grad_norm": 1.25,
      "learning_rate": 0.0004641364268065943,
      "loss": 5.1209,
      "mean_token_accuracy": 0.1894539326429367,
      "num_tokens": 42044565.0,
      "step": 24235
    },
    {
      "entropy": 5.564204502105713,
      "epoch": 1.8859754911495816,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00046412141331944644,
      "loss": 5.1985,
      "mean_token_accuracy": 0.18171575367450715,
      "num_tokens": 42053088.0,
      "step": 24240
    },
    {
      "entropy": 5.476246404647827,
      "epoch": 1.8863645205212993,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046410639696268206,
      "loss": 5.179,
      "mean_token_accuracy": 0.1850145012140274,
      "num_tokens": 42061880.0,
      "step": 24245
    },
    {
      "entropy": 5.396104431152343,
      "epoch": 1.886753549893017,
      "grad_norm": 1.25,
      "learning_rate": 0.0004640913777365292,
      "loss": 5.1479,
      "mean_token_accuracy": 0.1812047079205513,
      "num_tokens": 42070767.0,
      "step": 24250
    },
    {
      "entropy": 5.460137462615966,
      "epoch": 1.8871425792647345,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00046407635564121565,
      "loss": 5.1456,
      "mean_token_accuracy": 0.18472465127706528,
      "num_tokens": 42080037.0,
      "step": 24255
    },
    {
      "entropy": 5.489675855636596,
      "epoch": 1.887531608636452,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00046406133067696936,
      "loss": 5.119,
      "mean_token_accuracy": 0.18877738565206528,
      "num_tokens": 42088372.0,
      "step": 24260
    },
    {
      "entropy": 5.418736791610717,
      "epoch": 1.8879206380081697,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004640463028440182,
      "loss": 5.1241,
      "mean_token_accuracy": 0.1886160746216774,
      "num_tokens": 42096550.0,
      "step": 24265
    },
    {
      "entropy": 5.472839593887329,
      "epoch": 1.8883096673798871,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004640312721425904,
      "loss": 5.1559,
      "mean_token_accuracy": 0.19066636264324188,
      "num_tokens": 42105660.0,
      "step": 24270
    },
    {
      "entropy": 5.49428915977478,
      "epoch": 1.8886986967516046,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004640162385729139,
      "loss": 5.1125,
      "mean_token_accuracy": 0.19055162221193314,
      "num_tokens": 42114586.0,
      "step": 24275
    },
    {
      "entropy": 5.450196552276611,
      "epoch": 1.8890877261233223,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004640012021352168,
      "loss": 5.1452,
      "mean_token_accuracy": 0.19101667255163193,
      "num_tokens": 42123104.0,
      "step": 24280
    },
    {
      "entropy": 5.55902247428894,
      "epoch": 1.88947675549504,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004639861628297273,
      "loss": 5.2758,
      "mean_token_accuracy": 0.17615424245595931,
      "num_tokens": 42132127.0,
      "step": 24285
    },
    {
      "entropy": 5.5298919677734375,
      "epoch": 1.8898657848667575,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046397112065667354,
      "loss": 5.2409,
      "mean_token_accuracy": 0.17821563631296158,
      "num_tokens": 42141586.0,
      "step": 24290
    },
    {
      "entropy": 5.467160224914551,
      "epoch": 1.890254814238475,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046395607561628387,
      "loss": 5.1787,
      "mean_token_accuracy": 0.18007429987192153,
      "num_tokens": 42151003.0,
      "step": 24295
    },
    {
      "entropy": 5.493098258972168,
      "epoch": 1.8906438436101927,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00046394102770878643,
      "loss": 5.1853,
      "mean_token_accuracy": 0.18469370603561402,
      "num_tokens": 42159291.0,
      "step": 24300
    },
    {
      "entropy": 5.5369429111480715,
      "epoch": 1.8910328729819101,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046392597693440974,
      "loss": 5.2629,
      "mean_token_accuracy": 0.17496658116579056,
      "num_tokens": 42167925.0,
      "step": 24305
    },
    {
      "entropy": 5.500406694412232,
      "epoch": 1.8914219023536276,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004639109232933819,
      "loss": 5.1572,
      "mean_token_accuracy": 0.18934030085802078,
      "num_tokens": 42175866.0,
      "step": 24310
    },
    {
      "entropy": 5.455942058563233,
      "epoch": 1.8918109317253453,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004638958667859316,
      "loss": 5.1324,
      "mean_token_accuracy": 0.200722573697567,
      "num_tokens": 42184233.0,
      "step": 24315
    },
    {
      "entropy": 5.477989721298218,
      "epoch": 1.8921999610970628,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004638808074122872,
      "loss": 5.209,
      "mean_token_accuracy": 0.18311432600021363,
      "num_tokens": 42192972.0,
      "step": 24320
    },
    {
      "entropy": 5.436160755157471,
      "epoch": 1.8925889904687803,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004638657451726771,
      "loss": 5.0976,
      "mean_token_accuracy": 0.18958324790000916,
      "num_tokens": 42201380.0,
      "step": 24325
    },
    {
      "entropy": 5.444985818862915,
      "epoch": 1.892978019840498,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046385068006732995,
      "loss": 5.1742,
      "mean_token_accuracy": 0.19034308940172195,
      "num_tokens": 42209276.0,
      "step": 24330
    },
    {
      "entropy": 5.585738563537598,
      "epoch": 1.8933670492122157,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00046383561209647437,
      "loss": 5.3202,
      "mean_token_accuracy": 0.1718480780720711,
      "num_tokens": 42218682.0,
      "step": 24335
    },
    {
      "entropy": 5.660860776901245,
      "epoch": 1.8937560785839331,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046382054126033884,
      "loss": 5.3636,
      "mean_token_accuracy": 0.1705701857805252,
      "num_tokens": 42226692.0,
      "step": 24340
    },
    {
      "entropy": 5.539766120910644,
      "epoch": 1.8941451079556506,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004638054675591523,
      "loss": 5.2864,
      "mean_token_accuracy": 0.16784291565418244,
      "num_tokens": 42235582.0,
      "step": 24345
    },
    {
      "entropy": 5.549159574508667,
      "epoch": 1.8945341373273683,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004637903909931432,
      "loss": 5.1455,
      "mean_token_accuracy": 0.18435688316822052,
      "num_tokens": 42244354.0,
      "step": 24350
    },
    {
      "entropy": 5.465712451934815,
      "epoch": 1.8949231666990858,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004637753115625404,
      "loss": 5.0755,
      "mean_token_accuracy": 0.19005461782217026,
      "num_tokens": 42253062.0,
      "step": 24355
    },
    {
      "entropy": 5.432435417175293,
      "epoch": 1.8953121960708033,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004637602292675726,
      "loss": 5.2125,
      "mean_token_accuracy": 0.17695188224315644,
      "num_tokens": 42262380.0,
      "step": 24360
    },
    {
      "entropy": 5.502747201919556,
      "epoch": 1.895701225442521,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004637451441084689,
      "loss": 5.0929,
      "mean_token_accuracy": 0.19193921536207198,
      "num_tokens": 42270262.0,
      "step": 24365
    },
    {
      "entropy": 5.4889427661895756,
      "epoch": 1.8960902548142384,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004637300560854581,
      "loss": 5.2289,
      "mean_token_accuracy": 0.18317514657974243,
      "num_tokens": 42278935.0,
      "step": 24370
    },
    {
      "entropy": 5.4823219776153564,
      "epoch": 1.896479284185956,
      "grad_norm": 1.265625,
      "learning_rate": 0.000463714965198769,
      "loss": 5.1278,
      "mean_token_accuracy": 0.18473281264305114,
      "num_tokens": 42287354.0,
      "step": 24375
    },
    {
      "entropy": 5.51255464553833,
      "epoch": 1.8968683135576736,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004636998714486307,
      "loss": 5.1323,
      "mean_token_accuracy": 0.1845674231648445,
      "num_tokens": 42296475.0,
      "step": 24380
    },
    {
      "entropy": 5.545323514938355,
      "epoch": 1.8972573429293913,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046368477483527224,
      "loss": 5.2555,
      "mean_token_accuracy": 0.17985828667879106,
      "num_tokens": 42305306.0,
      "step": 24385
    },
    {
      "entropy": 5.4085352420806885,
      "epoch": 1.8976463723011088,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004636696753589226,
      "loss": 5.1971,
      "mean_token_accuracy": 0.17815522998571395,
      "num_tokens": 42314827.0,
      "step": 24390
    },
    {
      "entropy": 5.374692773818969,
      "epoch": 1.8980354016728263,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004636545730198109,
      "loss": 5.0799,
      "mean_token_accuracy": 0.19325724989175797,
      "num_tokens": 42323396.0,
      "step": 24395
    },
    {
      "entropy": 5.514501571655273,
      "epoch": 1.898424431044544,
      "grad_norm": 1.421875,
      "learning_rate": 0.00046363946781816643,
      "loss": 5.1626,
      "mean_token_accuracy": 0.18632536083459855,
      "num_tokens": 42331402.0,
      "step": 24400
    },
    {
      "entropy": 5.396797752380371,
      "epoch": 1.8988134604162614,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046362435975421816,
      "loss": 5.1078,
      "mean_token_accuracy": 0.19595541805028915,
      "num_tokens": 42339458.0,
      "step": 24405
    },
    {
      "entropy": 5.496005010604859,
      "epoch": 1.899202489787979,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046360924882819554,
      "loss": 5.2025,
      "mean_token_accuracy": 0.18168049454689025,
      "num_tokens": 42346713.0,
      "step": 24410
    },
    {
      "entropy": 5.41813178062439,
      "epoch": 1.8995915191596966,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004635941350403277,
      "loss": 5.0804,
      "mean_token_accuracy": 0.18585621565580368,
      "num_tokens": 42355429.0,
      "step": 24415
    },
    {
      "entropy": 5.395499038696289,
      "epoch": 1.899980548531414,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004635790183908442,
      "loss": 5.1331,
      "mean_token_accuracy": 0.18568074107170104,
      "num_tokens": 42364228.0,
      "step": 24420
    },
    {
      "entropy": 5.420674848556518,
      "epoch": 1.9003695779031315,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046356389887997415,
      "loss": 5.0443,
      "mean_token_accuracy": 0.18802448660135268,
      "num_tokens": 42372837.0,
      "step": 24425
    },
    {
      "entropy": 5.434771776199341,
      "epoch": 1.9007586072748492,
      "grad_norm": 1.25,
      "learning_rate": 0.00046354877650794707,
      "loss": 5.0797,
      "mean_token_accuracy": 0.19657609462738038,
      "num_tokens": 42381792.0,
      "step": 24430
    },
    {
      "entropy": 5.478389501571655,
      "epoch": 1.901147636646567,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046353365127499235,
      "loss": 5.211,
      "mean_token_accuracy": 0.17944760024547576,
      "num_tokens": 42390650.0,
      "step": 24435
    },
    {
      "entropy": 5.468466567993164,
      "epoch": 1.9015366660182844,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004635185231813396,
      "loss": 5.1482,
      "mean_token_accuracy": 0.19282734990119935,
      "num_tokens": 42399253.0,
      "step": 24440
    },
    {
      "entropy": 5.567887878417968,
      "epoch": 1.901925695390002,
      "grad_norm": 1.375,
      "learning_rate": 0.0004635033922272183,
      "loss": 5.365,
      "mean_token_accuracy": 0.1687123730778694,
      "num_tokens": 42407916.0,
      "step": 24445
    },
    {
      "entropy": 5.407737588882446,
      "epoch": 1.9023147247617196,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046348825841285813,
      "loss": 5.0707,
      "mean_token_accuracy": 0.19327960461378096,
      "num_tokens": 42416151.0,
      "step": 24450
    },
    {
      "entropy": 5.5105267524719235,
      "epoch": 1.902703754133437,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004634731217384886,
      "loss": 5.2865,
      "mean_token_accuracy": 0.17415828704833985,
      "num_tokens": 42424845.0,
      "step": 24455
    },
    {
      "entropy": 5.612914991378784,
      "epoch": 1.9030927835051545,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004634579822043394,
      "loss": 5.1847,
      "mean_token_accuracy": 0.17936229556798935,
      "num_tokens": 42434022.0,
      "step": 24460
    },
    {
      "entropy": 5.56455512046814,
      "epoch": 1.9034818128768722,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046344283981064035,
      "loss": 5.2286,
      "mean_token_accuracy": 0.18321893662214278,
      "num_tokens": 42443614.0,
      "step": 24465
    },
    {
      "entropy": 5.386825656890869,
      "epoch": 1.90387084224859,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046342769455762114,
      "loss": 5.0531,
      "mean_token_accuracy": 0.18732108920812607,
      "num_tokens": 42452389.0,
      "step": 24470
    },
    {
      "entropy": 5.42635440826416,
      "epoch": 1.9042598716203072,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004634125464455116,
      "loss": 5.1537,
      "mean_token_accuracy": 0.1788182109594345,
      "num_tokens": 42460796.0,
      "step": 24475
    },
    {
      "entropy": 5.5532256126403805,
      "epoch": 1.9046489009920249,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046339739547454146,
      "loss": 5.2471,
      "mean_token_accuracy": 0.18061573207378387,
      "num_tokens": 42470108.0,
      "step": 24480
    },
    {
      "entropy": 5.478749227523804,
      "epoch": 1.9050379303637426,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046338224164494074,
      "loss": 5.1837,
      "mean_token_accuracy": 0.1765371322631836,
      "num_tokens": 42478361.0,
      "step": 24485
    },
    {
      "entropy": 5.52775468826294,
      "epoch": 1.90542695973546,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046336708495693933,
      "loss": 5.2449,
      "mean_token_accuracy": 0.1775071680545807,
      "num_tokens": 42486254.0,
      "step": 24490
    },
    {
      "entropy": 5.452859020233154,
      "epoch": 1.9058159891071775,
      "grad_norm": 1.390625,
      "learning_rate": 0.00046335192541076725,
      "loss": 5.0929,
      "mean_token_accuracy": 0.19359275698661804,
      "num_tokens": 42495379.0,
      "step": 24495
    },
    {
      "entropy": 5.4394957542419435,
      "epoch": 1.9062050184788952,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004633367630066545,
      "loss": 5.1654,
      "mean_token_accuracy": 0.18104279637336732,
      "num_tokens": 42503777.0,
      "step": 24500
    },
    {
      "entropy": 5.51709017753601,
      "epoch": 1.9065940478506127,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046332159774483116,
      "loss": 5.242,
      "mean_token_accuracy": 0.17623042762279512,
      "num_tokens": 42512557.0,
      "step": 24505
    },
    {
      "entropy": 5.469807434082031,
      "epoch": 1.9069830772223302,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004633064296255273,
      "loss": 5.1528,
      "mean_token_accuracy": 0.18538284301757812,
      "num_tokens": 42521245.0,
      "step": 24510
    },
    {
      "entropy": 5.49872670173645,
      "epoch": 1.9073721065940479,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046329125864897307,
      "loss": 5.1296,
      "mean_token_accuracy": 0.1875286504626274,
      "num_tokens": 42529726.0,
      "step": 24515
    },
    {
      "entropy": 5.46023588180542,
      "epoch": 1.9077611359657656,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004632760848153987,
      "loss": 5.1425,
      "mean_token_accuracy": 0.18572385460138321,
      "num_tokens": 42538376.0,
      "step": 24520
    },
    {
      "entropy": 5.471598863601685,
      "epoch": 1.9081501653374828,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004632609081250345,
      "loss": 5.1805,
      "mean_token_accuracy": 0.1787942185997963,
      "num_tokens": 42546977.0,
      "step": 24525
    },
    {
      "entropy": 5.5511139869689945,
      "epoch": 1.9085391947092005,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046324572857811054,
      "loss": 5.1785,
      "mean_token_accuracy": 0.17790181338787078,
      "num_tokens": 42555531.0,
      "step": 24530
    },
    {
      "entropy": 5.515816307067871,
      "epoch": 1.9089282240809182,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004632305461748573,
      "loss": 5.2135,
      "mean_token_accuracy": 0.18481094837188722,
      "num_tokens": 42564803.0,
      "step": 24535
    },
    {
      "entropy": 5.518814992904663,
      "epoch": 1.9093172534526357,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046321536091550517,
      "loss": 5.2119,
      "mean_token_accuracy": 0.1798251450061798,
      "num_tokens": 42573560.0,
      "step": 24540
    },
    {
      "entropy": 5.49469313621521,
      "epoch": 1.9097062828243532,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004632001728002845,
      "loss": 5.2082,
      "mean_token_accuracy": 0.18028008043766022,
      "num_tokens": 42582177.0,
      "step": 24545
    },
    {
      "entropy": 5.475135850906372,
      "epoch": 1.9100953121960709,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004631849818294257,
      "loss": 5.1803,
      "mean_token_accuracy": 0.1802906021475792,
      "num_tokens": 42590874.0,
      "step": 24550
    },
    {
      "entropy": 5.43516354560852,
      "epoch": 1.9104843415677883,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046316978800315934,
      "loss": 5.0961,
      "mean_token_accuracy": 0.19092043191194535,
      "num_tokens": 42599644.0,
      "step": 24555
    },
    {
      "entropy": 5.4462066173553465,
      "epoch": 1.9108733709395058,
      "grad_norm": 1.265625,
      "learning_rate": 0.000463154591321716,
      "loss": 5.1619,
      "mean_token_accuracy": 0.1807519018650055,
      "num_tokens": 42609207.0,
      "step": 24560
    },
    {
      "entropy": 5.527137613296508,
      "epoch": 1.9112624003112235,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046313939178532624,
      "loss": 5.2717,
      "mean_token_accuracy": 0.17353625148534774,
      "num_tokens": 42618669.0,
      "step": 24565
    },
    {
      "entropy": 5.584531259536743,
      "epoch": 1.9116514296829412,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004631241893942206,
      "loss": 5.1929,
      "mean_token_accuracy": 0.17800243347883224,
      "num_tokens": 42628433.0,
      "step": 24570
    },
    {
      "entropy": 5.516142463684082,
      "epoch": 1.9120404590546585,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046310898414862983,
      "loss": 5.1395,
      "mean_token_accuracy": 0.1809336870908737,
      "num_tokens": 42636839.0,
      "step": 24575
    },
    {
      "entropy": 5.437991666793823,
      "epoch": 1.9124294884263762,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004630937760487847,
      "loss": 5.129,
      "mean_token_accuracy": 0.187970107793808,
      "num_tokens": 42645328.0,
      "step": 24580
    },
    {
      "entropy": 5.49188461303711,
      "epoch": 1.9128185177980939,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004630785650949158,
      "loss": 5.2445,
      "mean_token_accuracy": 0.17445521652698517,
      "num_tokens": 42653883.0,
      "step": 24585
    },
    {
      "entropy": 5.4548211097717285,
      "epoch": 1.9132075471698113,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004630633512872541,
      "loss": 5.1874,
      "mean_token_accuracy": 0.18122837394475938,
      "num_tokens": 42662364.0,
      "step": 24590
    },
    {
      "entropy": 5.453305912017822,
      "epoch": 1.9135965765415288,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046304813462603035,
      "loss": 5.051,
      "mean_token_accuracy": 0.19155173301696776,
      "num_tokens": 42670424.0,
      "step": 24595
    },
    {
      "entropy": 5.4290824890136715,
      "epoch": 1.9139856059132465,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004630329151114754,
      "loss": 5.1229,
      "mean_token_accuracy": 0.18964357376098634,
      "num_tokens": 42679494.0,
      "step": 24600
    },
    {
      "entropy": 5.596363544464111,
      "epoch": 1.914374635284964,
      "grad_norm": 1.6875,
      "learning_rate": 0.00046301769274382034,
      "loss": 5.3167,
      "mean_token_accuracy": 0.17971367239952088,
      "num_tokens": 42688042.0,
      "step": 24605
    },
    {
      "entropy": 5.603968667984009,
      "epoch": 1.9147636646566815,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004630024675232961,
      "loss": 5.2719,
      "mean_token_accuracy": 0.18046784549951553,
      "num_tokens": 42696859.0,
      "step": 24610
    },
    {
      "entropy": 5.53309440612793,
      "epoch": 1.9151526940283992,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046298723945013354,
      "loss": 5.1868,
      "mean_token_accuracy": 0.17964863777160645,
      "num_tokens": 42705291.0,
      "step": 24615
    },
    {
      "entropy": 5.423347854614258,
      "epoch": 1.9155417234001169,
      "grad_norm": 1.25,
      "learning_rate": 0.0004629720085245639,
      "loss": 5.2011,
      "mean_token_accuracy": 0.18143181949853898,
      "num_tokens": 42713777.0,
      "step": 24620
    },
    {
      "entropy": 5.494828701019287,
      "epoch": 1.9159307527718341,
      "grad_norm": 1.25,
      "learning_rate": 0.0004629567747468182,
      "loss": 5.2383,
      "mean_token_accuracy": 0.18515785932540893,
      "num_tokens": 42722373.0,
      "step": 24625
    },
    {
      "entropy": 5.534469270706177,
      "epoch": 1.9163197821435518,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004629415381171276,
      "loss": 5.1088,
      "mean_token_accuracy": 0.1884836584329605,
      "num_tokens": 42730791.0,
      "step": 24630
    },
    {
      "entropy": 5.438490390777588,
      "epoch": 1.9167088115152695,
      "grad_norm": 1.390625,
      "learning_rate": 0.00046292629863572325,
      "loss": 5.2066,
      "mean_token_accuracy": 0.18314601927995683,
      "num_tokens": 42739448.0,
      "step": 24635
    },
    {
      "entropy": 5.501352977752686,
      "epoch": 1.917097840886987,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004629110563028365,
      "loss": 5.2018,
      "mean_token_accuracy": 0.17880702912807464,
      "num_tokens": 42748757.0,
      "step": 24640
    },
    {
      "entropy": 5.512107419967651,
      "epoch": 1.9174868702587045,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004628958111186985,
      "loss": 5.1,
      "mean_token_accuracy": 0.18767209500074386,
      "num_tokens": 42757166.0,
      "step": 24645
    },
    {
      "entropy": 5.474408054351807,
      "epoch": 1.9178758996304222,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004628805630835407,
      "loss": 5.1389,
      "mean_token_accuracy": 0.18309541642665864,
      "num_tokens": 42765671.0,
      "step": 24650
    },
    {
      "entropy": 5.4812366485595705,
      "epoch": 1.9182649290021396,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004628653121975944,
      "loss": 5.1758,
      "mean_token_accuracy": 0.18782023191452027,
      "num_tokens": 42774499.0,
      "step": 24655
    },
    {
      "entropy": 5.484792852401734,
      "epoch": 1.918653958373857,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000462850058461091,
      "loss": 5.1525,
      "mean_token_accuracy": 0.18305207490921022,
      "num_tokens": 42783237.0,
      "step": 24660
    },
    {
      "entropy": 5.488727712631226,
      "epoch": 1.9190429877455748,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004628348018742619,
      "loss": 5.1134,
      "mean_token_accuracy": 0.18319776505231858,
      "num_tokens": 42791217.0,
      "step": 24665
    },
    {
      "entropy": 5.384256601333618,
      "epoch": 1.9194320171172925,
      "grad_norm": 1.5,
      "learning_rate": 0.0004628195424373387,
      "loss": 5.119,
      "mean_token_accuracy": 0.18912703692913055,
      "num_tokens": 42799885.0,
      "step": 24670
    },
    {
      "entropy": 5.423672819137574,
      "epoch": 1.91982104648901,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046280428015055286,
      "loss": 5.1474,
      "mean_token_accuracy": 0.18288756757974625,
      "num_tokens": 42809141.0,
      "step": 24675
    },
    {
      "entropy": 5.556517219543457,
      "epoch": 1.9202100758607274,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046278901501413606,
      "loss": 5.2235,
      "mean_token_accuracy": 0.18359093219041825,
      "num_tokens": 42817809.0,
      "step": 24680
    },
    {
      "entropy": 5.436758327484131,
      "epoch": 1.9205991052324451,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00046277374702831983,
      "loss": 5.1055,
      "mean_token_accuracy": 0.18228615075349808,
      "num_tokens": 42826813.0,
      "step": 24685
    },
    {
      "entropy": 5.4531676292419435,
      "epoch": 1.9209881346041626,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046275847619333576,
      "loss": 5.1598,
      "mean_token_accuracy": 0.1791495278477669,
      "num_tokens": 42835325.0,
      "step": 24690
    },
    {
      "entropy": 5.530086278915405,
      "epoch": 1.92137716397588,
      "grad_norm": 1.375,
      "learning_rate": 0.00046274320250941576,
      "loss": 5.238,
      "mean_token_accuracy": 0.18319268226623536,
      "num_tokens": 42843975.0,
      "step": 24695
    },
    {
      "entropy": 5.424391269683838,
      "epoch": 1.9217661933475978,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046272792597679146,
      "loss": 5.0294,
      "mean_token_accuracy": 0.1888217270374298,
      "num_tokens": 42852146.0,
      "step": 24700
    },
    {
      "entropy": 5.459353923797607,
      "epoch": 1.9221552227193153,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046271264659569474,
      "loss": 5.1883,
      "mean_token_accuracy": 0.18058166205883025,
      "num_tokens": 42860958.0,
      "step": 24705
    },
    {
      "entropy": 5.512636327743531,
      "epoch": 1.9225442520910327,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004626973643663573,
      "loss": 5.0598,
      "mean_token_accuracy": 0.19570437222719192,
      "num_tokens": 42869281.0,
      "step": 24710
    },
    {
      "entropy": 5.451256513595581,
      "epoch": 1.9229332814627504,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004626820792890112,
      "loss": 5.2038,
      "mean_token_accuracy": 0.1903185099363327,
      "num_tokens": 42878359.0,
      "step": 24715
    },
    {
      "entropy": 5.418535757064819,
      "epoch": 1.9233223108344681,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004626667913638882,
      "loss": 5.0038,
      "mean_token_accuracy": 0.19892269521951675,
      "num_tokens": 42886156.0,
      "step": 24720
    },
    {
      "entropy": 5.358740997314453,
      "epoch": 1.9237113402061856,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004626515005912203,
      "loss": 5.0377,
      "mean_token_accuracy": 0.19124755412340164,
      "num_tokens": 42894970.0,
      "step": 24725
    },
    {
      "entropy": 5.435474300384522,
      "epoch": 1.924100369577903,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004626362069712397,
      "loss": 5.1916,
      "mean_token_accuracy": 0.18195788711309432,
      "num_tokens": 42903319.0,
      "step": 24730
    },
    {
      "entropy": 5.408607864379883,
      "epoch": 1.9244893989496208,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004626209105041782,
      "loss": 5.1895,
      "mean_token_accuracy": 0.18333597332239152,
      "num_tokens": 42911912.0,
      "step": 24735
    },
    {
      "entropy": 5.517161178588867,
      "epoch": 1.9248784283213383,
      "grad_norm": 1.3125,
      "learning_rate": 0.000462605611190268,
      "loss": 5.1225,
      "mean_token_accuracy": 0.19310240298509598,
      "num_tokens": 42920329.0,
      "step": 24740
    },
    {
      "entropy": 5.461654901504517,
      "epoch": 1.9252674576930557,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046259030902974133,
      "loss": 5.1434,
      "mean_token_accuracy": 0.18402402400970458,
      "num_tokens": 42928422.0,
      "step": 24745
    },
    {
      "entropy": 5.490858936309815,
      "epoch": 1.9256564870647734,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004625750040228302,
      "loss": 5.1313,
      "mean_token_accuracy": 0.18420308083295822,
      "num_tokens": 42937016.0,
      "step": 24750
    },
    {
      "entropy": 5.432418537139893,
      "epoch": 1.926045516436491,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046255969616976704,
      "loss": 5.1098,
      "mean_token_accuracy": 0.18644391596317292,
      "num_tokens": 42945209.0,
      "step": 24755
    },
    {
      "entropy": 5.484076738357544,
      "epoch": 1.9264345458082084,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000462544385470784,
      "loss": 5.1579,
      "mean_token_accuracy": 0.17972705364227295,
      "num_tokens": 42953798.0,
      "step": 24760
    },
    {
      "entropy": 5.473036479949951,
      "epoch": 1.926823575179926,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004625290719261134,
      "loss": 5.1362,
      "mean_token_accuracy": 0.18376094251871108,
      "num_tokens": 42961780.0,
      "step": 24765
    },
    {
      "entropy": 5.464092159271241,
      "epoch": 1.9272126045516438,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004625137555359876,
      "loss": 5.212,
      "mean_token_accuracy": 0.1846457153558731,
      "num_tokens": 42970087.0,
      "step": 24770
    },
    {
      "entropy": 5.4487128257751465,
      "epoch": 1.9276016339233613,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046249843630063905,
      "loss": 5.2009,
      "mean_token_accuracy": 0.1820971041917801,
      "num_tokens": 42978732.0,
      "step": 24775
    },
    {
      "entropy": 5.505410099029541,
      "epoch": 1.9279906632950787,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004624831142203001,
      "loss": 5.2219,
      "mean_token_accuracy": 0.18028130978345872,
      "num_tokens": 42986899.0,
      "step": 24780
    },
    {
      "entropy": 5.5234497547149655,
      "epoch": 1.9283796926667964,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046246778929520346,
      "loss": 5.1298,
      "mean_token_accuracy": 0.18116574585437775,
      "num_tokens": 42995047.0,
      "step": 24785
    },
    {
      "entropy": 5.463469409942627,
      "epoch": 1.928768722038514,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004624524615255814,
      "loss": 5.2048,
      "mean_token_accuracy": 0.18570882827043533,
      "num_tokens": 43004377.0,
      "step": 24790
    },
    {
      "entropy": 5.495899438858032,
      "epoch": 1.9291577514102314,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046243713091166655,
      "loss": 5.2328,
      "mean_token_accuracy": 0.1802689641714096,
      "num_tokens": 43013610.0,
      "step": 24795
    },
    {
      "entropy": 5.47877869606018,
      "epoch": 1.929546780781949,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004624217974536916,
      "loss": 5.1571,
      "mean_token_accuracy": 0.18857490122318268,
      "num_tokens": 43022512.0,
      "step": 24800
    },
    {
      "entropy": 5.400929546356201,
      "epoch": 1.9299358101536666,
      "grad_norm": 1.25,
      "learning_rate": 0.00046240646115188925,
      "loss": 5.0671,
      "mean_token_accuracy": 0.18943429738283157,
      "num_tokens": 43030588.0,
      "step": 24805
    },
    {
      "entropy": 5.542538022994995,
      "epoch": 1.930324839525384,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004623911220064921,
      "loss": 5.1958,
      "mean_token_accuracy": 0.1839794099330902,
      "num_tokens": 43038717.0,
      "step": 24810
    },
    {
      "entropy": 5.479157590866089,
      "epoch": 1.9307138688971017,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046237578001773297,
      "loss": 5.042,
      "mean_token_accuracy": 0.19117599427700044,
      "num_tokens": 43046736.0,
      "step": 24815
    },
    {
      "entropy": 5.363182115554809,
      "epoch": 1.9311028982688194,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046236043518584454,
      "loss": 4.9983,
      "mean_token_accuracy": 0.1980949565768242,
      "num_tokens": 43054773.0,
      "step": 24820
    },
    {
      "entropy": 5.445832061767578,
      "epoch": 1.931491927640537,
      "grad_norm": 1.25,
      "learning_rate": 0.0004623450875110597,
      "loss": 5.1518,
      "mean_token_accuracy": 0.18135943859815598,
      "num_tokens": 43062983.0,
      "step": 24825
    },
    {
      "entropy": 5.472748517990112,
      "epoch": 1.9318809570122544,
      "grad_norm": 1.34375,
      "learning_rate": 0.00046232973699361147,
      "loss": 5.093,
      "mean_token_accuracy": 0.1839778333902359,
      "num_tokens": 43071161.0,
      "step": 24830
    },
    {
      "entropy": 5.550309514999389,
      "epoch": 1.932269986383972,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004623143836337326,
      "loss": 5.3224,
      "mean_token_accuracy": 0.1758324071764946,
      "num_tokens": 43080514.0,
      "step": 24835
    },
    {
      "entropy": 5.471392488479614,
      "epoch": 1.9326590157556895,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046229902743165607,
      "loss": 5.0756,
      "mean_token_accuracy": 0.1939949944615364,
      "num_tokens": 43089177.0,
      "step": 24840
    },
    {
      "entropy": 5.497359657287598,
      "epoch": 1.933048045127407,
      "grad_norm": 1.375,
      "learning_rate": 0.0004622836683876149,
      "loss": 5.2299,
      "mean_token_accuracy": 0.1805806577205658,
      "num_tokens": 43097722.0,
      "step": 24845
    },
    {
      "entropy": 5.474992847442627,
      "epoch": 1.9334370744991247,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004622683065018422,
      "loss": 5.1161,
      "mean_token_accuracy": 0.1871684804558754,
      "num_tokens": 43106203.0,
      "step": 24850
    },
    {
      "entropy": 5.415040302276611,
      "epoch": 1.9338261038708422,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046225294177457105,
      "loss": 5.1944,
      "mean_token_accuracy": 0.18180101811885835,
      "num_tokens": 43114607.0,
      "step": 24855
    },
    {
      "entropy": 5.596254777908325,
      "epoch": 1.9342151332425597,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046223757420603445,
      "loss": 5.2964,
      "mean_token_accuracy": 0.17320114225149155,
      "num_tokens": 43123179.0,
      "step": 24860
    },
    {
      "entropy": 5.5797360897064205,
      "epoch": 1.9346041626142774,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004622222037964658,
      "loss": 5.1363,
      "mean_token_accuracy": 0.18786537647247314,
      "num_tokens": 43130883.0,
      "step": 24865
    },
    {
      "entropy": 5.347901010513306,
      "epoch": 1.934993191985995,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046220683054609815,
      "loss": 4.898,
      "mean_token_accuracy": 0.20393352657556535,
      "num_tokens": 43139812.0,
      "step": 24870
    },
    {
      "entropy": 5.465852975845337,
      "epoch": 1.9353822213577125,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00046219145445516483,
      "loss": 5.2255,
      "mean_token_accuracy": 0.17730635553598403,
      "num_tokens": 43148756.0,
      "step": 24875
    },
    {
      "entropy": 5.497779655456543,
      "epoch": 1.93577125072943,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046217607552389905,
      "loss": 5.2146,
      "mean_token_accuracy": 0.17359041273593903,
      "num_tokens": 43158131.0,
      "step": 24880
    },
    {
      "entropy": 5.455423974990845,
      "epoch": 1.9361602801011477,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046216069375253435,
      "loss": 4.9887,
      "mean_token_accuracy": 0.1951142132282257,
      "num_tokens": 43166400.0,
      "step": 24885
    },
    {
      "entropy": 5.480981111526489,
      "epoch": 1.9365493094728652,
      "grad_norm": 1.25,
      "learning_rate": 0.000462145309141304,
      "loss": 5.1182,
      "mean_token_accuracy": 0.18427708446979524,
      "num_tokens": 43175265.0,
      "step": 24890
    },
    {
      "entropy": 5.396628522872925,
      "epoch": 1.9369383388445827,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004621299216904414,
      "loss": 5.1811,
      "mean_token_accuracy": 0.17596068531274794,
      "num_tokens": 43183735.0,
      "step": 24895
    },
    {
      "entropy": 5.5798238754272464,
      "epoch": 1.9373273682163004,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046211453140018006,
      "loss": 5.2497,
      "mean_token_accuracy": 0.1837347626686096,
      "num_tokens": 43192217.0,
      "step": 24900
    },
    {
      "entropy": 5.515407609939575,
      "epoch": 1.937716397588018,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004620991382707537,
      "loss": 5.1815,
      "mean_token_accuracy": 0.1841086506843567,
      "num_tokens": 43200241.0,
      "step": 24905
    },
    {
      "entropy": 5.478782749176025,
      "epoch": 1.9381054269597353,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046208374230239556,
      "loss": 5.1517,
      "mean_token_accuracy": 0.18634021729230882,
      "num_tokens": 43209111.0,
      "step": 24910
    },
    {
      "entropy": 5.494665193557739,
      "epoch": 1.938494456331453,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004620683434953394,
      "loss": 5.2349,
      "mean_token_accuracy": 0.17851362377405167,
      "num_tokens": 43219091.0,
      "step": 24915
    },
    {
      "entropy": 5.464920425415039,
      "epoch": 1.9388834857031707,
      "grad_norm": 1.25,
      "learning_rate": 0.00046205294184981896,
      "loss": 5.0911,
      "mean_token_accuracy": 0.19249077141284943,
      "num_tokens": 43227140.0,
      "step": 24920
    },
    {
      "entropy": 5.468114042282105,
      "epoch": 1.9392725150748882,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046203753736606787,
      "loss": 5.2459,
      "mean_token_accuracy": 0.17988266795873642,
      "num_tokens": 43236356.0,
      "step": 24925
    },
    {
      "entropy": 5.424195194244385,
      "epoch": 1.9396615444466057,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004620221300443197,
      "loss": 5.0968,
      "mean_token_accuracy": 0.19473231136798858,
      "num_tokens": 43245083.0,
      "step": 24930
    },
    {
      "entropy": 5.49233341217041,
      "epoch": 1.9400505738183234,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004620067198848086,
      "loss": 5.1954,
      "mean_token_accuracy": 0.18348480612039567,
      "num_tokens": 43253042.0,
      "step": 24935
    },
    {
      "entropy": 5.408527851104736,
      "epoch": 1.9404396031900408,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046199130688776805,
      "loss": 5.1683,
      "mean_token_accuracy": 0.18580490946769715,
      "num_tokens": 43261783.0,
      "step": 24940
    },
    {
      "entropy": 5.529404163360596,
      "epoch": 1.9408286325617583,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046197589105343204,
      "loss": 5.1488,
      "mean_token_accuracy": 0.18931350409984588,
      "num_tokens": 43270286.0,
      "step": 24945
    },
    {
      "entropy": 5.41760516166687,
      "epoch": 1.941217661933476,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004619604723820345,
      "loss": 5.1864,
      "mean_token_accuracy": 0.18728157877922058,
      "num_tokens": 43278402.0,
      "step": 24950
    },
    {
      "entropy": 5.435387325286865,
      "epoch": 1.9416066913051937,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004619450508738094,
      "loss": 5.1687,
      "mean_token_accuracy": 0.1881063312292099,
      "num_tokens": 43286791.0,
      "step": 24955
    },
    {
      "entropy": 5.484977197647095,
      "epoch": 1.941995720676911,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046192962652899056,
      "loss": 5.1544,
      "mean_token_accuracy": 0.18187614977359773,
      "num_tokens": 43295317.0,
      "step": 24960
    },
    {
      "entropy": 5.526894474029541,
      "epoch": 1.9423847500486286,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00046191419934781236,
      "loss": 5.2108,
      "mean_token_accuracy": 0.18590932637453078,
      "num_tokens": 43303545.0,
      "step": 24965
    },
    {
      "entropy": 5.493436574935913,
      "epoch": 1.9427737794203463,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004618987693305086,
      "loss": 5.2122,
      "mean_token_accuracy": 0.18237805664539336,
      "num_tokens": 43311964.0,
      "step": 24970
    },
    {
      "entropy": 5.498593425750732,
      "epoch": 1.9431628087920638,
      "grad_norm": 1.375,
      "learning_rate": 0.0004618833364773135,
      "loss": 5.176,
      "mean_token_accuracy": 0.18283374160528182,
      "num_tokens": 43320137.0,
      "step": 24975
    },
    {
      "entropy": 5.5316297054290775,
      "epoch": 1.9435518381637813,
      "grad_norm": 1.484375,
      "learning_rate": 0.00046186790078846127,
      "loss": 5.2257,
      "mean_token_accuracy": 0.18245524019002915,
      "num_tokens": 43328894.0,
      "step": 24980
    },
    {
      "entropy": 5.4991593837738035,
      "epoch": 1.943940867535499,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046185246226418603,
      "loss": 5.1266,
      "mean_token_accuracy": 0.18938162177801132,
      "num_tokens": 43337269.0,
      "step": 24985
    },
    {
      "entropy": 5.513243675231934,
      "epoch": 1.9443298969072165,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046183702090472206,
      "loss": 5.1854,
      "mean_token_accuracy": 0.18574079871177673,
      "num_tokens": 43345838.0,
      "step": 24990
    },
    {
      "entropy": 5.468989610671997,
      "epoch": 1.944718926278934,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004618215767103037,
      "loss": 5.2347,
      "mean_token_accuracy": 0.18680340349674224,
      "num_tokens": 43354395.0,
      "step": 24995
    },
    {
      "entropy": 5.480809116363526,
      "epoch": 1.9451079556506516,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004618061296811653,
      "loss": 5.1397,
      "mean_token_accuracy": 0.18767135888338088,
      "num_tokens": 43363320.0,
      "step": 25000
    },
    {
      "entropy": 5.514970827102661,
      "epoch": 1.9454969850223693,
      "grad_norm": 1.25,
      "learning_rate": 0.00046179067981754124,
      "loss": 5.1652,
      "mean_token_accuracy": 0.187693490087986,
      "num_tokens": 43371528.0,
      "step": 25005
    },
    {
      "entropy": 5.496232414245606,
      "epoch": 1.9458860143940866,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046177522711966584,
      "loss": 5.1473,
      "mean_token_accuracy": 0.1826369732618332,
      "num_tokens": 43380177.0,
      "step": 25010
    },
    {
      "entropy": 5.328496837615967,
      "epoch": 1.9462750437658043,
      "grad_norm": 1.25,
      "learning_rate": 0.00046175977158777377,
      "loss": 5.0053,
      "mean_token_accuracy": 0.19852996468544007,
      "num_tokens": 43387992.0,
      "step": 25015
    },
    {
      "entropy": 5.466800355911255,
      "epoch": 1.946664073137522,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004617443132220993,
      "loss": 5.2669,
      "mean_token_accuracy": 0.17822102457284927,
      "num_tokens": 43397620.0,
      "step": 25020
    },
    {
      "entropy": 5.478514671325684,
      "epoch": 1.9470531025092395,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046172885202287717,
      "loss": 5.0533,
      "mean_token_accuracy": 0.18851378411054612,
      "num_tokens": 43407023.0,
      "step": 25025
    },
    {
      "entropy": 5.498660039901734,
      "epoch": 1.947442131880957,
      "grad_norm": 1.34375,
      "learning_rate": 0.00046171338799034194,
      "loss": 5.1313,
      "mean_token_accuracy": 0.18287546187639236,
      "num_tokens": 43415713.0,
      "step": 25030
    },
    {
      "entropy": 5.367970085144043,
      "epoch": 1.9478311612526746,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004616979211247282,
      "loss": 5.1163,
      "mean_token_accuracy": 0.1881483241915703,
      "num_tokens": 43424383.0,
      "step": 25035
    },
    {
      "entropy": 5.423070287704467,
      "epoch": 1.9482201906243921,
      "grad_norm": 1.25,
      "learning_rate": 0.00046168245142627065,
      "loss": 5.1347,
      "mean_token_accuracy": 0.19027692824602127,
      "num_tokens": 43433090.0,
      "step": 25040
    },
    {
      "entropy": 5.482160902023315,
      "epoch": 1.9486092199961096,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004616669788952041,
      "loss": 5.1284,
      "mean_token_accuracy": 0.18089062571525574,
      "num_tokens": 43442091.0,
      "step": 25045
    },
    {
      "entropy": 5.512087774276734,
      "epoch": 1.9489982493678273,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046165150353176315,
      "loss": 5.1387,
      "mean_token_accuracy": 0.19666801244020463,
      "num_tokens": 43450392.0,
      "step": 25050
    },
    {
      "entropy": 5.405067873001099,
      "epoch": 1.949387278739545,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004616360253361828,
      "loss": 5.08,
      "mean_token_accuracy": 0.18863831907510759,
      "num_tokens": 43459252.0,
      "step": 25055
    },
    {
      "entropy": 5.472977972030639,
      "epoch": 1.9497763081112622,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046162054430869777,
      "loss": 5.2537,
      "mean_token_accuracy": 0.17645413130521775,
      "num_tokens": 43467580.0,
      "step": 25060
    },
    {
      "entropy": 5.643496417999268,
      "epoch": 1.95016533748298,
      "grad_norm": 1.203125,
      "learning_rate": 0.000461605060449543,
      "loss": 5.2674,
      "mean_token_accuracy": 0.18107384145259858,
      "num_tokens": 43476879.0,
      "step": 25065
    },
    {
      "entropy": 5.526113700866699,
      "epoch": 1.9505543668546976,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046158957375895353,
      "loss": 5.16,
      "mean_token_accuracy": 0.17806796729564667,
      "num_tokens": 43485881.0,
      "step": 25070
    },
    {
      "entropy": 5.392427444458008,
      "epoch": 1.950943396226415,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004615740842371643,
      "loss": 5.1071,
      "mean_token_accuracy": 0.1893598437309265,
      "num_tokens": 43494598.0,
      "step": 25075
    },
    {
      "entropy": 5.509137916564941,
      "epoch": 1.9513324255981326,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046155859188441023,
      "loss": 5.1881,
      "mean_token_accuracy": 0.17653862982988358,
      "num_tokens": 43504638.0,
      "step": 25080
    },
    {
      "entropy": 5.546071195602417,
      "epoch": 1.9517214549698503,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004615430967009265,
      "loss": 5.2117,
      "mean_token_accuracy": 0.18373489379882812,
      "num_tokens": 43512950.0,
      "step": 25085
    },
    {
      "entropy": 5.418782806396484,
      "epoch": 1.9521104843415678,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046152759868694815,
      "loss": 5.2165,
      "mean_token_accuracy": 0.18723226934671403,
      "num_tokens": 43521383.0,
      "step": 25090
    },
    {
      "entropy": 5.36074743270874,
      "epoch": 1.9524995137132852,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004615120978427104,
      "loss": 5.0095,
      "mean_token_accuracy": 0.19643847793340682,
      "num_tokens": 43529442.0,
      "step": 25095
    },
    {
      "entropy": 5.478216218948364,
      "epoch": 1.952888543085003,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004614965941684485,
      "loss": 5.1795,
      "mean_token_accuracy": 0.19091429561376572,
      "num_tokens": 43536938.0,
      "step": 25100
    },
    {
      "entropy": 5.39796552658081,
      "epoch": 1.9532775724567206,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004614810876643975,
      "loss": 5.1791,
      "mean_token_accuracy": 0.18348912000656128,
      "num_tokens": 43545337.0,
      "step": 25105
    },
    {
      "entropy": 5.378086519241333,
      "epoch": 1.953666601828438,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046146557833079286,
      "loss": 5.038,
      "mean_token_accuracy": 0.1872916415333748,
      "num_tokens": 43553699.0,
      "step": 25110
    },
    {
      "entropy": 5.501021814346314,
      "epoch": 1.9540556312001556,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004614500661678698,
      "loss": 5.2748,
      "mean_token_accuracy": 0.17575658112764359,
      "num_tokens": 43562545.0,
      "step": 25115
    },
    {
      "entropy": 5.53188796043396,
      "epoch": 1.9544446605718733,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004614345511758639,
      "loss": 5.2346,
      "mean_token_accuracy": 0.18346190303564072,
      "num_tokens": 43571392.0,
      "step": 25120
    },
    {
      "entropy": 5.431199789047241,
      "epoch": 1.9548336899435907,
      "grad_norm": 1.515625,
      "learning_rate": 0.00046141903335501034,
      "loss": 5.1522,
      "mean_token_accuracy": 0.18521708697080613,
      "num_tokens": 43579864.0,
      "step": 25125
    },
    {
      "entropy": 5.527487087249756,
      "epoch": 1.9552227193153082,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004614035127055447,
      "loss": 5.2984,
      "mean_token_accuracy": 0.17568491399288177,
      "num_tokens": 43588546.0,
      "step": 25130
    },
    {
      "entropy": 5.474833822250366,
      "epoch": 1.955611748687026,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004613879892277024,
      "loss": 5.0982,
      "mean_token_accuracy": 0.18994525521993638,
      "num_tokens": 43596582.0,
      "step": 25135
    },
    {
      "entropy": 5.387027025222778,
      "epoch": 1.9560007780587434,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000461372462921719,
      "loss": 5.2151,
      "mean_token_accuracy": 0.18022741228342057,
      "num_tokens": 43605727.0,
      "step": 25140
    },
    {
      "entropy": 5.453405570983887,
      "epoch": 1.9563898074304609,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004613569337878302,
      "loss": 5.1868,
      "mean_token_accuracy": 0.17774559408426285,
      "num_tokens": 43614889.0,
      "step": 25145
    },
    {
      "entropy": 5.449581241607666,
      "epoch": 1.9567788368021786,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004613414018262715,
      "loss": 5.1228,
      "mean_token_accuracy": 0.18792954385280608,
      "num_tokens": 43624059.0,
      "step": 25150
    },
    {
      "entropy": 5.579153728485108,
      "epoch": 1.9571678661738963,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004613258670372786,
      "loss": 5.297,
      "mean_token_accuracy": 0.18181269913911818,
      "num_tokens": 43633824.0,
      "step": 25155
    },
    {
      "entropy": 5.387596845626831,
      "epoch": 1.9575568955456137,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004613103294210873,
      "loss": 5.0741,
      "mean_token_accuracy": 0.19260424971580506,
      "num_tokens": 43642010.0,
      "step": 25160
    },
    {
      "entropy": 5.559239864349365,
      "epoch": 1.9579459249173312,
      "grad_norm": 1.421875,
      "learning_rate": 0.00046129478897793323,
      "loss": 5.306,
      "mean_token_accuracy": 0.1761642038822174,
      "num_tokens": 43650828.0,
      "step": 25165
    },
    {
      "entropy": 5.634287643432617,
      "epoch": 1.958334954289049,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046127924570805236,
      "loss": 5.1833,
      "mean_token_accuracy": 0.1872473731637001,
      "num_tokens": 43659310.0,
      "step": 25170
    },
    {
      "entropy": 5.4130027294158936,
      "epoch": 1.9587239836607664,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004612636996116803,
      "loss": 5.0938,
      "mean_token_accuracy": 0.18490393906831742,
      "num_tokens": 43668238.0,
      "step": 25175
    },
    {
      "entropy": 5.370238208770752,
      "epoch": 1.9591130130324839,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004612481506890532,
      "loss": 5.0466,
      "mean_token_accuracy": 0.20276857614517213,
      "num_tokens": 43675748.0,
      "step": 25180
    },
    {
      "entropy": 5.440761518478394,
      "epoch": 1.9595020424042016,
      "grad_norm": 1.25,
      "learning_rate": 0.00046123259894040677,
      "loss": 5.2315,
      "mean_token_accuracy": 0.17397599667310715,
      "num_tokens": 43685750.0,
      "step": 25185
    },
    {
      "entropy": 5.477128410339356,
      "epoch": 1.959891071775919,
      "grad_norm": 1.21875,
      "learning_rate": 0.000461217044365977,
      "loss": 5.1477,
      "mean_token_accuracy": 0.18405405133962632,
      "num_tokens": 43694941.0,
      "step": 25190
    },
    {
      "entropy": 5.537965488433838,
      "epoch": 1.9602801011476365,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004612014869660002,
      "loss": 5.2287,
      "mean_token_accuracy": 0.17738770246505736,
      "num_tokens": 43703500.0,
      "step": 25195
    },
    {
      "entropy": 5.528535270690918,
      "epoch": 1.9606691305193542,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046118592674071197,
      "loss": 5.3071,
      "mean_token_accuracy": 0.18057193160057067,
      "num_tokens": 43712235.0,
      "step": 25200
    },
    {
      "entropy": 5.5984885692596436,
      "epoch": 1.961058159891072,
      "grad_norm": 1.25,
      "learning_rate": 0.0004611703636903488,
      "loss": 5.2721,
      "mean_token_accuracy": 0.1782281592488289,
      "num_tokens": 43721047.0,
      "step": 25205
    },
    {
      "entropy": 5.546380043029785,
      "epoch": 1.9614471892627894,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004611547978151466,
      "loss": 5.2382,
      "mean_token_accuracy": 0.18203096091747284,
      "num_tokens": 43729635.0,
      "step": 25210
    },
    {
      "entropy": 5.488947486877441,
      "epoch": 1.9618362186345069,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046113922911534167,
      "loss": 5.1541,
      "mean_token_accuracy": 0.18466203659772873,
      "num_tokens": 43737748.0,
      "step": 25215
    },
    {
      "entropy": 5.48777437210083,
      "epoch": 1.9622252480062246,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004611236575911702,
      "loss": 5.1269,
      "mean_token_accuracy": 0.18876243382692337,
      "num_tokens": 43745999.0,
      "step": 25220
    },
    {
      "entropy": 5.393968677520752,
      "epoch": 1.962614277377942,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046110808324286844,
      "loss": 4.9745,
      "mean_token_accuracy": 0.19572101384401322,
      "num_tokens": 43754029.0,
      "step": 25225
    },
    {
      "entropy": 5.441364336013794,
      "epoch": 1.9630033067496595,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046109250607067275,
      "loss": 5.1244,
      "mean_token_accuracy": 0.1899360716342926,
      "num_tokens": 43762289.0,
      "step": 25230
    },
    {
      "entropy": 5.442074108123779,
      "epoch": 1.9633923361213772,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004610769260748196,
      "loss": 5.2412,
      "mean_token_accuracy": 0.18013914078474044,
      "num_tokens": 43771479.0,
      "step": 25235
    },
    {
      "entropy": 5.401347732543945,
      "epoch": 1.9637813654930947,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004610613432555451,
      "loss": 5.0586,
      "mean_token_accuracy": 0.19501953721046447,
      "num_tokens": 43779889.0,
      "step": 25240
    },
    {
      "entropy": 5.503679800033569,
      "epoch": 1.9641703948648122,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046104575761308597,
      "loss": 5.1271,
      "mean_token_accuracy": 0.1913775995373726,
      "num_tokens": 43788394.0,
      "step": 25245
    },
    {
      "entropy": 5.465291118621826,
      "epoch": 1.9645594242365298,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004610301691476786,
      "loss": 5.1863,
      "mean_token_accuracy": 0.17730966806411744,
      "num_tokens": 43797188.0,
      "step": 25250
    },
    {
      "entropy": 5.5290687561035154,
      "epoch": 1.9649484536082475,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004610145778595594,
      "loss": 5.2678,
      "mean_token_accuracy": 0.17698303759098052,
      "num_tokens": 43806174.0,
      "step": 25255
    },
    {
      "entropy": 5.424228000640869,
      "epoch": 1.965337482979965,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004609989837489651,
      "loss": 5.0863,
      "mean_token_accuracy": 0.19475599229335785,
      "num_tokens": 43814704.0,
      "step": 25260
    },
    {
      "entropy": 5.3893218517303465,
      "epoch": 1.9657265123516825,
      "grad_norm": 1.4375,
      "learning_rate": 0.00046098338681613234,
      "loss": 5.0495,
      "mean_token_accuracy": 0.18727365136146545,
      "num_tokens": 43823122.0,
      "step": 25265
    },
    {
      "entropy": 5.472195863723755,
      "epoch": 1.9661155417234002,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004609677870612976,
      "loss": 5.1251,
      "mean_token_accuracy": 0.192566579580307,
      "num_tokens": 43832281.0,
      "step": 25270
    },
    {
      "entropy": 5.432113838195801,
      "epoch": 1.9665045710951177,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004609521844846978,
      "loss": 5.0906,
      "mean_token_accuracy": 0.18610522150993347,
      "num_tokens": 43840351.0,
      "step": 25275
    },
    {
      "entropy": 5.512760353088379,
      "epoch": 1.9668936004668351,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004609365790865695,
      "loss": 5.2177,
      "mean_token_accuracy": 0.1776440218091011,
      "num_tokens": 43849469.0,
      "step": 25280
    },
    {
      "entropy": 5.499239826202393,
      "epoch": 1.9672826298385528,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046092097086714956,
      "loss": 5.1288,
      "mean_token_accuracy": 0.18410846441984177,
      "num_tokens": 43858013.0,
      "step": 25285
    },
    {
      "entropy": 5.482743549346924,
      "epoch": 1.9676716592102705,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046090535982667486,
      "loss": 5.1301,
      "mean_token_accuracy": 0.1877378985285759,
      "num_tokens": 43867056.0,
      "step": 25290
    },
    {
      "entropy": 5.429666042327881,
      "epoch": 1.9680606885819878,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046088974596538216,
      "loss": 5.0926,
      "mean_token_accuracy": 0.19599473029375075,
      "num_tokens": 43875160.0,
      "step": 25295
    },
    {
      "entropy": 5.384914541244507,
      "epoch": 1.9684497179537055,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004608741292835085,
      "loss": 5.1068,
      "mean_token_accuracy": 0.1888219952583313,
      "num_tokens": 43884243.0,
      "step": 25300
    },
    {
      "entropy": 5.446265316009521,
      "epoch": 1.9688387473254232,
      "grad_norm": 1.25,
      "learning_rate": 0.0004608585097812907,
      "loss": 5.1339,
      "mean_token_accuracy": 0.18835226595401763,
      "num_tokens": 43893378.0,
      "step": 25305
    },
    {
      "entropy": 5.409064245223999,
      "epoch": 1.9692277766971407,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004608428874589659,
      "loss": 5.06,
      "mean_token_accuracy": 0.19077397435903548,
      "num_tokens": 43902095.0,
      "step": 25310
    },
    {
      "entropy": 5.461485576629639,
      "epoch": 1.9696168060688581,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004608272623167711,
      "loss": 5.1727,
      "mean_token_accuracy": 0.18362220525741577,
      "num_tokens": 43911452.0,
      "step": 25315
    },
    {
      "entropy": 5.416750860214234,
      "epoch": 1.9700058354405758,
      "grad_norm": 1.390625,
      "learning_rate": 0.00046081163435494335,
      "loss": 5.0649,
      "mean_token_accuracy": 0.18266730159521102,
      "num_tokens": 43920208.0,
      "step": 25320
    },
    {
      "entropy": 5.50468111038208,
      "epoch": 1.9703948648122933,
      "grad_norm": 1.25,
      "learning_rate": 0.0004607960035737198,
      "loss": 5.2038,
      "mean_token_accuracy": 0.1859483003616333,
      "num_tokens": 43929069.0,
      "step": 25325
    },
    {
      "entropy": 5.546693277359009,
      "epoch": 1.9707838941840108,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004607803699733376,
      "loss": 5.1999,
      "mean_token_accuracy": 0.18293482661247254,
      "num_tokens": 43937444.0,
      "step": 25330
    },
    {
      "entropy": 5.556116390228271,
      "epoch": 1.9711729235557285,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000460764733554034,
      "loss": 5.2387,
      "mean_token_accuracy": 0.1804982379078865,
      "num_tokens": 43946744.0,
      "step": 25335
    },
    {
      "entropy": 5.505416297912598,
      "epoch": 1.9715619529274462,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046074909431604623,
      "loss": 5.193,
      "mean_token_accuracy": 0.1787700042128563,
      "num_tokens": 43954810.0,
      "step": 25340
    },
    {
      "entropy": 5.481557846069336,
      "epoch": 1.9719509822991634,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004607334522596116,
      "loss": 5.167,
      "mean_token_accuracy": 0.18779837787151338,
      "num_tokens": 43963366.0,
      "step": 25345
    },
    {
      "entropy": 5.472129201889038,
      "epoch": 1.9723400116708811,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004607178073849675,
      "loss": 5.1213,
      "mean_token_accuracy": 0.18650121986865997,
      "num_tokens": 43972170.0,
      "step": 25350
    },
    {
      "entropy": 5.48205795288086,
      "epoch": 1.9727290410425988,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004607021596923512,
      "loss": 5.1964,
      "mean_token_accuracy": 0.18093065768480301,
      "num_tokens": 43980743.0,
      "step": 25355
    },
    {
      "entropy": 5.47332911491394,
      "epoch": 1.9731180704143163,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004606865091820003,
      "loss": 5.1125,
      "mean_token_accuracy": 0.1856198564171791,
      "num_tokens": 43990332.0,
      "step": 25360
    },
    {
      "entropy": 5.484947776794433,
      "epoch": 1.9735070997860338,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004606708558541521,
      "loss": 5.1527,
      "mean_token_accuracy": 0.1871538445353508,
      "num_tokens": 43999058.0,
      "step": 25365
    },
    {
      "entropy": 5.505348777770996,
      "epoch": 1.9738961291577515,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004606551997090442,
      "loss": 5.1674,
      "mean_token_accuracy": 0.19216561913490296,
      "num_tokens": 44008310.0,
      "step": 25370
    },
    {
      "entropy": 5.442971229553223,
      "epoch": 1.974285158529469,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004606395407469141,
      "loss": 5.1393,
      "mean_token_accuracy": 0.1868462786078453,
      "num_tokens": 44016818.0,
      "step": 25375
    },
    {
      "entropy": 5.473188304901123,
      "epoch": 1.9746741879011864,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004606238789679995,
      "loss": 5.2363,
      "mean_token_accuracy": 0.1797398880124092,
      "num_tokens": 44024755.0,
      "step": 25380
    },
    {
      "entropy": 5.526695680618286,
      "epoch": 1.9750632172729041,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004606082143725381,
      "loss": 5.201,
      "mean_token_accuracy": 0.18505900949239731,
      "num_tokens": 44033416.0,
      "step": 25385
    },
    {
      "entropy": 5.463621377944946,
      "epoch": 1.9754522466446218,
      "grad_norm": 1.25,
      "learning_rate": 0.0004605925469607673,
      "loss": 5.1404,
      "mean_token_accuracy": 0.1915491119027138,
      "num_tokens": 44041487.0,
      "step": 25390
    },
    {
      "entropy": 5.450394153594971,
      "epoch": 1.975841276016339,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046057687673292506,
      "loss": 5.2043,
      "mean_token_accuracy": 0.1864388704299927,
      "num_tokens": 44050766.0,
      "step": 25395
    },
    {
      "entropy": 5.52214879989624,
      "epoch": 1.9762303053880568,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046056120368924907,
      "loss": 5.2703,
      "mean_token_accuracy": 0.1834894075989723,
      "num_tokens": 44060173.0,
      "step": 25400
    },
    {
      "entropy": 5.442066860198975,
      "epoch": 1.9766193347597745,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004605455278299772,
      "loss": 5.0504,
      "mean_token_accuracy": 0.19249071925878525,
      "num_tokens": 44068472.0,
      "step": 25405
    },
    {
      "entropy": 5.411147212982177,
      "epoch": 1.977008364131492,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004605298491553472,
      "loss": 5.1764,
      "mean_token_accuracy": 0.18433773666620254,
      "num_tokens": 44077209.0,
      "step": 25410
    },
    {
      "entropy": 5.538371324539185,
      "epoch": 1.9773973935032094,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004605141676655971,
      "loss": 5.295,
      "mean_token_accuracy": 0.17790180295705796,
      "num_tokens": 44086525.0,
      "step": 25415
    },
    {
      "entropy": 5.532166814804077,
      "epoch": 1.9777864228749271,
      "grad_norm": 1.34375,
      "learning_rate": 0.00046049848336096485,
      "loss": 5.174,
      "mean_token_accuracy": 0.18911987990140916,
      "num_tokens": 44095101.0,
      "step": 25420
    },
    {
      "entropy": 5.457807826995849,
      "epoch": 1.9781754522466446,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004604827962416883,
      "loss": 5.0736,
      "mean_token_accuracy": 0.19483690112829208,
      "num_tokens": 44103469.0,
      "step": 25425
    },
    {
      "entropy": 5.395337963104248,
      "epoch": 1.978564481618362,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004604671063080055,
      "loss": 5.0696,
      "mean_token_accuracy": 0.18762816190719606,
      "num_tokens": 44112161.0,
      "step": 25430
    },
    {
      "entropy": 5.524976539611816,
      "epoch": 1.9789535109900798,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004604514135601546,
      "loss": 5.1604,
      "mean_token_accuracy": 0.18970298618078232,
      "num_tokens": 44120452.0,
      "step": 25435
    },
    {
      "entropy": 5.527880907058716,
      "epoch": 1.9793425403617975,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004604357179983736,
      "loss": 5.2369,
      "mean_token_accuracy": 0.17993032336235046,
      "num_tokens": 44129581.0,
      "step": 25440
    },
    {
      "entropy": 5.544310235977173,
      "epoch": 1.9797315697335147,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004604200196229009,
      "loss": 5.1671,
      "mean_token_accuracy": 0.18491467237472534,
      "num_tokens": 44138086.0,
      "step": 25445
    },
    {
      "entropy": 5.494865465164184,
      "epoch": 1.9801205991052324,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004604043184339744,
      "loss": 5.1221,
      "mean_token_accuracy": 0.19335436522960664,
      "num_tokens": 44146549.0,
      "step": 25450
    },
    {
      "entropy": 5.508090686798096,
      "epoch": 1.9805096284769501,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004603886144318325,
      "loss": 5.2236,
      "mean_token_accuracy": 0.17968143075704573,
      "num_tokens": 44156142.0,
      "step": 25455
    },
    {
      "entropy": 5.465776681900024,
      "epoch": 1.9808986578486676,
      "grad_norm": 1.25,
      "learning_rate": 0.00046037290761671346,
      "loss": 5.1653,
      "mean_token_accuracy": 0.18690854609012603,
      "num_tokens": 44165023.0,
      "step": 25460
    },
    {
      "entropy": 5.501298236846924,
      "epoch": 1.981287687220385,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004603571979888556,
      "loss": 5.1738,
      "mean_token_accuracy": 0.18056414127349854,
      "num_tokens": 44174034.0,
      "step": 25465
    },
    {
      "entropy": 5.497203826904297,
      "epoch": 1.9816767165921028,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004603414855484973,
      "loss": 5.2376,
      "mean_token_accuracy": 0.17781688272953033,
      "num_tokens": 44183072.0,
      "step": 25470
    },
    {
      "entropy": 5.491177701950074,
      "epoch": 1.9820657459638202,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000460325770295877,
      "loss": 5.1591,
      "mean_token_accuracy": 0.18131774067878723,
      "num_tokens": 44192181.0,
      "step": 25475
    },
    {
      "entropy": 5.4706768035888675,
      "epoch": 1.9824547753355377,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046031005223123297,
      "loss": 5.1025,
      "mean_token_accuracy": 0.1920669823884964,
      "num_tokens": 44200967.0,
      "step": 25480
    },
    {
      "entropy": 5.4509875774383545,
      "epoch": 1.9828438047072554,
      "grad_norm": 1.28125,
      "learning_rate": 0.000460294331354804,
      "loss": 5.1519,
      "mean_token_accuracy": 0.18777865469455718,
      "num_tokens": 44209432.0,
      "step": 25485
    },
    {
      "entropy": 5.388992166519165,
      "epoch": 1.983232834078973,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004602786076668283,
      "loss": 5.1342,
      "mean_token_accuracy": 0.18712625950574874,
      "num_tokens": 44218503.0,
      "step": 25490
    },
    {
      "entropy": 5.436001253128052,
      "epoch": 1.9836218634506906,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046026288116754477,
      "loss": 5.0745,
      "mean_token_accuracy": 0.1878805637359619,
      "num_tokens": 44227164.0,
      "step": 25495
    },
    {
      "entropy": 5.512021541595459,
      "epoch": 1.984010892822408,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004602471518571919,
      "loss": 5.1626,
      "mean_token_accuracy": 0.18564372658729553,
      "num_tokens": 44235950.0,
      "step": 25500
    },
    {
      "entropy": 5.421997356414795,
      "epoch": 1.9843999221941258,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004602314197360083,
      "loss": 5.088,
      "mean_token_accuracy": 0.19366597831249238,
      "num_tokens": 44244098.0,
      "step": 25505
    },
    {
      "entropy": 5.443205404281616,
      "epoch": 1.9847889515658432,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004602156848042328,
      "loss": 5.2004,
      "mean_token_accuracy": 0.17902885526418685,
      "num_tokens": 44252798.0,
      "step": 25510
    },
    {
      "entropy": 5.518117427825928,
      "epoch": 1.9851779809375607,
      "grad_norm": 1.1875,
      "learning_rate": 0.000460199947062104,
      "loss": 5.2325,
      "mean_token_accuracy": 0.18336304128170014,
      "num_tokens": 44261662.0,
      "step": 25515
    },
    {
      "entropy": 5.589923477172851,
      "epoch": 1.9855670103092784,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046018420650986074,
      "loss": 5.2423,
      "mean_token_accuracy": 0.18122700452804566,
      "num_tokens": 44270480.0,
      "step": 25520
    },
    {
      "entropy": 5.497834825515747,
      "epoch": 1.9859560396809959,
      "grad_norm": 1.21875,
      "learning_rate": 0.000460168463147742,
      "loss": 5.1554,
      "mean_token_accuracy": 0.1871015712618828,
      "num_tokens": 44278673.0,
      "step": 25525
    },
    {
      "entropy": 5.359372520446778,
      "epoch": 1.9863450690527134,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004601527169759865,
      "loss": 5.0349,
      "mean_token_accuracy": 0.19219622761011124,
      "num_tokens": 44287465.0,
      "step": 25530
    },
    {
      "entropy": 5.460550975799561,
      "epoch": 1.986734098424431,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004601369679948333,
      "loss": 5.2416,
      "mean_token_accuracy": 0.17935955077409743,
      "num_tokens": 44296204.0,
      "step": 25535
    },
    {
      "entropy": 5.5692850112915036,
      "epoch": 1.9871231277961487,
      "grad_norm": 2.15625,
      "learning_rate": 0.00046012121620452127,
      "loss": 5.2181,
      "mean_token_accuracy": 0.18243835121393204,
      "num_tokens": 44305338.0,
      "step": 25540
    },
    {
      "entropy": 5.4553173065185545,
      "epoch": 1.9875121571678662,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004601054616052893,
      "loss": 5.044,
      "mean_token_accuracy": 0.19335984587669372,
      "num_tokens": 44312960.0,
      "step": 25545
    },
    {
      "entropy": 5.40078387260437,
      "epoch": 1.9879011865395837,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046008970419737674,
      "loss": 5.1064,
      "mean_token_accuracy": 0.18838817179203032,
      "num_tokens": 44321367.0,
      "step": 25550
    },
    {
      "entropy": 5.449056243896484,
      "epoch": 1.9882902159113014,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004600739439810225,
      "loss": 5.1867,
      "mean_token_accuracy": 0.18482354283332825,
      "num_tokens": 44330294.0,
      "step": 25555
    },
    {
      "entropy": 5.571813154220581,
      "epoch": 1.9886792452830189,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00046005818095646564,
      "loss": 5.1885,
      "mean_token_accuracy": 0.180316698551178,
      "num_tokens": 44338639.0,
      "step": 25560
    },
    {
      "entropy": 5.399201679229736,
      "epoch": 1.9890682746547363,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046004241512394544,
      "loss": 5.0717,
      "mean_token_accuracy": 0.1873680755496025,
      "num_tokens": 44346911.0,
      "step": 25565
    },
    {
      "entropy": 5.446534538269043,
      "epoch": 1.989457304026454,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004600266464837012,
      "loss": 5.169,
      "mean_token_accuracy": 0.18562185317277907,
      "num_tokens": 44355580.0,
      "step": 25570
    },
    {
      "entropy": 5.505319404602051,
      "epoch": 1.9898463333981715,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004600108750359721,
      "loss": 5.1648,
      "mean_token_accuracy": 0.1830005392432213,
      "num_tokens": 44363465.0,
      "step": 25575
    },
    {
      "entropy": 5.481562757492066,
      "epoch": 1.990235362769889,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045999510078099736,
      "loss": 5.1514,
      "mean_token_accuracy": 0.19406624138355255,
      "num_tokens": 44372797.0,
      "step": 25580
    },
    {
      "entropy": 5.454525423049927,
      "epoch": 1.9906243921416067,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045997932371901645,
      "loss": 5.1261,
      "mean_token_accuracy": 0.18497560322284698,
      "num_tokens": 44381537.0,
      "step": 25585
    },
    {
      "entropy": 5.468609571456909,
      "epoch": 1.9910134215133244,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004599635438502687,
      "loss": 5.1471,
      "mean_token_accuracy": 0.19010152518749238,
      "num_tokens": 44390170.0,
      "step": 25590
    },
    {
      "entropy": 5.460803842544555,
      "epoch": 1.9914024508850419,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045994776117499363,
      "loss": 5.2141,
      "mean_token_accuracy": 0.1841409534215927,
      "num_tokens": 44398545.0,
      "step": 25595
    },
    {
      "entropy": 5.460994720458984,
      "epoch": 1.9917914802567593,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045993197569343063,
      "loss": 5.1644,
      "mean_token_accuracy": 0.18038479536771773,
      "num_tokens": 44407587.0,
      "step": 25600
    },
    {
      "entropy": 5.449535799026489,
      "epoch": 1.992180509628477,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045991618740581926,
      "loss": 5.0513,
      "mean_token_accuracy": 0.18754464387893677,
      "num_tokens": 44416005.0,
      "step": 25605
    },
    {
      "entropy": 5.426014137268067,
      "epoch": 1.9925695390001945,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004599003963123991,
      "loss": 5.0699,
      "mean_token_accuracy": 0.19097606092691422,
      "num_tokens": 44425450.0,
      "step": 25610
    },
    {
      "entropy": 5.42979621887207,
      "epoch": 1.992958568371912,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045988460241340966,
      "loss": 5.1467,
      "mean_token_accuracy": 0.18691133856773376,
      "num_tokens": 44434657.0,
      "step": 25615
    },
    {
      "entropy": 5.518140554428101,
      "epoch": 1.9933475977436297,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045986880570909075,
      "loss": 5.1693,
      "mean_token_accuracy": 0.17848251312971114,
      "num_tokens": 44443160.0,
      "step": 25620
    },
    {
      "entropy": 5.446799325942993,
      "epoch": 1.9937366271153472,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045985300619968186,
      "loss": 5.1221,
      "mean_token_accuracy": 0.18127365857362748,
      "num_tokens": 44452097.0,
      "step": 25625
    },
    {
      "entropy": 5.40854320526123,
      "epoch": 1.9941256564870646,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045983720388542286,
      "loss": 5.0774,
      "mean_token_accuracy": 0.18011631965637206,
      "num_tokens": 44459674.0,
      "step": 25630
    },
    {
      "entropy": 5.391352796554566,
      "epoch": 1.9945146858587823,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004598213987665535,
      "loss": 5.1638,
      "mean_token_accuracy": 0.18987185060977935,
      "num_tokens": 44468315.0,
      "step": 25635
    },
    {
      "entropy": 5.543141937255859,
      "epoch": 1.9949037152305,
      "grad_norm": 1.25,
      "learning_rate": 0.00045980559084331354,
      "loss": 5.1987,
      "mean_token_accuracy": 0.18342542946338652,
      "num_tokens": 44477436.0,
      "step": 25640
    },
    {
      "entropy": 5.551082754135132,
      "epoch": 1.9952927446022175,
      "grad_norm": 1.375,
      "learning_rate": 0.000459789780115943,
      "loss": 5.187,
      "mean_token_accuracy": 0.1835482209920883,
      "num_tokens": 44485518.0,
      "step": 25645
    },
    {
      "entropy": 5.385361051559448,
      "epoch": 1.995681773973935,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045977396658468156,
      "loss": 5.0896,
      "mean_token_accuracy": 0.18648289144039154,
      "num_tokens": 44493734.0,
      "step": 25650
    },
    {
      "entropy": 5.418517112731934,
      "epoch": 1.9960708033456527,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004597581502497693,
      "loss": 5.1352,
      "mean_token_accuracy": 0.18446629047393798,
      "num_tokens": 44502615.0,
      "step": 25655
    },
    {
      "entropy": 5.523607635498047,
      "epoch": 1.9964598327173702,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004597423311114462,
      "loss": 5.2338,
      "mean_token_accuracy": 0.18076692819595336,
      "num_tokens": 44511792.0,
      "step": 25660
    },
    {
      "entropy": 5.500126695632934,
      "epoch": 1.9968488620890876,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004597265091699522,
      "loss": 5.1795,
      "mean_token_accuracy": 0.1797320932149887,
      "num_tokens": 44520627.0,
      "step": 25665
    },
    {
      "entropy": 5.452884006500244,
      "epoch": 1.9972378914608053,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004597106844255276,
      "loss": 5.1147,
      "mean_token_accuracy": 0.18461923450231552,
      "num_tokens": 44528844.0,
      "step": 25670
    },
    {
      "entropy": 5.514548444747925,
      "epoch": 1.9976269208325228,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045969485687841227,
      "loss": 5.2346,
      "mean_token_accuracy": 0.18351624757051468,
      "num_tokens": 44537602.0,
      "step": 25675
    },
    {
      "entropy": 5.5059802532196045,
      "epoch": 1.9980159502042403,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045967902652884645,
      "loss": 5.1869,
      "mean_token_accuracy": 0.1845601737499237,
      "num_tokens": 44546261.0,
      "step": 25680
    },
    {
      "entropy": 5.405078601837158,
      "epoch": 1.998404979575958,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004596631933770704,
      "loss": 5.092,
      "mean_token_accuracy": 0.18809015601873397,
      "num_tokens": 44554291.0,
      "step": 25685
    },
    {
      "entropy": 5.4805034637451175,
      "epoch": 1.9987940089476757,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045964735742332427,
      "loss": 5.1248,
      "mean_token_accuracy": 0.190411713719368,
      "num_tokens": 44562355.0,
      "step": 25690
    },
    {
      "entropy": 5.483749437332153,
      "epoch": 1.9991830383193931,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004596315186678484,
      "loss": 5.0853,
      "mean_token_accuracy": 0.19052745401859283,
      "num_tokens": 44570517.0,
      "step": 25695
    },
    {
      "entropy": 5.39281153678894,
      "epoch": 1.9995720676911106,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045961567711088307,
      "loss": 5.0488,
      "mean_token_accuracy": 0.18237481713294984,
      "num_tokens": 44578871.0,
      "step": 25700
    },
    {
      "entropy": 5.497526407241821,
      "epoch": 1.9999610970628283,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045959983275266873,
      "loss": 5.1604,
      "mean_token_accuracy": 0.18408870995044707,
      "num_tokens": 44588081.0,
      "step": 25705
    },
    {
      "entropy": 5.581485642327203,
      "epoch": 2.000311223497374,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00045958398559344573,
      "loss": 5.2352,
      "mean_token_accuracy": 0.18249769177701738,
      "num_tokens": 44596153.0,
      "step": 25710
    },
    {
      "entropy": 5.444728708267212,
      "epoch": 2.0007002528690916,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045956813563345454,
      "loss": 5.0338,
      "mean_token_accuracy": 0.18786338567733765,
      "num_tokens": 44604760.0,
      "step": 25715
    },
    {
      "entropy": 5.376680564880371,
      "epoch": 2.0010892822408093,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004595522828729357,
      "loss": 4.9761,
      "mean_token_accuracy": 0.19338334798812867,
      "num_tokens": 44613251.0,
      "step": 25720
    },
    {
      "entropy": 5.388321208953857,
      "epoch": 2.0014783116125265,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004595364273121296,
      "loss": 5.0736,
      "mean_token_accuracy": 0.19962829798460008,
      "num_tokens": 44621914.0,
      "step": 25725
    },
    {
      "entropy": 5.308832311630249,
      "epoch": 2.0018673409842442,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004595205689512771,
      "loss": 4.9465,
      "mean_token_accuracy": 0.20085244476795197,
      "num_tokens": 44630096.0,
      "step": 25730
    },
    {
      "entropy": 5.416966962814331,
      "epoch": 2.002256370355962,
      "grad_norm": 1.25,
      "learning_rate": 0.00045950470779061855,
      "loss": 5.062,
      "mean_token_accuracy": 0.1934100419282913,
      "num_tokens": 44638175.0,
      "step": 25735
    },
    {
      "entropy": 5.542482137680054,
      "epoch": 2.0026453997276796,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045948884383039475,
      "loss": 5.0923,
      "mean_token_accuracy": 0.18295737504959106,
      "num_tokens": 44646970.0,
      "step": 25740
    },
    {
      "entropy": 5.510114479064941,
      "epoch": 2.003034429099397,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045947297707084637,
      "loss": 5.1766,
      "mean_token_accuracy": 0.18045057803392411,
      "num_tokens": 44655789.0,
      "step": 25745
    },
    {
      "entropy": 5.395209074020386,
      "epoch": 2.0034234584711146,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004594571075122142,
      "loss": 4.9918,
      "mean_token_accuracy": 0.20173296928405762,
      "num_tokens": 44664190.0,
      "step": 25750
    },
    {
      "entropy": 5.419756984710693,
      "epoch": 2.0038124878428323,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045944123515473905,
      "loss": 5.0446,
      "mean_token_accuracy": 0.1888587787747383,
      "num_tokens": 44673295.0,
      "step": 25755
    },
    {
      "entropy": 5.519266653060913,
      "epoch": 2.0042015172145495,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00045942535999866175,
      "loss": 5.1451,
      "mean_token_accuracy": 0.18339453041553497,
      "num_tokens": 44681890.0,
      "step": 25760
    },
    {
      "entropy": 5.5163695335388185,
      "epoch": 2.0045905465862672,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004594094820442231,
      "loss": 5.1522,
      "mean_token_accuracy": 0.1822892665863037,
      "num_tokens": 44690915.0,
      "step": 25765
    },
    {
      "entropy": 5.489056491851807,
      "epoch": 2.004979575957985,
      "grad_norm": 1.3125,
      "learning_rate": 0.000459393601291664,
      "loss": 5.1054,
      "mean_token_accuracy": 0.18322639763355256,
      "num_tokens": 44700075.0,
      "step": 25770
    },
    {
      "entropy": 5.364870166778564,
      "epoch": 2.005368605329702,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00045937771774122563,
      "loss": 4.8824,
      "mean_token_accuracy": 0.20090241730213165,
      "num_tokens": 44708732.0,
      "step": 25775
    },
    {
      "entropy": 5.462747383117676,
      "epoch": 2.00575763470142,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004593618313931488,
      "loss": 5.0249,
      "mean_token_accuracy": 0.19056898653507232,
      "num_tokens": 44717019.0,
      "step": 25780
    },
    {
      "entropy": 5.473164653778076,
      "epoch": 2.0061466640731376,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045934594224767463,
      "loss": 5.0781,
      "mean_token_accuracy": 0.19051527827978135,
      "num_tokens": 44725359.0,
      "step": 25785
    },
    {
      "entropy": 5.472438335418701,
      "epoch": 2.0065356934448553,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045933005030504424,
      "loss": 5.0448,
      "mean_token_accuracy": 0.1917184591293335,
      "num_tokens": 44733913.0,
      "step": 25790
    },
    {
      "entropy": 5.5343818187713625,
      "epoch": 2.0069247228165725,
      "grad_norm": 1.375,
      "learning_rate": 0.00045931415556549863,
      "loss": 5.22,
      "mean_token_accuracy": 0.1775995224714279,
      "num_tokens": 44743019.0,
      "step": 25795
    },
    {
      "entropy": 5.485415983200073,
      "epoch": 2.0073137521882902,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004592982580292792,
      "loss": 5.1332,
      "mean_token_accuracy": 0.1870163843035698,
      "num_tokens": 44753140.0,
      "step": 25800
    },
    {
      "entropy": 5.4236150741577145,
      "epoch": 2.007702781560008,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045928235769662697,
      "loss": 5.0646,
      "mean_token_accuracy": 0.1901487648487091,
      "num_tokens": 44761470.0,
      "step": 25805
    },
    {
      "entropy": 5.338818120956421,
      "epoch": 2.008091810931725,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045926645456778327,
      "loss": 4.9969,
      "mean_token_accuracy": 0.19235067665576935,
      "num_tokens": 44770570.0,
      "step": 25810
    },
    {
      "entropy": 5.486411762237549,
      "epoch": 2.008480840303443,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00045925054864298946,
      "loss": 5.0966,
      "mean_token_accuracy": 0.18307472616434098,
      "num_tokens": 44779551.0,
      "step": 25815
    },
    {
      "entropy": 5.464999485015869,
      "epoch": 2.0088698696751606,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00045923463992248684,
      "loss": 5.0734,
      "mean_token_accuracy": 0.1903066396713257,
      "num_tokens": 44787633.0,
      "step": 25820
    },
    {
      "entropy": 5.431466054916382,
      "epoch": 2.009258899046878,
      "grad_norm": 1.25,
      "learning_rate": 0.00045921872840651675,
      "loss": 5.0716,
      "mean_token_accuracy": 0.18642838895320893,
      "num_tokens": 44796030.0,
      "step": 25825
    },
    {
      "entropy": 5.508851051330566,
      "epoch": 2.0096479284185955,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045920281409532064,
      "loss": 5.1566,
      "mean_token_accuracy": 0.19014932066202164,
      "num_tokens": 44805055.0,
      "step": 25830
    },
    {
      "entropy": 5.404452848434448,
      "epoch": 2.010036957790313,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004591868969891401,
      "loss": 5.0101,
      "mean_token_accuracy": 0.1900265708565712,
      "num_tokens": 44813767.0,
      "step": 25835
    },
    {
      "entropy": 5.495506429672242,
      "epoch": 2.010425987162031,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004591709770882165,
      "loss": 5.137,
      "mean_token_accuracy": 0.18695679903030396,
      "num_tokens": 44821955.0,
      "step": 25840
    },
    {
      "entropy": 5.47645001411438,
      "epoch": 2.010815016533748,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004591550543927915,
      "loss": 5.1209,
      "mean_token_accuracy": 0.18559882491827012,
      "num_tokens": 44830565.0,
      "step": 25845
    },
    {
      "entropy": 5.422461366653442,
      "epoch": 2.011204045905466,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004591391289031067,
      "loss": 5.0134,
      "mean_token_accuracy": 0.19362840056419373,
      "num_tokens": 44839224.0,
      "step": 25850
    },
    {
      "entropy": 5.344720411300659,
      "epoch": 2.0115930752771836,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004591232006194036,
      "loss": 4.9259,
      "mean_token_accuracy": 0.1979294165968895,
      "num_tokens": 44847756.0,
      "step": 25855
    },
    {
      "entropy": 5.520812129974365,
      "epoch": 2.011982104648901,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045910726954192404,
      "loss": 5.1609,
      "mean_token_accuracy": 0.19344780445098878,
      "num_tokens": 44855739.0,
      "step": 25860
    },
    {
      "entropy": 5.470018815994263,
      "epoch": 2.0123711340206185,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004590913356709097,
      "loss": 5.1066,
      "mean_token_accuracy": 0.1839215040206909,
      "num_tokens": 44864402.0,
      "step": 25865
    },
    {
      "entropy": 5.43224458694458,
      "epoch": 2.012760163392336,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045907539900660237,
      "loss": 5.1302,
      "mean_token_accuracy": 0.18525513410568237,
      "num_tokens": 44873369.0,
      "step": 25870
    },
    {
      "entropy": 5.454889726638794,
      "epoch": 2.013149192764054,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004590594595492438,
      "loss": 5.1507,
      "mean_token_accuracy": 0.18207110166549684,
      "num_tokens": 44882464.0,
      "step": 25875
    },
    {
      "entropy": 5.466670274734497,
      "epoch": 2.013538222135771,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045904351729907593,
      "loss": 5.1088,
      "mean_token_accuracy": 0.1833759665489197,
      "num_tokens": 44891729.0,
      "step": 25880
    },
    {
      "entropy": 5.48141040802002,
      "epoch": 2.013927251507489,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00045902757225634066,
      "loss": 5.1041,
      "mean_token_accuracy": 0.18325370252132417,
      "num_tokens": 44901752.0,
      "step": 25885
    },
    {
      "entropy": 5.483792734146118,
      "epoch": 2.0143162808792066,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004590116244212799,
      "loss": 5.1015,
      "mean_token_accuracy": 0.19141349494457244,
      "num_tokens": 44909899.0,
      "step": 25890
    },
    {
      "entropy": 5.488071727752685,
      "epoch": 2.014705310250924,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004589956737941355,
      "loss": 5.0922,
      "mean_token_accuracy": 0.18420880883932114,
      "num_tokens": 44918472.0,
      "step": 25895
    },
    {
      "entropy": 5.412648487091064,
      "epoch": 2.0150943396226415,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004589797203751497,
      "loss": 4.9924,
      "mean_token_accuracy": 0.19799681454896928,
      "num_tokens": 44926799.0,
      "step": 25900
    },
    {
      "entropy": 5.35377459526062,
      "epoch": 2.015483368994359,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004589637641645645,
      "loss": 5.0084,
      "mean_token_accuracy": 0.1928921714425087,
      "num_tokens": 44935867.0,
      "step": 25905
    },
    {
      "entropy": 5.4387894630432125,
      "epoch": 2.0158723983660765,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00045894780516262193,
      "loss": 5.0925,
      "mean_token_accuracy": 0.18745613247156143,
      "num_tokens": 44944705.0,
      "step": 25910
    },
    {
      "entropy": 5.370718479156494,
      "epoch": 2.016261427737794,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004589318433695642,
      "loss": 4.9483,
      "mean_token_accuracy": 0.19626040905714034,
      "num_tokens": 44954301.0,
      "step": 25915
    },
    {
      "entropy": 5.406123352050781,
      "epoch": 2.016650457109512,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004589158787856336,
      "loss": 5.0294,
      "mean_token_accuracy": 0.18952097594738007,
      "num_tokens": 44963005.0,
      "step": 25920
    },
    {
      "entropy": 5.422341775894165,
      "epoch": 2.0170394864812295,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004588999114110721,
      "loss": 5.0268,
      "mean_token_accuracy": 0.19401021152734757,
      "num_tokens": 44970935.0,
      "step": 25925
    },
    {
      "entropy": 5.447469234466553,
      "epoch": 2.017428515852947,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004588839412461223,
      "loss": 5.0758,
      "mean_token_accuracy": 0.18731636852025985,
      "num_tokens": 44979915.0,
      "step": 25930
    },
    {
      "entropy": 5.466852521896362,
      "epoch": 2.0178175452246645,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004588679682910264,
      "loss": 5.1262,
      "mean_token_accuracy": 0.187038554251194,
      "num_tokens": 44988934.0,
      "step": 25935
    },
    {
      "entropy": 5.5741759777069095,
      "epoch": 2.018206574596382,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004588519925460266,
      "loss": 5.1703,
      "mean_token_accuracy": 0.18277468979358674,
      "num_tokens": 44997578.0,
      "step": 25940
    },
    {
      "entropy": 5.4935530662536625,
      "epoch": 2.0185956039680995,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004588360140113655,
      "loss": 5.1047,
      "mean_token_accuracy": 0.18689310550689697,
      "num_tokens": 45005806.0,
      "step": 25945
    },
    {
      "entropy": 5.4236218452453615,
      "epoch": 2.018984633339817,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004588200326872855,
      "loss": 5.0751,
      "mean_token_accuracy": 0.18582610040903091,
      "num_tokens": 45013948.0,
      "step": 25950
    },
    {
      "entropy": 5.458466720581055,
      "epoch": 2.019373662711535,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004588040485740291,
      "loss": 5.0598,
      "mean_token_accuracy": 0.19267645478248596,
      "num_tokens": 45022636.0,
      "step": 25955
    },
    {
      "entropy": 5.571482706069946,
      "epoch": 2.019762692083252,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004587880616718387,
      "loss": 5.1493,
      "mean_token_accuracy": 0.18160705119371415,
      "num_tokens": 45031194.0,
      "step": 25960
    },
    {
      "entropy": 5.517333269119263,
      "epoch": 2.02015172145497,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004587720719809572,
      "loss": 5.1446,
      "mean_token_accuracy": 0.18359096050262452,
      "num_tokens": 45039436.0,
      "step": 25965
    },
    {
      "entropy": 5.435524034500122,
      "epoch": 2.0205407508266875,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004587560795016269,
      "loss": 5.058,
      "mean_token_accuracy": 0.1918371707201004,
      "num_tokens": 45048173.0,
      "step": 25970
    },
    {
      "entropy": 5.444525098800659,
      "epoch": 2.020929780198405,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004587400842340905,
      "loss": 5.1909,
      "mean_token_accuracy": 0.18015470802783967,
      "num_tokens": 45057896.0,
      "step": 25975
    },
    {
      "entropy": 5.509703636169434,
      "epoch": 2.0213188095701224,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045872408617859083,
      "loss": 5.1107,
      "mean_token_accuracy": 0.18852764070034028,
      "num_tokens": 45066451.0,
      "step": 25980
    },
    {
      "entropy": 5.429889440536499,
      "epoch": 2.02170783894184,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045870808533537066,
      "loss": 4.961,
      "mean_token_accuracy": 0.19611795097589493,
      "num_tokens": 45075143.0,
      "step": 25985
    },
    {
      "entropy": 5.477476787567139,
      "epoch": 2.022096868313558,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045869208170467256,
      "loss": 5.0906,
      "mean_token_accuracy": 0.1898880898952484,
      "num_tokens": 45083370.0,
      "step": 25990
    },
    {
      "entropy": 5.395438241958618,
      "epoch": 2.022485897685275,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004586760752867396,
      "loss": 5.0369,
      "mean_token_accuracy": 0.1846868947148323,
      "num_tokens": 45091765.0,
      "step": 25995
    },
    {
      "entropy": 5.474615669250488,
      "epoch": 2.022874927056993,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045866006608181456,
      "loss": 5.0316,
      "mean_token_accuracy": 0.187349633872509,
      "num_tokens": 45100206.0,
      "step": 26000
    },
    {
      "entropy": 5.415533256530762,
      "epoch": 2.0232639564287105,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004586440540901403,
      "loss": 5.0245,
      "mean_token_accuracy": 0.18898305147886277,
      "num_tokens": 45108308.0,
      "step": 26005
    },
    {
      "entropy": 5.485188388824463,
      "epoch": 2.0236529858004277,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045862803931195976,
      "loss": 5.122,
      "mean_token_accuracy": 0.18841476291418074,
      "num_tokens": 45116473.0,
      "step": 26010
    },
    {
      "entropy": 5.400331974029541,
      "epoch": 2.0240420151721454,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004586120217475161,
      "loss": 5.1042,
      "mean_token_accuracy": 0.18988458812236786,
      "num_tokens": 45125913.0,
      "step": 26015
    },
    {
      "entropy": 5.414013957977295,
      "epoch": 2.024431044543863,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004585960013970522,
      "loss": 5.0189,
      "mean_token_accuracy": 0.18993803411722182,
      "num_tokens": 45134406.0,
      "step": 26020
    },
    {
      "entropy": 5.417628955841065,
      "epoch": 2.024820073915581,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004585799782608112,
      "loss": 5.0484,
      "mean_token_accuracy": 0.18064669519662857,
      "num_tokens": 45142736.0,
      "step": 26025
    },
    {
      "entropy": 5.485821533203125,
      "epoch": 2.025209103287298,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00045856395233903624,
      "loss": 5.0938,
      "mean_token_accuracy": 0.1913389578461647,
      "num_tokens": 45151148.0,
      "step": 26030
    },
    {
      "entropy": 5.439710187911987,
      "epoch": 2.025598132659016,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004585479236319705,
      "loss": 4.9809,
      "mean_token_accuracy": 0.19658254384994506,
      "num_tokens": 45160091.0,
      "step": 26035
    },
    {
      "entropy": 5.43528733253479,
      "epoch": 2.0259871620307335,
      "grad_norm": 1.25,
      "learning_rate": 0.00045853189213985714,
      "loss": 5.0918,
      "mean_token_accuracy": 0.18603586107492448,
      "num_tokens": 45168796.0,
      "step": 26040
    },
    {
      "entropy": 5.409790897369385,
      "epoch": 2.0263761914024507,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045851585786293943,
      "loss": 5.0175,
      "mean_token_accuracy": 0.1993498235940933,
      "num_tokens": 45176891.0,
      "step": 26045
    },
    {
      "entropy": 5.472186803817749,
      "epoch": 2.0267652207741684,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045849982080146067,
      "loss": 5.0964,
      "mean_token_accuracy": 0.1854078933596611,
      "num_tokens": 45184768.0,
      "step": 26050
    },
    {
      "entropy": 5.481969261169434,
      "epoch": 2.027154250145886,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004584837809556642,
      "loss": 5.1394,
      "mean_token_accuracy": 0.18682254999876022,
      "num_tokens": 45193554.0,
      "step": 26055
    },
    {
      "entropy": 5.410466575622559,
      "epoch": 2.0275432795176034,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045846773832579346,
      "loss": 5.112,
      "mean_token_accuracy": 0.1870773747563362,
      "num_tokens": 45202517.0,
      "step": 26060
    },
    {
      "entropy": 5.4473748207092285,
      "epoch": 2.027932308889321,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045845169291209176,
      "loss": 5.0427,
      "mean_token_accuracy": 0.18514372706413268,
      "num_tokens": 45211260.0,
      "step": 26065
    },
    {
      "entropy": 5.451929950714112,
      "epoch": 2.0283213382610388,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045843564471480263,
      "loss": 5.1107,
      "mean_token_accuracy": 0.18812189996242523,
      "num_tokens": 45219828.0,
      "step": 26070
    },
    {
      "entropy": 5.535624027252197,
      "epoch": 2.0287103676327565,
      "grad_norm": 1.5,
      "learning_rate": 0.0004584195937341695,
      "loss": 5.1505,
      "mean_token_accuracy": 0.18370659202337264,
      "num_tokens": 45228977.0,
      "step": 26075
    },
    {
      "entropy": 5.465624761581421,
      "epoch": 2.0290993970044737,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045840353997043606,
      "loss": 5.0588,
      "mean_token_accuracy": 0.19447536021471024,
      "num_tokens": 45237191.0,
      "step": 26080
    },
    {
      "entropy": 5.400886201858521,
      "epoch": 2.0294884263761914,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004583874834238458,
      "loss": 4.9899,
      "mean_token_accuracy": 0.2003542050719261,
      "num_tokens": 45245201.0,
      "step": 26085
    },
    {
      "entropy": 5.4844322681427,
      "epoch": 2.029877455747909,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004583714240946423,
      "loss": 5.1693,
      "mean_token_accuracy": 0.18563995659351348,
      "num_tokens": 45253795.0,
      "step": 26090
    },
    {
      "entropy": 5.569413709640503,
      "epoch": 2.0302664851196264,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045835536198306936,
      "loss": 5.2393,
      "mean_token_accuracy": 0.17743299901485443,
      "num_tokens": 45263403.0,
      "step": 26095
    },
    {
      "entropy": 5.48958911895752,
      "epoch": 2.030655514491344,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045833929708937067,
      "loss": 5.0255,
      "mean_token_accuracy": 0.1974035069346428,
      "num_tokens": 45272085.0,
      "step": 26100
    },
    {
      "entropy": 5.4285706043243405,
      "epoch": 2.0310445438630618,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004583232294137899,
      "loss": 5.0595,
      "mean_token_accuracy": 0.1926318734884262,
      "num_tokens": 45281004.0,
      "step": 26105
    },
    {
      "entropy": 5.430748224258423,
      "epoch": 2.031433573234779,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004583071589565709,
      "loss": 4.9935,
      "mean_token_accuracy": 0.1875285968184471,
      "num_tokens": 45289950.0,
      "step": 26110
    },
    {
      "entropy": 5.476962327957153,
      "epoch": 2.0318226026064967,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004582910857179576,
      "loss": 5.1284,
      "mean_token_accuracy": 0.1811276063323021,
      "num_tokens": 45298155.0,
      "step": 26115
    },
    {
      "entropy": 5.479234266281128,
      "epoch": 2.0322116319782144,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004582750096981938,
      "loss": 5.1491,
      "mean_token_accuracy": 0.18668553978204727,
      "num_tokens": 45306879.0,
      "step": 26120
    },
    {
      "entropy": 5.4478682518005375,
      "epoch": 2.032600661349932,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004582589308975234,
      "loss": 5.1534,
      "mean_token_accuracy": 0.1817260429263115,
      "num_tokens": 45315403.0,
      "step": 26125
    },
    {
      "entropy": 5.447778654098511,
      "epoch": 2.0329896907216494,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045824284931619044,
      "loss": 5.0138,
      "mean_token_accuracy": 0.19322607070207595,
      "num_tokens": 45324351.0,
      "step": 26130
    },
    {
      "entropy": 5.46865553855896,
      "epoch": 2.033378720093367,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045822676495443893,
      "loss": 5.1107,
      "mean_token_accuracy": 0.1877137914299965,
      "num_tokens": 45332513.0,
      "step": 26135
    },
    {
      "entropy": 5.493714332580566,
      "epoch": 2.0337677494650848,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045821067781251284,
      "loss": 5.134,
      "mean_token_accuracy": 0.18351096510887147,
      "num_tokens": 45341002.0,
      "step": 26140
    },
    {
      "entropy": 5.552550315856934,
      "epoch": 2.034156778836802,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045819458789065633,
      "loss": 5.1533,
      "mean_token_accuracy": 0.18525945395231247,
      "num_tokens": 45349650.0,
      "step": 26145
    },
    {
      "entropy": 5.484904813766479,
      "epoch": 2.0345458082085197,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004581784951891135,
      "loss": 5.2299,
      "mean_token_accuracy": 0.17485384047031402,
      "num_tokens": 45359189.0,
      "step": 26150
    },
    {
      "entropy": 5.418768787384034,
      "epoch": 2.0349348375802374,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004581623997081286,
      "loss": 4.9962,
      "mean_token_accuracy": 0.1977767750620842,
      "num_tokens": 45367274.0,
      "step": 26155
    },
    {
      "entropy": 5.442041730880737,
      "epoch": 2.0353238669519547,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004581463014479459,
      "loss": 5.06,
      "mean_token_accuracy": 0.19644575119018554,
      "num_tokens": 45375424.0,
      "step": 26160
    },
    {
      "entropy": 5.488533067703247,
      "epoch": 2.0357128963236724,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004581302004088095,
      "loss": 5.1059,
      "mean_token_accuracy": 0.18596280217170716,
      "num_tokens": 45384333.0,
      "step": 26165
    },
    {
      "entropy": 5.509543514251709,
      "epoch": 2.03610192569539,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004581140965909638,
      "loss": 5.2042,
      "mean_token_accuracy": 0.18016870319843292,
      "num_tokens": 45393309.0,
      "step": 26170
    },
    {
      "entropy": 5.501273250579834,
      "epoch": 2.0364909550671078,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004580979899946531,
      "loss": 5.1504,
      "mean_token_accuracy": 0.18365611284971237,
      "num_tokens": 45401590.0,
      "step": 26175
    },
    {
      "entropy": 5.457723617553711,
      "epoch": 2.036879984438825,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004580818806201219,
      "loss": 5.0504,
      "mean_token_accuracy": 0.1897273913025856,
      "num_tokens": 45409458.0,
      "step": 26180
    },
    {
      "entropy": 5.41987419128418,
      "epoch": 2.0372690138105427,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045806576846761453,
      "loss": 5.0283,
      "mean_token_accuracy": 0.1898242086172104,
      "num_tokens": 45417777.0,
      "step": 26185
    },
    {
      "entropy": 5.450578069686889,
      "epoch": 2.0376580431822604,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045804965353737556,
      "loss": 5.1302,
      "mean_token_accuracy": 0.18693762123584748,
      "num_tokens": 45425489.0,
      "step": 26190
    },
    {
      "entropy": 5.416676998138428,
      "epoch": 2.0380470725539777,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004580335358296494,
      "loss": 5.0398,
      "mean_token_accuracy": 0.19506158530712128,
      "num_tokens": 45433770.0,
      "step": 26195
    },
    {
      "entropy": 5.452640628814697,
      "epoch": 2.0384361019256954,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045801741534468065,
      "loss": 5.1445,
      "mean_token_accuracy": 0.1862376183271408,
      "num_tokens": 45442607.0,
      "step": 26200
    },
    {
      "entropy": 5.423161029815674,
      "epoch": 2.038825131297413,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004580012920827139,
      "loss": 5.0732,
      "mean_token_accuracy": 0.18894716948270798,
      "num_tokens": 45451671.0,
      "step": 26205
    },
    {
      "entropy": 5.503476619720459,
      "epoch": 2.0392141606691303,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004579851660439939,
      "loss": 5.1051,
      "mean_token_accuracy": 0.18642035722732545,
      "num_tokens": 45459861.0,
      "step": 26210
    },
    {
      "entropy": 5.404365110397339,
      "epoch": 2.039603190040848,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045796903722876523,
      "loss": 4.9803,
      "mean_token_accuracy": 0.19433657228946685,
      "num_tokens": 45468623.0,
      "step": 26215
    },
    {
      "entropy": 5.381502723693847,
      "epoch": 2.0399922194125657,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004579529056372726,
      "loss": 4.9651,
      "mean_token_accuracy": 0.19460041224956512,
      "num_tokens": 45476763.0,
      "step": 26220
    },
    {
      "entropy": 5.368109512329101,
      "epoch": 2.0403812487842834,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004579367712697609,
      "loss": 5.0203,
      "mean_token_accuracy": 0.19235813915729522,
      "num_tokens": 45485069.0,
      "step": 26225
    },
    {
      "entropy": 5.4267431735992435,
      "epoch": 2.0407702781560006,
      "grad_norm": 1.390625,
      "learning_rate": 0.00045792063412647475,
      "loss": 5.1842,
      "mean_token_accuracy": 0.1847541570663452,
      "num_tokens": 45494236.0,
      "step": 26230
    },
    {
      "entropy": 5.436605930328369,
      "epoch": 2.0411593075277183,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045790449420765925,
      "loss": 5.0143,
      "mean_token_accuracy": 0.18702832609415054,
      "num_tokens": 45503257.0,
      "step": 26235
    },
    {
      "entropy": 5.509296178817749,
      "epoch": 2.041548336899436,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045788835151355914,
      "loss": 5.0965,
      "mean_token_accuracy": 0.18745123893022536,
      "num_tokens": 45511544.0,
      "step": 26240
    },
    {
      "entropy": 5.418730354309082,
      "epoch": 2.0419373662711533,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045787220604441933,
      "loss": 5.0611,
      "mean_token_accuracy": 0.1919870436191559,
      "num_tokens": 45519951.0,
      "step": 26245
    },
    {
      "entropy": 5.496915102005005,
      "epoch": 2.042326395642871,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045785605780048497,
      "loss": 5.1016,
      "mean_token_accuracy": 0.19065672904253006,
      "num_tokens": 45527978.0,
      "step": 26250
    },
    {
      "entropy": 5.53138689994812,
      "epoch": 2.0427154250145887,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045783990678200086,
      "loss": 5.1572,
      "mean_token_accuracy": 0.1834472596645355,
      "num_tokens": 45537103.0,
      "step": 26255
    },
    {
      "entropy": 5.434952020645142,
      "epoch": 2.043104454386306,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045782375298921227,
      "loss": 5.0444,
      "mean_token_accuracy": 0.1881848692893982,
      "num_tokens": 45545136.0,
      "step": 26260
    },
    {
      "entropy": 5.5000511646270756,
      "epoch": 2.0434934837580236,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004578075964223642,
      "loss": 5.1969,
      "mean_token_accuracy": 0.17813641875982283,
      "num_tokens": 45553709.0,
      "step": 26265
    },
    {
      "entropy": 5.3977649211883545,
      "epoch": 2.0438825131297413,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004577914370817018,
      "loss": 5.0871,
      "mean_token_accuracy": 0.1889914482831955,
      "num_tokens": 45562947.0,
      "step": 26270
    },
    {
      "entropy": 5.4389242172241214,
      "epoch": 2.044271542501459,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045777527496747034,
      "loss": 5.0646,
      "mean_token_accuracy": 0.1866018906235695,
      "num_tokens": 45571319.0,
      "step": 26275
    },
    {
      "entropy": 5.477866172790527,
      "epoch": 2.0446605718731763,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045775911007991493,
      "loss": 5.1864,
      "mean_token_accuracy": 0.18381573110818863,
      "num_tokens": 45580071.0,
      "step": 26280
    },
    {
      "entropy": 5.408913040161133,
      "epoch": 2.045049601244894,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045774294241928093,
      "loss": 4.9924,
      "mean_token_accuracy": 0.19837945997714995,
      "num_tokens": 45587998.0,
      "step": 26285
    },
    {
      "entropy": 5.411702585220337,
      "epoch": 2.0454386306166117,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004577267719858137,
      "loss": 5.1164,
      "mean_token_accuracy": 0.19005484580993653,
      "num_tokens": 45597109.0,
      "step": 26290
    },
    {
      "entropy": 5.507308483123779,
      "epoch": 2.045827659988329,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045771059877975853,
      "loss": 5.0761,
      "mean_token_accuracy": 0.19275854676961898,
      "num_tokens": 45605699.0,
      "step": 26295
    },
    {
      "entropy": 5.581901597976684,
      "epoch": 2.0462166893600466,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004576944228013608,
      "loss": 5.2023,
      "mean_token_accuracy": 0.18384405821561814,
      "num_tokens": 45613634.0,
      "step": 26300
    },
    {
      "entropy": 5.455799531936646,
      "epoch": 2.0466057187317643,
      "grad_norm": 1.203125,
      "learning_rate": 0.000457678244050866,
      "loss": 4.9829,
      "mean_token_accuracy": 0.19694255739450456,
      "num_tokens": 45621814.0,
      "step": 26305
    },
    {
      "entropy": 5.457636404037475,
      "epoch": 2.046994748103482,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004576620625285196,
      "loss": 5.1607,
      "mean_token_accuracy": 0.18184020817279817,
      "num_tokens": 45629386.0,
      "step": 26310
    },
    {
      "entropy": 5.373083591461182,
      "epoch": 2.0473837774751993,
      "grad_norm": 1.390625,
      "learning_rate": 0.00045764587823456717,
      "loss": 4.973,
      "mean_token_accuracy": 0.19928345829248428,
      "num_tokens": 45637565.0,
      "step": 26315
    },
    {
      "entropy": 5.432145118713379,
      "epoch": 2.047772806846917,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045762969116925424,
      "loss": 5.0847,
      "mean_token_accuracy": 0.18930524885654448,
      "num_tokens": 45646876.0,
      "step": 26320
    },
    {
      "entropy": 5.399530553817749,
      "epoch": 2.0481618362186347,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045761350133282643,
      "loss": 4.9632,
      "mean_token_accuracy": 0.1875717118382454,
      "num_tokens": 45655445.0,
      "step": 26325
    },
    {
      "entropy": 5.4098467350006105,
      "epoch": 2.048550865590352,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045759730872552937,
      "loss": 5.0816,
      "mean_token_accuracy": 0.18415096700191497,
      "num_tokens": 45663908.0,
      "step": 26330
    },
    {
      "entropy": 5.505611610412598,
      "epoch": 2.0489398949620696,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004575811133476088,
      "loss": 5.1333,
      "mean_token_accuracy": 0.1881997600197792,
      "num_tokens": 45672842.0,
      "step": 26335
    },
    {
      "entropy": 5.475399303436279,
      "epoch": 2.0493289243337873,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045756491519931047,
      "loss": 5.0688,
      "mean_token_accuracy": 0.18538758009672165,
      "num_tokens": 45680988.0,
      "step": 26340
    },
    {
      "entropy": 5.4382446765899655,
      "epoch": 2.0497179537055046,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004575487142808801,
      "loss": 5.0648,
      "mean_token_accuracy": 0.18918948918581008,
      "num_tokens": 45690165.0,
      "step": 26345
    },
    {
      "entropy": 5.487147951126099,
      "epoch": 2.0501069830772223,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004575325105925636,
      "loss": 5.1278,
      "mean_token_accuracy": 0.1850400447845459,
      "num_tokens": 45698721.0,
      "step": 26350
    },
    {
      "entropy": 5.489384365081787,
      "epoch": 2.05049601244894,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045751630413460665,
      "loss": 5.0933,
      "mean_token_accuracy": 0.18805812895298005,
      "num_tokens": 45707374.0,
      "step": 26355
    },
    {
      "entropy": 5.424233770370483,
      "epoch": 2.0508850418206577,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004575000949072554,
      "loss": 4.969,
      "mean_token_accuracy": 0.19701954424381257,
      "num_tokens": 45715756.0,
      "step": 26360
    },
    {
      "entropy": 5.360181283950806,
      "epoch": 2.051274071192375,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004574838829107557,
      "loss": 5.0651,
      "mean_token_accuracy": 0.19640946239233018,
      "num_tokens": 45724173.0,
      "step": 26365
    },
    {
      "entropy": 5.450315952301025,
      "epoch": 2.0516631005640926,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004574676681453535,
      "loss": 5.0814,
      "mean_token_accuracy": 0.1922634795308113,
      "num_tokens": 45732330.0,
      "step": 26370
    },
    {
      "entropy": 5.481554746627808,
      "epoch": 2.0520521299358103,
      "grad_norm": 1.25,
      "learning_rate": 0.00045745145061129485,
      "loss": 5.1041,
      "mean_token_accuracy": 0.18633596003055572,
      "num_tokens": 45741489.0,
      "step": 26375
    },
    {
      "entropy": 5.381827068328858,
      "epoch": 2.0524411593075276,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045743523030882584,
      "loss": 4.9855,
      "mean_token_accuracy": 0.19808474630117417,
      "num_tokens": 45750816.0,
      "step": 26380
    },
    {
      "entropy": 5.5357691764831545,
      "epoch": 2.0528301886792453,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004574190072381926,
      "loss": 5.1708,
      "mean_token_accuracy": 0.18435940593481065,
      "num_tokens": 45759393.0,
      "step": 26385
    },
    {
      "entropy": 5.523303127288818,
      "epoch": 2.053219218050963,
      "grad_norm": 1.25,
      "learning_rate": 0.0004574027813996413,
      "loss": 5.0721,
      "mean_token_accuracy": 0.18744754791259766,
      "num_tokens": 45767749.0,
      "step": 26390
    },
    {
      "entropy": 5.400083065032959,
      "epoch": 2.05360824742268,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004573865527934181,
      "loss": 5.1404,
      "mean_token_accuracy": 0.18763604164123535,
      "num_tokens": 45776455.0,
      "step": 26395
    },
    {
      "entropy": 5.4435841083526615,
      "epoch": 2.053997276794398,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004573703214197692,
      "loss": 5.0468,
      "mean_token_accuracy": 0.1912248447537422,
      "num_tokens": 45784869.0,
      "step": 26400
    },
    {
      "entropy": 5.431168842315674,
      "epoch": 2.0543863061661156,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045735408727894095,
      "loss": 5.0834,
      "mean_token_accuracy": 0.19728166610002518,
      "num_tokens": 45793956.0,
      "step": 26405
    },
    {
      "entropy": 5.4899486064910885,
      "epoch": 2.0547753355378333,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045733785037117977,
      "loss": 5.0555,
      "mean_token_accuracy": 0.18795484751462938,
      "num_tokens": 45801918.0,
      "step": 26410
    },
    {
      "entropy": 5.443356323242187,
      "epoch": 2.0551643649095506,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004573216106967319,
      "loss": 5.0576,
      "mean_token_accuracy": 0.18777785897254945,
      "num_tokens": 45810026.0,
      "step": 26415
    },
    {
      "entropy": 5.351675224304199,
      "epoch": 2.0555533942812683,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045730536825584365,
      "loss": 5.0211,
      "mean_token_accuracy": 0.18824243396520615,
      "num_tokens": 45818101.0,
      "step": 26420
    },
    {
      "entropy": 5.462250757217407,
      "epoch": 2.055942423652986,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00045728912304876176,
      "loss": 5.1066,
      "mean_token_accuracy": 0.19235883206129073,
      "num_tokens": 45826467.0,
      "step": 26425
    },
    {
      "entropy": 5.461958646774292,
      "epoch": 2.056331453024703,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004572728750757326,
      "loss": 5.0408,
      "mean_token_accuracy": 0.19003259390592575,
      "num_tokens": 45835329.0,
      "step": 26430
    },
    {
      "entropy": 5.413185358047485,
      "epoch": 2.056720482396421,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004572566243370025,
      "loss": 5.1016,
      "mean_token_accuracy": 0.19288131296634675,
      "num_tokens": 45844449.0,
      "step": 26435
    },
    {
      "entropy": 5.4415202140808105,
      "epoch": 2.0571095117681386,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004572403708328183,
      "loss": 5.0519,
      "mean_token_accuracy": 0.18573687523603438,
      "num_tokens": 45854042.0,
      "step": 26440
    },
    {
      "entropy": 5.433035564422608,
      "epoch": 2.057498541139856,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004572241145634266,
      "loss": 5.0416,
      "mean_token_accuracy": 0.19421865195035934,
      "num_tokens": 45863164.0,
      "step": 26445
    },
    {
      "entropy": 5.497725915908814,
      "epoch": 2.0578875705115736,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000457207855529074,
      "loss": 5.1331,
      "mean_token_accuracy": 0.19196872264146805,
      "num_tokens": 45872054.0,
      "step": 26450
    },
    {
      "entropy": 5.443451309204102,
      "epoch": 2.0582765998832913,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045719159373000713,
      "loss": 5.0514,
      "mean_token_accuracy": 0.19789825826883317,
      "num_tokens": 45880887.0,
      "step": 26455
    },
    {
      "entropy": 5.501171827316284,
      "epoch": 2.058665629255009,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004571753291664729,
      "loss": 5.2103,
      "mean_token_accuracy": 0.17678143531084062,
      "num_tokens": 45890078.0,
      "step": 26460
    },
    {
      "entropy": 5.47356276512146,
      "epoch": 2.059054658626726,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004571590618387179,
      "loss": 5.1348,
      "mean_token_accuracy": 0.1892164722084999,
      "num_tokens": 45899803.0,
      "step": 26465
    },
    {
      "entropy": 5.531498432159424,
      "epoch": 2.059443687998444,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004571427917469892,
      "loss": 5.0921,
      "mean_token_accuracy": 0.1861984297633171,
      "num_tokens": 45908851.0,
      "step": 26470
    },
    {
      "entropy": 5.407815551757812,
      "epoch": 2.0598327173701616,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045712651889153345,
      "loss": 5.0732,
      "mean_token_accuracy": 0.19155399799346923,
      "num_tokens": 45917891.0,
      "step": 26475
    },
    {
      "entropy": 5.467431974411011,
      "epoch": 2.060221746741879,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045711024327259764,
      "loss": 5.0599,
      "mean_token_accuracy": 0.18578184992074967,
      "num_tokens": 45926185.0,
      "step": 26480
    },
    {
      "entropy": 5.473532247543335,
      "epoch": 2.0606107761135966,
      "grad_norm": 1.640625,
      "learning_rate": 0.00045709396489042884,
      "loss": 5.1841,
      "mean_token_accuracy": 0.18667600452899932,
      "num_tokens": 45935064.0,
      "step": 26485
    },
    {
      "entropy": 5.4500171661376955,
      "epoch": 2.0609998054853143,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045707768374527385,
      "loss": 5.0569,
      "mean_token_accuracy": 0.18398671746253967,
      "num_tokens": 45943216.0,
      "step": 26490
    },
    {
      "entropy": 5.529759693145752,
      "epoch": 2.0613888348570315,
      "grad_norm": 1.25,
      "learning_rate": 0.0004570613998373799,
      "loss": 5.175,
      "mean_token_accuracy": 0.18865414410829545,
      "num_tokens": 45952178.0,
      "step": 26495
    },
    {
      "entropy": 5.564587593078613,
      "epoch": 2.061777864228749,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045704511316699395,
      "loss": 5.1737,
      "mean_token_accuracy": 0.17856037020683288,
      "num_tokens": 45960962.0,
      "step": 26500
    },
    {
      "entropy": 5.36931529045105,
      "epoch": 2.062166893600467,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045702882373436317,
      "loss": 4.898,
      "mean_token_accuracy": 0.19553142488002778,
      "num_tokens": 45969527.0,
      "step": 26505
    },
    {
      "entropy": 5.43325023651123,
      "epoch": 2.0625559229721846,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004570125315397347,
      "loss": 5.0719,
      "mean_token_accuracy": 0.1851261094212532,
      "num_tokens": 45978258.0,
      "step": 26510
    },
    {
      "entropy": 5.3452150344848635,
      "epoch": 2.062944952343902,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004569962365833558,
      "loss": 5.0274,
      "mean_token_accuracy": 0.19989991188049316,
      "num_tokens": 45986647.0,
      "step": 26515
    },
    {
      "entropy": 5.443856430053711,
      "epoch": 2.0633339817156195,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045697993886547374,
      "loss": 5.0437,
      "mean_token_accuracy": 0.19464795887470246,
      "num_tokens": 45995458.0,
      "step": 26520
    },
    {
      "entropy": 5.523579454421997,
      "epoch": 2.0637230110873372,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045696363838633566,
      "loss": 5.1614,
      "mean_token_accuracy": 0.18642623573541642,
      "num_tokens": 46004293.0,
      "step": 26525
    },
    {
      "entropy": 5.474764680862426,
      "epoch": 2.0641120404590545,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00045694733514618904,
      "loss": 5.0873,
      "mean_token_accuracy": 0.1744238555431366,
      "num_tokens": 46013752.0,
      "step": 26530
    },
    {
      "entropy": 5.4697774887084964,
      "epoch": 2.064501069830772,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004569310291452812,
      "loss": 5.1361,
      "mean_token_accuracy": 0.18623490184545516,
      "num_tokens": 46021935.0,
      "step": 26535
    },
    {
      "entropy": 5.37326717376709,
      "epoch": 2.06489009920249,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004569147203838595,
      "loss": 5.0118,
      "mean_token_accuracy": 0.19864792227745057,
      "num_tokens": 46030567.0,
      "step": 26540
    },
    {
      "entropy": 5.471074819564819,
      "epoch": 2.065279128574207,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045689840886217157,
      "loss": 5.1515,
      "mean_token_accuracy": 0.17549849897623063,
      "num_tokens": 46038764.0,
      "step": 26545
    },
    {
      "entropy": 5.347313499450683,
      "epoch": 2.065668157945925,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045688209458046475,
      "loss": 4.9205,
      "mean_token_accuracy": 0.19911065995693206,
      "num_tokens": 46047393.0,
      "step": 26550
    },
    {
      "entropy": 5.405013465881348,
      "epoch": 2.0660571873176425,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045686577753898663,
      "loss": 5.048,
      "mean_token_accuracy": 0.1798100233078003,
      "num_tokens": 46055748.0,
      "step": 26555
    },
    {
      "entropy": 5.460014057159424,
      "epoch": 2.0664462166893602,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045684945773798483,
      "loss": 5.0269,
      "mean_token_accuracy": 0.1878630682826042,
      "num_tokens": 46063888.0,
      "step": 26560
    },
    {
      "entropy": 5.405939340591431,
      "epoch": 2.0668352460610775,
      "grad_norm": 1.1875,
      "learning_rate": 0.000456833135177707,
      "loss": 5.0308,
      "mean_token_accuracy": 0.19919078499078752,
      "num_tokens": 46073031.0,
      "step": 26565
    },
    {
      "entropy": 5.4378499507904055,
      "epoch": 2.067224275432795,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004568168098584007,
      "loss": 5.1053,
      "mean_token_accuracy": 0.1878155365586281,
      "num_tokens": 46081204.0,
      "step": 26570
    },
    {
      "entropy": 5.48003740310669,
      "epoch": 2.067613304804513,
      "grad_norm": 1.375,
      "learning_rate": 0.0004568004817803137,
      "loss": 5.0864,
      "mean_token_accuracy": 0.19692007452249527,
      "num_tokens": 46089842.0,
      "step": 26575
    },
    {
      "entropy": 5.430802488327027,
      "epoch": 2.06800233417623,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004567841509436937,
      "loss": 5.0851,
      "mean_token_accuracy": 0.1911781370639801,
      "num_tokens": 46098400.0,
      "step": 26580
    },
    {
      "entropy": 5.4463705062866214,
      "epoch": 2.068391363547948,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004567678173487887,
      "loss": 5.1226,
      "mean_token_accuracy": 0.1812913402915001,
      "num_tokens": 46106948.0,
      "step": 26585
    },
    {
      "entropy": 5.417099380493164,
      "epoch": 2.0687803929196655,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004567514809958462,
      "loss": 5.0407,
      "mean_token_accuracy": 0.1880608394742012,
      "num_tokens": 46115361.0,
      "step": 26590
    },
    {
      "entropy": 5.442030429840088,
      "epoch": 2.069169422291383,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004567351418851144,
      "loss": 5.0635,
      "mean_token_accuracy": 0.19073839038610457,
      "num_tokens": 46123658.0,
      "step": 26595
    },
    {
      "entropy": 5.529055643081665,
      "epoch": 2.0695584516631005,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045671880001684113,
      "loss": 5.1086,
      "mean_token_accuracy": 0.1870644211769104,
      "num_tokens": 46132511.0,
      "step": 26600
    },
    {
      "entropy": 5.442026615142822,
      "epoch": 2.069947481034818,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00045670245539127413,
      "loss": 5.1247,
      "mean_token_accuracy": 0.18608764111995696,
      "num_tokens": 46141337.0,
      "step": 26605
    },
    {
      "entropy": 5.437988996505737,
      "epoch": 2.070336510406536,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045668610800866173,
      "loss": 5.1138,
      "mean_token_accuracy": 0.1860414445400238,
      "num_tokens": 46149161.0,
      "step": 26610
    },
    {
      "entropy": 5.538371276855469,
      "epoch": 2.070725539778253,
      "grad_norm": 1.25,
      "learning_rate": 0.00045666975786925177,
      "loss": 5.1825,
      "mean_token_accuracy": 0.17995938807725906,
      "num_tokens": 46157497.0,
      "step": 26615
    },
    {
      "entropy": 5.4514062881469725,
      "epoch": 2.071114569149971,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004566534049732923,
      "loss": 5.0488,
      "mean_token_accuracy": 0.1849229320883751,
      "num_tokens": 46165989.0,
      "step": 26620
    },
    {
      "entropy": 5.440406608581543,
      "epoch": 2.0715035985216885,
      "grad_norm": 1.125,
      "learning_rate": 0.00045663704932103166,
      "loss": 5.0722,
      "mean_token_accuracy": 0.19279370307922364,
      "num_tokens": 46174847.0,
      "step": 26625
    },
    {
      "entropy": 5.436432933807373,
      "epoch": 2.071892627893406,
      "grad_norm": 1.25,
      "learning_rate": 0.00045662069091271785,
      "loss": 5.0611,
      "mean_token_accuracy": 0.19166813045740128,
      "num_tokens": 46183680.0,
      "step": 26630
    },
    {
      "entropy": 5.443097829818726,
      "epoch": 2.0722816572651235,
      "grad_norm": 1.453125,
      "learning_rate": 0.00045660432974859924,
      "loss": 5.0708,
      "mean_token_accuracy": 0.1911827564239502,
      "num_tokens": 46192878.0,
      "step": 26635
    },
    {
      "entropy": 5.476205205917358,
      "epoch": 2.072670686636841,
      "grad_norm": 1.328125,
      "learning_rate": 0.00045658796582892383,
      "loss": 5.1433,
      "mean_token_accuracy": 0.18601690828800202,
      "num_tokens": 46200891.0,
      "step": 26640
    },
    {
      "entropy": 5.409503316879272,
      "epoch": 2.073059716008559,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045657159915394013,
      "loss": 5.024,
      "mean_token_accuracy": 0.1971341699361801,
      "num_tokens": 46210138.0,
      "step": 26645
    },
    {
      "entropy": 5.485125350952148,
      "epoch": 2.073448745380276,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004565552297238964,
      "loss": 5.0592,
      "mean_token_accuracy": 0.18856684118509293,
      "num_tokens": 46218729.0,
      "step": 26650
    },
    {
      "entropy": 5.380882406234742,
      "epoch": 2.073837774751994,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000456538857539041,
      "loss": 5.0714,
      "mean_token_accuracy": 0.18923247158527373,
      "num_tokens": 46228462.0,
      "step": 26655
    },
    {
      "entropy": 5.435634899139404,
      "epoch": 2.0742268041237115,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00045652248259962254,
      "loss": 5.0747,
      "mean_token_accuracy": 0.19677155166864396,
      "num_tokens": 46237417.0,
      "step": 26660
    },
    {
      "entropy": 5.415677404403686,
      "epoch": 2.0746158334954288,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004565061049058892,
      "loss": 5.1254,
      "mean_token_accuracy": 0.17920970618724824,
      "num_tokens": 46246344.0,
      "step": 26665
    },
    {
      "entropy": 5.469440650939942,
      "epoch": 2.0750048628671465,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00045648972445808963,
      "loss": 5.1355,
      "mean_token_accuracy": 0.1877238243818283,
      "num_tokens": 46255286.0,
      "step": 26670
    },
    {
      "entropy": 5.509819650650025,
      "epoch": 2.075393892238864,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045647334125647235,
      "loss": 5.1122,
      "mean_token_accuracy": 0.1903544098138809,
      "num_tokens": 46264737.0,
      "step": 26675
    },
    {
      "entropy": 5.404733943939209,
      "epoch": 2.0757829216105814,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000456456955301286,
      "loss": 5.0166,
      "mean_token_accuracy": 0.19794097989797593,
      "num_tokens": 46273159.0,
      "step": 26680
    },
    {
      "entropy": 5.452092266082763,
      "epoch": 2.076171950982299,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004564405665927791,
      "loss": 5.0906,
      "mean_token_accuracy": 0.1851406380534172,
      "num_tokens": 46281559.0,
      "step": 26685
    },
    {
      "entropy": 5.394826936721802,
      "epoch": 2.076560980354017,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045642417513120043,
      "loss": 5.0289,
      "mean_token_accuracy": 0.19696059226989746,
      "num_tokens": 46289941.0,
      "step": 26690
    },
    {
      "entropy": 5.40030345916748,
      "epoch": 2.076950009725734,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045640778091679876,
      "loss": 5.0354,
      "mean_token_accuracy": 0.1922016590833664,
      "num_tokens": 46298711.0,
      "step": 26695
    },
    {
      "entropy": 5.399805974960327,
      "epoch": 2.0773390390974518,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004563913839498226,
      "loss": 4.9964,
      "mean_token_accuracy": 0.190868778526783,
      "num_tokens": 46306956.0,
      "step": 26700
    },
    {
      "entropy": 5.474187707901001,
      "epoch": 2.0777280684691695,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000456374984230521,
      "loss": 5.0144,
      "mean_token_accuracy": 0.1913638561964035,
      "num_tokens": 46315165.0,
      "step": 26705
    },
    {
      "entropy": 5.4113133430480955,
      "epoch": 2.078117097840887,
      "grad_norm": 1.25,
      "learning_rate": 0.0004563585817591427,
      "loss": 4.947,
      "mean_token_accuracy": 0.20016194581985475,
      "num_tokens": 46323347.0,
      "step": 26710
    },
    {
      "entropy": 5.4486222743988035,
      "epoch": 2.0785061272126044,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004563421765359365,
      "loss": 5.075,
      "mean_token_accuracy": 0.19261564761400224,
      "num_tokens": 46331302.0,
      "step": 26715
    },
    {
      "entropy": 5.436268615722656,
      "epoch": 2.078895156584322,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045632576856115156,
      "loss": 5.0971,
      "mean_token_accuracy": 0.1845285639166832,
      "num_tokens": 46339050.0,
      "step": 26720
    },
    {
      "entropy": 5.4048937320709225,
      "epoch": 2.07928418595604,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045630935783503657,
      "loss": 5.073,
      "mean_token_accuracy": 0.19236867129802704,
      "num_tokens": 46347350.0,
      "step": 26725
    },
    {
      "entropy": 5.523930406570434,
      "epoch": 2.079673215327757,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004562929443578407,
      "loss": 5.1607,
      "mean_token_accuracy": 0.17551758736371995,
      "num_tokens": 46356951.0,
      "step": 26730
    },
    {
      "entropy": 5.569313859939575,
      "epoch": 2.0800622446994748,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004562765281298129,
      "loss": 5.1806,
      "mean_token_accuracy": 0.18447115272283554,
      "num_tokens": 46367056.0,
      "step": 26735
    },
    {
      "entropy": 5.418645620346069,
      "epoch": 2.0804512740711925,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004562601091512024,
      "loss": 5.062,
      "mean_token_accuracy": 0.1939232885837555,
      "num_tokens": 46375668.0,
      "step": 26740
    },
    {
      "entropy": 5.414381265640259,
      "epoch": 2.08084030344291,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004562436874222582,
      "loss": 5.086,
      "mean_token_accuracy": 0.18924905955791474,
      "num_tokens": 46384460.0,
      "step": 26745
    },
    {
      "entropy": 5.321546459197998,
      "epoch": 2.0812293328146274,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045622726294322955,
      "loss": 4.9072,
      "mean_token_accuracy": 0.19551345109939575,
      "num_tokens": 46393349.0,
      "step": 26750
    },
    {
      "entropy": 5.435777044296264,
      "epoch": 2.081618362186345,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045621083571436563,
      "loss": 5.0909,
      "mean_token_accuracy": 0.18395120948553084,
      "num_tokens": 46403081.0,
      "step": 26755
    },
    {
      "entropy": 5.3761180400848385,
      "epoch": 2.082007391558063,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004561944057359157,
      "loss": 4.944,
      "mean_token_accuracy": 0.199859519302845,
      "num_tokens": 46412219.0,
      "step": 26760
    },
    {
      "entropy": 5.471138668060303,
      "epoch": 2.08239642092978,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004561779730081291,
      "loss": 5.1175,
      "mean_token_accuracy": 0.19303781688213348,
      "num_tokens": 46420811.0,
      "step": 26765
    },
    {
      "entropy": 5.33242883682251,
      "epoch": 2.0827854503014978,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004561615375312551,
      "loss": 5.0116,
      "mean_token_accuracy": 0.1910262480378151,
      "num_tokens": 46429431.0,
      "step": 26770
    },
    {
      "entropy": 5.425367927551269,
      "epoch": 2.0831744796732155,
      "grad_norm": 1.25,
      "learning_rate": 0.00045614509930554304,
      "loss": 5.0834,
      "mean_token_accuracy": 0.18684103786945344,
      "num_tokens": 46438344.0,
      "step": 26775
    },
    {
      "entropy": 5.499448537826538,
      "epoch": 2.0835635090449327,
      "grad_norm": 1.25,
      "learning_rate": 0.00045612865833124253,
      "loss": 5.1484,
      "mean_token_accuracy": 0.1878812789916992,
      "num_tokens": 46446960.0,
      "step": 26780
    },
    {
      "entropy": 5.4102109432220455,
      "epoch": 2.0839525384166504,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004561122146086029,
      "loss": 5.0397,
      "mean_token_accuracy": 0.19318843483924866,
      "num_tokens": 46454862.0,
      "step": 26785
    },
    {
      "entropy": 5.435419607162475,
      "epoch": 2.084341567788368,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004560957681378737,
      "loss": 5.1569,
      "mean_token_accuracy": 0.18789760768413544,
      "num_tokens": 46463208.0,
      "step": 26790
    },
    {
      "entropy": 5.526937437057495,
      "epoch": 2.084730597160086,
      "grad_norm": 1.25,
      "learning_rate": 0.00045607931891930445,
      "loss": 5.1728,
      "mean_token_accuracy": 0.18427389115095139,
      "num_tokens": 46472253.0,
      "step": 26795
    },
    {
      "entropy": 5.575931739807129,
      "epoch": 2.085119626531803,
      "grad_norm": 1.25,
      "learning_rate": 0.0004560628669531447,
      "loss": 5.1762,
      "mean_token_accuracy": 0.18843009322881699,
      "num_tokens": 46482133.0,
      "step": 26800
    },
    {
      "entropy": 5.43471245765686,
      "epoch": 2.0855086559035207,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045604641223964416,
      "loss": 5.1283,
      "mean_token_accuracy": 0.1859818920493126,
      "num_tokens": 46491566.0,
      "step": 26805
    },
    {
      "entropy": 5.483510065078735,
      "epoch": 2.0858976852752384,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045602995477905247,
      "loss": 5.1576,
      "mean_token_accuracy": 0.17420623004436492,
      "num_tokens": 46501283.0,
      "step": 26810
    },
    {
      "entropy": 5.5487672805786135,
      "epoch": 2.0862867146469557,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004560134945716194,
      "loss": 5.1095,
      "mean_token_accuracy": 0.18795526921749114,
      "num_tokens": 46510384.0,
      "step": 26815
    },
    {
      "entropy": 5.519675445556641,
      "epoch": 2.0866757440186734,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045599703161759464,
      "loss": 5.0839,
      "mean_token_accuracy": 0.18378926813602448,
      "num_tokens": 46519202.0,
      "step": 26820
    },
    {
      "entropy": 5.416194581985474,
      "epoch": 2.087064773390391,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045598056591722805,
      "loss": 5.0521,
      "mean_token_accuracy": 0.19471883177757263,
      "num_tokens": 46528030.0,
      "step": 26825
    },
    {
      "entropy": 5.476199293136597,
      "epoch": 2.0874538027621083,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045596409747076936,
      "loss": 5.1469,
      "mean_token_accuracy": 0.19041065871715546,
      "num_tokens": 46537222.0,
      "step": 26830
    },
    {
      "entropy": 5.498877620697021,
      "epoch": 2.087842832133826,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004559476262784686,
      "loss": 5.0657,
      "mean_token_accuracy": 0.19220069497823716,
      "num_tokens": 46546486.0,
      "step": 26835
    },
    {
      "entropy": 5.50741868019104,
      "epoch": 2.0882318615055437,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004559311523405755,
      "loss": 5.1267,
      "mean_token_accuracy": 0.18929739892482758,
      "num_tokens": 46554518.0,
      "step": 26840
    },
    {
      "entropy": 5.5014581203460695,
      "epoch": 2.0886208908772614,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004559146756573402,
      "loss": 5.093,
      "mean_token_accuracy": 0.18957652896642685,
      "num_tokens": 46563183.0,
      "step": 26845
    },
    {
      "entropy": 5.418142318725586,
      "epoch": 2.0890099202489787,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045589819622901274,
      "loss": 4.9733,
      "mean_token_accuracy": 0.19585610330104827,
      "num_tokens": 46571514.0,
      "step": 26850
    },
    {
      "entropy": 5.427031993865967,
      "epoch": 2.0893989496206964,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000455881714055843,
      "loss": 5.0884,
      "mean_token_accuracy": 0.1900682717561722,
      "num_tokens": 46580827.0,
      "step": 26855
    },
    {
      "entropy": 5.500057506561279,
      "epoch": 2.089787978992414,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045586522913808114,
      "loss": 5.1467,
      "mean_token_accuracy": 0.18281211853027343,
      "num_tokens": 46591254.0,
      "step": 26860
    },
    {
      "entropy": 5.4590778827667235,
      "epoch": 2.0901770083641313,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004558487414759773,
      "loss": 5.0245,
      "mean_token_accuracy": 0.19076938182115555,
      "num_tokens": 46599576.0,
      "step": 26865
    },
    {
      "entropy": 5.408864831924438,
      "epoch": 2.090566037735849,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045583225106978175,
      "loss": 5.0925,
      "mean_token_accuracy": 0.1896760046482086,
      "num_tokens": 46607607.0,
      "step": 26870
    },
    {
      "entropy": 5.35610785484314,
      "epoch": 2.0909550671075667,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004558157579197446,
      "loss": 5.0217,
      "mean_token_accuracy": 0.19266801029443742,
      "num_tokens": 46616446.0,
      "step": 26875
    },
    {
      "entropy": 5.461592102050782,
      "epoch": 2.091344096479284,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045579926202611603,
      "loss": 5.0055,
      "mean_token_accuracy": 0.18817213773727418,
      "num_tokens": 46624875.0,
      "step": 26880
    },
    {
      "entropy": 5.4236732006073,
      "epoch": 2.0917331258510017,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004557827633891465,
      "loss": 5.0104,
      "mean_token_accuracy": 0.19401447474956512,
      "num_tokens": 46633354.0,
      "step": 26885
    },
    {
      "entropy": 5.440175342559814,
      "epoch": 2.0921221552227194,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004557662620090863,
      "loss": 5.0953,
      "mean_token_accuracy": 0.1941651865839958,
      "num_tokens": 46642034.0,
      "step": 26890
    },
    {
      "entropy": 5.438977003097534,
      "epoch": 2.092511184594437,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045574975788618564,
      "loss": 5.0705,
      "mean_token_accuracy": 0.18808974027633668,
      "num_tokens": 46650675.0,
      "step": 26895
    },
    {
      "entropy": 5.405484294891357,
      "epoch": 2.0929002139661543,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004557332510206953,
      "loss": 4.9798,
      "mean_token_accuracy": 0.19802161902189255,
      "num_tokens": 46658834.0,
      "step": 26900
    },
    {
      "entropy": 5.3337541103363035,
      "epoch": 2.093289243337872,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004557167414128654,
      "loss": 4.9514,
      "mean_token_accuracy": 0.20096097737550736,
      "num_tokens": 46667386.0,
      "step": 26905
    },
    {
      "entropy": 5.372097730636597,
      "epoch": 2.0936782727095897,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004557002290629467,
      "loss": 4.9626,
      "mean_token_accuracy": 0.20097084492444992,
      "num_tokens": 46675492.0,
      "step": 26910
    },
    {
      "entropy": 5.394836950302124,
      "epoch": 2.094067302081307,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045568371397118954,
      "loss": 5.0709,
      "mean_token_accuracy": 0.19426148980855942,
      "num_tokens": 46684079.0,
      "step": 26915
    },
    {
      "entropy": 5.4286980628967285,
      "epoch": 2.0944563314530247,
      "grad_norm": 1.25,
      "learning_rate": 0.0004556671961378446,
      "loss": 5.1026,
      "mean_token_accuracy": 0.1844624862074852,
      "num_tokens": 46693059.0,
      "step": 26920
    },
    {
      "entropy": 5.51028356552124,
      "epoch": 2.0948453608247424,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045565067556316264,
      "loss": 5.0375,
      "mean_token_accuracy": 0.19147275537252426,
      "num_tokens": 46701639.0,
      "step": 26925
    },
    {
      "entropy": 5.446740913391113,
      "epoch": 2.0952343901964596,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004556341522473941,
      "loss": 5.0944,
      "mean_token_accuracy": 0.19343752562999725,
      "num_tokens": 46710870.0,
      "step": 26930
    },
    {
      "entropy": 5.504725217819214,
      "epoch": 2.0956234195681773,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004556176261907899,
      "loss": 5.1207,
      "mean_token_accuracy": 0.1879246711730957,
      "num_tokens": 46719236.0,
      "step": 26935
    },
    {
      "entropy": 5.439628601074219,
      "epoch": 2.096012448939895,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004556010973936006,
      "loss": 5.0609,
      "mean_token_accuracy": 0.1904678836464882,
      "num_tokens": 46727595.0,
      "step": 26940
    },
    {
      "entropy": 5.458144617080689,
      "epoch": 2.0964014783116127,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004555845658560772,
      "loss": 5.0648,
      "mean_token_accuracy": 0.19215698689222335,
      "num_tokens": 46735874.0,
      "step": 26945
    },
    {
      "entropy": 5.404890298843384,
      "epoch": 2.09679050768333,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00045556803157847036,
      "loss": 5.0841,
      "mean_token_accuracy": 0.1939731389284134,
      "num_tokens": 46744738.0,
      "step": 26950
    },
    {
      "entropy": 5.475359582901001,
      "epoch": 2.0971795370550477,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045555149456103113,
      "loss": 5.0566,
      "mean_token_accuracy": 0.18531084954738616,
      "num_tokens": 46753252.0,
      "step": 26955
    },
    {
      "entropy": 5.410978078842163,
      "epoch": 2.0975685664267654,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004555349548040102,
      "loss": 4.9936,
      "mean_token_accuracy": 0.192803356051445,
      "num_tokens": 46761234.0,
      "step": 26960
    },
    {
      "entropy": 5.438618946075439,
      "epoch": 2.0979575957984826,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004555184123076589,
      "loss": 5.1332,
      "mean_token_accuracy": 0.17934600114822388,
      "num_tokens": 46770087.0,
      "step": 26965
    },
    {
      "entropy": 5.403347969055176,
      "epoch": 2.0983466251702003,
      "grad_norm": 1.25,
      "learning_rate": 0.00045550186707222785,
      "loss": 5.0017,
      "mean_token_accuracy": 0.19087327569723128,
      "num_tokens": 46778985.0,
      "step": 26970
    },
    {
      "entropy": 5.431661319732666,
      "epoch": 2.098735654541918,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004554853190979683,
      "loss": 5.0994,
      "mean_token_accuracy": 0.19005565643310546,
      "num_tokens": 46787402.0,
      "step": 26975
    },
    {
      "entropy": 5.4861914157867435,
      "epoch": 2.0991246839136353,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045546876838513134,
      "loss": 5.1137,
      "mean_token_accuracy": 0.17927255779504775,
      "num_tokens": 46796254.0,
      "step": 26980
    },
    {
      "entropy": 5.4192259311676025,
      "epoch": 2.099513713285353,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045545221493396805,
      "loss": 5.0464,
      "mean_token_accuracy": 0.1902633339166641,
      "num_tokens": 46804979.0,
      "step": 26985
    },
    {
      "entropy": 5.3654101371765135,
      "epoch": 2.0999027426570707,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045543565874472963,
      "loss": 4.9886,
      "mean_token_accuracy": 0.1958442971110344,
      "num_tokens": 46813718.0,
      "step": 26990
    },
    {
      "entropy": 5.475353765487671,
      "epoch": 2.1002917720287884,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045541909981766725,
      "loss": 5.0951,
      "mean_token_accuracy": 0.1835153505206108,
      "num_tokens": 46822265.0,
      "step": 26995
    },
    {
      "entropy": 5.432806825637817,
      "epoch": 2.1006808014005056,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004554025381530322,
      "loss": 5.0497,
      "mean_token_accuracy": 0.1957593619823456,
      "num_tokens": 46831292.0,
      "step": 27000
    },
    {
      "epoch": 2.1006808014005056,
      "eval_entropy": 5.212705510848984,
      "eval_loss": 5.211482524871826,
      "eval_mean_token_accuracy": 0.1925574202302044,
      "eval_num_tokens": 46831292.0,
      "eval_runtime": 28.7252,
      "eval_samples_per_second": 1446.745,
      "eval_steps_per_second": 180.852,
      "step": 27000
    },
    {
      "entropy": 5.319648027420044,
      "epoch": 2.1010698307722233,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004553859737510758,
      "loss": 5.0331,
      "mean_token_accuracy": 0.20153597891330718,
      "num_tokens": 46840029.0,
      "step": 27005
    },
    {
      "entropy": 5.408263206481934,
      "epoch": 2.101458860143941,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004553694066120493,
      "loss": 4.9651,
      "mean_token_accuracy": 0.19719047248363494,
      "num_tokens": 46848475.0,
      "step": 27010
    },
    {
      "entropy": 5.497275447845459,
      "epoch": 2.1018478895156583,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045535283673620426,
      "loss": 5.1135,
      "mean_token_accuracy": 0.1872415155172348,
      "num_tokens": 46857124.0,
      "step": 27015
    },
    {
      "entropy": 5.560809373855591,
      "epoch": 2.102236918887376,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045533626412379194,
      "loss": 5.1999,
      "mean_token_accuracy": 0.18556116968393327,
      "num_tokens": 46865809.0,
      "step": 27020
    },
    {
      "entropy": 5.440860414505005,
      "epoch": 2.1026259482590937,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004553196887750638,
      "loss": 5.1073,
      "mean_token_accuracy": 0.18493400663137435,
      "num_tokens": 46874845.0,
      "step": 27025
    },
    {
      "entropy": 5.4391368389129635,
      "epoch": 2.103014977630811,
      "grad_norm": 1.25,
      "learning_rate": 0.0004553031106902714,
      "loss": 5.0385,
      "mean_token_accuracy": 0.19123439937829972,
      "num_tokens": 46883740.0,
      "step": 27030
    },
    {
      "entropy": 5.388023614883423,
      "epoch": 2.1034040070025286,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045528652986966627,
      "loss": 5.0239,
      "mean_token_accuracy": 0.20042583495378494,
      "num_tokens": 46891936.0,
      "step": 27035
    },
    {
      "entropy": 5.312006092071533,
      "epoch": 2.1037930363742463,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045526994631350006,
      "loss": 5.0463,
      "mean_token_accuracy": 0.18921314924955368,
      "num_tokens": 46900918.0,
      "step": 27040
    },
    {
      "entropy": 5.4372230052948,
      "epoch": 2.104182065745964,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004552533600220243,
      "loss": 5.0856,
      "mean_token_accuracy": 0.18699978440999984,
      "num_tokens": 46910785.0,
      "step": 27045
    },
    {
      "entropy": 5.510256624221801,
      "epoch": 2.1045710951176813,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045523677099549073,
      "loss": 5.1338,
      "mean_token_accuracy": 0.1900095283985138,
      "num_tokens": 46919191.0,
      "step": 27050
    },
    {
      "entropy": 5.340344429016113,
      "epoch": 2.104960124489399,
      "grad_norm": 1.109375,
      "learning_rate": 0.000455220179234151,
      "loss": 5.0214,
      "mean_token_accuracy": 0.19161920696496965,
      "num_tokens": 46927574.0,
      "step": 27055
    },
    {
      "entropy": 5.371052503585815,
      "epoch": 2.1053491538611167,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004552035847382569,
      "loss": 5.0104,
      "mean_token_accuracy": 0.19595623165369033,
      "num_tokens": 46935694.0,
      "step": 27060
    },
    {
      "entropy": 5.485352230072022,
      "epoch": 2.105738183232834,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00045518698750806024,
      "loss": 5.1502,
      "mean_token_accuracy": 0.18573129624128343,
      "num_tokens": 46943929.0,
      "step": 27065
    },
    {
      "entropy": 5.403091287612915,
      "epoch": 2.1061272126045516,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045517038754381287,
      "loss": 4.942,
      "mean_token_accuracy": 0.1995285466313362,
      "num_tokens": 46952130.0,
      "step": 27070
    },
    {
      "entropy": 5.396305942535401,
      "epoch": 2.1065162419762693,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004551537848457666,
      "loss": 5.1003,
      "mean_token_accuracy": 0.1918123722076416,
      "num_tokens": 46961010.0,
      "step": 27075
    },
    {
      "entropy": 5.45666356086731,
      "epoch": 2.106905271347987,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004551371794141734,
      "loss": 5.0937,
      "mean_token_accuracy": 0.18980918377637862,
      "num_tokens": 46969789.0,
      "step": 27080
    },
    {
      "entropy": 5.515809869766235,
      "epoch": 2.1072943007197042,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004551205712492852,
      "loss": 5.1191,
      "mean_token_accuracy": 0.1911153793334961,
      "num_tokens": 46977436.0,
      "step": 27085
    },
    {
      "entropy": 5.430668735504151,
      "epoch": 2.107683330091422,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004551039603513541,
      "loss": 5.0727,
      "mean_token_accuracy": 0.1939419686794281,
      "num_tokens": 46986065.0,
      "step": 27090
    },
    {
      "entropy": 5.4206853866577145,
      "epoch": 2.1080723594631396,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045508734672063204,
      "loss": 5.0802,
      "mean_token_accuracy": 0.18355799168348313,
      "num_tokens": 46994355.0,
      "step": 27095
    },
    {
      "entropy": 5.3749549865722654,
      "epoch": 2.108461388834857,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004550707303573711,
      "loss": 5.0283,
      "mean_token_accuracy": 0.18941378593444824,
      "num_tokens": 47003028.0,
      "step": 27100
    },
    {
      "entropy": 5.45003752708435,
      "epoch": 2.1088504182065746,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045505411126182347,
      "loss": 5.1741,
      "mean_token_accuracy": 0.18125323504209517,
      "num_tokens": 47012351.0,
      "step": 27105
    },
    {
      "entropy": 5.516589403152466,
      "epoch": 2.1092394475782923,
      "grad_norm": 1.25,
      "learning_rate": 0.0004550374894342412,
      "loss": 5.1012,
      "mean_token_accuracy": 0.2013810694217682,
      "num_tokens": 47020600.0,
      "step": 27110
    },
    {
      "entropy": 5.386406469345093,
      "epoch": 2.1096284769500095,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004550208648748767,
      "loss": 5.0605,
      "mean_token_accuracy": 0.18609558343887328,
      "num_tokens": 47029276.0,
      "step": 27115
    },
    {
      "entropy": 5.379668283462524,
      "epoch": 2.1100175063217272,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000455004237583982,
      "loss": 5.0106,
      "mean_token_accuracy": 0.19206558167934418,
      "num_tokens": 47037894.0,
      "step": 27120
    },
    {
      "entropy": 5.3908305168151855,
      "epoch": 2.110406535693445,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004549876075618096,
      "loss": 5.0097,
      "mean_token_accuracy": 0.1887384682893753,
      "num_tokens": 47046602.0,
      "step": 27125
    },
    {
      "entropy": 5.444711303710937,
      "epoch": 2.110795565065162,
      "grad_norm": 1.25,
      "learning_rate": 0.0004549709748086117,
      "loss": 5.1698,
      "mean_token_accuracy": 0.18722872734069823,
      "num_tokens": 47056091.0,
      "step": 27130
    },
    {
      "entropy": 5.517124319076538,
      "epoch": 2.11118459443688,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045495433932464063,
      "loss": 5.1146,
      "mean_token_accuracy": 0.17795937955379487,
      "num_tokens": 47065420.0,
      "step": 27135
    },
    {
      "entropy": 5.478368425369263,
      "epoch": 2.1115736238085976,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000454937701110149,
      "loss": 5.0554,
      "mean_token_accuracy": 0.19001957029104233,
      "num_tokens": 47074247.0,
      "step": 27140
    },
    {
      "entropy": 5.492049312591552,
      "epoch": 2.1119626531803153,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045492106016538917,
      "loss": 5.1624,
      "mean_token_accuracy": 0.17680409401655198,
      "num_tokens": 47083589.0,
      "step": 27145
    },
    {
      "entropy": 5.448315763473511,
      "epoch": 2.1123516825520325,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00045490441649061354,
      "loss": 5.0281,
      "mean_token_accuracy": 0.18793509155511856,
      "num_tokens": 47091861.0,
      "step": 27150
    },
    {
      "entropy": 5.51703462600708,
      "epoch": 2.1127407119237502,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004548877700860747,
      "loss": 5.2389,
      "mean_token_accuracy": 0.17979475408792495,
      "num_tokens": 47101436.0,
      "step": 27155
    },
    {
      "entropy": 5.447565650939941,
      "epoch": 2.113129741295468,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045487112095202544,
      "loss": 5.0732,
      "mean_token_accuracy": 0.19703569412231445,
      "num_tokens": 47110390.0,
      "step": 27160
    },
    {
      "entropy": 5.518107795715332,
      "epoch": 2.113518770667185,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004548544690887181,
      "loss": 5.227,
      "mean_token_accuracy": 0.1833069920539856,
      "num_tokens": 47119001.0,
      "step": 27165
    },
    {
      "entropy": 5.410075855255127,
      "epoch": 2.113907800038903,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004548378144964054,
      "loss": 5.0382,
      "mean_token_accuracy": 0.19441525787115096,
      "num_tokens": 47127067.0,
      "step": 27170
    },
    {
      "entropy": 5.503660440444946,
      "epoch": 2.1142968294106206,
      "grad_norm": 1.125,
      "learning_rate": 0.00045482115717534024,
      "loss": 5.2423,
      "mean_token_accuracy": 0.17909064143896103,
      "num_tokens": 47136747.0,
      "step": 27175
    },
    {
      "entropy": 5.482768297195435,
      "epoch": 2.1146858587823383,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004548044971257752,
      "loss": 5.0258,
      "mean_token_accuracy": 0.1966236039996147,
      "num_tokens": 47145783.0,
      "step": 27180
    },
    {
      "entropy": 5.4686445713043215,
      "epoch": 2.1150748881540555,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045478783434796307,
      "loss": 5.0753,
      "mean_token_accuracy": 0.18881845027208327,
      "num_tokens": 47154344.0,
      "step": 27185
    },
    {
      "entropy": 5.386042785644531,
      "epoch": 2.1154639175257732,
      "grad_norm": 1.25,
      "learning_rate": 0.00045477116884215675,
      "loss": 5.0962,
      "mean_token_accuracy": 0.18970223963260652,
      "num_tokens": 47163733.0,
      "step": 27190
    },
    {
      "entropy": 5.414426374435425,
      "epoch": 2.115852946897491,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000454754500608609,
      "loss": 5.0815,
      "mean_token_accuracy": 0.19444743543863297,
      "num_tokens": 47172341.0,
      "step": 27195
    },
    {
      "entropy": 5.2976878643035885,
      "epoch": 2.116241976269208,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004547378296475729,
      "loss": 4.882,
      "mean_token_accuracy": 0.20876674503087997,
      "num_tokens": 47181118.0,
      "step": 27200
    },
    {
      "entropy": 5.422082853317261,
      "epoch": 2.116631005640926,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045472115595930124,
      "loss": 5.1567,
      "mean_token_accuracy": 0.18738051801919936,
      "num_tokens": 47190348.0,
      "step": 27205
    },
    {
      "entropy": 5.466672468185425,
      "epoch": 2.1170200350126436,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045470447954404716,
      "loss": 5.1683,
      "mean_token_accuracy": 0.18718502968549727,
      "num_tokens": 47199812.0,
      "step": 27210
    },
    {
      "entropy": 5.512445497512817,
      "epoch": 2.117409064384361,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00045468780040206364,
      "loss": 5.0687,
      "mean_token_accuracy": 0.19043071269989015,
      "num_tokens": 47208547.0,
      "step": 27215
    },
    {
      "entropy": 5.396255111694336,
      "epoch": 2.1177980937560785,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004546711185336037,
      "loss": 5.0087,
      "mean_token_accuracy": 0.19955634772777558,
      "num_tokens": 47216980.0,
      "step": 27220
    },
    {
      "entropy": 5.435706949234008,
      "epoch": 2.1181871231277962,
      "grad_norm": 1.25,
      "learning_rate": 0.00045465443393892047,
      "loss": 5.1168,
      "mean_token_accuracy": 0.18398870825767516,
      "num_tokens": 47225701.0,
      "step": 27225
    },
    {
      "entropy": 5.478459644317627,
      "epoch": 2.118576152499514,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004546377466182673,
      "loss": 5.1166,
      "mean_token_accuracy": 0.1901984393596649,
      "num_tokens": 47234760.0,
      "step": 27230
    },
    {
      "entropy": 5.497698640823364,
      "epoch": 2.118965181871231,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004546210565718972,
      "loss": 5.1146,
      "mean_token_accuracy": 0.18418630361557006,
      "num_tokens": 47243392.0,
      "step": 27235
    },
    {
      "entropy": 5.440277814865112,
      "epoch": 2.119354211242949,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045460436380006337,
      "loss": 5.0706,
      "mean_token_accuracy": 0.19175945669412614,
      "num_tokens": 47251297.0,
      "step": 27240
    },
    {
      "entropy": 5.323189783096313,
      "epoch": 2.1197432406146666,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004545876683030192,
      "loss": 4.9474,
      "mean_token_accuracy": 0.19277252554893493,
      "num_tokens": 47259859.0,
      "step": 27245
    },
    {
      "entropy": 5.479398107528686,
      "epoch": 2.120132269986384,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004545709700810181,
      "loss": 5.1164,
      "mean_token_accuracy": 0.19212204664945604,
      "num_tokens": 47268653.0,
      "step": 27250
    },
    {
      "entropy": 5.485763931274414,
      "epoch": 2.1205212993581015,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004545542691343133,
      "loss": 5.0414,
      "mean_token_accuracy": 0.18908853381872176,
      "num_tokens": 47277254.0,
      "step": 27255
    },
    {
      "entropy": 5.458822727203369,
      "epoch": 2.120910328729819,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004545375654631582,
      "loss": 5.1261,
      "mean_token_accuracy": 0.18795017302036285,
      "num_tokens": 47285885.0,
      "step": 27260
    },
    {
      "entropy": 5.3482067584991455,
      "epoch": 2.1212993581015365,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045452085906780645,
      "loss": 5.0457,
      "mean_token_accuracy": 0.19111974090337752,
      "num_tokens": 47294025.0,
      "step": 27265
    },
    {
      "entropy": 5.349497747421265,
      "epoch": 2.121688387473254,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045450414994851123,
      "loss": 5.0149,
      "mean_token_accuracy": 0.19197070002555847,
      "num_tokens": 47303531.0,
      "step": 27270
    },
    {
      "entropy": 5.4431891441345215,
      "epoch": 2.122077416844972,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045448743810552636,
      "loss": 5.1466,
      "mean_token_accuracy": 0.18833398669958115,
      "num_tokens": 47311753.0,
      "step": 27275
    },
    {
      "entropy": 5.384894466400146,
      "epoch": 2.1224664462166896,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045447072353910537,
      "loss": 5.0455,
      "mean_token_accuracy": 0.18789823949337006,
      "num_tokens": 47320736.0,
      "step": 27280
    },
    {
      "entropy": 5.505587005615235,
      "epoch": 2.122855475588407,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004544540062495016,
      "loss": 5.0716,
      "mean_token_accuracy": 0.1924387976527214,
      "num_tokens": 47329498.0,
      "step": 27285
    },
    {
      "entropy": 5.459635448455811,
      "epoch": 2.1232445049601245,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045443728623696905,
      "loss": 5.1251,
      "mean_token_accuracy": 0.18720201402902603,
      "num_tokens": 47338375.0,
      "step": 27290
    },
    {
      "entropy": 5.359905052185058,
      "epoch": 2.123633534331842,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045442056350176123,
      "loss": 5.0777,
      "mean_token_accuracy": 0.1917021319270134,
      "num_tokens": 47346608.0,
      "step": 27295
    },
    {
      "entropy": 5.472007989883423,
      "epoch": 2.1240225637035595,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000454403838044132,
      "loss": 5.1358,
      "mean_token_accuracy": 0.1864597961306572,
      "num_tokens": 47355375.0,
      "step": 27300
    },
    {
      "entropy": 5.417171621322632,
      "epoch": 2.124411593075277,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000454387109864335,
      "loss": 5.0552,
      "mean_token_accuracy": 0.19496077299118042,
      "num_tokens": 47363138.0,
      "step": 27305
    },
    {
      "entropy": 5.518266725540161,
      "epoch": 2.124800622446995,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045437037896262425,
      "loss": 5.1801,
      "mean_token_accuracy": 0.17953557819128035,
      "num_tokens": 47371431.0,
      "step": 27310
    },
    {
      "entropy": 5.419029664993286,
      "epoch": 2.125189651818712,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004543536453392534,
      "loss": 5.0451,
      "mean_token_accuracy": 0.1967315211892128,
      "num_tokens": 47379739.0,
      "step": 27315
    },
    {
      "entropy": 5.448039960861206,
      "epoch": 2.12557868119043,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045433690899447643,
      "loss": 5.0944,
      "mean_token_accuracy": 0.18771040290594102,
      "num_tokens": 47389240.0,
      "step": 27320
    },
    {
      "entropy": 5.429193830490112,
      "epoch": 2.1259677105621475,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045432016992854734,
      "loss": 4.985,
      "mean_token_accuracy": 0.19559376835823059,
      "num_tokens": 47397666.0,
      "step": 27325
    },
    {
      "entropy": 5.403881788253784,
      "epoch": 2.126356739933865,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004543034281417201,
      "loss": 5.1311,
      "mean_token_accuracy": 0.1848018705844879,
      "num_tokens": 47407419.0,
      "step": 27330
    },
    {
      "entropy": 5.42021803855896,
      "epoch": 2.1267457693055825,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004542866836342488,
      "loss": 5.1238,
      "mean_token_accuracy": 0.18618351072072983,
      "num_tokens": 47416750.0,
      "step": 27335
    },
    {
      "entropy": 5.486727333068847,
      "epoch": 2.1271347986773,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004542699364063873,
      "loss": 5.0872,
      "mean_token_accuracy": 0.19028878808021546,
      "num_tokens": 47425833.0,
      "step": 27340
    },
    {
      "entropy": 5.373021125793457,
      "epoch": 2.127523828049018,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004542531864583899,
      "loss": 5.0213,
      "mean_token_accuracy": 0.18876579254865647,
      "num_tokens": 47434470.0,
      "step": 27345
    },
    {
      "entropy": 5.5226404666900635,
      "epoch": 2.127912857420735,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004542364337905108,
      "loss": 5.0957,
      "mean_token_accuracy": 0.1855452060699463,
      "num_tokens": 47442904.0,
      "step": 27350
    },
    {
      "entropy": 5.492037534713745,
      "epoch": 2.128301886792453,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045421967840300403,
      "loss": 5.1327,
      "mean_token_accuracy": 0.19095717668533324,
      "num_tokens": 47452373.0,
      "step": 27355
    },
    {
      "entropy": 5.402172946929932,
      "epoch": 2.1286909161641705,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004542029202961239,
      "loss": 4.9473,
      "mean_token_accuracy": 0.19280818551778794,
      "num_tokens": 47460504.0,
      "step": 27360
    },
    {
      "entropy": 5.4122096538543705,
      "epoch": 2.1290799455358878,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045418615947012464,
      "loss": 5.0633,
      "mean_token_accuracy": 0.18990051746368408,
      "num_tokens": 47469526.0,
      "step": 27365
    },
    {
      "entropy": 5.409101057052612,
      "epoch": 2.1294689749076054,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004541693959252607,
      "loss": 5.0344,
      "mean_token_accuracy": 0.19243221431970597,
      "num_tokens": 47478241.0,
      "step": 27370
    },
    {
      "entropy": 5.407530307769775,
      "epoch": 2.129858004279323,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045415262966178635,
      "loss": 5.0323,
      "mean_token_accuracy": 0.19911688715219497,
      "num_tokens": 47486956.0,
      "step": 27375
    },
    {
      "entropy": 5.3898216724395756,
      "epoch": 2.130247033651041,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004541358606799559,
      "loss": 4.9619,
      "mean_token_accuracy": 0.20635183602571489,
      "num_tokens": 47494817.0,
      "step": 27380
    },
    {
      "entropy": 5.355669021606445,
      "epoch": 2.130636063022758,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000454119088980024,
      "loss": 5.0647,
      "mean_token_accuracy": 0.18386910259723663,
      "num_tokens": 47504124.0,
      "step": 27385
    },
    {
      "entropy": 5.516399478912353,
      "epoch": 2.131025092394476,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004541023145622449,
      "loss": 5.0609,
      "mean_token_accuracy": 0.18649635165929795,
      "num_tokens": 47513160.0,
      "step": 27390
    },
    {
      "entropy": 5.4230626106262205,
      "epoch": 2.1314141217661935,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045408553742687336,
      "loss": 4.9707,
      "mean_token_accuracy": 0.19775646328926086,
      "num_tokens": 47521298.0,
      "step": 27395
    },
    {
      "entropy": 5.293092393875122,
      "epoch": 2.1318031511379107,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004540687575741638,
      "loss": 5.0,
      "mean_token_accuracy": 0.1913327768445015,
      "num_tokens": 47529516.0,
      "step": 27400
    },
    {
      "entropy": 5.374094772338867,
      "epoch": 2.1321921805096284,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045405197500437077,
      "loss": 5.0333,
      "mean_token_accuracy": 0.18559107333421707,
      "num_tokens": 47538311.0,
      "step": 27405
    },
    {
      "entropy": 5.480267953872681,
      "epoch": 2.132581209881346,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045403518971774915,
      "loss": 5.1555,
      "mean_token_accuracy": 0.18266653269529343,
      "num_tokens": 47547186.0,
      "step": 27410
    },
    {
      "entropy": 5.403508234024048,
      "epoch": 2.132970239253064,
      "grad_norm": 1.25,
      "learning_rate": 0.00045401840171455343,
      "loss": 4.9908,
      "mean_token_accuracy": 0.19451460838317872,
      "num_tokens": 47556173.0,
      "step": 27415
    },
    {
      "entropy": 5.435228538513184,
      "epoch": 2.133359268624781,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004540016109950384,
      "loss": 5.0031,
      "mean_token_accuracy": 0.19427998661994933,
      "num_tokens": 47564595.0,
      "step": 27420
    },
    {
      "entropy": 5.38707857131958,
      "epoch": 2.133748297996499,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004539848175594589,
      "loss": 5.0327,
      "mean_token_accuracy": 0.18725859820842744,
      "num_tokens": 47573559.0,
      "step": 27425
    },
    {
      "entropy": 5.449259519577026,
      "epoch": 2.1341373273682165,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004539680214080695,
      "loss": 4.9788,
      "mean_token_accuracy": 0.19199885427951813,
      "num_tokens": 47582178.0,
      "step": 27430
    },
    {
      "entropy": 5.3611307621002195,
      "epoch": 2.1345263567399337,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045395122254112536,
      "loss": 4.9233,
      "mean_token_accuracy": 0.19895289242267608,
      "num_tokens": 47590738.0,
      "step": 27435
    },
    {
      "entropy": 5.465515899658203,
      "epoch": 2.1349153861116514,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004539344209588812,
      "loss": 5.1783,
      "mean_token_accuracy": 0.17806089371442796,
      "num_tokens": 47599339.0,
      "step": 27440
    },
    {
      "entropy": 5.359873580932617,
      "epoch": 2.135304415483369,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045391761666159204,
      "loss": 5.0164,
      "mean_token_accuracy": 0.19119245558977127,
      "num_tokens": 47608250.0,
      "step": 27445
    },
    {
      "entropy": 5.466415548324585,
      "epoch": 2.1356934448550864,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045390080964951294,
      "loss": 5.0789,
      "mean_token_accuracy": 0.1916109636425972,
      "num_tokens": 47616823.0,
      "step": 27450
    },
    {
      "entropy": 5.466552305221557,
      "epoch": 2.136082474226804,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045388399992289864,
      "loss": 5.1181,
      "mean_token_accuracy": 0.18937683701515198,
      "num_tokens": 47627002.0,
      "step": 27455
    },
    {
      "entropy": 5.581350088119507,
      "epoch": 2.136471503598522,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045386718748200436,
      "loss": 5.178,
      "mean_token_accuracy": 0.18116298466920852,
      "num_tokens": 47635588.0,
      "step": 27460
    },
    {
      "entropy": 5.533190441131592,
      "epoch": 2.136860532970239,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045385037232708534,
      "loss": 5.1392,
      "mean_token_accuracy": 0.19071145355701447,
      "num_tokens": 47645089.0,
      "step": 27465
    },
    {
      "entropy": 5.456801128387451,
      "epoch": 2.1372495623419567,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004538335544583964,
      "loss": 5.1238,
      "mean_token_accuracy": 0.1788894295692444,
      "num_tokens": 47654088.0,
      "step": 27470
    },
    {
      "entropy": 5.474858522415161,
      "epoch": 2.1376385917136744,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045381673387619306,
      "loss": 5.1431,
      "mean_token_accuracy": 0.1856838881969452,
      "num_tokens": 47662367.0,
      "step": 27475
    },
    {
      "entropy": 5.47779107093811,
      "epoch": 2.138027621085392,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045379991058073035,
      "loss": 5.0762,
      "mean_token_accuracy": 0.1819026991724968,
      "num_tokens": 47670197.0,
      "step": 27480
    },
    {
      "entropy": 5.386983633041382,
      "epoch": 2.1384166504571094,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004537830845722636,
      "loss": 5.1166,
      "mean_token_accuracy": 0.19046351164579392,
      "num_tokens": 47678396.0,
      "step": 27485
    },
    {
      "entropy": 5.370035600662232,
      "epoch": 2.138805679828827,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004537662558510481,
      "loss": 4.9871,
      "mean_token_accuracy": 0.1928339794278145,
      "num_tokens": 47686671.0,
      "step": 27490
    },
    {
      "entropy": 5.474032306671143,
      "epoch": 2.1391947092005448,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045374942441733925,
      "loss": 5.08,
      "mean_token_accuracy": 0.18277022540569304,
      "num_tokens": 47694923.0,
      "step": 27495
    },
    {
      "entropy": 5.364909315109253,
      "epoch": 2.139583738572262,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004537325902713923,
      "loss": 5.0362,
      "mean_token_accuracy": 0.19324759840965272,
      "num_tokens": 47702944.0,
      "step": 27500
    },
    {
      "entropy": 5.461622619628907,
      "epoch": 2.1399727679439797,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004537157534134629,
      "loss": 5.093,
      "mean_token_accuracy": 0.20239340513944626,
      "num_tokens": 47711075.0,
      "step": 27505
    },
    {
      "entropy": 5.430045557022095,
      "epoch": 2.1403617973156974,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045369891384380623,
      "loss": 4.9761,
      "mean_token_accuracy": 0.19512924253940583,
      "num_tokens": 47719338.0,
      "step": 27510
    },
    {
      "entropy": 5.480515623092652,
      "epoch": 2.140750826687415,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004536820715626781,
      "loss": 5.1222,
      "mean_token_accuracy": 0.18416400402784347,
      "num_tokens": 47728035.0,
      "step": 27515
    },
    {
      "entropy": 5.4787016868591305,
      "epoch": 2.1411398560591324,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045366522657033385,
      "loss": 5.0818,
      "mean_token_accuracy": 0.1825898125767708,
      "num_tokens": 47736438.0,
      "step": 27520
    },
    {
      "entropy": 5.428909015655518,
      "epoch": 2.14152888543085,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045364837886702924,
      "loss": 5.0501,
      "mean_token_accuracy": 0.19144577533006668,
      "num_tokens": 47744661.0,
      "step": 27525
    },
    {
      "entropy": 5.436733198165894,
      "epoch": 2.1419179148025678,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004536315284530198,
      "loss": 4.9586,
      "mean_token_accuracy": 0.19463801085948945,
      "num_tokens": 47752550.0,
      "step": 27530
    },
    {
      "entropy": 5.430111217498779,
      "epoch": 2.142306944174285,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004536146753285612,
      "loss": 5.0545,
      "mean_token_accuracy": 0.1974322348833084,
      "num_tokens": 47761116.0,
      "step": 27535
    },
    {
      "entropy": 5.472072792053223,
      "epoch": 2.1426959735460027,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004535978194939093,
      "loss": 5.1541,
      "mean_token_accuracy": 0.1904282122850418,
      "num_tokens": 47769804.0,
      "step": 27540
    },
    {
      "entropy": 5.41582236289978,
      "epoch": 2.1430850029177204,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045358096094931966,
      "loss": 5.0132,
      "mean_token_accuracy": 0.19200569838285447,
      "num_tokens": 47777684.0,
      "step": 27545
    },
    {
      "entropy": 5.383411455154419,
      "epoch": 2.1434740322894377,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004535640996950482,
      "loss": 5.0008,
      "mean_token_accuracy": 0.1942657321691513,
      "num_tokens": 47786443.0,
      "step": 27550
    },
    {
      "entropy": 5.438593578338623,
      "epoch": 2.1438630616611554,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004535472357313507,
      "loss": 5.1642,
      "mean_token_accuracy": 0.18987253159284592,
      "num_tokens": 47795408.0,
      "step": 27555
    },
    {
      "entropy": 5.396753978729248,
      "epoch": 2.144252091032873,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045353036905848316,
      "loss": 5.0014,
      "mean_token_accuracy": 0.1942703381180763,
      "num_tokens": 47804108.0,
      "step": 27560
    },
    {
      "entropy": 5.448256731033325,
      "epoch": 2.1446411204045903,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004535134996767014,
      "loss": 5.0605,
      "mean_token_accuracy": 0.19769213497638702,
      "num_tokens": 47811972.0,
      "step": 27565
    },
    {
      "entropy": 5.3705707550048825,
      "epoch": 2.145030149776308,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004534966275862614,
      "loss": 5.1092,
      "mean_token_accuracy": 0.18637319207191466,
      "num_tokens": 47820615.0,
      "step": 27570
    },
    {
      "entropy": 5.462299108505249,
      "epoch": 2.1454191791480257,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004534797527874192,
      "loss": 5.1255,
      "mean_token_accuracy": 0.18862320631742477,
      "num_tokens": 47828773.0,
      "step": 27575
    },
    {
      "entropy": 5.422725343704224,
      "epoch": 2.1458082085197434,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045346287528043083,
      "loss": 5.0298,
      "mean_token_accuracy": 0.18736000806093217,
      "num_tokens": 47837197.0,
      "step": 27580
    },
    {
      "entropy": 5.4021069526672365,
      "epoch": 2.1461972378914607,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045344599506555244,
      "loss": 5.1668,
      "mean_token_accuracy": 0.18078043162822724,
      "num_tokens": 47846127.0,
      "step": 27585
    },
    {
      "entropy": 5.436137723922729,
      "epoch": 2.1465862672631784,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045342911214304,
      "loss": 5.0606,
      "mean_token_accuracy": 0.19557226896286012,
      "num_tokens": 47854832.0,
      "step": 27590
    },
    {
      "entropy": 5.4818675994873045,
      "epoch": 2.146975296634896,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004534122265131498,
      "loss": 5.1008,
      "mean_token_accuracy": 0.194764843583107,
      "num_tokens": 47862621.0,
      "step": 27595
    },
    {
      "entropy": 5.4536529064178465,
      "epoch": 2.1473643260066133,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00045339533817613807,
      "loss": 5.1035,
      "mean_token_accuracy": 0.1872079625725746,
      "num_tokens": 47871406.0,
      "step": 27600
    },
    {
      "entropy": 5.384205102920532,
      "epoch": 2.147753355378331,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045337844713226106,
      "loss": 5.0086,
      "mean_token_accuracy": 0.1961725875735283,
      "num_tokens": 47880361.0,
      "step": 27605
    },
    {
      "entropy": 5.464580917358399,
      "epoch": 2.1481423847500487,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045336155338177497,
      "loss": 5.1101,
      "mean_token_accuracy": 0.18749476671218873,
      "num_tokens": 47889632.0,
      "step": 27610
    },
    {
      "entropy": 5.395323276519775,
      "epoch": 2.1485314141217664,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004533446569249362,
      "loss": 4.9934,
      "mean_token_accuracy": 0.19297121465206146,
      "num_tokens": 47898612.0,
      "step": 27615
    },
    {
      "entropy": 5.385963201522827,
      "epoch": 2.1489204434934837,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00045332775776200116,
      "loss": 5.0866,
      "mean_token_accuracy": 0.18658825904130935,
      "num_tokens": 47906628.0,
      "step": 27620
    },
    {
      "entropy": 5.360697221755982,
      "epoch": 2.1493094728652014,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045331085589322626,
      "loss": 5.096,
      "mean_token_accuracy": 0.19419705271720886,
      "num_tokens": 47915831.0,
      "step": 27625
    },
    {
      "entropy": 5.523052835464478,
      "epoch": 2.149698502236919,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004532939513188679,
      "loss": 5.0888,
      "mean_token_accuracy": 0.19227644801139832,
      "num_tokens": 47924289.0,
      "step": 27630
    },
    {
      "entropy": 5.514377498626709,
      "epoch": 2.1500875316086363,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004532770440391826,
      "loss": 5.157,
      "mean_token_accuracy": 0.18228931874036788,
      "num_tokens": 47932438.0,
      "step": 27635
    },
    {
      "entropy": 5.410264539718628,
      "epoch": 2.150476560980354,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004532601340544269,
      "loss": 5.0279,
      "mean_token_accuracy": 0.18553309738636017,
      "num_tokens": 47940634.0,
      "step": 27640
    },
    {
      "entropy": 5.467669105529785,
      "epoch": 2.1508655903520717,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004532432213648574,
      "loss": 5.1205,
      "mean_token_accuracy": 0.18643853366374968,
      "num_tokens": 47949871.0,
      "step": 27645
    },
    {
      "entropy": 5.403092241287231,
      "epoch": 2.151254619723789,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045322630597073067,
      "loss": 5.0496,
      "mean_token_accuracy": 0.19111949503421782,
      "num_tokens": 47957847.0,
      "step": 27650
    },
    {
      "entropy": 5.3359763622283936,
      "epoch": 2.1516436490955066,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045320938787230355,
      "loss": 5.1035,
      "mean_token_accuracy": 0.1912636250257492,
      "num_tokens": 47965978.0,
      "step": 27655
    },
    {
      "entropy": 5.506854724884033,
      "epoch": 2.1520326784672243,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045319246706983257,
      "loss": 5.2264,
      "mean_token_accuracy": 0.18004737347364425,
      "num_tokens": 47975407.0,
      "step": 27660
    },
    {
      "entropy": 5.443437004089356,
      "epoch": 2.152421707838942,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045317554356357446,
      "loss": 5.0131,
      "mean_token_accuracy": 0.19553641080856324,
      "num_tokens": 47983652.0,
      "step": 27665
    },
    {
      "entropy": 5.466415357589722,
      "epoch": 2.1528107372106593,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045315861735378615,
      "loss": 5.083,
      "mean_token_accuracy": 0.1845807120203972,
      "num_tokens": 47992750.0,
      "step": 27670
    },
    {
      "entropy": 5.346964168548584,
      "epoch": 2.153199766582377,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045314168844072435,
      "loss": 5.0052,
      "mean_token_accuracy": 0.1911673754453659,
      "num_tokens": 48000930.0,
      "step": 27675
    },
    {
      "entropy": 5.397316646575928,
      "epoch": 2.1535887959540947,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045312475682464604,
      "loss": 5.0619,
      "mean_token_accuracy": 0.18944141119718552,
      "num_tokens": 48009285.0,
      "step": 27680
    },
    {
      "entropy": 5.448140430450439,
      "epoch": 2.153977825325812,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045310782250580794,
      "loss": 5.1217,
      "mean_token_accuracy": 0.18256623446941375,
      "num_tokens": 48017806.0,
      "step": 27685
    },
    {
      "entropy": 5.430179119110107,
      "epoch": 2.1543668546975296,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004530908854844672,
      "loss": 5.0969,
      "mean_token_accuracy": 0.1865006670355797,
      "num_tokens": 48026930.0,
      "step": 27690
    },
    {
      "entropy": 5.489858102798462,
      "epoch": 2.1547558840692473,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045307394576088076,
      "loss": 5.1772,
      "mean_token_accuracy": 0.18242247253656388,
      "num_tokens": 48036039.0,
      "step": 27695
    },
    {
      "entropy": 5.4185450077056885,
      "epoch": 2.1551449134409646,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004530570033353056,
      "loss": 4.9781,
      "mean_token_accuracy": 0.20122114568948746,
      "num_tokens": 48045278.0,
      "step": 27700
    },
    {
      "entropy": 5.46020770072937,
      "epoch": 2.1555339428126823,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045304005820799874,
      "loss": 5.1161,
      "mean_token_accuracy": 0.1817747250199318,
      "num_tokens": 48054448.0,
      "step": 27705
    },
    {
      "entropy": 5.475445222854614,
      "epoch": 2.1559229721844,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004530231103792175,
      "loss": 5.122,
      "mean_token_accuracy": 0.18852549493312837,
      "num_tokens": 48062672.0,
      "step": 27710
    },
    {
      "entropy": 5.431478118896484,
      "epoch": 2.1563120015561177,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004530061598492188,
      "loss": 5.0305,
      "mean_token_accuracy": 0.19085882753133773,
      "num_tokens": 48072140.0,
      "step": 27715
    },
    {
      "entropy": 5.482812356948853,
      "epoch": 2.156701030927835,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045298920661826004,
      "loss": 5.1903,
      "mean_token_accuracy": 0.18097154647111893,
      "num_tokens": 48080394.0,
      "step": 27720
    },
    {
      "entropy": 5.460504579544067,
      "epoch": 2.1570900602995526,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004529722506865984,
      "loss": 5.0983,
      "mean_token_accuracy": 0.19175025671720505,
      "num_tokens": 48088947.0,
      "step": 27725
    },
    {
      "entropy": 5.360564470291138,
      "epoch": 2.1574790896712703,
      "grad_norm": 1.171875,
      "learning_rate": 0.000452955292054491,
      "loss": 5.0106,
      "mean_token_accuracy": 0.190871225297451,
      "num_tokens": 48097156.0,
      "step": 27730
    },
    {
      "entropy": 5.400822353363037,
      "epoch": 2.1578681190429876,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00045293833072219535,
      "loss": 5.0355,
      "mean_token_accuracy": 0.19466277807950974,
      "num_tokens": 48105692.0,
      "step": 27735
    },
    {
      "entropy": 5.453074026107788,
      "epoch": 2.1582571484147053,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045292136668996876,
      "loss": 5.047,
      "mean_token_accuracy": 0.1868092894554138,
      "num_tokens": 48114967.0,
      "step": 27740
    },
    {
      "entropy": 5.427117967605591,
      "epoch": 2.158646177786423,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004529043999580686,
      "loss": 5.0436,
      "mean_token_accuracy": 0.19195033013820648,
      "num_tokens": 48123933.0,
      "step": 27745
    },
    {
      "entropy": 5.393619537353516,
      "epoch": 2.1590352071581402,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045288743052675234,
      "loss": 5.0351,
      "mean_token_accuracy": 0.19574755430221558,
      "num_tokens": 48131998.0,
      "step": 27750
    },
    {
      "entropy": 5.4142295837402346,
      "epoch": 2.159424236529858,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045287045839627744,
      "loss": 5.0044,
      "mean_token_accuracy": 0.20303898602724074,
      "num_tokens": 48140111.0,
      "step": 27755
    },
    {
      "entropy": 5.511449193954467,
      "epoch": 2.1598132659015756,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00045285348356690155,
      "loss": 5.0843,
      "mean_token_accuracy": 0.19149483889341354,
      "num_tokens": 48148826.0,
      "step": 27760
    },
    {
      "entropy": 5.459193468093872,
      "epoch": 2.1602022952732933,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000452836506038882,
      "loss": 5.0543,
      "mean_token_accuracy": 0.19005140513181687,
      "num_tokens": 48157242.0,
      "step": 27765
    },
    {
      "entropy": 5.438192081451416,
      "epoch": 2.1605913246450106,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045281952581247654,
      "loss": 5.1845,
      "mean_token_accuracy": 0.17860541641712188,
      "num_tokens": 48166571.0,
      "step": 27770
    },
    {
      "entropy": 5.485002183914185,
      "epoch": 2.1609803540167283,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045280254288794286,
      "loss": 5.0796,
      "mean_token_accuracy": 0.19285960197448732,
      "num_tokens": 48175280.0,
      "step": 27775
    },
    {
      "entropy": 5.52338056564331,
      "epoch": 2.161369383388446,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045278555726553855,
      "loss": 5.1208,
      "mean_token_accuracy": 0.1875110000371933,
      "num_tokens": 48184931.0,
      "step": 27780
    },
    {
      "entropy": 5.338022518157959,
      "epoch": 2.1617584127601632,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045276856894552143,
      "loss": 5.0056,
      "mean_token_accuracy": 0.19658712595701217,
      "num_tokens": 48192945.0,
      "step": 27785
    },
    {
      "entropy": 5.438974857330322,
      "epoch": 2.162147442131881,
      "grad_norm": 1.25,
      "learning_rate": 0.0004527515779281492,
      "loss": 5.117,
      "mean_token_accuracy": 0.18399262577295303,
      "num_tokens": 48201963.0,
      "step": 27790
    },
    {
      "entropy": 5.484705209732056,
      "epoch": 2.1625364715035986,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045273458421367976,
      "loss": 5.0852,
      "mean_token_accuracy": 0.19040601551532746,
      "num_tokens": 48211178.0,
      "step": 27795
    },
    {
      "entropy": 5.431409931182861,
      "epoch": 2.162925500875316,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004527175878023708,
      "loss": 4.9472,
      "mean_token_accuracy": 0.19549288600683212,
      "num_tokens": 48219294.0,
      "step": 27800
    },
    {
      "entropy": 5.357197999954224,
      "epoch": 2.1633145302470336,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004527005886944803,
      "loss": 5.0502,
      "mean_token_accuracy": 0.19317653477191926,
      "num_tokens": 48227782.0,
      "step": 27805
    },
    {
      "entropy": 5.46384711265564,
      "epoch": 2.1637035596187513,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045268358689026626,
      "loss": 5.0779,
      "mean_token_accuracy": 0.18082280158996583,
      "num_tokens": 48236850.0,
      "step": 27810
    },
    {
      "entropy": 5.475611686706543,
      "epoch": 2.164092588990469,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004526665823899866,
      "loss": 5.1384,
      "mean_token_accuracy": 0.1877797469496727,
      "num_tokens": 48246095.0,
      "step": 27815
    },
    {
      "entropy": 5.45708384513855,
      "epoch": 2.164481618362186,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045264957519389936,
      "loss": 5.0694,
      "mean_token_accuracy": 0.1892252594232559,
      "num_tokens": 48254943.0,
      "step": 27820
    },
    {
      "entropy": 5.437220335006714,
      "epoch": 2.164870647733904,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045263256530226253,
      "loss": 5.0129,
      "mean_token_accuracy": 0.1956775367259979,
      "num_tokens": 48263954.0,
      "step": 27825
    },
    {
      "entropy": 5.439289712905884,
      "epoch": 2.1652596771056216,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004526155527153343,
      "loss": 5.1197,
      "mean_token_accuracy": 0.1849600851535797,
      "num_tokens": 48272708.0,
      "step": 27830
    },
    {
      "entropy": 5.4651939392089846,
      "epoch": 2.165648706477339,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004525985374333727,
      "loss": 5.0992,
      "mean_token_accuracy": 0.18519890159368516,
      "num_tokens": 48280782.0,
      "step": 27835
    },
    {
      "entropy": 5.424927711486816,
      "epoch": 2.1660377358490566,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000452581519456636,
      "loss": 4.9683,
      "mean_token_accuracy": 0.19105817526578903,
      "num_tokens": 48289105.0,
      "step": 27840
    },
    {
      "entropy": 5.428668403625489,
      "epoch": 2.1664267652207743,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045256449878538243,
      "loss": 5.0737,
      "mean_token_accuracy": 0.193313392996788,
      "num_tokens": 48297552.0,
      "step": 27845
    },
    {
      "entropy": 5.358900260925293,
      "epoch": 2.166815794592492,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045254747541987017,
      "loss": 5.0532,
      "mean_token_accuracy": 0.19058728367090225,
      "num_tokens": 48307541.0,
      "step": 27850
    },
    {
      "entropy": 5.4484563827514645,
      "epoch": 2.167204823964209,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004525304493603576,
      "loss": 5.0648,
      "mean_token_accuracy": 0.1963243991136551,
      "num_tokens": 48316078.0,
      "step": 27855
    },
    {
      "entropy": 5.4707286834716795,
      "epoch": 2.167593853335927,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045251342060710295,
      "loss": 5.1205,
      "mean_token_accuracy": 0.18670790642499924,
      "num_tokens": 48324221.0,
      "step": 27860
    },
    {
      "entropy": 5.464034175872802,
      "epoch": 2.1679828827076446,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004524963891603647,
      "loss": 5.1963,
      "mean_token_accuracy": 0.18476096540689468,
      "num_tokens": 48333047.0,
      "step": 27865
    },
    {
      "entropy": 5.471977853775025,
      "epoch": 2.168371912079362,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045247935502040136,
      "loss": 5.1441,
      "mean_token_accuracy": 0.18933992236852645,
      "num_tokens": 48342255.0,
      "step": 27870
    },
    {
      "entropy": 5.435747861862183,
      "epoch": 2.1687609414510796,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004524623181874712,
      "loss": 5.1413,
      "mean_token_accuracy": 0.18017890602350234,
      "num_tokens": 48350877.0,
      "step": 27875
    },
    {
      "entropy": 5.564033651351929,
      "epoch": 2.1691499708227973,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004524452786618329,
      "loss": 5.2507,
      "mean_token_accuracy": 0.17731962949037552,
      "num_tokens": 48360537.0,
      "step": 27880
    },
    {
      "entropy": 5.565032911300659,
      "epoch": 2.1695390001945145,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045242823644374484,
      "loss": 5.1578,
      "mean_token_accuracy": 0.18690354377031326,
      "num_tokens": 48370518.0,
      "step": 27885
    },
    {
      "entropy": 5.408589887619018,
      "epoch": 2.169928029566232,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004524111915334658,
      "loss": 4.9778,
      "mean_token_accuracy": 0.20465470999479293,
      "num_tokens": 48379379.0,
      "step": 27890
    },
    {
      "entropy": 5.331674385070801,
      "epoch": 2.17031705893795,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045239414393125424,
      "loss": 4.9693,
      "mean_token_accuracy": 0.19495616853237152,
      "num_tokens": 48388090.0,
      "step": 27895
    },
    {
      "entropy": 5.456219577789307,
      "epoch": 2.170706088309667,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004523770936373689,
      "loss": 5.0926,
      "mean_token_accuracy": 0.18139129132032394,
      "num_tokens": 48396527.0,
      "step": 27900
    },
    {
      "entropy": 5.444386339187622,
      "epoch": 2.171095117681385,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004523600406520685,
      "loss": 5.083,
      "mean_token_accuracy": 0.1933899700641632,
      "num_tokens": 48404632.0,
      "step": 27905
    },
    {
      "entropy": 5.37218074798584,
      "epoch": 2.1714841470531026,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004523429849756118,
      "loss": 4.9919,
      "mean_token_accuracy": 0.2036385640501976,
      "num_tokens": 48413573.0,
      "step": 27910
    },
    {
      "entropy": 5.510362815856934,
      "epoch": 2.1718731764248203,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004523259266082576,
      "loss": 5.1678,
      "mean_token_accuracy": 0.18552769124507903,
      "num_tokens": 48421252.0,
      "step": 27915
    },
    {
      "entropy": 5.453756952285767,
      "epoch": 2.1722622057965375,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045230886555026455,
      "loss": 5.1085,
      "mean_token_accuracy": 0.18361749351024628,
      "num_tokens": 48430322.0,
      "step": 27920
    },
    {
      "entropy": 5.466126108169556,
      "epoch": 2.172651235168255,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045229180180189176,
      "loss": 5.0396,
      "mean_token_accuracy": 0.19721615314483643,
      "num_tokens": 48439570.0,
      "step": 27925
    },
    {
      "entropy": 5.40469765663147,
      "epoch": 2.173040264539973,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045227473536339817,
      "loss": 5.0727,
      "mean_token_accuracy": 0.18553993105888367,
      "num_tokens": 48447803.0,
      "step": 27930
    },
    {
      "entropy": 5.474167013168335,
      "epoch": 2.17342929391169,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004522576662350425,
      "loss": 5.1549,
      "mean_token_accuracy": 0.1882725864648819,
      "num_tokens": 48456694.0,
      "step": 27935
    },
    {
      "entropy": 5.449743461608887,
      "epoch": 2.173818323283408,
      "grad_norm": 1.234375,
      "learning_rate": 0.000452240594417084,
      "loss": 5.126,
      "mean_token_accuracy": 0.18421879410743713,
      "num_tokens": 48465321.0,
      "step": 27940
    },
    {
      "entropy": 5.465271854400635,
      "epoch": 2.1742073526551255,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004522235199097815,
      "loss": 5.1231,
      "mean_token_accuracy": 0.187352155148983,
      "num_tokens": 48474299.0,
      "step": 27945
    },
    {
      "entropy": 5.402745485305786,
      "epoch": 2.1745963820268432,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004522064427133942,
      "loss": 4.9873,
      "mean_token_accuracy": 0.19438674449920654,
      "num_tokens": 48483427.0,
      "step": 27950
    },
    {
      "entropy": 5.390454721450806,
      "epoch": 2.1749854113985605,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045218936282818116,
      "loss": 4.9983,
      "mean_token_accuracy": 0.20087809711694718,
      "num_tokens": 48492610.0,
      "step": 27955
    },
    {
      "entropy": 5.41278829574585,
      "epoch": 2.175374440770278,
      "grad_norm": 1.125,
      "learning_rate": 0.0004521722802544016,
      "loss": 5.2019,
      "mean_token_accuracy": 0.18960207849740982,
      "num_tokens": 48502531.0,
      "step": 27960
    },
    {
      "entropy": 5.511424541473389,
      "epoch": 2.175763470141996,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045215519499231465,
      "loss": 5.0973,
      "mean_token_accuracy": 0.19403574168682097,
      "num_tokens": 48511167.0,
      "step": 27965
    },
    {
      "entropy": 5.4764515399932865,
      "epoch": 2.176152499513713,
      "grad_norm": 1.328125,
      "learning_rate": 0.00045213810704217963,
      "loss": 5.1409,
      "mean_token_accuracy": 0.18980092257261277,
      "num_tokens": 48520355.0,
      "step": 27970
    },
    {
      "entropy": 5.416035413742065,
      "epoch": 2.176541528885431,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004521210164042557,
      "loss": 5.1243,
      "mean_token_accuracy": 0.18617822825908661,
      "num_tokens": 48529269.0,
      "step": 27975
    },
    {
      "entropy": 5.478228807449341,
      "epoch": 2.1769305582571485,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004521039230788024,
      "loss": 5.0341,
      "mean_token_accuracy": 0.1973237156867981,
      "num_tokens": 48537764.0,
      "step": 27980
    },
    {
      "entropy": 5.390616607666016,
      "epoch": 2.177319587628866,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004520868270660789,
      "loss": 5.0744,
      "mean_token_accuracy": 0.18631513863801957,
      "num_tokens": 48547168.0,
      "step": 27985
    },
    {
      "entropy": 5.362071084976196,
      "epoch": 2.1777086170005835,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004520697283663446,
      "loss": 5.0106,
      "mean_token_accuracy": 0.1936831995844841,
      "num_tokens": 48555809.0,
      "step": 27990
    },
    {
      "entropy": 5.518894243240356,
      "epoch": 2.178097646372301,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045205262697985897,
      "loss": 5.1004,
      "mean_token_accuracy": 0.184615258872509,
      "num_tokens": 48564754.0,
      "step": 27995
    },
    {
      "entropy": 5.381280851364136,
      "epoch": 2.1784866757440184,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004520355229068816,
      "loss": 5.0092,
      "mean_token_accuracy": 0.19297883659601212,
      "num_tokens": 48573732.0,
      "step": 28000
    },
    {
      "entropy": 5.35177206993103,
      "epoch": 2.178875705115736,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045201841614767203,
      "loss": 4.9911,
      "mean_token_accuracy": 0.1921868994832039,
      "num_tokens": 48581676.0,
      "step": 28005
    },
    {
      "entropy": 5.395832204818726,
      "epoch": 2.179264734487454,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045200130670248974,
      "loss": 5.0634,
      "mean_token_accuracy": 0.18696290254592896,
      "num_tokens": 48590310.0,
      "step": 28010
    },
    {
      "entropy": 5.41760287284851,
      "epoch": 2.1796537638591715,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045198419457159424,
      "loss": 5.0558,
      "mean_token_accuracy": 0.194281929731369,
      "num_tokens": 48599306.0,
      "step": 28015
    },
    {
      "entropy": 5.512250185012817,
      "epoch": 2.180042793230889,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004519670797552453,
      "loss": 5.1716,
      "mean_token_accuracy": 0.18363481909036636,
      "num_tokens": 48608264.0,
      "step": 28020
    },
    {
      "entropy": 5.461954069137573,
      "epoch": 2.1804318226026065,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045194996225370266,
      "loss": 5.08,
      "mean_token_accuracy": 0.19424430131912232,
      "num_tokens": 48617102.0,
      "step": 28025
    },
    {
      "entropy": 5.5104724884033205,
      "epoch": 2.180820851974324,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045193284206722595,
      "loss": 5.1614,
      "mean_token_accuracy": 0.18416298031806946,
      "num_tokens": 48626207.0,
      "step": 28030
    },
    {
      "entropy": 5.4618144035339355,
      "epoch": 2.1812098813460414,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004519157191960749,
      "loss": 5.008,
      "mean_token_accuracy": 0.18832229673862458,
      "num_tokens": 48634322.0,
      "step": 28035
    },
    {
      "entropy": 5.381257724761963,
      "epoch": 2.181598910717759,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004518985936405096,
      "loss": 5.0378,
      "mean_token_accuracy": 0.19622325003147126,
      "num_tokens": 48642026.0,
      "step": 28040
    },
    {
      "entropy": 5.3979002952575685,
      "epoch": 2.181987940089477,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004518814654007896,
      "loss": 4.9549,
      "mean_token_accuracy": 0.19703333377838134,
      "num_tokens": 48650114.0,
      "step": 28045
    },
    {
      "entropy": 5.491523885726929,
      "epoch": 2.1823769694611945,
      "grad_norm": 1.25,
      "learning_rate": 0.00045186433447717486,
      "loss": 5.114,
      "mean_token_accuracy": 0.1812614843249321,
      "num_tokens": 48660185.0,
      "step": 28050
    },
    {
      "entropy": 5.470400094985962,
      "epoch": 2.182765998832912,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045184720086992536,
      "loss": 5.1207,
      "mean_token_accuracy": 0.1929317981004715,
      "num_tokens": 48668603.0,
      "step": 28055
    },
    {
      "entropy": 5.443241262435913,
      "epoch": 2.1831550282046295,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045183006457930107,
      "loss": 5.1027,
      "mean_token_accuracy": 0.19116961807012559,
      "num_tokens": 48676962.0,
      "step": 28060
    },
    {
      "entropy": 5.422074317932129,
      "epoch": 2.183544057576347,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004518129256055621,
      "loss": 5.0979,
      "mean_token_accuracy": 0.19456173330545426,
      "num_tokens": 48685823.0,
      "step": 28065
    },
    {
      "entropy": 5.415818929672241,
      "epoch": 2.1839330869480644,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004517957839489683,
      "loss": 4.9975,
      "mean_token_accuracy": 0.1962646797299385,
      "num_tokens": 48695116.0,
      "step": 28070
    },
    {
      "entropy": 5.349808883666992,
      "epoch": 2.184322116319782,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045177863960977986,
      "loss": 5.0531,
      "mean_token_accuracy": 0.1934889853000641,
      "num_tokens": 48702725.0,
      "step": 28075
    },
    {
      "entropy": 5.391039180755615,
      "epoch": 2.1847111456915,
      "grad_norm": 1.265625,
      "learning_rate": 0.000451761492588257,
      "loss": 5.0769,
      "mean_token_accuracy": 0.19466860592365265,
      "num_tokens": 48711104.0,
      "step": 28080
    },
    {
      "entropy": 5.42861533164978,
      "epoch": 2.185100175063217,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045174434288465986,
      "loss": 5.0308,
      "mean_token_accuracy": 0.19239652305841445,
      "num_tokens": 48720177.0,
      "step": 28085
    },
    {
      "entropy": 5.363170957565307,
      "epoch": 2.1854892044349348,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004517271904992486,
      "loss": 4.9788,
      "mean_token_accuracy": 0.19360375553369522,
      "num_tokens": 48728729.0,
      "step": 28090
    },
    {
      "entropy": 5.346538162231445,
      "epoch": 2.1858782338066525,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004517100354322835,
      "loss": 4.9866,
      "mean_token_accuracy": 0.1936441034078598,
      "num_tokens": 48736876.0,
      "step": 28095
    },
    {
      "entropy": 5.350077342987061,
      "epoch": 2.18626726317837,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045169287768402497,
      "loss": 4.9866,
      "mean_token_accuracy": 0.193696229159832,
      "num_tokens": 48745957.0,
      "step": 28100
    },
    {
      "entropy": 5.475774240493775,
      "epoch": 2.1866562925500874,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004516757172547331,
      "loss": 5.1631,
      "mean_token_accuracy": 0.18738242834806443,
      "num_tokens": 48754488.0,
      "step": 28105
    },
    {
      "entropy": 5.442249965667725,
      "epoch": 2.187045321921805,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004516585541446685,
      "loss": 5.0483,
      "mean_token_accuracy": 0.19367127269506454,
      "num_tokens": 48762797.0,
      "step": 28110
    },
    {
      "entropy": 5.350644779205322,
      "epoch": 2.187434351293523,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045164138835409153,
      "loss": 5.0792,
      "mean_token_accuracy": 0.19054704010486603,
      "num_tokens": 48771639.0,
      "step": 28115
    },
    {
      "entropy": 5.383282136917114,
      "epoch": 2.18782338066524,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045162421988326266,
      "loss": 5.0733,
      "mean_token_accuracy": 0.19311015903949738,
      "num_tokens": 48779978.0,
      "step": 28120
    },
    {
      "entropy": 5.425003433227539,
      "epoch": 2.1882124100369578,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004516070487324424,
      "loss": 5.0928,
      "mean_token_accuracy": 0.1860708624124527,
      "num_tokens": 48788202.0,
      "step": 28125
    },
    {
      "entropy": 5.4426717281341555,
      "epoch": 2.1886014394086755,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045158987490189124,
      "loss": 5.0739,
      "mean_token_accuracy": 0.18856469243764878,
      "num_tokens": 48796533.0,
      "step": 28130
    },
    {
      "entropy": 5.482371520996094,
      "epoch": 2.1889904687803927,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004515726983918697,
      "loss": 5.1296,
      "mean_token_accuracy": 0.1837695598602295,
      "num_tokens": 48805627.0,
      "step": 28135
    },
    {
      "entropy": 5.457989692687988,
      "epoch": 2.1893794981521104,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045155551920263864,
      "loss": 5.1042,
      "mean_token_accuracy": 0.18310075402259826,
      "num_tokens": 48814132.0,
      "step": 28140
    },
    {
      "entropy": 5.419422626495361,
      "epoch": 2.189768527523828,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045153833733445857,
      "loss": 5.0242,
      "mean_token_accuracy": 0.18546267002820968,
      "num_tokens": 48823210.0,
      "step": 28145
    },
    {
      "entropy": 5.368236017227173,
      "epoch": 2.190157556895546,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004515211527875903,
      "loss": 5.0761,
      "mean_token_accuracy": 0.19146686941385269,
      "num_tokens": 48831835.0,
      "step": 28150
    },
    {
      "entropy": 5.441962671279907,
      "epoch": 2.190546586267263,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004515039655622943,
      "loss": 5.1303,
      "mean_token_accuracy": 0.18572519421577455,
      "num_tokens": 48840917.0,
      "step": 28155
    },
    {
      "entropy": 5.452027940750122,
      "epoch": 2.1909356156389808,
      "grad_norm": 1.25,
      "learning_rate": 0.00045148677565883167,
      "loss": 5.0772,
      "mean_token_accuracy": 0.18325790613889695,
      "num_tokens": 48849328.0,
      "step": 28160
    },
    {
      "entropy": 5.4154456615447994,
      "epoch": 2.1913246450106985,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004514695830774631,
      "loss": 5.0645,
      "mean_token_accuracy": 0.19069142788648605,
      "num_tokens": 48858511.0,
      "step": 28165
    },
    {
      "entropy": 5.451482534408569,
      "epoch": 2.1917136743824157,
      "grad_norm": 1.4375,
      "learning_rate": 0.00045145238781844953,
      "loss": 5.131,
      "mean_token_accuracy": 0.19776459634304047,
      "num_tokens": 48867152.0,
      "step": 28170
    },
    {
      "entropy": 5.425117921829224,
      "epoch": 2.1921027037541334,
      "grad_norm": 1.25,
      "learning_rate": 0.00045143518988205184,
      "loss": 5.0355,
      "mean_token_accuracy": 0.18443744778633117,
      "num_tokens": 48875233.0,
      "step": 28175
    },
    {
      "entropy": 5.373607349395752,
      "epoch": 2.192491733125851,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045141798926853094,
      "loss": 5.0893,
      "mean_token_accuracy": 0.18793631196022034,
      "num_tokens": 48884569.0,
      "step": 28180
    },
    {
      "entropy": 5.496940612792969,
      "epoch": 2.1928807624975684,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004514007859781478,
      "loss": 5.138,
      "mean_token_accuracy": 0.18603332340717316,
      "num_tokens": 48894402.0,
      "step": 28185
    },
    {
      "entropy": 5.5778134822845455,
      "epoch": 2.193269791869286,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004513835800111635,
      "loss": 5.0586,
      "mean_token_accuracy": 0.1922069936990738,
      "num_tokens": 48902105.0,
      "step": 28190
    },
    {
      "entropy": 5.427403831481934,
      "epoch": 2.1936588212410038,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004513663713678392,
      "loss": 5.1712,
      "mean_token_accuracy": 0.1806142359972,
      "num_tokens": 48911650.0,
      "step": 28195
    },
    {
      "entropy": 5.4093098640441895,
      "epoch": 2.1940478506127215,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00045134916004843596,
      "loss": 5.037,
      "mean_token_accuracy": 0.19125207364559174,
      "num_tokens": 48920035.0,
      "step": 28200
    },
    {
      "entropy": 5.4056295394897464,
      "epoch": 2.1944368799844387,
      "grad_norm": 1.125,
      "learning_rate": 0.0004513319460532148,
      "loss": 5.0286,
      "mean_token_accuracy": 0.19497465193271638,
      "num_tokens": 48928460.0,
      "step": 28205
    },
    {
      "entropy": 5.4268535614013675,
      "epoch": 2.1948259093561564,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045131472938243707,
      "loss": 5.0615,
      "mean_token_accuracy": 0.1875137984752655,
      "num_tokens": 48937606.0,
      "step": 28210
    },
    {
      "entropy": 5.376762104034424,
      "epoch": 2.195214938727874,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004512975100363639,
      "loss": 4.9684,
      "mean_token_accuracy": 0.19636159092187883,
      "num_tokens": 48945782.0,
      "step": 28215
    },
    {
      "entropy": 5.440301895141602,
      "epoch": 2.1956039680995914,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045128028801525675,
      "loss": 5.1213,
      "mean_token_accuracy": 0.19429844617843628,
      "num_tokens": 48954243.0,
      "step": 28220
    },
    {
      "entropy": 5.470314168930054,
      "epoch": 2.195992997471309,
      "grad_norm": 1.328125,
      "learning_rate": 0.00045126306331937674,
      "loss": 5.0536,
      "mean_token_accuracy": 0.1940629467368126,
      "num_tokens": 48962578.0,
      "step": 28225
    },
    {
      "entropy": 5.402642869949341,
      "epoch": 2.1963820268430267,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004512458359489853,
      "loss": 5.0628,
      "mean_token_accuracy": 0.19591874182224273,
      "num_tokens": 48971237.0,
      "step": 28230
    },
    {
      "entropy": 5.400737762451172,
      "epoch": 2.196771056214744,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004512286059043437,
      "loss": 5.0121,
      "mean_token_accuracy": 0.19429802894592285,
      "num_tokens": 48979954.0,
      "step": 28235
    },
    {
      "entropy": 5.454825687408447,
      "epoch": 2.1971600855864617,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045121137318571364,
      "loss": 5.1496,
      "mean_token_accuracy": 0.18409615457057954,
      "num_tokens": 48989174.0,
      "step": 28240
    },
    {
      "entropy": 5.43331995010376,
      "epoch": 2.1975491149581794,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045119413779335645,
      "loss": 5.0816,
      "mean_token_accuracy": 0.18207580596208572,
      "num_tokens": 48998341.0,
      "step": 28245
    },
    {
      "entropy": 5.372051811218261,
      "epoch": 2.197938144329897,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004511768997275336,
      "loss": 5.0181,
      "mean_token_accuracy": 0.19545132368803025,
      "num_tokens": 49007024.0,
      "step": 28250
    },
    {
      "entropy": 5.370613479614258,
      "epoch": 2.1983271737016143,
      "grad_norm": 1.25,
      "learning_rate": 0.0004511596589885068,
      "loss": 5.0834,
      "mean_token_accuracy": 0.18728319257497789,
      "num_tokens": 49015632.0,
      "step": 28255
    },
    {
      "entropy": 5.44011697769165,
      "epoch": 2.198716203073332,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004511424155765375,
      "loss": 5.0817,
      "mean_token_accuracy": 0.19601324647665025,
      "num_tokens": 49024056.0,
      "step": 28260
    },
    {
      "entropy": 5.38209080696106,
      "epoch": 2.1991052324450497,
      "grad_norm": 1.5,
      "learning_rate": 0.00045112516949188735,
      "loss": 5.0247,
      "mean_token_accuracy": 0.19042078107595445,
      "num_tokens": 49033095.0,
      "step": 28265
    },
    {
      "entropy": 5.378571844100952,
      "epoch": 2.199494261816767,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004511079207348182,
      "loss": 5.0163,
      "mean_token_accuracy": 0.19554224312305452,
      "num_tokens": 49041632.0,
      "step": 28270
    },
    {
      "entropy": 5.450714015960694,
      "epoch": 2.1998832911884847,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004510906693055916,
      "loss": 5.1248,
      "mean_token_accuracy": 0.18695131242275237,
      "num_tokens": 49050336.0,
      "step": 28275
    },
    {
      "entropy": 5.407855749130249,
      "epoch": 2.2002723205602024,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004510734152044694,
      "loss": 5.0327,
      "mean_token_accuracy": 0.19591775834560393,
      "num_tokens": 49058567.0,
      "step": 28280
    },
    {
      "entropy": 5.377872467041016,
      "epoch": 2.20066134993192,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045105615843171336,
      "loss": 4.9006,
      "mean_token_accuracy": 0.20400691777467728,
      "num_tokens": 49067003.0,
      "step": 28285
    },
    {
      "entropy": 5.407643508911133,
      "epoch": 2.2010503793036373,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045103889898758534,
      "loss": 5.0529,
      "mean_token_accuracy": 0.18717027008533477,
      "num_tokens": 49074965.0,
      "step": 28290
    },
    {
      "entropy": 5.401415538787842,
      "epoch": 2.201439408675355,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004510216368723472,
      "loss": 5.0604,
      "mean_token_accuracy": 0.19194337129592895,
      "num_tokens": 49085106.0,
      "step": 28295
    },
    {
      "entropy": 5.443338251113891,
      "epoch": 2.2018284380470727,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045100437208626086,
      "loss": 5.1253,
      "mean_token_accuracy": 0.18340777456760407,
      "num_tokens": 49093125.0,
      "step": 28300
    },
    {
      "entropy": 5.4146322250366214,
      "epoch": 2.20221746741879,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045098710462958844,
      "loss": 5.0288,
      "mean_token_accuracy": 0.19164495468139647,
      "num_tokens": 49101956.0,
      "step": 28305
    },
    {
      "entropy": 5.42751407623291,
      "epoch": 2.2026064967905077,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045096983450259166,
      "loss": 5.0988,
      "mean_token_accuracy": 0.1869824394583702,
      "num_tokens": 49110942.0,
      "step": 28310
    },
    {
      "entropy": 5.481998062133789,
      "epoch": 2.2029955261622254,
      "grad_norm": 1.25,
      "learning_rate": 0.00045095256170553285,
      "loss": 5.1297,
      "mean_token_accuracy": 0.18799692690372466,
      "num_tokens": 49119181.0,
      "step": 28315
    },
    {
      "entropy": 5.3771728515625,
      "epoch": 2.2033845555339426,
      "grad_norm": 1.25,
      "learning_rate": 0.00045093528623867384,
      "loss": 5.0198,
      "mean_token_accuracy": 0.19963831305503846,
      "num_tokens": 49127451.0,
      "step": 28320
    },
    {
      "entropy": 5.448152685165406,
      "epoch": 2.2037735849056603,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045091800810227706,
      "loss": 5.1034,
      "mean_token_accuracy": 0.18977164775133132,
      "num_tokens": 49135550.0,
      "step": 28325
    },
    {
      "entropy": 5.3321741104125975,
      "epoch": 2.204162614277378,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004509007272966044,
      "loss": 4.8991,
      "mean_token_accuracy": 0.19848773926496505,
      "num_tokens": 49144551.0,
      "step": 28330
    },
    {
      "entropy": 5.419197082519531,
      "epoch": 2.2045516436490953,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004508834438219182,
      "loss": 5.0417,
      "mean_token_accuracy": 0.19688933491706848,
      "num_tokens": 49153079.0,
      "step": 28335
    },
    {
      "entropy": 5.38760290145874,
      "epoch": 2.204940673020813,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004508661576784807,
      "loss": 5.0919,
      "mean_token_accuracy": 0.19122634530067445,
      "num_tokens": 49161253.0,
      "step": 28340
    },
    {
      "entropy": 5.372367525100708,
      "epoch": 2.2053297023925307,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004508488688665541,
      "loss": 5.0188,
      "mean_token_accuracy": 0.1935473635792732,
      "num_tokens": 49170192.0,
      "step": 28345
    },
    {
      "entropy": 5.493615961074829,
      "epoch": 2.2057187317642484,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004508315773864009,
      "loss": 5.0925,
      "mean_token_accuracy": 0.19202068746089934,
      "num_tokens": 49179512.0,
      "step": 28350
    },
    {
      "entropy": 5.409499740600586,
      "epoch": 2.2061077611359656,
      "grad_norm": 1.125,
      "learning_rate": 0.00045081428323828337,
      "loss": 4.9658,
      "mean_token_accuracy": 0.20301949232816696,
      "num_tokens": 49188431.0,
      "step": 28355
    },
    {
      "entropy": 5.354013252258301,
      "epoch": 2.2064967905076833,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004507969864224639,
      "loss": 5.0229,
      "mean_token_accuracy": 0.193155437707901,
      "num_tokens": 49196852.0,
      "step": 28360
    },
    {
      "entropy": 5.41056079864502,
      "epoch": 2.206885819879401,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000450779686939205,
      "loss": 5.071,
      "mean_token_accuracy": 0.19184329956769944,
      "num_tokens": 49206007.0,
      "step": 28365
    },
    {
      "entropy": 5.4016162872314455,
      "epoch": 2.2072748492511183,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045076238478876916,
      "loss": 5.0128,
      "mean_token_accuracy": 0.19359312057495118,
      "num_tokens": 49214016.0,
      "step": 28370
    },
    {
      "entropy": 5.369148063659668,
      "epoch": 2.207663878622836,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004507450799714188,
      "loss": 5.0214,
      "mean_token_accuracy": 0.19977216571569442,
      "num_tokens": 49222558.0,
      "step": 28375
    },
    {
      "entropy": 5.449032592773437,
      "epoch": 2.2080529079945537,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045072777248741664,
      "loss": 5.059,
      "mean_token_accuracy": 0.18759192675352096,
      "num_tokens": 49230983.0,
      "step": 28380
    },
    {
      "entropy": 5.468269968032837,
      "epoch": 2.2084419373662714,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004507104623370252,
      "loss": 5.1339,
      "mean_token_accuracy": 0.18264946192502976,
      "num_tokens": 49239429.0,
      "step": 28385
    },
    {
      "entropy": 5.4634058475494385,
      "epoch": 2.2088309667379886,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004506931495205072,
      "loss": 5.0516,
      "mean_token_accuracy": 0.19091873168945311,
      "num_tokens": 49247316.0,
      "step": 28390
    },
    {
      "entropy": 5.417049789428711,
      "epoch": 2.2092199961097063,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004506758340381253,
      "loss": 5.0182,
      "mean_token_accuracy": 0.19499534070491792,
      "num_tokens": 49256293.0,
      "step": 28395
    },
    {
      "entropy": 5.502082347869873,
      "epoch": 2.209609025481424,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004506585158901422,
      "loss": 5.1269,
      "mean_token_accuracy": 0.19277836382389069,
      "num_tokens": 49264735.0,
      "step": 28400
    },
    {
      "entropy": 5.471552181243896,
      "epoch": 2.2099980548531413,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004506411950768207,
      "loss": 5.2091,
      "mean_token_accuracy": 0.17960940003395082,
      "num_tokens": 49273370.0,
      "step": 28405
    },
    {
      "entropy": 5.455255222320557,
      "epoch": 2.210387084224859,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045062387159842363,
      "loss": 5.1606,
      "mean_token_accuracy": 0.18633482605218887,
      "num_tokens": 49282850.0,
      "step": 28410
    },
    {
      "entropy": 5.438771152496338,
      "epoch": 2.2107761135965767,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045060654545521386,
      "loss": 5.0558,
      "mean_token_accuracy": 0.19522936046123504,
      "num_tokens": 49290546.0,
      "step": 28415
    },
    {
      "entropy": 5.381656265258789,
      "epoch": 2.211165142968294,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045058921664745436,
      "loss": 5.0716,
      "mean_token_accuracy": 0.1923053815960884,
      "num_tokens": 49298648.0,
      "step": 28420
    },
    {
      "entropy": 5.404882574081421,
      "epoch": 2.2115541723400116,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045057188517540783,
      "loss": 5.0737,
      "mean_token_accuracy": 0.19062569588422776,
      "num_tokens": 49307134.0,
      "step": 28425
    },
    {
      "entropy": 5.447901821136474,
      "epoch": 2.2119432017117293,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004505545510393374,
      "loss": 5.0586,
      "mean_token_accuracy": 0.18720521926879882,
      "num_tokens": 49315617.0,
      "step": 28430
    },
    {
      "entropy": 5.409620571136474,
      "epoch": 2.2123322310834466,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045053721423950617,
      "loss": 5.0896,
      "mean_token_accuracy": 0.18790622651576996,
      "num_tokens": 49324435.0,
      "step": 28435
    },
    {
      "entropy": 5.339232730865478,
      "epoch": 2.2127212604551643,
      "grad_norm": 1.125,
      "learning_rate": 0.0004505198747761771,
      "loss": 5.0046,
      "mean_token_accuracy": 0.19634063243865968,
      "num_tokens": 49333758.0,
      "step": 28440
    },
    {
      "entropy": 5.372382593154907,
      "epoch": 2.213110289826882,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004505025326496132,
      "loss": 4.9331,
      "mean_token_accuracy": 0.19533218890428544,
      "num_tokens": 49342643.0,
      "step": 28445
    },
    {
      "entropy": 5.405748128890991,
      "epoch": 2.2134993191985997,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045048518786007784,
      "loss": 5.0402,
      "mean_token_accuracy": 0.20078994929790497,
      "num_tokens": 49351009.0,
      "step": 28450
    },
    {
      "entropy": 5.412637042999267,
      "epoch": 2.213888348570317,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045046784040783395,
      "loss": 5.0826,
      "mean_token_accuracy": 0.19002192169427873,
      "num_tokens": 49359415.0,
      "step": 28455
    },
    {
      "entropy": 5.427623462677002,
      "epoch": 2.2142773779420346,
      "grad_norm": 1.21875,
      "learning_rate": 0.000450450490293145,
      "loss": 5.1282,
      "mean_token_accuracy": 0.18631070405244826,
      "num_tokens": 49367898.0,
      "step": 28460
    },
    {
      "entropy": 5.459657192230225,
      "epoch": 2.2146664073137523,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000450433137516274,
      "loss": 5.0226,
      "mean_token_accuracy": 0.19263877719640732,
      "num_tokens": 49376651.0,
      "step": 28465
    },
    {
      "entropy": 5.346524047851562,
      "epoch": 2.2150554366854696,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004504157820774844,
      "loss": 4.9978,
      "mean_token_accuracy": 0.19707806408405304,
      "num_tokens": 49385259.0,
      "step": 28470
    },
    {
      "entropy": 5.432752656936645,
      "epoch": 2.2154444660571873,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004503984239770396,
      "loss": 5.034,
      "mean_token_accuracy": 0.19362403750419616,
      "num_tokens": 49393868.0,
      "step": 28475
    },
    {
      "entropy": 5.513563251495361,
      "epoch": 2.215833495428905,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004503810632152028,
      "loss": 5.0747,
      "mean_token_accuracy": 0.194164177775383,
      "num_tokens": 49402062.0,
      "step": 28480
    },
    {
      "entropy": 5.5048255443573,
      "epoch": 2.2162225248006227,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004503636997922375,
      "loss": 5.1958,
      "mean_token_accuracy": 0.18427686840295793,
      "num_tokens": 49411231.0,
      "step": 28485
    },
    {
      "entropy": 5.453837156295776,
      "epoch": 2.21661155417234,
      "grad_norm": 1.25,
      "learning_rate": 0.0004503463337084072,
      "loss": 5.1438,
      "mean_token_accuracy": 0.18478370010852813,
      "num_tokens": 49419410.0,
      "step": 28490
    },
    {
      "entropy": 5.46797924041748,
      "epoch": 2.2170005835440576,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004503289649639754,
      "loss": 5.0016,
      "mean_token_accuracy": 0.19264817535877227,
      "num_tokens": 49427365.0,
      "step": 28495
    },
    {
      "entropy": 5.424461507797242,
      "epoch": 2.2173896129157753,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004503115935592056,
      "loss": 5.0423,
      "mean_token_accuracy": 0.18813115507364273,
      "num_tokens": 49436395.0,
      "step": 28500
    },
    {
      "entropy": 5.4797711849212645,
      "epoch": 2.2177786422874926,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004502942194943614,
      "loss": 5.1259,
      "mean_token_accuracy": 0.18891520500183107,
      "num_tokens": 49446021.0,
      "step": 28505
    },
    {
      "entropy": 5.482440614700318,
      "epoch": 2.2181676716592102,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045027684276970645,
      "loss": 5.1841,
      "mean_token_accuracy": 0.17942865788936616,
      "num_tokens": 49454306.0,
      "step": 28510
    },
    {
      "entropy": 5.431322050094605,
      "epoch": 2.218556701030928,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045025946338550434,
      "loss": 5.0751,
      "mean_token_accuracy": 0.1899477317929268,
      "num_tokens": 49462735.0,
      "step": 28515
    },
    {
      "entropy": 5.456733369827271,
      "epoch": 2.218945730402645,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004502420813420189,
      "loss": 5.1175,
      "mean_token_accuracy": 0.1880837544798851,
      "num_tokens": 49471156.0,
      "step": 28520
    },
    {
      "entropy": 5.468562746047974,
      "epoch": 2.219334759774363,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004502246966395137,
      "loss": 5.1296,
      "mean_token_accuracy": 0.18950305134058,
      "num_tokens": 49479505.0,
      "step": 28525
    },
    {
      "entropy": 5.457633638381958,
      "epoch": 2.2197237891460806,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045020730927825274,
      "loss": 5.1434,
      "mean_token_accuracy": 0.18986338824033738,
      "num_tokens": 49487879.0,
      "step": 28530
    },
    {
      "entropy": 5.421398019790649,
      "epoch": 2.2201128185177983,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004501899192584997,
      "loss": 4.9773,
      "mean_token_accuracy": 0.19717420041561126,
      "num_tokens": 49496321.0,
      "step": 28535
    },
    {
      "entropy": 5.547014236450195,
      "epoch": 2.2205018478895155,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004501725265805185,
      "loss": 5.3122,
      "mean_token_accuracy": 0.17927169501781465,
      "num_tokens": 49504955.0,
      "step": 28540
    },
    {
      "entropy": 5.416229104995727,
      "epoch": 2.2208908772612332,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045015513124457303,
      "loss": 4.9933,
      "mean_token_accuracy": 0.1946098431944847,
      "num_tokens": 49513678.0,
      "step": 28545
    },
    {
      "entropy": 5.388616132736206,
      "epoch": 2.221279906632951,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004501377332509272,
      "loss": 5.0373,
      "mean_token_accuracy": 0.1920957610011101,
      "num_tokens": 49522315.0,
      "step": 28550
    },
    {
      "entropy": 5.430415630340576,
      "epoch": 2.221668936004668,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045012033259984494,
      "loss": 5.1308,
      "mean_token_accuracy": 0.18593131452798844,
      "num_tokens": 49531077.0,
      "step": 28555
    },
    {
      "entropy": 5.5044420719146725,
      "epoch": 2.222057965376386,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004501029292915905,
      "loss": 5.0978,
      "mean_token_accuracy": 0.18727333545684816,
      "num_tokens": 49540327.0,
      "step": 28560
    },
    {
      "entropy": 5.407759284973144,
      "epoch": 2.2224469947481036,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045008552332642774,
      "loss": 5.0943,
      "mean_token_accuracy": 0.18644336014986038,
      "num_tokens": 49549440.0,
      "step": 28565
    },
    {
      "entropy": 5.457917594909668,
      "epoch": 2.222836024119821,
      "grad_norm": 1.328125,
      "learning_rate": 0.00045006811470462083,
      "loss": 5.0699,
      "mean_token_accuracy": 0.19020809531211852,
      "num_tokens": 49558066.0,
      "step": 28570
    },
    {
      "entropy": 5.500832271575928,
      "epoch": 2.2232250534915385,
      "grad_norm": 1.21875,
      "learning_rate": 0.000450050703426434,
      "loss": 5.0361,
      "mean_token_accuracy": 0.20039083808660507,
      "num_tokens": 49566970.0,
      "step": 28575
    },
    {
      "entropy": 5.30230860710144,
      "epoch": 2.2236140828632562,
      "grad_norm": 1.25,
      "learning_rate": 0.0004500332894921313,
      "loss": 4.9489,
      "mean_token_accuracy": 0.19905491471290587,
      "num_tokens": 49575524.0,
      "step": 28580
    },
    {
      "entropy": 5.485428524017334,
      "epoch": 2.224003112234974,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004500158729019771,
      "loss": 5.2087,
      "mean_token_accuracy": 0.17952203750610352,
      "num_tokens": 49584040.0,
      "step": 28585
    },
    {
      "entropy": 5.540430545806885,
      "epoch": 2.224392141606691,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004499984536562355,
      "loss": 5.1361,
      "mean_token_accuracy": 0.18530689626932145,
      "num_tokens": 49592991.0,
      "step": 28590
    },
    {
      "entropy": 5.420074081420898,
      "epoch": 2.224781170978409,
      "grad_norm": 1.21875,
      "learning_rate": 0.000449981031755171,
      "loss": 5.0994,
      "mean_token_accuracy": 0.18820798248052598,
      "num_tokens": 49601958.0,
      "step": 28595
    },
    {
      "entropy": 5.381592464447022,
      "epoch": 2.2251702003501266,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00044996360719904777,
      "loss": 5.0187,
      "mean_token_accuracy": 0.19493012875318527,
      "num_tokens": 49610624.0,
      "step": 28600
    },
    {
      "entropy": 5.412230062484741,
      "epoch": 2.225559229721844,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004499461799881303,
      "loss": 5.0745,
      "mean_token_accuracy": 0.19116441756486893,
      "num_tokens": 49619308.0,
      "step": 28605
    },
    {
      "entropy": 5.42528133392334,
      "epoch": 2.2259482590935615,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004499287501226831,
      "loss": 5.0531,
      "mean_token_accuracy": 0.18971817791461945,
      "num_tokens": 49627842.0,
      "step": 28610
    },
    {
      "entropy": 5.4649186611175535,
      "epoch": 2.2263372884652792,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044991131760297045,
      "loss": 5.1492,
      "mean_token_accuracy": 0.18584991842508317,
      "num_tokens": 49636712.0,
      "step": 28615
    },
    {
      "entropy": 5.498069715499878,
      "epoch": 2.226726317836997,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044989388242925693,
      "loss": 5.1396,
      "mean_token_accuracy": 0.1903313785791397,
      "num_tokens": 49645324.0,
      "step": 28620
    },
    {
      "entropy": 5.420678806304932,
      "epoch": 2.227115347208714,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004498764446018073,
      "loss": 5.0833,
      "mean_token_accuracy": 0.19233340322971343,
      "num_tokens": 49654203.0,
      "step": 28625
    },
    {
      "entropy": 5.476797962188721,
      "epoch": 2.227504376580432,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004498590041208857,
      "loss": 5.1332,
      "mean_token_accuracy": 0.17837443500757216,
      "num_tokens": 49663197.0,
      "step": 28630
    },
    {
      "entropy": 5.39409122467041,
      "epoch": 2.2278934059521496,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004498415609867572,
      "loss": 5.0118,
      "mean_token_accuracy": 0.18957238495349885,
      "num_tokens": 49671993.0,
      "step": 28635
    },
    {
      "entropy": 5.324062442779541,
      "epoch": 2.228282435323867,
      "grad_norm": 1.25,
      "learning_rate": 0.00044982411519968625,
      "loss": 4.9067,
      "mean_token_accuracy": 0.19392403662204744,
      "num_tokens": 49680403.0,
      "step": 28640
    },
    {
      "entropy": 5.370967531204224,
      "epoch": 2.2286714646955845,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004498066667599377,
      "loss": 4.9371,
      "mean_token_accuracy": 0.20088836550712585,
      "num_tokens": 49688667.0,
      "step": 28645
    },
    {
      "entropy": 5.339514923095703,
      "epoch": 2.2290604940673022,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044978921566777604,
      "loss": 5.0718,
      "mean_token_accuracy": 0.1877834066748619,
      "num_tokens": 49696753.0,
      "step": 28650
    },
    {
      "entropy": 5.4572125434875485,
      "epoch": 2.2294495234390195,
      "grad_norm": 1.375,
      "learning_rate": 0.0004497717619234664,
      "loss": 5.0418,
      "mean_token_accuracy": 0.18978618532419206,
      "num_tokens": 49705222.0,
      "step": 28655
    },
    {
      "entropy": 5.419244575500488,
      "epoch": 2.229838552810737,
      "grad_norm": 1.5625,
      "learning_rate": 0.00044975430552727337,
      "loss": 5.0502,
      "mean_token_accuracy": 0.19420383125543594,
      "num_tokens": 49714133.0,
      "step": 28660
    },
    {
      "entropy": 5.34322075843811,
      "epoch": 2.230227582182455,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004497368464794619,
      "loss": 4.9956,
      "mean_token_accuracy": 0.1938999339938164,
      "num_tokens": 49722743.0,
      "step": 28665
    },
    {
      "entropy": 5.392423868179321,
      "epoch": 2.230616611554172,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00044971938478029693,
      "loss": 4.9794,
      "mean_token_accuracy": 0.20266159474849701,
      "num_tokens": 49731721.0,
      "step": 28670
    },
    {
      "entropy": 5.373440980911255,
      "epoch": 2.23100564092589,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044970192043004343,
      "loss": 4.9583,
      "mean_token_accuracy": 0.1940787300467491,
      "num_tokens": 49740373.0,
      "step": 28675
    },
    {
      "entropy": 5.344507455825806,
      "epoch": 2.2313946702976075,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004496844534289663,
      "loss": 5.0543,
      "mean_token_accuracy": 0.19620848894119264,
      "num_tokens": 49748544.0,
      "step": 28680
    },
    {
      "entropy": 5.545773124694824,
      "epoch": 2.231783699669325,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044966698377733064,
      "loss": 5.1485,
      "mean_token_accuracy": 0.1830441474914551,
      "num_tokens": 49757738.0,
      "step": 28685
    },
    {
      "entropy": 5.40210976600647,
      "epoch": 2.2321727290410425,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004496495114754015,
      "loss": 4.9624,
      "mean_token_accuracy": 0.19104286432266235,
      "num_tokens": 49766203.0,
      "step": 28690
    },
    {
      "entropy": 5.348806571960449,
      "epoch": 2.23256175841276,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044963203652344404,
      "loss": 5.0225,
      "mean_token_accuracy": 0.1862621620297432,
      "num_tokens": 49774951.0,
      "step": 28695
    },
    {
      "entropy": 5.315744638442993,
      "epoch": 2.232950787784478,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004496145589217234,
      "loss": 5.0177,
      "mean_token_accuracy": 0.18839140385389327,
      "num_tokens": 49783250.0,
      "step": 28700
    },
    {
      "entropy": 5.391508865356445,
      "epoch": 2.233339817156195,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044959707867050467,
      "loss": 4.9774,
      "mean_token_accuracy": 0.19507981687784196,
      "num_tokens": 49792832.0,
      "step": 28705
    },
    {
      "entropy": 5.383459138870239,
      "epoch": 2.233728846527913,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004495795957700532,
      "loss": 4.9905,
      "mean_token_accuracy": 0.1987452045083046,
      "num_tokens": 49800745.0,
      "step": 28710
    },
    {
      "entropy": 5.358058118820191,
      "epoch": 2.2341178758996305,
      "grad_norm": 1.125,
      "learning_rate": 0.00044956211022063424,
      "loss": 4.9483,
      "mean_token_accuracy": 0.20443753749132157,
      "num_tokens": 49809458.0,
      "step": 28715
    },
    {
      "entropy": 5.432945203781128,
      "epoch": 2.234506905271348,
      "grad_norm": 1.5,
      "learning_rate": 0.00044954462202251316,
      "loss": 5.0485,
      "mean_token_accuracy": 0.19775092601776123,
      "num_tokens": 49818122.0,
      "step": 28720
    },
    {
      "entropy": 5.371125364303589,
      "epoch": 2.2348959346430655,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044952713117595516,
      "loss": 4.9877,
      "mean_token_accuracy": 0.19033448845148088,
      "num_tokens": 49826451.0,
      "step": 28725
    },
    {
      "entropy": 5.484125709533691,
      "epoch": 2.235284964014783,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044950963768122576,
      "loss": 5.0815,
      "mean_token_accuracy": 0.18741485923528672,
      "num_tokens": 49835463.0,
      "step": 28730
    },
    {
      "entropy": 5.393840980529785,
      "epoch": 2.235673993386501,
      "grad_norm": 1.25,
      "learning_rate": 0.0004494921415385905,
      "loss": 5.0058,
      "mean_token_accuracy": 0.1955430656671524,
      "num_tokens": 49843896.0,
      "step": 28735
    },
    {
      "entropy": 5.426755666732788,
      "epoch": 2.236063022758218,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044947464274831464,
      "loss": 5.0618,
      "mean_token_accuracy": 0.18466218709945678,
      "num_tokens": 49852380.0,
      "step": 28740
    },
    {
      "entropy": 5.453187656402588,
      "epoch": 2.236452052129936,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004494571413106637,
      "loss": 5.0169,
      "mean_token_accuracy": 0.1939795657992363,
      "num_tokens": 49861290.0,
      "step": 28745
    },
    {
      "entropy": 5.468720436096191,
      "epoch": 2.2368410815016535,
      "grad_norm": 1.3125,
      "learning_rate": 0.00044943963722590344,
      "loss": 5.0676,
      "mean_token_accuracy": 0.19270249754190444,
      "num_tokens": 49869295.0,
      "step": 28750
    },
    {
      "entropy": 5.416519832611084,
      "epoch": 2.2372301108733708,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044942213049429917,
      "loss": 5.1548,
      "mean_token_accuracy": 0.19087494760751725,
      "num_tokens": 49877771.0,
      "step": 28755
    },
    {
      "entropy": 5.422810077667236,
      "epoch": 2.2376191402450885,
      "grad_norm": 1.296875,
      "learning_rate": 0.00044940462111611685,
      "loss": 5.1267,
      "mean_token_accuracy": 0.18862754702568055,
      "num_tokens": 49886531.0,
      "step": 28760
    },
    {
      "entropy": 5.498632097244263,
      "epoch": 2.238008169616806,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044938710909162195,
      "loss": 5.1466,
      "mean_token_accuracy": 0.19681938141584396,
      "num_tokens": 49895323.0,
      "step": 28765
    },
    {
      "entropy": 5.472868013381958,
      "epoch": 2.2383971989885234,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004493695944210802,
      "loss": 5.0793,
      "mean_token_accuracy": 0.1913537636399269,
      "num_tokens": 49903918.0,
      "step": 28770
    },
    {
      "entropy": 5.372855377197266,
      "epoch": 2.238786228360241,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004493520771047574,
      "loss": 5.0133,
      "mean_token_accuracy": 0.1948249414563179,
      "num_tokens": 49912754.0,
      "step": 28775
    },
    {
      "entropy": 5.438584613800049,
      "epoch": 2.239175257731959,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044933455714291944,
      "loss": 5.1382,
      "mean_token_accuracy": 0.18726300746202468,
      "num_tokens": 49922188.0,
      "step": 28780
    },
    {
      "entropy": 5.499379587173462,
      "epoch": 2.2395642871036765,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044931703453583195,
      "loss": 5.1224,
      "mean_token_accuracy": 0.18636402487754822,
      "num_tokens": 49930718.0,
      "step": 28785
    },
    {
      "entropy": 5.410118627548218,
      "epoch": 2.2399533164753938,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044929950928376095,
      "loss": 5.1274,
      "mean_token_accuracy": 0.19219346046447755,
      "num_tokens": 49939095.0,
      "step": 28790
    },
    {
      "entropy": 5.431325531005859,
      "epoch": 2.2403423458471114,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004492819813869723,
      "loss": 5.0525,
      "mean_token_accuracy": 0.19916273951530455,
      "num_tokens": 49946996.0,
      "step": 28795
    },
    {
      "entropy": 5.491430044174194,
      "epoch": 2.240731375218829,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044926445084573206,
      "loss": 5.1067,
      "mean_token_accuracy": 0.19212830662727357,
      "num_tokens": 49956262.0,
      "step": 28800
    },
    {
      "entropy": 5.426499891281128,
      "epoch": 2.2411204045905464,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000449246917660306,
      "loss": 5.0266,
      "mean_token_accuracy": 0.19874854981899262,
      "num_tokens": 49964581.0,
      "step": 28805
    },
    {
      "entropy": 5.286220932006836,
      "epoch": 2.241509433962264,
      "grad_norm": 1.25,
      "learning_rate": 0.0004492293818309604,
      "loss": 4.9045,
      "mean_token_accuracy": 0.1973341479897499,
      "num_tokens": 49972876.0,
      "step": 28810
    },
    {
      "entropy": 5.4111871242523195,
      "epoch": 2.241898463333982,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004492118433579613,
      "loss": 5.0521,
      "mean_token_accuracy": 0.1908900886774063,
      "num_tokens": 49981605.0,
      "step": 28815
    },
    {
      "entropy": 5.543170976638794,
      "epoch": 2.2422874927056995,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044919430224157463,
      "loss": 5.1933,
      "mean_token_accuracy": 0.18579653054475784,
      "num_tokens": 49990576.0,
      "step": 28820
    },
    {
      "entropy": 5.406686449050904,
      "epoch": 2.2426765220774167,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044917675848206684,
      "loss": 5.0313,
      "mean_token_accuracy": 0.19062115699052812,
      "num_tokens": 49999350.0,
      "step": 28825
    },
    {
      "entropy": 5.374755716323852,
      "epoch": 2.2430655514491344,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044915921207970387,
      "loss": 5.0278,
      "mean_token_accuracy": 0.19166239798069,
      "num_tokens": 50008202.0,
      "step": 28830
    },
    {
      "entropy": 5.358312320709229,
      "epoch": 2.243454580820852,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044914166303475206,
      "loss": 4.9353,
      "mean_token_accuracy": 0.19787064790725709,
      "num_tokens": 50016600.0,
      "step": 28835
    },
    {
      "entropy": 5.351320123672485,
      "epoch": 2.2438436101925694,
      "grad_norm": 1.25,
      "learning_rate": 0.00044912411134747764,
      "loss": 4.9857,
      "mean_token_accuracy": 0.20012570917606354,
      "num_tokens": 50025029.0,
      "step": 28840
    },
    {
      "entropy": 5.283933448791504,
      "epoch": 2.244232639564287,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044910655701814705,
      "loss": 4.9761,
      "mean_token_accuracy": 0.1964990586042404,
      "num_tokens": 50033692.0,
      "step": 28845
    },
    {
      "entropy": 5.447194290161133,
      "epoch": 2.244621668936005,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004490890000470266,
      "loss": 5.0877,
      "mean_token_accuracy": 0.1911623165011406,
      "num_tokens": 50042572.0,
      "step": 28850
    },
    {
      "entropy": 5.535709571838379,
      "epoch": 2.245010698307722,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044907144043438265,
      "loss": 5.1375,
      "mean_token_accuracy": 0.19437725096940994,
      "num_tokens": 50050823.0,
      "step": 28855
    },
    {
      "entropy": 5.39934458732605,
      "epoch": 2.2453997276794397,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004490538781804816,
      "loss": 4.9785,
      "mean_token_accuracy": 0.19421305060386657,
      "num_tokens": 50058596.0,
      "step": 28860
    },
    {
      "entropy": 5.423104906082154,
      "epoch": 2.2457887570511574,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044903631328559,
      "loss": 4.9741,
      "mean_token_accuracy": 0.19655991196632386,
      "num_tokens": 50067482.0,
      "step": 28865
    },
    {
      "entropy": 5.437716484069824,
      "epoch": 2.246177786422875,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00044901874574997444,
      "loss": 5.1002,
      "mean_token_accuracy": 0.19409218430519104,
      "num_tokens": 50075581.0,
      "step": 28870
    },
    {
      "entropy": 5.374107122421265,
      "epoch": 2.2465668157945924,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00044900117557390137,
      "loss": 4.994,
      "mean_token_accuracy": 0.19663966596126556,
      "num_tokens": 50083906.0,
      "step": 28875
    },
    {
      "entropy": 5.430074167251587,
      "epoch": 2.24695584516631,
      "grad_norm": 1.296875,
      "learning_rate": 0.00044898360275763733,
      "loss": 5.0866,
      "mean_token_accuracy": 0.1902938663959503,
      "num_tokens": 50091494.0,
      "step": 28880
    },
    {
      "entropy": 5.460253763198852,
      "epoch": 2.247344874538028,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044896602730144905,
      "loss": 5.1619,
      "mean_token_accuracy": 0.19280390590429305,
      "num_tokens": 50100576.0,
      "step": 28885
    },
    {
      "entropy": 5.5161497592926025,
      "epoch": 2.247733903909745,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044894844920560323,
      "loss": 5.2076,
      "mean_token_accuracy": 0.17911085188388826,
      "num_tokens": 50109832.0,
      "step": 28890
    },
    {
      "entropy": 5.454965972900391,
      "epoch": 2.2481229332814627,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004489308684703666,
      "loss": 5.1924,
      "mean_token_accuracy": 0.17899487018585206,
      "num_tokens": 50120038.0,
      "step": 28895
    },
    {
      "entropy": 5.441544055938721,
      "epoch": 2.2485119626531804,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004489132850960059,
      "loss": 5.074,
      "mean_token_accuracy": 0.1898920103907585,
      "num_tokens": 50128580.0,
      "step": 28900
    },
    {
      "entropy": 5.41399564743042,
      "epoch": 2.2489009920248977,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004488956990827878,
      "loss": 5.1446,
      "mean_token_accuracy": 0.18947150111198424,
      "num_tokens": 50137540.0,
      "step": 28905
    },
    {
      "entropy": 5.4897847175598145,
      "epoch": 2.2492900213966154,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004488781104309793,
      "loss": 5.1146,
      "mean_token_accuracy": 0.18805892169475555,
      "num_tokens": 50146292.0,
      "step": 28910
    },
    {
      "entropy": 5.390196943283081,
      "epoch": 2.249679050768333,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044886051914084726,
      "loss": 4.9595,
      "mean_token_accuracy": 0.19698358178138733,
      "num_tokens": 50154893.0,
      "step": 28915
    },
    {
      "entropy": 5.385761737823486,
      "epoch": 2.2500680801400508,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044884292521265846,
      "loss": 5.1037,
      "mean_token_accuracy": 0.18638947308063508,
      "num_tokens": 50163707.0,
      "step": 28920
    },
    {
      "entropy": 5.42761263847351,
      "epoch": 2.250457109511768,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004488253286466801,
      "loss": 5.0519,
      "mean_token_accuracy": 0.1898472025990486,
      "num_tokens": 50172062.0,
      "step": 28925
    },
    {
      "entropy": 5.415374374389648,
      "epoch": 2.2508461388834857,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044880772944317905,
      "loss": 5.0075,
      "mean_token_accuracy": 0.19126275479793547,
      "num_tokens": 50180362.0,
      "step": 28930
    },
    {
      "entropy": 5.3909505844116214,
      "epoch": 2.2512351682552034,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044879012760242224,
      "loss": 5.0867,
      "mean_token_accuracy": 0.1932997301220894,
      "num_tokens": 50189038.0,
      "step": 28935
    },
    {
      "entropy": 5.447734737396241,
      "epoch": 2.2516241976269207,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044877252312467694,
      "loss": 5.0367,
      "mean_token_accuracy": 0.18395038396120073,
      "num_tokens": 50198288.0,
      "step": 28940
    },
    {
      "entropy": 5.34761757850647,
      "epoch": 2.2520132269986384,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004487549160102101,
      "loss": 5.036,
      "mean_token_accuracy": 0.19511747658252715,
      "num_tokens": 50207442.0,
      "step": 28945
    },
    {
      "entropy": 5.447969007492065,
      "epoch": 2.252402256370356,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044873730625928914,
      "loss": 5.1624,
      "mean_token_accuracy": 0.18647781759500504,
      "num_tokens": 50216794.0,
      "step": 28950
    },
    {
      "entropy": 5.439132499694824,
      "epoch": 2.2527912857420738,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044871969387218094,
      "loss": 5.0052,
      "mean_token_accuracy": 0.1937894269824028,
      "num_tokens": 50224764.0,
      "step": 28955
    },
    {
      "entropy": 5.399581575393677,
      "epoch": 2.253180315113791,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000448702078849153,
      "loss": 5.1171,
      "mean_token_accuracy": 0.18378374129533767,
      "num_tokens": 50233256.0,
      "step": 28960
    },
    {
      "entropy": 5.414939117431641,
      "epoch": 2.2535693444855087,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044868446119047234,
      "loss": 5.0409,
      "mean_token_accuracy": 0.1940825402736664,
      "num_tokens": 50242862.0,
      "step": 28965
    },
    {
      "entropy": 5.410463571548462,
      "epoch": 2.253958373857226,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004486668408964065,
      "loss": 5.048,
      "mean_token_accuracy": 0.18888943642377853,
      "num_tokens": 50252048.0,
      "step": 28970
    },
    {
      "entropy": 5.455747842788696,
      "epoch": 2.2543474032289437,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004486492179672228,
      "loss": 5.162,
      "mean_token_accuracy": 0.18574519455432892,
      "num_tokens": 50260545.0,
      "step": 28975
    },
    {
      "entropy": 5.404513311386109,
      "epoch": 2.2547364326006614,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044863159240318863,
      "loss": 4.9761,
      "mean_token_accuracy": 0.19828768372535704,
      "num_tokens": 50268714.0,
      "step": 28980
    },
    {
      "entropy": 5.3327226638793945,
      "epoch": 2.255125461972379,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004486139642045714,
      "loss": 4.9714,
      "mean_token_accuracy": 0.2000202015042305,
      "num_tokens": 50276876.0,
      "step": 28985
    },
    {
      "entropy": 5.441846609115601,
      "epoch": 2.2555144913440963,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004485963333716385,
      "loss": 5.1219,
      "mean_token_accuracy": 0.18513816595077515,
      "num_tokens": 50285924.0,
      "step": 28990
    },
    {
      "entropy": 5.518434953689575,
      "epoch": 2.255903520715814,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004485786999046576,
      "loss": 5.1112,
      "mean_token_accuracy": 0.182501682639122,
      "num_tokens": 50295358.0,
      "step": 28995
    },
    {
      "entropy": 5.444577646255493,
      "epoch": 2.2562925500875317,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044856106380389624,
      "loss": 5.097,
      "mean_token_accuracy": 0.18781280219554902,
      "num_tokens": 50303572.0,
      "step": 29000
    },
    {
      "entropy": 5.404143285751343,
      "epoch": 2.256681579459249,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000448543425069622,
      "loss": 5.021,
      "mean_token_accuracy": 0.19773191213607788,
      "num_tokens": 50312533.0,
      "step": 29005
    },
    {
      "entropy": 5.401978158950806,
      "epoch": 2.2570706088309667,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004485257837021025,
      "loss": 5.0128,
      "mean_token_accuracy": 0.1936962679028511,
      "num_tokens": 50321049.0,
      "step": 29010
    },
    {
      "entropy": 5.340402030944825,
      "epoch": 2.2574596382026844,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044850813970160534,
      "loss": 4.9734,
      "mean_token_accuracy": 0.19698132276535035,
      "num_tokens": 50329308.0,
      "step": 29015
    },
    {
      "entropy": 5.417828845977783,
      "epoch": 2.257848667574402,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004484904930683984,
      "loss": 4.9817,
      "mean_token_accuracy": 0.19921959489583968,
      "num_tokens": 50337311.0,
      "step": 29020
    },
    {
      "entropy": 5.369078588485718,
      "epoch": 2.2582376969461193,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004484728438027494,
      "loss": 5.0673,
      "mean_token_accuracy": 0.19529227763414383,
      "num_tokens": 50346557.0,
      "step": 29025
    },
    {
      "entropy": 5.316523790359497,
      "epoch": 2.258626726317837,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004484551919049261,
      "loss": 5.0134,
      "mean_token_accuracy": 0.19744038879871367,
      "num_tokens": 50354982.0,
      "step": 29030
    },
    {
      "entropy": 5.407460021972656,
      "epoch": 2.2590157556895547,
      "grad_norm": 1.34375,
      "learning_rate": 0.00044843753737519625,
      "loss": 5.0306,
      "mean_token_accuracy": 0.18970168083906175,
      "num_tokens": 50363414.0,
      "step": 29035
    },
    {
      "entropy": 5.4611560821533205,
      "epoch": 2.259404785061272,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004484198802138279,
      "loss": 4.9862,
      "mean_token_accuracy": 0.1963168665766716,
      "num_tokens": 50371803.0,
      "step": 29040
    },
    {
      "entropy": 5.421046543121338,
      "epoch": 2.2597938144329897,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004484022204210889,
      "loss": 5.0969,
      "mean_token_accuracy": 0.19071290791034698,
      "num_tokens": 50380125.0,
      "step": 29045
    },
    {
      "entropy": 5.305266714096069,
      "epoch": 2.2601828438047074,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044838455799724717,
      "loss": 4.966,
      "mean_token_accuracy": 0.18888027369976043,
      "num_tokens": 50388916.0,
      "step": 29050
    },
    {
      "entropy": 5.4966155052185055,
      "epoch": 2.260571873176425,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00044836689294257075,
      "loss": 5.1424,
      "mean_token_accuracy": 0.1934264823794365,
      "num_tokens": 50397353.0,
      "step": 29055
    },
    {
      "entropy": 5.408742189407349,
      "epoch": 2.2609609025481423,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004483492252573276,
      "loss": 5.0104,
      "mean_token_accuracy": 0.1963096097111702,
      "num_tokens": 50405176.0,
      "step": 29060
    },
    {
      "entropy": 5.316344499588013,
      "epoch": 2.26134993191986,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000448331554941786,
      "loss": 5.0289,
      "mean_token_accuracy": 0.19149352014064788,
      "num_tokens": 50413845.0,
      "step": 29065
    },
    {
      "entropy": 5.403061866760254,
      "epoch": 2.2617389612915777,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044831388199621385,
      "loss": 5.1175,
      "mean_token_accuracy": 0.18719917237758638,
      "num_tokens": 50422343.0,
      "step": 29070
    },
    {
      "entropy": 5.361449766159057,
      "epoch": 2.262127990663295,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044829620642087946,
      "loss": 4.978,
      "mean_token_accuracy": 0.20042907148599626,
      "num_tokens": 50430438.0,
      "step": 29075
    },
    {
      "entropy": 5.436171627044677,
      "epoch": 2.2625170200350126,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004482785282160509,
      "loss": 5.0388,
      "mean_token_accuracy": 0.19168473333120345,
      "num_tokens": 50438865.0,
      "step": 29080
    },
    {
      "entropy": 5.456004333496094,
      "epoch": 2.2629060494067303,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044826084738199657,
      "loss": 5.1194,
      "mean_token_accuracy": 0.18938930779695512,
      "num_tokens": 50447784.0,
      "step": 29085
    },
    {
      "entropy": 5.496140670776367,
      "epoch": 2.2632950787784476,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044824316391898464,
      "loss": 5.1129,
      "mean_token_accuracy": 0.1836037814617157,
      "num_tokens": 50457266.0,
      "step": 29090
    },
    {
      "entropy": 5.5526509284973145,
      "epoch": 2.2636841081501653,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044822547782728334,
      "loss": 5.2366,
      "mean_token_accuracy": 0.18093956559896468,
      "num_tokens": 50466171.0,
      "step": 29095
    },
    {
      "entropy": 5.449325704574585,
      "epoch": 2.264073137521883,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004482077891071612,
      "loss": 5.0149,
      "mean_token_accuracy": 0.20048542469739913,
      "num_tokens": 50474127.0,
      "step": 29100
    },
    {
      "entropy": 5.331634569168091,
      "epoch": 2.2644621668936002,
      "grad_norm": 1.125,
      "learning_rate": 0.00044819009775888655,
      "loss": 4.9716,
      "mean_token_accuracy": 0.19020601660013198,
      "num_tokens": 50482724.0,
      "step": 29105
    },
    {
      "entropy": 5.482419681549072,
      "epoch": 2.264851196265318,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044817240378272784,
      "loss": 5.1956,
      "mean_token_accuracy": 0.17646986842155457,
      "num_tokens": 50491412.0,
      "step": 29110
    },
    {
      "entropy": 5.540668392181397,
      "epoch": 2.2652402256370356,
      "grad_norm": 1.390625,
      "learning_rate": 0.00044815470717895345,
      "loss": 5.1991,
      "mean_token_accuracy": 0.18348639756441115,
      "num_tokens": 50499843.0,
      "step": 29115
    },
    {
      "entropy": 5.535886526107788,
      "epoch": 2.2656292550087533,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004481370079478321,
      "loss": 5.1177,
      "mean_token_accuracy": 0.19067116230726242,
      "num_tokens": 50509501.0,
      "step": 29120
    },
    {
      "entropy": 5.437737131118775,
      "epoch": 2.2660182843804706,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004481193060896322,
      "loss": 5.0878,
      "mean_token_accuracy": 0.1952413260936737,
      "num_tokens": 50518438.0,
      "step": 29125
    },
    {
      "entropy": 5.425158166885376,
      "epoch": 2.2664073137521883,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004481016016046223,
      "loss": 5.0654,
      "mean_token_accuracy": 0.18836312741041183,
      "num_tokens": 50527516.0,
      "step": 29130
    },
    {
      "entropy": 5.434842777252197,
      "epoch": 2.266796343123906,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004480838944930712,
      "loss": 5.0138,
      "mean_token_accuracy": 0.18986533731222152,
      "num_tokens": 50536179.0,
      "step": 29135
    },
    {
      "entropy": 5.448123502731323,
      "epoch": 2.2671853724956232,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004480661847552474,
      "loss": 5.0894,
      "mean_token_accuracy": 0.19332789927721022,
      "num_tokens": 50545231.0,
      "step": 29140
    },
    {
      "entropy": 5.374819087982178,
      "epoch": 2.267574401867341,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004480484723914198,
      "loss": 5.0919,
      "mean_token_accuracy": 0.18890254199504852,
      "num_tokens": 50553873.0,
      "step": 29145
    },
    {
      "entropy": 5.475753450393677,
      "epoch": 2.2679634312390586,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004480307574018571,
      "loss": 5.1105,
      "mean_token_accuracy": 0.19271204024553298,
      "num_tokens": 50563733.0,
      "step": 29150
    },
    {
      "entropy": 5.466091203689575,
      "epoch": 2.2683524606107763,
      "grad_norm": 1.140625,
      "learning_rate": 0.000448013039786828,
      "loss": 5.036,
      "mean_token_accuracy": 0.19168713688850403,
      "num_tokens": 50573187.0,
      "step": 29155
    },
    {
      "entropy": 5.378099346160889,
      "epoch": 2.2687414899824936,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044799531954660133,
      "loss": 5.0041,
      "mean_token_accuracy": 0.1955648735165596,
      "num_tokens": 50581580.0,
      "step": 29160
    },
    {
      "entropy": 5.410655355453491,
      "epoch": 2.2691305193542113,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004479775966814461,
      "loss": 5.0454,
      "mean_token_accuracy": 0.19516540914773942,
      "num_tokens": 50589471.0,
      "step": 29165
    },
    {
      "entropy": 5.460061407089233,
      "epoch": 2.269519548725929,
      "grad_norm": 1.296875,
      "learning_rate": 0.00044795987119163115,
      "loss": 5.0984,
      "mean_token_accuracy": 0.19677649885416032,
      "num_tokens": 50597923.0,
      "step": 29170
    },
    {
      "entropy": 5.476475620269776,
      "epoch": 2.2699085780976462,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004479421430774255,
      "loss": 5.1775,
      "mean_token_accuracy": 0.1796620413661003,
      "num_tokens": 50606981.0,
      "step": 29175
    },
    {
      "entropy": 5.497299385070801,
      "epoch": 2.270297607469364,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004479244123390981,
      "loss": 5.0737,
      "mean_token_accuracy": 0.18257933408021926,
      "num_tokens": 50615680.0,
      "step": 29180
    },
    {
      "entropy": 5.472466850280762,
      "epoch": 2.2706866368410816,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044790667897691796,
      "loss": 5.0384,
      "mean_token_accuracy": 0.19708404839038848,
      "num_tokens": 50624053.0,
      "step": 29185
    },
    {
      "entropy": 5.388488245010376,
      "epoch": 2.271075666212799,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044788894299115415,
      "loss": 5.0828,
      "mean_token_accuracy": 0.1889742434024811,
      "num_tokens": 50632671.0,
      "step": 29190
    },
    {
      "entropy": 5.446991729736328,
      "epoch": 2.2714646955845166,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004478712043820758,
      "loss": 5.0804,
      "mean_token_accuracy": 0.19289014488458633,
      "num_tokens": 50641734.0,
      "step": 29195
    },
    {
      "entropy": 5.387019300460816,
      "epoch": 2.2718537249562343,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004478534631499521,
      "loss": 4.9959,
      "mean_token_accuracy": 0.1988363042473793,
      "num_tokens": 50650711.0,
      "step": 29200
    },
    {
      "entropy": 5.411226081848144,
      "epoch": 2.2722427543279515,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044783571929505223,
      "loss": 5.074,
      "mean_token_accuracy": 0.18043090850114823,
      "num_tokens": 50659814.0,
      "step": 29205
    },
    {
      "entropy": 5.456407642364502,
      "epoch": 2.2726317836996692,
      "grad_norm": 1.25,
      "learning_rate": 0.0004478179728176454,
      "loss": 5.0283,
      "mean_token_accuracy": 0.19659682214260102,
      "num_tokens": 50668340.0,
      "step": 29210
    },
    {
      "entropy": 5.423078441619873,
      "epoch": 2.273020813071387,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004478002237180009,
      "loss": 5.0006,
      "mean_token_accuracy": 0.19361726194620132,
      "num_tokens": 50676847.0,
      "step": 29215
    },
    {
      "entropy": 5.2886981010437015,
      "epoch": 2.2734098424431046,
      "grad_norm": 1.25,
      "learning_rate": 0.000447782471996388,
      "loss": 5.021,
      "mean_token_accuracy": 0.19512952417135238,
      "num_tokens": 50685074.0,
      "step": 29220
    },
    {
      "entropy": 5.401895570755005,
      "epoch": 2.273798871814822,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004477647176530762,
      "loss": 5.0852,
      "mean_token_accuracy": 0.19187178760766982,
      "num_tokens": 50693386.0,
      "step": 29225
    },
    {
      "entropy": 5.459427404403686,
      "epoch": 2.2741879011865396,
      "grad_norm": 1.375,
      "learning_rate": 0.0004477469606883347,
      "loss": 5.0488,
      "mean_token_accuracy": 0.18817493170499802,
      "num_tokens": 50701966.0,
      "step": 29230
    },
    {
      "entropy": 5.355460453033447,
      "epoch": 2.2745769305582573,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004477292011024331,
      "loss": 5.0032,
      "mean_token_accuracy": 0.19927186965942384,
      "num_tokens": 50710630.0,
      "step": 29235
    },
    {
      "entropy": 5.374324131011963,
      "epoch": 2.2749659599299745,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044771143889564075,
      "loss": 4.9883,
      "mean_token_accuracy": 0.19673231542110442,
      "num_tokens": 50719415.0,
      "step": 29240
    },
    {
      "entropy": 5.434623575210571,
      "epoch": 2.275354989301692,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044769367406822727,
      "loss": 4.9667,
      "mean_token_accuracy": 0.20096562951803207,
      "num_tokens": 50728064.0,
      "step": 29245
    },
    {
      "entropy": 5.34841103553772,
      "epoch": 2.27574401867341,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004476759066204621,
      "loss": 5.075,
      "mean_token_accuracy": 0.1870547741651535,
      "num_tokens": 50736978.0,
      "step": 29250
    },
    {
      "entropy": 5.440932989120483,
      "epoch": 2.2761330480451276,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004476581365526149,
      "loss": 5.0319,
      "mean_token_accuracy": 0.18871684968471528,
      "num_tokens": 50745368.0,
      "step": 29255
    },
    {
      "entropy": 5.427804183959961,
      "epoch": 2.276522077416845,
      "grad_norm": 1.125,
      "learning_rate": 0.0004476403638649553,
      "loss": 5.0307,
      "mean_token_accuracy": 0.20431438386440276,
      "num_tokens": 50753765.0,
      "step": 29260
    },
    {
      "entropy": 5.41740460395813,
      "epoch": 2.2769111067885626,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044762258855775294,
      "loss": 5.1041,
      "mean_token_accuracy": 0.18782347738742827,
      "num_tokens": 50762165.0,
      "step": 29265
    },
    {
      "entropy": 5.459023284912109,
      "epoch": 2.2773001361602803,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044760481063127754,
      "loss": 5.1308,
      "mean_token_accuracy": 0.19137006849050522,
      "num_tokens": 50771837.0,
      "step": 29270
    },
    {
      "entropy": 5.488284969329834,
      "epoch": 2.2776891655319975,
      "grad_norm": 1.203125,
      "learning_rate": 0.000447587030085799,
      "loss": 5.1033,
      "mean_token_accuracy": 0.18859454095363617,
      "num_tokens": 50781518.0,
      "step": 29275
    },
    {
      "entropy": 5.442433023452759,
      "epoch": 2.278078194903715,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044756924692158684,
      "loss": 5.0603,
      "mean_token_accuracy": 0.18570311814546586,
      "num_tokens": 50790708.0,
      "step": 29280
    },
    {
      "entropy": 5.375849294662475,
      "epoch": 2.278467224275433,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004475514611389111,
      "loss": 5.076,
      "mean_token_accuracy": 0.18676667958498,
      "num_tokens": 50798783.0,
      "step": 29285
    },
    {
      "entropy": 5.391128492355347,
      "epoch": 2.2788562536471506,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004475336727380415,
      "loss": 4.9943,
      "mean_token_accuracy": 0.19031715095043183,
      "num_tokens": 50807033.0,
      "step": 29290
    },
    {
      "entropy": 5.357386207580566,
      "epoch": 2.279245283018868,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004475158817192481,
      "loss": 4.9354,
      "mean_token_accuracy": 0.19494040310382843,
      "num_tokens": 50815548.0,
      "step": 29295
    },
    {
      "entropy": 5.479776620864868,
      "epoch": 2.2796343123905856,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044749808808280077,
      "loss": 5.1051,
      "mean_token_accuracy": 0.18730299919843674,
      "num_tokens": 50824268.0,
      "step": 29300
    },
    {
      "entropy": 5.391113042831421,
      "epoch": 2.280023341762303,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044748029182896956,
      "loss": 4.9366,
      "mean_token_accuracy": 0.19704650491476058,
      "num_tokens": 50833136.0,
      "step": 29305
    },
    {
      "entropy": 5.447116756439209,
      "epoch": 2.2804123711340205,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004474624929580243,
      "loss": 5.124,
      "mean_token_accuracy": 0.1866270437836647,
      "num_tokens": 50841911.0,
      "step": 29310
    },
    {
      "entropy": 5.482091236114502,
      "epoch": 2.280801400505738,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004474446914702354,
      "loss": 5.2607,
      "mean_token_accuracy": 0.1794453963637352,
      "num_tokens": 50850402.0,
      "step": 29315
    },
    {
      "entropy": 5.497564172744751,
      "epoch": 2.281190429877456,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004474268873658727,
      "loss": 5.0668,
      "mean_token_accuracy": 0.19331685453653336,
      "num_tokens": 50858473.0,
      "step": 29320
    },
    {
      "entropy": 5.40937180519104,
      "epoch": 2.281579459249173,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004474090806452064,
      "loss": 4.9292,
      "mean_token_accuracy": 0.1995991811156273,
      "num_tokens": 50866798.0,
      "step": 29325
    },
    {
      "entropy": 5.3183125972747805,
      "epoch": 2.281968488620891,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044739127130850675,
      "loss": 5.0383,
      "mean_token_accuracy": 0.19435769617557525,
      "num_tokens": 50875279.0,
      "step": 29330
    },
    {
      "entropy": 5.352911949157715,
      "epoch": 2.2823575179926086,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044737345935604397,
      "loss": 4.9255,
      "mean_token_accuracy": 0.20354053527116775,
      "num_tokens": 50884021.0,
      "step": 29335
    },
    {
      "entropy": 5.337819623947143,
      "epoch": 2.282746547364326,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004473556447880883,
      "loss": 4.9876,
      "mean_token_accuracy": 0.1977727770805359,
      "num_tokens": 50891722.0,
      "step": 29340
    },
    {
      "entropy": 5.389324903488159,
      "epoch": 2.2831355767360435,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044733782760490997,
      "loss": 5.0941,
      "mean_token_accuracy": 0.19295429140329362,
      "num_tokens": 50900985.0,
      "step": 29345
    },
    {
      "entropy": 5.518482208251953,
      "epoch": 2.283524606107761,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004473200078067795,
      "loss": 5.147,
      "mean_token_accuracy": 0.17880460619926453,
      "num_tokens": 50909750.0,
      "step": 29350
    },
    {
      "entropy": 5.420819664001465,
      "epoch": 2.283913635479479,
      "grad_norm": 1.125,
      "learning_rate": 0.00044730218539396717,
      "loss": 5.0991,
      "mean_token_accuracy": 0.18782745748758317,
      "num_tokens": 50918563.0,
      "step": 29355
    },
    {
      "entropy": 5.40172176361084,
      "epoch": 2.284302664851196,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004472843603667434,
      "loss": 5.1234,
      "mean_token_accuracy": 0.18483064621686934,
      "num_tokens": 50927503.0,
      "step": 29360
    },
    {
      "entropy": 5.392805337905884,
      "epoch": 2.284691694222914,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044726653272537866,
      "loss": 5.0061,
      "mean_token_accuracy": 0.19408462941646576,
      "num_tokens": 50936198.0,
      "step": 29365
    },
    {
      "entropy": 5.384798192977906,
      "epoch": 2.2850807235946315,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004472487024701435,
      "loss": 5.0254,
      "mean_token_accuracy": 0.19249866902828217,
      "num_tokens": 50945589.0,
      "step": 29370
    },
    {
      "entropy": 5.373767375946045,
      "epoch": 2.285469752966349,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004472308696013085,
      "loss": 5.0031,
      "mean_token_accuracy": 0.19697571247816087,
      "num_tokens": 50953524.0,
      "step": 29375
    },
    {
      "entropy": 5.413288354873657,
      "epoch": 2.2858587823380665,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004472130341191441,
      "loss": 5.061,
      "mean_token_accuracy": 0.18685273677110673,
      "num_tokens": 50962127.0,
      "step": 29380
    },
    {
      "entropy": 5.409551286697388,
      "epoch": 2.286247811709784,
      "grad_norm": 1.25,
      "learning_rate": 0.0004471951960239211,
      "loss": 5.0506,
      "mean_token_accuracy": 0.198875330388546,
      "num_tokens": 50970662.0,
      "step": 29385
    },
    {
      "entropy": 5.373150777816773,
      "epoch": 2.286636841081502,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044717735531591,
      "loss": 4.9783,
      "mean_token_accuracy": 0.19755344539880754,
      "num_tokens": 50979182.0,
      "step": 29390
    },
    {
      "entropy": 5.402038145065307,
      "epoch": 2.287025870453219,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044715951199538156,
      "loss": 5.118,
      "mean_token_accuracy": 0.1900848016142845,
      "num_tokens": 50987285.0,
      "step": 29395
    },
    {
      "entropy": 5.432511949539185,
      "epoch": 2.287414899824937,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044714166606260657,
      "loss": 5.0629,
      "mean_token_accuracy": 0.1923697456717491,
      "num_tokens": 50995988.0,
      "step": 29400
    },
    {
      "entropy": 5.361797094345093,
      "epoch": 2.2878039291966545,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004471238175178559,
      "loss": 4.9582,
      "mean_token_accuracy": 0.19657040387392044,
      "num_tokens": 51003963.0,
      "step": 29405
    },
    {
      "entropy": 5.380674934387207,
      "epoch": 2.288192958568372,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004471059663614002,
      "loss": 5.0138,
      "mean_token_accuracy": 0.1858625367283821,
      "num_tokens": 51012429.0,
      "step": 29410
    },
    {
      "entropy": 5.4750550270080565,
      "epoch": 2.2885819879400895,
      "grad_norm": 1.25,
      "learning_rate": 0.0004470881125935103,
      "loss": 5.0615,
      "mean_token_accuracy": 0.1930341601371765,
      "num_tokens": 51021321.0,
      "step": 29415
    },
    {
      "entropy": 5.36868691444397,
      "epoch": 2.288971017311807,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044707025621445735,
      "loss": 4.988,
      "mean_token_accuracy": 0.19266353100538253,
      "num_tokens": 51029713.0,
      "step": 29420
    },
    {
      "entropy": 5.349739027023316,
      "epoch": 2.2893600466835244,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044705239722451194,
      "loss": 5.072,
      "mean_token_accuracy": 0.18422227799892427,
      "num_tokens": 51038832.0,
      "step": 29425
    },
    {
      "entropy": 5.4558004379272464,
      "epoch": 2.289749076055242,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044703453562394544,
      "loss": 4.977,
      "mean_token_accuracy": 0.20158940851688384,
      "num_tokens": 51046786.0,
      "step": 29430
    },
    {
      "entropy": 5.391645002365112,
      "epoch": 2.29013810542696,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004470166714130286,
      "loss": 5.0462,
      "mean_token_accuracy": 0.18849603831768036,
      "num_tokens": 51054855.0,
      "step": 29435
    },
    {
      "entropy": 5.461908960342408,
      "epoch": 2.290527134798677,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044699880459203264,
      "loss": 5.1694,
      "mean_token_accuracy": 0.18931534737348557,
      "num_tokens": 51063098.0,
      "step": 29440
    },
    {
      "entropy": 5.427787113189697,
      "epoch": 2.290916164170395,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044698093516122844,
      "loss": 4.9331,
      "mean_token_accuracy": 0.19889785796403886,
      "num_tokens": 51071140.0,
      "step": 29445
    },
    {
      "entropy": 5.400129556655884,
      "epoch": 2.2913051935421125,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004469630631208874,
      "loss": 5.0684,
      "mean_token_accuracy": 0.19501245617866517,
      "num_tokens": 51079698.0,
      "step": 29450
    },
    {
      "entropy": 5.395448827743531,
      "epoch": 2.29169422291383,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004469451884712805,
      "loss": 5.0734,
      "mean_token_accuracy": 0.1888278603553772,
      "num_tokens": 51088900.0,
      "step": 29455
    },
    {
      "entropy": 5.398367261886596,
      "epoch": 2.2920832522855474,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004469273112126791,
      "loss": 4.9981,
      "mean_token_accuracy": 0.19329816550016404,
      "num_tokens": 51098289.0,
      "step": 29460
    },
    {
      "entropy": 5.475690460205078,
      "epoch": 2.292472281657265,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044690943134535445,
      "loss": 5.0153,
      "mean_token_accuracy": 0.19418433606624602,
      "num_tokens": 51106795.0,
      "step": 29465
    },
    {
      "entropy": 5.367531061172485,
      "epoch": 2.292861311028983,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004468915488695777,
      "loss": 5.0219,
      "mean_token_accuracy": 0.19653323590755462,
      "num_tokens": 51115329.0,
      "step": 29470
    },
    {
      "entropy": 5.277588558197022,
      "epoch": 2.2932503404007,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044687366378562036,
      "loss": 5.0074,
      "mean_token_accuracy": 0.19630371034145355,
      "num_tokens": 51124215.0,
      "step": 29475
    },
    {
      "entropy": 5.407614517211914,
      "epoch": 2.293639369772418,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044685577609375373,
      "loss": 5.09,
      "mean_token_accuracy": 0.1888623610138893,
      "num_tokens": 51132528.0,
      "step": 29480
    },
    {
      "entropy": 5.442210483551025,
      "epoch": 2.2940283991441355,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004468378857942492,
      "loss": 5.1283,
      "mean_token_accuracy": 0.18535656332969666,
      "num_tokens": 51140726.0,
      "step": 29485
    },
    {
      "entropy": 5.368334627151489,
      "epoch": 2.294417428515853,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044681999288737826,
      "loss": 5.024,
      "mean_token_accuracy": 0.19480169117450713,
      "num_tokens": 51149312.0,
      "step": 29490
    },
    {
      "entropy": 5.479760932922363,
      "epoch": 2.2948064578875704,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044680209737341244,
      "loss": 5.1455,
      "mean_token_accuracy": 0.1908273845911026,
      "num_tokens": 51157464.0,
      "step": 29495
    },
    {
      "entropy": 5.3976658344268795,
      "epoch": 2.295195487259288,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004467841992526233,
      "loss": 4.9731,
      "mean_token_accuracy": 0.1964695170521736,
      "num_tokens": 51166331.0,
      "step": 29500
    },
    {
      "entropy": 5.420523071289063,
      "epoch": 2.295584516631006,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044676629852528226,
      "loss": 5.048,
      "mean_token_accuracy": 0.2003919556736946,
      "num_tokens": 51174404.0,
      "step": 29505
    },
    {
      "entropy": 5.273918390274048,
      "epoch": 2.295973546002723,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044674839519166104,
      "loss": 4.8827,
      "mean_token_accuracy": 0.20315038412809372,
      "num_tokens": 51182650.0,
      "step": 29510
    },
    {
      "entropy": 5.408922147750855,
      "epoch": 2.2963625753744408,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044673048925203136,
      "loss": 5.0708,
      "mean_token_accuracy": 0.18883897513151168,
      "num_tokens": 51192107.0,
      "step": 29515
    },
    {
      "entropy": 5.3842743873596195,
      "epoch": 2.2967516047461585,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044671258070666476,
      "loss": 4.9988,
      "mean_token_accuracy": 0.19348547756671905,
      "num_tokens": 51200506.0,
      "step": 29520
    },
    {
      "entropy": 5.35621280670166,
      "epoch": 2.2971406341178757,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004466946695558331,
      "loss": 4.9545,
      "mean_token_accuracy": 0.19873870611190797,
      "num_tokens": 51209442.0,
      "step": 29525
    },
    {
      "entropy": 5.386664867401123,
      "epoch": 2.2975296634895934,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044667675579980817,
      "loss": 5.0264,
      "mean_token_accuracy": 0.19772400856018066,
      "num_tokens": 51217520.0,
      "step": 29530
    },
    {
      "entropy": 5.448230361938476,
      "epoch": 2.297918692861311,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004466588394388616,
      "loss": 5.1389,
      "mean_token_accuracy": 0.18627449721097947,
      "num_tokens": 51226228.0,
      "step": 29535
    },
    {
      "entropy": 5.468140697479248,
      "epoch": 2.2983077222330284,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004466409204732654,
      "loss": 5.1084,
      "mean_token_accuracy": 0.18486340939998627,
      "num_tokens": 51235271.0,
      "step": 29540
    },
    {
      "entropy": 5.391859865188598,
      "epoch": 2.298696751604746,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044662299890329144,
      "loss": 5.0028,
      "mean_token_accuracy": 0.19454284608364106,
      "num_tokens": 51243927.0,
      "step": 29545
    },
    {
      "entropy": 5.458694171905518,
      "epoch": 2.2990857809764638,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004466050747292116,
      "loss": 5.0669,
      "mean_token_accuracy": 0.19687831699848174,
      "num_tokens": 51253097.0,
      "step": 29550
    },
    {
      "entropy": 5.411312627792358,
      "epoch": 2.2994748103481815,
      "grad_norm": 1.171875,
      "learning_rate": 0.000446587147951298,
      "loss": 5.0481,
      "mean_token_accuracy": 0.19691979885101318,
      "num_tokens": 51261275.0,
      "step": 29555
    },
    {
      "entropy": 5.438805675506591,
      "epoch": 2.2998638397198987,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004465692185698224,
      "loss": 5.1066,
      "mean_token_accuracy": 0.18948965072631835,
      "num_tokens": 51269698.0,
      "step": 29560
    },
    {
      "entropy": 5.359121990203858,
      "epoch": 2.3002528690916164,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044655128658505717,
      "loss": 4.9722,
      "mean_token_accuracy": 0.1997550517320633,
      "num_tokens": 51277850.0,
      "step": 29565
    },
    {
      "entropy": 5.4082457542419435,
      "epoch": 2.300641898463334,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004465333519972741,
      "loss": 4.9819,
      "mean_token_accuracy": 0.19506940990686417,
      "num_tokens": 51285919.0,
      "step": 29570
    },
    {
      "entropy": 5.375007820129395,
      "epoch": 2.3010309278350514,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004465154148067454,
      "loss": 5.0549,
      "mean_token_accuracy": 0.19366700798273087,
      "num_tokens": 51294403.0,
      "step": 29575
    },
    {
      "entropy": 5.489139699935913,
      "epoch": 2.301419957206769,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044649747501374336,
      "loss": 5.0687,
      "mean_token_accuracy": 0.1923191502690315,
      "num_tokens": 51302820.0,
      "step": 29580
    },
    {
      "entropy": 5.356007146835327,
      "epoch": 2.3018089865784868,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004464795326185401,
      "loss": 5.0477,
      "mean_token_accuracy": 0.19427828639745712,
      "num_tokens": 51311385.0,
      "step": 29585
    },
    {
      "entropy": 5.4268226146698,
      "epoch": 2.3021980159502045,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004464615876214078,
      "loss": 5.0875,
      "mean_token_accuracy": 0.18926291018724442,
      "num_tokens": 51320928.0,
      "step": 29590
    },
    {
      "entropy": 5.354275894165039,
      "epoch": 2.3025870453219217,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004464436400226188,
      "loss": 4.9639,
      "mean_token_accuracy": 0.20075487792491914,
      "num_tokens": 51329384.0,
      "step": 29595
    },
    {
      "entropy": 5.3405256271362305,
      "epoch": 2.3029760746936394,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004464256898224455,
      "loss": 5.0366,
      "mean_token_accuracy": 0.1833977073431015,
      "num_tokens": 51337765.0,
      "step": 29600
    },
    {
      "entropy": 5.43039493560791,
      "epoch": 2.303365104065357,
      "grad_norm": 1.25,
      "learning_rate": 0.0004464077370211602,
      "loss": 5.0644,
      "mean_token_accuracy": 0.18582940697669983,
      "num_tokens": 51345947.0,
      "step": 29605
    },
    {
      "entropy": 5.372714805603027,
      "epoch": 2.3037541334370744,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044638978161903533,
      "loss": 4.9782,
      "mean_token_accuracy": 0.18902389109134674,
      "num_tokens": 51354742.0,
      "step": 29610
    },
    {
      "entropy": 5.328454923629761,
      "epoch": 2.304143162808792,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004463718236163433,
      "loss": 4.9964,
      "mean_token_accuracy": 0.19618382900953293,
      "num_tokens": 51364208.0,
      "step": 29615
    },
    {
      "entropy": 5.424306154251099,
      "epoch": 2.3045321921805098,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044635386301335666,
      "loss": 5.1717,
      "mean_token_accuracy": 0.18312696367502213,
      "num_tokens": 51373081.0,
      "step": 29620
    },
    {
      "entropy": 5.4239577293396,
      "epoch": 2.304921221552227,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004463358998103478,
      "loss": 5.0664,
      "mean_token_accuracy": 0.19712869822978973,
      "num_tokens": 51382561.0,
      "step": 29625
    },
    {
      "entropy": 5.480073022842407,
      "epoch": 2.3053102509239447,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004463179340075894,
      "loss": 5.1312,
      "mean_token_accuracy": 0.18804681301116943,
      "num_tokens": 51391671.0,
      "step": 29630
    },
    {
      "entropy": 5.382198143005371,
      "epoch": 2.3056992802956624,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004462999656053541,
      "loss": 4.977,
      "mean_token_accuracy": 0.1994564101099968,
      "num_tokens": 51400727.0,
      "step": 29635
    },
    {
      "entropy": 5.4393714427947994,
      "epoch": 2.3060883096673797,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044628199460391445,
      "loss": 5.1006,
      "mean_token_accuracy": 0.18588894605636597,
      "num_tokens": 51409943.0,
      "step": 29640
    },
    {
      "entropy": 5.432221412658691,
      "epoch": 2.3064773390390974,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044626402100354307,
      "loss": 5.1651,
      "mean_token_accuracy": 0.18300071805715562,
      "num_tokens": 51418511.0,
      "step": 29645
    },
    {
      "entropy": 5.375193166732788,
      "epoch": 2.306866368410815,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004462460448045129,
      "loss": 5.0003,
      "mean_token_accuracy": 0.1999703198671341,
      "num_tokens": 51426304.0,
      "step": 29650
    },
    {
      "entropy": 5.475950288772583,
      "epoch": 2.3072553977825327,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044622806600709645,
      "loss": 5.1028,
      "mean_token_accuracy": 0.1893684148788452,
      "num_tokens": 51434779.0,
      "step": 29655
    },
    {
      "entropy": 5.426788377761841,
      "epoch": 2.30764442715425,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004462100846115667,
      "loss": 5.0398,
      "mean_token_accuracy": 0.19934193342924117,
      "num_tokens": 51444275.0,
      "step": 29660
    },
    {
      "entropy": 5.452623319625855,
      "epoch": 2.3080334565259677,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004461921006181964,
      "loss": 5.0929,
      "mean_token_accuracy": 0.1869780734181404,
      "num_tokens": 51453149.0,
      "step": 29665
    },
    {
      "entropy": 5.398927831649781,
      "epoch": 2.3084224858976854,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004461741140272584,
      "loss": 5.0648,
      "mean_token_accuracy": 0.20120845288038253,
      "num_tokens": 51462302.0,
      "step": 29670
    },
    {
      "entropy": 5.465122270584106,
      "epoch": 2.3088115152694026,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004461561248390257,
      "loss": 5.1392,
      "mean_token_accuracy": 0.1798361822962761,
      "num_tokens": 51471186.0,
      "step": 29675
    },
    {
      "entropy": 5.434913778305054,
      "epoch": 2.3092005446411203,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004461381330537713,
      "loss": 5.0938,
      "mean_token_accuracy": 0.18915633112192154,
      "num_tokens": 51479868.0,
      "step": 29680
    },
    {
      "entropy": 5.342674350738525,
      "epoch": 2.309589574012838,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000446120138671768,
      "loss": 4.9715,
      "mean_token_accuracy": 0.19977009147405625,
      "num_tokens": 51489118.0,
      "step": 29685
    },
    {
      "entropy": 5.394464635848999,
      "epoch": 2.3099786033845557,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044610214169328913,
      "loss": 5.0252,
      "mean_token_accuracy": 0.19371685832738877,
      "num_tokens": 51497933.0,
      "step": 29690
    },
    {
      "entropy": 5.528307294845581,
      "epoch": 2.310367632756273,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004460841421186075,
      "loss": 5.0989,
      "mean_token_accuracy": 0.18722251057624817,
      "num_tokens": 51507046.0,
      "step": 29695
    },
    {
      "entropy": 5.474293518066406,
      "epoch": 2.3107566621279907,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004460661399479963,
      "loss": 5.0884,
      "mean_token_accuracy": 0.19430774599313735,
      "num_tokens": 51516558.0,
      "step": 29700
    },
    {
      "entropy": 5.313049745559693,
      "epoch": 2.3111456914997084,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044604813518172876,
      "loss": 4.9448,
      "mean_token_accuracy": 0.1992548882961273,
      "num_tokens": 51524411.0,
      "step": 29705
    },
    {
      "entropy": 5.362483978271484,
      "epoch": 2.3115347208714256,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044603012782007796,
      "loss": 5.0447,
      "mean_token_accuracy": 0.180847629904747,
      "num_tokens": 51533121.0,
      "step": 29710
    },
    {
      "entropy": 5.406411361694336,
      "epoch": 2.3119237502431433,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004460121178633172,
      "loss": 5.0855,
      "mean_token_accuracy": 0.18723346292972565,
      "num_tokens": 51541417.0,
      "step": 29715
    },
    {
      "entropy": 5.451214551925659,
      "epoch": 2.312312779614861,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044599410531171986,
      "loss": 4.9983,
      "mean_token_accuracy": 0.20454004406929016,
      "num_tokens": 51549444.0,
      "step": 29720
    },
    {
      "entropy": 5.42308292388916,
      "epoch": 2.3127018089865787,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000445976090165559,
      "loss": 5.1523,
      "mean_token_accuracy": 0.1887134313583374,
      "num_tokens": 51558090.0,
      "step": 29725
    },
    {
      "entropy": 5.4823503494262695,
      "epoch": 2.313090838358296,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004459580724251082,
      "loss": 5.1415,
      "mean_token_accuracy": 0.18770647943019866,
      "num_tokens": 51566794.0,
      "step": 29730
    },
    {
      "entropy": 5.348286151885986,
      "epoch": 2.3134798677300137,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044594005209064064,
      "loss": 4.9789,
      "mean_token_accuracy": 0.19568664878606795,
      "num_tokens": 51575411.0,
      "step": 29735
    },
    {
      "entropy": 5.419104719161988,
      "epoch": 2.313868897101731,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044592202916243003,
      "loss": 5.0688,
      "mean_token_accuracy": 0.1897635743021965,
      "num_tokens": 51584438.0,
      "step": 29740
    },
    {
      "entropy": 5.401909589767456,
      "epoch": 2.3142579264734486,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004459040036407495,
      "loss": 5.058,
      "mean_token_accuracy": 0.19565843939781188,
      "num_tokens": 51593517.0,
      "step": 29745
    },
    {
      "entropy": 5.422411870956421,
      "epoch": 2.3146469558451663,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004458859755258729,
      "loss": 5.0912,
      "mean_token_accuracy": 0.18768022060394288,
      "num_tokens": 51602312.0,
      "step": 29750
    },
    {
      "entropy": 5.365849733352661,
      "epoch": 2.315035985216884,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044586794481807357,
      "loss": 4.981,
      "mean_token_accuracy": 0.20302048325538635,
      "num_tokens": 51611021.0,
      "step": 29755
    },
    {
      "entropy": 5.384206056594849,
      "epoch": 2.3154250145886013,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044584991151762506,
      "loss": 5.056,
      "mean_token_accuracy": 0.1870503842830658,
      "num_tokens": 51620322.0,
      "step": 29760
    },
    {
      "entropy": 5.4131112575531,
      "epoch": 2.315814043960319,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004458318756248011,
      "loss": 5.0361,
      "mean_token_accuracy": 0.1950208678841591,
      "num_tokens": 51628626.0,
      "step": 29765
    },
    {
      "entropy": 5.364333391189575,
      "epoch": 2.3162030733320367,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044581383713987547,
      "loss": 5.0153,
      "mean_token_accuracy": 0.19559272676706313,
      "num_tokens": 51637868.0,
      "step": 29770
    },
    {
      "entropy": 5.37861967086792,
      "epoch": 2.316592102703754,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004457957960631216,
      "loss": 4.9896,
      "mean_token_accuracy": 0.1930706560611725,
      "num_tokens": 51645756.0,
      "step": 29775
    },
    {
      "entropy": 5.418609142303467,
      "epoch": 2.3169811320754716,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004457777523948134,
      "loss": 5.0896,
      "mean_token_accuracy": 0.18789514899253845,
      "num_tokens": 51654959.0,
      "step": 29780
    },
    {
      "entropy": 5.49388952255249,
      "epoch": 2.3173701614471893,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004457597061352247,
      "loss": 5.0949,
      "mean_token_accuracy": 0.19222895354032515,
      "num_tokens": 51663208.0,
      "step": 29785
    },
    {
      "entropy": 5.404866695404053,
      "epoch": 2.317759190818907,
      "grad_norm": 1.125,
      "learning_rate": 0.0004457416572846291,
      "loss": 5.1012,
      "mean_token_accuracy": 0.18278964459896088,
      "num_tokens": 51671627.0,
      "step": 29790
    },
    {
      "entropy": 5.306273508071899,
      "epoch": 2.3181482201906243,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004457236058433008,
      "loss": 4.9733,
      "mean_token_accuracy": 0.19995750486850739,
      "num_tokens": 51680870.0,
      "step": 29795
    },
    {
      "entropy": 5.45002121925354,
      "epoch": 2.318537249562342,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044570555181151333,
      "loss": 5.1707,
      "mean_token_accuracy": 0.18445187360048293,
      "num_tokens": 51690566.0,
      "step": 29800
    },
    {
      "entropy": 5.472688150405884,
      "epoch": 2.3189262789340597,
      "grad_norm": 1.203125,
      "learning_rate": 0.000445687495189541,
      "loss": 5.1004,
      "mean_token_accuracy": 0.19017809629440308,
      "num_tokens": 51698792.0,
      "step": 29805
    },
    {
      "entropy": 5.412845611572266,
      "epoch": 2.319315308305777,
      "grad_norm": 1.25,
      "learning_rate": 0.00044566943597765745,
      "loss": 5.0687,
      "mean_token_accuracy": 0.19169116914272308,
      "num_tokens": 51707821.0,
      "step": 29810
    },
    {
      "entropy": 5.347038888931275,
      "epoch": 2.3197043376774946,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044565137417613694,
      "loss": 5.02,
      "mean_token_accuracy": 0.1953651562333107,
      "num_tokens": 51716802.0,
      "step": 29815
    },
    {
      "entropy": 5.4139646053314205,
      "epoch": 2.3200933670492123,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004456333097852534,
      "loss": 5.0539,
      "mean_token_accuracy": 0.19388844668865204,
      "num_tokens": 51725644.0,
      "step": 29820
    },
    {
      "entropy": 5.4817342281341555,
      "epoch": 2.32048239642093,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044561524280528104,
      "loss": 5.1295,
      "mean_token_accuracy": 0.19017827957868577,
      "num_tokens": 51734072.0,
      "step": 29825
    },
    {
      "entropy": 5.459099435806275,
      "epoch": 2.3208714257926473,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044559717323649383,
      "loss": 5.0789,
      "mean_token_accuracy": 0.19487529695034028,
      "num_tokens": 51743112.0,
      "step": 29830
    },
    {
      "entropy": 5.428200149536133,
      "epoch": 2.321260455164365,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004455791010791662,
      "loss": 5.095,
      "mean_token_accuracy": 0.18801604807376862,
      "num_tokens": 51752535.0,
      "step": 29835
    },
    {
      "entropy": 5.415243005752563,
      "epoch": 2.3216494845360827,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004455610263335721,
      "loss": 5.0179,
      "mean_token_accuracy": 0.1962419182062149,
      "num_tokens": 51760595.0,
      "step": 29840
    },
    {
      "entropy": 5.4521159648895265,
      "epoch": 2.3220385139078,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004455429489999859,
      "loss": 5.1121,
      "mean_token_accuracy": 0.19558138251304627,
      "num_tokens": 51769156.0,
      "step": 29845
    },
    {
      "entropy": 5.452931213378906,
      "epoch": 2.3224275432795176,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044552486907868194,
      "loss": 5.0334,
      "mean_token_accuracy": 0.18486237823963164,
      "num_tokens": 51778091.0,
      "step": 29850
    },
    {
      "entropy": 5.44926438331604,
      "epoch": 2.3228165726512353,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044550678656993454,
      "loss": 5.0588,
      "mean_token_accuracy": 0.19071279764175414,
      "num_tokens": 51787207.0,
      "step": 29855
    },
    {
      "entropy": 5.309511661529541,
      "epoch": 2.3232056020229526,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000445488701474018,
      "loss": 4.943,
      "mean_token_accuracy": 0.20208046436309815,
      "num_tokens": 51796158.0,
      "step": 29860
    },
    {
      "entropy": 5.346232080459595,
      "epoch": 2.3235946313946703,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004454706137912067,
      "loss": 4.8967,
      "mean_token_accuracy": 0.2046591207385063,
      "num_tokens": 51803762.0,
      "step": 29865
    },
    {
      "entropy": 5.471810150146484,
      "epoch": 2.323983660766388,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004454525235217753,
      "loss": 5.0812,
      "mean_token_accuracy": 0.18785480111837388,
      "num_tokens": 51812610.0,
      "step": 29870
    },
    {
      "entropy": 5.393780612945557,
      "epoch": 2.324372690138105,
      "grad_norm": 1.25,
      "learning_rate": 0.00044543443066599807,
      "loss": 5.0745,
      "mean_token_accuracy": 0.18588633239269256,
      "num_tokens": 51821398.0,
      "step": 29875
    },
    {
      "entropy": 5.400189685821533,
      "epoch": 2.324761719509823,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004454163352241498,
      "loss": 5.015,
      "mean_token_accuracy": 0.1869928225874901,
      "num_tokens": 51829945.0,
      "step": 29880
    },
    {
      "entropy": 5.420876598358154,
      "epoch": 2.3251507488815406,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044539823719650463,
      "loss": 5.0554,
      "mean_token_accuracy": 0.19794220626354217,
      "num_tokens": 51838816.0,
      "step": 29885
    },
    {
      "entropy": 5.3814856052398685,
      "epoch": 2.3255397782532583,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004453801365833376,
      "loss": 5.0635,
      "mean_token_accuracy": 0.19703920781612397,
      "num_tokens": 51847775.0,
      "step": 29890
    },
    {
      "entropy": 5.459873628616333,
      "epoch": 2.3259288076249756,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044536203338492317,
      "loss": 5.0187,
      "mean_token_accuracy": 0.1896474301815033,
      "num_tokens": 51856934.0,
      "step": 29895
    },
    {
      "entropy": 5.514960908889771,
      "epoch": 2.3263178369966933,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044534392760153596,
      "loss": 5.1112,
      "mean_token_accuracy": 0.17878314852714539,
      "num_tokens": 51865393.0,
      "step": 29900
    },
    {
      "entropy": 5.354064416885376,
      "epoch": 2.326706866368411,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004453258192334508,
      "loss": 4.9932,
      "mean_token_accuracy": 0.20275797992944716,
      "num_tokens": 51873783.0,
      "step": 29905
    },
    {
      "entropy": 5.343626546859741,
      "epoch": 2.327095895740128,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004453077082809425,
      "loss": 5.1015,
      "mean_token_accuracy": 0.18707127571105958,
      "num_tokens": 51882934.0,
      "step": 29910
    },
    {
      "entropy": 5.397631454467773,
      "epoch": 2.327484925111846,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044528959474428575,
      "loss": 5.0088,
      "mean_token_accuracy": 0.19322898387908935,
      "num_tokens": 51891217.0,
      "step": 29915
    },
    {
      "entropy": 5.460697174072266,
      "epoch": 2.3278739544835636,
      "grad_norm": 1.125,
      "learning_rate": 0.0004452714786237555,
      "loss": 5.1068,
      "mean_token_accuracy": 0.18826145976781844,
      "num_tokens": 51900550.0,
      "step": 29920
    },
    {
      "entropy": 5.418945217132569,
      "epoch": 2.3282629838552813,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004452533599196265,
      "loss": 5.0507,
      "mean_token_accuracy": 0.19013779610395432,
      "num_tokens": 51908436.0,
      "step": 29925
    },
    {
      "entropy": 5.4354567527771,
      "epoch": 2.3286520132269986,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044523523863217374,
      "loss": 5.1102,
      "mean_token_accuracy": 0.1913825437426567,
      "num_tokens": 51917687.0,
      "step": 29930
    },
    {
      "entropy": 5.433904933929443,
      "epoch": 2.3290410425987162,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004452171147616723,
      "loss": 5.1344,
      "mean_token_accuracy": 0.18713640421628952,
      "num_tokens": 51927011.0,
      "step": 29935
    },
    {
      "entropy": 5.526001119613648,
      "epoch": 2.329430071970434,
      "grad_norm": 1.125,
      "learning_rate": 0.000445198988308397,
      "loss": 5.1336,
      "mean_token_accuracy": 0.1826961010694504,
      "num_tokens": 51936255.0,
      "step": 29940
    },
    {
      "entropy": 5.364567136764526,
      "epoch": 2.329819101342151,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044518085927262293,
      "loss": 5.0752,
      "mean_token_accuracy": 0.19084919840097428,
      "num_tokens": 51944919.0,
      "step": 29945
    },
    {
      "entropy": 5.459301471710205,
      "epoch": 2.330208130713869,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004451627276546252,
      "loss": 5.1167,
      "mean_token_accuracy": 0.19054297655820845,
      "num_tokens": 51952963.0,
      "step": 29950
    },
    {
      "entropy": 5.411231231689453,
      "epoch": 2.3305971600855866,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000445144593454679,
      "loss": 5.0086,
      "mean_token_accuracy": 0.1937964752316475,
      "num_tokens": 51961493.0,
      "step": 29955
    },
    {
      "entropy": 5.354438543319702,
      "epoch": 2.330986189457304,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004451264566730593,
      "loss": 5.103,
      "mean_token_accuracy": 0.18868838101625443,
      "num_tokens": 51969911.0,
      "step": 29960
    },
    {
      "entropy": 5.415657043457031,
      "epoch": 2.3313752188290215,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044510831731004146,
      "loss": 5.0021,
      "mean_token_accuracy": 0.19032670110464095,
      "num_tokens": 51978936.0,
      "step": 29965
    },
    {
      "entropy": 5.521789026260376,
      "epoch": 2.3317642482007392,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004450901753659007,
      "loss": 5.0717,
      "mean_token_accuracy": 0.1853356584906578,
      "num_tokens": 51987373.0,
      "step": 29970
    },
    {
      "entropy": 5.3963470458984375,
      "epoch": 2.3321532775724565,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044507203084091215,
      "loss": 5.1096,
      "mean_token_accuracy": 0.18220414519309996,
      "num_tokens": 51995948.0,
      "step": 29975
    },
    {
      "entropy": 5.404459857940674,
      "epoch": 2.332542306944174,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004450538837353513,
      "loss": 5.082,
      "mean_token_accuracy": 0.18861262649297714,
      "num_tokens": 52004623.0,
      "step": 29980
    },
    {
      "entropy": 5.489960956573486,
      "epoch": 2.332931336315892,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044503573404949343,
      "loss": 5.0578,
      "mean_token_accuracy": 0.1924934834241867,
      "num_tokens": 52012971.0,
      "step": 29985
    },
    {
      "entropy": 5.438990879058838,
      "epoch": 2.3333203656876096,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004450175817836139,
      "loss": 5.0913,
      "mean_token_accuracy": 0.1909508928656578,
      "num_tokens": 52021462.0,
      "step": 29990
    },
    {
      "entropy": 5.344920682907104,
      "epoch": 2.333709395059327,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044499942693798823,
      "loss": 4.9899,
      "mean_token_accuracy": 0.1946091026067734,
      "num_tokens": 52030494.0,
      "step": 29995
    },
    {
      "entropy": 5.410649108886719,
      "epoch": 2.3340984244310445,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004449812695128918,
      "loss": 5.0144,
      "mean_token_accuracy": 0.19331903457641603,
      "num_tokens": 52039120.0,
      "step": 30000
    },
    {
      "epoch": 2.3340984244310445,
      "eval_entropy": 5.278062728187921,
      "eval_loss": 5.173376083374023,
      "eval_mean_token_accuracy": 0.19548090663994014,
      "eval_num_tokens": 52039120.0,
      "eval_runtime": 29.4602,
      "eval_samples_per_second": 1410.649,
      "eval_steps_per_second": 176.34,
      "step": 30000
    },
    {
      "entropy": 5.531785535812378,
      "epoch": 2.3344874538027622,
      "grad_norm": 1.25,
      "learning_rate": 0.00044496310950860015,
      "loss": 5.134,
      "mean_token_accuracy": 0.18347583562135697,
      "num_tokens": 52047248.0,
      "step": 30005
    },
    {
      "entropy": 5.412385892868042,
      "epoch": 2.3348764831744795,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044494494692538886,
      "loss": 5.0592,
      "mean_token_accuracy": 0.1891682118177414,
      "num_tokens": 52056359.0,
      "step": 30010
    },
    {
      "entropy": 5.458208751678467,
      "epoch": 2.335265512546197,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044492678176353347,
      "loss": 5.0523,
      "mean_token_accuracy": 0.19029513299465178,
      "num_tokens": 52065042.0,
      "step": 30015
    },
    {
      "entropy": 5.350721406936645,
      "epoch": 2.335654541917915,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044490861402330967,
      "loss": 4.926,
      "mean_token_accuracy": 0.203727188706398,
      "num_tokens": 52073830.0,
      "step": 30020
    },
    {
      "entropy": 5.285924482345581,
      "epoch": 2.3360435712896326,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004448904437049931,
      "loss": 4.91,
      "mean_token_accuracy": 0.1948888823390007,
      "num_tokens": 52082605.0,
      "step": 30025
    },
    {
      "entropy": 5.42687578201294,
      "epoch": 2.33643260066135,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004448722708088594,
      "loss": 5.209,
      "mean_token_accuracy": 0.17946571111679077,
      "num_tokens": 52091997.0,
      "step": 30030
    },
    {
      "entropy": 5.425217008590698,
      "epoch": 2.3368216300330675,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004448540953351844,
      "loss": 5.0481,
      "mean_token_accuracy": 0.19460586160421373,
      "num_tokens": 52100690.0,
      "step": 30035
    },
    {
      "entropy": 5.398977279663086,
      "epoch": 2.3372106594047852,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004448359172842439,
      "loss": 5.0657,
      "mean_token_accuracy": 0.19714615792036055,
      "num_tokens": 52109310.0,
      "step": 30040
    },
    {
      "entropy": 5.484135532379151,
      "epoch": 2.3375996887765025,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044481773665631355,
      "loss": 5.0951,
      "mean_token_accuracy": 0.20113715529441833,
      "num_tokens": 52118662.0,
      "step": 30045
    },
    {
      "entropy": 5.309747982025146,
      "epoch": 2.33798871814822,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004447995534516695,
      "loss": 4.9766,
      "mean_token_accuracy": 0.2025471344590187,
      "num_tokens": 52127098.0,
      "step": 30050
    },
    {
      "entropy": 5.3607337951660154,
      "epoch": 2.338377747519938,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044478136767058733,
      "loss": 5.0899,
      "mean_token_accuracy": 0.18626188784837722,
      "num_tokens": 52135539.0,
      "step": 30055
    },
    {
      "entropy": 5.522019386291504,
      "epoch": 2.338766776891655,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004447631793133432,
      "loss": 5.0718,
      "mean_token_accuracy": 0.19150737971067427,
      "num_tokens": 52144169.0,
      "step": 30060
    },
    {
      "entropy": 5.432363653182984,
      "epoch": 2.339155806263373,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004447449883802131,
      "loss": 5.0807,
      "mean_token_accuracy": 0.18884508907794953,
      "num_tokens": 52152679.0,
      "step": 30065
    },
    {
      "entropy": 5.332374334335327,
      "epoch": 2.3395448356350905,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044472679487147295,
      "loss": 5.0168,
      "mean_token_accuracy": 0.19738177061080933,
      "num_tokens": 52161407.0,
      "step": 30070
    },
    {
      "entropy": 5.539660453796387,
      "epoch": 2.339933865006808,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044470859878739877,
      "loss": 5.0821,
      "mean_token_accuracy": 0.18288897275924682,
      "num_tokens": 52170285.0,
      "step": 30075
    },
    {
      "entropy": 5.42182879447937,
      "epoch": 2.3403228943785255,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044469040012826676,
      "loss": 4.9325,
      "mean_token_accuracy": 0.20403649359941484,
      "num_tokens": 52178813.0,
      "step": 30080
    },
    {
      "entropy": 5.327577924728393,
      "epoch": 2.340711923750243,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044467219889435304,
      "loss": 4.9349,
      "mean_token_accuracy": 0.19863856881856917,
      "num_tokens": 52187205.0,
      "step": 30085
    },
    {
      "entropy": 5.393527030944824,
      "epoch": 2.341100953121961,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004446539950859337,
      "loss": 5.0701,
      "mean_token_accuracy": 0.18757230639457703,
      "num_tokens": 52195318.0,
      "step": 30090
    },
    {
      "entropy": 5.340764045715332,
      "epoch": 2.341489982493678,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044463578870328506,
      "loss": 5.002,
      "mean_token_accuracy": 0.19450665563344954,
      "num_tokens": 52204093.0,
      "step": 30095
    },
    {
      "entropy": 5.369998455047607,
      "epoch": 2.341879011865396,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004446175797466834,
      "loss": 5.0883,
      "mean_token_accuracy": 0.19455251097679138,
      "num_tokens": 52212319.0,
      "step": 30100
    },
    {
      "entropy": 5.418159151077271,
      "epoch": 2.3422680412371135,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044459936821640485,
      "loss": 5.0361,
      "mean_token_accuracy": 0.1892719805240631,
      "num_tokens": 52221878.0,
      "step": 30105
    },
    {
      "entropy": 5.3445008277893065,
      "epoch": 2.3426570706088308,
      "grad_norm": 1.125,
      "learning_rate": 0.0004445811541127258,
      "loss": 4.9596,
      "mean_token_accuracy": 0.19711434245109558,
      "num_tokens": 52230952.0,
      "step": 30110
    },
    {
      "entropy": 5.385547924041748,
      "epoch": 2.3430460999805485,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044456293743592274,
      "loss": 5.0551,
      "mean_token_accuracy": 0.19889726340770722,
      "num_tokens": 52239461.0,
      "step": 30115
    },
    {
      "entropy": 5.421613073348999,
      "epoch": 2.343435129352266,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044454471818627196,
      "loss": 5.0408,
      "mean_token_accuracy": 0.19032577723264693,
      "num_tokens": 52248478.0,
      "step": 30120
    },
    {
      "entropy": 5.433068895339966,
      "epoch": 2.343824158723984,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044452649636404995,
      "loss": 5.0923,
      "mean_token_accuracy": 0.19105210900306702,
      "num_tokens": 52256901.0,
      "step": 30125
    },
    {
      "entropy": 5.436404514312744,
      "epoch": 2.344213188095701,
      "grad_norm": 1.125,
      "learning_rate": 0.0004445082719695332,
      "loss": 5.0501,
      "mean_token_accuracy": 0.1933584541082382,
      "num_tokens": 52265249.0,
      "step": 30130
    },
    {
      "entropy": 5.440950012207031,
      "epoch": 2.344602217467419,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044449004500299815,
      "loss": 5.0823,
      "mean_token_accuracy": 0.18953050225973128,
      "num_tokens": 52274777.0,
      "step": 30135
    },
    {
      "entropy": 5.447582387924195,
      "epoch": 2.3449912468391365,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044447181546472146,
      "loss": 5.1117,
      "mean_token_accuracy": 0.18741219937801362,
      "num_tokens": 52283392.0,
      "step": 30140
    },
    {
      "entropy": 5.339155817031861,
      "epoch": 2.3453802762108538,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044445358335497975,
      "loss": 4.9898,
      "mean_token_accuracy": 0.19315367192029953,
      "num_tokens": 52291875.0,
      "step": 30145
    },
    {
      "entropy": 5.363594579696655,
      "epoch": 2.3457693055825715,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004444353486740496,
      "loss": 5.1246,
      "mean_token_accuracy": 0.19235342890024185,
      "num_tokens": 52300194.0,
      "step": 30150
    },
    {
      "entropy": 5.4511157989501955,
      "epoch": 2.346158334954289,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004444171114222078,
      "loss": 5.0726,
      "mean_token_accuracy": 0.19239448457956315,
      "num_tokens": 52309147.0,
      "step": 30155
    },
    {
      "entropy": 5.513557815551758,
      "epoch": 2.346547364326007,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044439887159973096,
      "loss": 5.155,
      "mean_token_accuracy": 0.18252138346433638,
      "num_tokens": 52317502.0,
      "step": 30160
    },
    {
      "entropy": 5.37524151802063,
      "epoch": 2.346936393697724,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004443806292068958,
      "loss": 4.949,
      "mean_token_accuracy": 0.20163785368204118,
      "num_tokens": 52325539.0,
      "step": 30165
    },
    {
      "entropy": 5.42692379951477,
      "epoch": 2.347325423069442,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004443623842439792,
      "loss": 5.0311,
      "mean_token_accuracy": 0.1920914828777313,
      "num_tokens": 52333676.0,
      "step": 30170
    },
    {
      "entropy": 5.413420486450195,
      "epoch": 2.347714452441159,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000444344136711258,
      "loss": 4.9939,
      "mean_token_accuracy": 0.20066285282373428,
      "num_tokens": 52342706.0,
      "step": 30175
    },
    {
      "entropy": 5.377599811553955,
      "epoch": 2.3481034818128768,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044432588660900905,
      "loss": 5.0697,
      "mean_token_accuracy": 0.1919751435518265,
      "num_tokens": 52351409.0,
      "step": 30180
    },
    {
      "entropy": 5.331076145172119,
      "epoch": 2.3484925111845945,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004443076339375093,
      "loss": 5.008,
      "mean_token_accuracy": 0.19617386013269425,
      "num_tokens": 52359781.0,
      "step": 30185
    },
    {
      "entropy": 5.386291790008545,
      "epoch": 2.348881540556312,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044428937869703575,
      "loss": 5.0125,
      "mean_token_accuracy": 0.19278571754693985,
      "num_tokens": 52368236.0,
      "step": 30190
    },
    {
      "entropy": 5.419387674331665,
      "epoch": 2.3492705699280294,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044427112088786525,
      "loss": 5.0603,
      "mean_token_accuracy": 0.19301276206970214,
      "num_tokens": 52376449.0,
      "step": 30195
    },
    {
      "entropy": 5.425244474411011,
      "epoch": 2.349659599299747,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004442528605102749,
      "loss": 5.0057,
      "mean_token_accuracy": 0.19385803788900374,
      "num_tokens": 52384844.0,
      "step": 30200
    },
    {
      "entropy": 5.4809102535247805,
      "epoch": 2.350048628671465,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004442345975645418,
      "loss": 5.166,
      "mean_token_accuracy": 0.1900211215019226,
      "num_tokens": 52393810.0,
      "step": 30205
    },
    {
      "entropy": 5.473064804077149,
      "epoch": 2.350437658043182,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004442163320509431,
      "loss": 5.0292,
      "mean_token_accuracy": 0.19601163417100906,
      "num_tokens": 52402081.0,
      "step": 30210
    },
    {
      "entropy": 5.3551130294799805,
      "epoch": 2.3508266874148998,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004441980639697558,
      "loss": 5.0179,
      "mean_token_accuracy": 0.19238125234842302,
      "num_tokens": 52411671.0,
      "step": 30215
    },
    {
      "entropy": 5.331557321548462,
      "epoch": 2.3512157167866174,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004441797933212573,
      "loss": 5.0608,
      "mean_token_accuracy": 0.1930033251643181,
      "num_tokens": 52420255.0,
      "step": 30220
    },
    {
      "entropy": 5.49574728012085,
      "epoch": 2.351604746158335,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004441615201057247,
      "loss": 5.2019,
      "mean_token_accuracy": 0.1814395323395729,
      "num_tokens": 52429682.0,
      "step": 30225
    },
    {
      "entropy": 5.505518436431885,
      "epoch": 2.3519937755300524,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004441432443234352,
      "loss": 5.0533,
      "mean_token_accuracy": 0.19414226710796356,
      "num_tokens": 52438719.0,
      "step": 30230
    },
    {
      "entropy": 5.326932907104492,
      "epoch": 2.35238280490177,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044412496597466626,
      "loss": 4.948,
      "mean_token_accuracy": 0.20036731660366058,
      "num_tokens": 52447543.0,
      "step": 30235
    },
    {
      "entropy": 5.374056482315064,
      "epoch": 2.352771834273488,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004441066850596951,
      "loss": 4.9984,
      "mean_token_accuracy": 0.19825717359781264,
      "num_tokens": 52456606.0,
      "step": 30240
    },
    {
      "entropy": 5.449829244613648,
      "epoch": 2.353160863645205,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004440884015787992,
      "loss": 5.0833,
      "mean_token_accuracy": 0.19108119159936904,
      "num_tokens": 52465222.0,
      "step": 30245
    },
    {
      "entropy": 5.421074914932251,
      "epoch": 2.3535498930169227,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004440701155322558,
      "loss": 5.0942,
      "mean_token_accuracy": 0.19033259600400926,
      "num_tokens": 52473955.0,
      "step": 30250
    },
    {
      "entropy": 5.393550205230713,
      "epoch": 2.3539389223886404,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004440518269203427,
      "loss": 5.1059,
      "mean_token_accuracy": 0.19346344023942946,
      "num_tokens": 52483786.0,
      "step": 30255
    },
    {
      "entropy": 5.495344448089599,
      "epoch": 2.354327951760358,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044403353574333715,
      "loss": 5.0923,
      "mean_token_accuracy": 0.18991378843784332,
      "num_tokens": 52492292.0,
      "step": 30260
    },
    {
      "entropy": 5.478783893585205,
      "epoch": 2.3547169811320754,
      "grad_norm": 1.25,
      "learning_rate": 0.00044401524200151667,
      "loss": 5.0639,
      "mean_token_accuracy": 0.18241613060235978,
      "num_tokens": 52501267.0,
      "step": 30265
    },
    {
      "entropy": 5.488956165313721,
      "epoch": 2.355106010503793,
      "grad_norm": 1.25,
      "learning_rate": 0.00044399694569515895,
      "loss": 5.1135,
      "mean_token_accuracy": 0.19408838003873824,
      "num_tokens": 52509515.0,
      "step": 30270
    },
    {
      "entropy": 5.399058723449707,
      "epoch": 2.355495039875511,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044397864682454156,
      "loss": 5.0789,
      "mean_token_accuracy": 0.19373828768730164,
      "num_tokens": 52517757.0,
      "step": 30275
    },
    {
      "entropy": 5.433415031433105,
      "epoch": 2.355884069247228,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004439603453899421,
      "loss": 5.1406,
      "mean_token_accuracy": 0.19351146817207338,
      "num_tokens": 52525996.0,
      "step": 30280
    },
    {
      "entropy": 5.42435507774353,
      "epoch": 2.3562730986189457,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004439420413916384,
      "loss": 4.9952,
      "mean_token_accuracy": 0.2008293554186821,
      "num_tokens": 52534457.0,
      "step": 30285
    },
    {
      "entropy": 5.44274411201477,
      "epoch": 2.3566621279906634,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044392373482990817,
      "loss": 5.1355,
      "mean_token_accuracy": 0.1857638865709305,
      "num_tokens": 52543754.0,
      "step": 30290
    },
    {
      "entropy": 5.482491159439087,
      "epoch": 2.3570511573623807,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004439054257050291,
      "loss": 5.0827,
      "mean_token_accuracy": 0.18308962881565094,
      "num_tokens": 52552857.0,
      "step": 30295
    },
    {
      "entropy": 5.501236200332642,
      "epoch": 2.3574401867340984,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004438871140172789,
      "loss": 5.1743,
      "mean_token_accuracy": 0.18104863166809082,
      "num_tokens": 52562760.0,
      "step": 30300
    },
    {
      "entropy": 5.346593236923217,
      "epoch": 2.357829216105816,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004438687997669357,
      "loss": 4.9831,
      "mean_token_accuracy": 0.19781598597764968,
      "num_tokens": 52571111.0,
      "step": 30305
    },
    {
      "entropy": 5.365532875061035,
      "epoch": 2.3582182454775333,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044385048295427724,
      "loss": 4.9684,
      "mean_token_accuracy": 0.19764413982629775,
      "num_tokens": 52579541.0,
      "step": 30310
    },
    {
      "entropy": 5.38826355934143,
      "epoch": 2.358607274849251,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004438321635795814,
      "loss": 5.0541,
      "mean_token_accuracy": 0.2004312038421631,
      "num_tokens": 52588253.0,
      "step": 30315
    },
    {
      "entropy": 5.356149673461914,
      "epoch": 2.3589963042209687,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004438138416431262,
      "loss": 5.0171,
      "mean_token_accuracy": 0.19001378268003463,
      "num_tokens": 52597922.0,
      "step": 30320
    },
    {
      "entropy": 5.4275548458099365,
      "epoch": 2.3593853335926864,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044379551714518966,
      "loss": 4.9358,
      "mean_token_accuracy": 0.19580564498901368,
      "num_tokens": 52606167.0,
      "step": 30325
    },
    {
      "entropy": 5.422748756408692,
      "epoch": 2.3597743629644037,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044377719008604984,
      "loss": 5.165,
      "mean_token_accuracy": 0.183042074739933,
      "num_tokens": 52614534.0,
      "step": 30330
    },
    {
      "entropy": 5.437446403503418,
      "epoch": 2.3601633923361214,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004437588604659848,
      "loss": 5.0549,
      "mean_token_accuracy": 0.18915238827466965,
      "num_tokens": 52623225.0,
      "step": 30335
    },
    {
      "entropy": 5.444179058074951,
      "epoch": 2.360552421707839,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004437405282852726,
      "loss": 5.0968,
      "mean_token_accuracy": 0.1912812516093254,
      "num_tokens": 52632025.0,
      "step": 30340
    },
    {
      "entropy": 5.36928973197937,
      "epoch": 2.3609414510795563,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044372219354419154,
      "loss": 4.9113,
      "mean_token_accuracy": 0.21283333599567414,
      "num_tokens": 52640731.0,
      "step": 30345
    },
    {
      "entropy": 5.3932994365692135,
      "epoch": 2.361330480451274,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004437038562430197,
      "loss": 5.029,
      "mean_token_accuracy": 0.19691042751073837,
      "num_tokens": 52648831.0,
      "step": 30350
    },
    {
      "entropy": 5.428686571121216,
      "epoch": 2.3617195098229917,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044368551638203536,
      "loss": 5.0808,
      "mean_token_accuracy": 0.1870300754904747,
      "num_tokens": 52657699.0,
      "step": 30355
    },
    {
      "entropy": 5.355464887619019,
      "epoch": 2.3621085391947094,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004436671739615168,
      "loss": 4.9579,
      "mean_token_accuracy": 0.19669124335050583,
      "num_tokens": 52666418.0,
      "step": 30360
    },
    {
      "entropy": 5.365160179138184,
      "epoch": 2.3624975685664267,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044364882898174235,
      "loss": 5.0406,
      "mean_token_accuracy": 0.19268265813589097,
      "num_tokens": 52675014.0,
      "step": 30365
    },
    {
      "entropy": 5.462120914459229,
      "epoch": 2.3628865979381444,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004436304814429903,
      "loss": 5.0696,
      "mean_token_accuracy": 0.18719243556261062,
      "num_tokens": 52683691.0,
      "step": 30370
    },
    {
      "entropy": 5.47960524559021,
      "epoch": 2.363275627309862,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004436121313455391,
      "loss": 5.0045,
      "mean_token_accuracy": 0.18897877186536788,
      "num_tokens": 52692317.0,
      "step": 30375
    },
    {
      "entropy": 5.365514755249023,
      "epoch": 2.3636646566815793,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004435937786896673,
      "loss": 5.0533,
      "mean_token_accuracy": 0.1879978820681572,
      "num_tokens": 52701523.0,
      "step": 30380
    },
    {
      "entropy": 5.461100244522095,
      "epoch": 2.364053686053297,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044357542347565323,
      "loss": 5.1415,
      "mean_token_accuracy": 0.18612246662378312,
      "num_tokens": 52710436.0,
      "step": 30385
    },
    {
      "entropy": 5.454371929168701,
      "epoch": 2.3644427154250147,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004435570657037753,
      "loss": 5.11,
      "mean_token_accuracy": 0.19020272493362428,
      "num_tokens": 52719186.0,
      "step": 30390
    },
    {
      "entropy": 5.396678686141968,
      "epoch": 2.364831744796732,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004435387053743123,
      "loss": 5.0598,
      "mean_token_accuracy": 0.19319379776716233,
      "num_tokens": 52728185.0,
      "step": 30395
    },
    {
      "entropy": 5.43155083656311,
      "epoch": 2.3652207741684497,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044352034248754265,
      "loss": 5.124,
      "mean_token_accuracy": 0.194413959980011,
      "num_tokens": 52737938.0,
      "step": 30400
    },
    {
      "entropy": 5.400070381164551,
      "epoch": 2.3656098035401674,
      "grad_norm": 1.1875,
      "learning_rate": 0.000443501977043745,
      "loss": 5.0107,
      "mean_token_accuracy": 0.1964032232761383,
      "num_tokens": 52747005.0,
      "step": 30405
    },
    {
      "entropy": 5.348212671279907,
      "epoch": 2.3659988329118846,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004434836090431981,
      "loss": 5.003,
      "mean_token_accuracy": 0.19062833487987518,
      "num_tokens": 52756251.0,
      "step": 30410
    },
    {
      "entropy": 5.40240888595581,
      "epoch": 2.3663878622836023,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004434652384861806,
      "loss": 5.0218,
      "mean_token_accuracy": 0.19201731234788894,
      "num_tokens": 52764443.0,
      "step": 30415
    },
    {
      "entropy": 5.355072069168091,
      "epoch": 2.36677689165532,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004434468653729712,
      "loss": 4.9232,
      "mean_token_accuracy": 0.2031481072306633,
      "num_tokens": 52772824.0,
      "step": 30420
    },
    {
      "entropy": 5.373838472366333,
      "epoch": 2.3671659210270377,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044342848970384876,
      "loss": 5.0582,
      "mean_token_accuracy": 0.1894853338599205,
      "num_tokens": 52782743.0,
      "step": 30425
    },
    {
      "entropy": 5.486165142059326,
      "epoch": 2.367554950398755,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044341011147909206,
      "loss": 5.2017,
      "mean_token_accuracy": 0.1820221722126007,
      "num_tokens": 52791564.0,
      "step": 30430
    },
    {
      "entropy": 5.415500211715698,
      "epoch": 2.3679439797704727,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044339173069897996,
      "loss": 4.9991,
      "mean_token_accuracy": 0.19910618364810945,
      "num_tokens": 52800229.0,
      "step": 30435
    },
    {
      "entropy": 5.406668567657471,
      "epoch": 2.3683330091421904,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044337334736379143,
      "loss": 4.9958,
      "mean_token_accuracy": 0.19860171526670456,
      "num_tokens": 52808342.0,
      "step": 30440
    },
    {
      "entropy": 5.386832523345947,
      "epoch": 2.3687220385139076,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004433549614738053,
      "loss": 5.0748,
      "mean_token_accuracy": 0.1946864977478981,
      "num_tokens": 52817462.0,
      "step": 30445
    },
    {
      "entropy": 5.351207065582275,
      "epoch": 2.3691110678856253,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044333657302930056,
      "loss": 4.9747,
      "mean_token_accuracy": 0.1973209038376808,
      "num_tokens": 52825567.0,
      "step": 30450
    },
    {
      "entropy": 5.400139951705933,
      "epoch": 2.369500097257343,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004433181820305564,
      "loss": 5.052,
      "mean_token_accuracy": 0.1825938493013382,
      "num_tokens": 52835771.0,
      "step": 30455
    },
    {
      "entropy": 5.372045516967773,
      "epoch": 2.3698891266290607,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044329978847785156,
      "loss": 5.0288,
      "mean_token_accuracy": 0.19163314253091812,
      "num_tokens": 52844344.0,
      "step": 30460
    },
    {
      "entropy": 5.451956558227539,
      "epoch": 2.370278156000778,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004432813923714654,
      "loss": 5.0553,
      "mean_token_accuracy": 0.19553797096014022,
      "num_tokens": 52852918.0,
      "step": 30465
    },
    {
      "entropy": 5.393067455291748,
      "epoch": 2.3706671853724957,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044326299371167695,
      "loss": 5.092,
      "mean_token_accuracy": 0.19259517192840575,
      "num_tokens": 52862142.0,
      "step": 30470
    },
    {
      "entropy": 5.427189826965332,
      "epoch": 2.3710562147442134,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044324459249876536,
      "loss": 5.0915,
      "mean_token_accuracy": 0.18962078988552095,
      "num_tokens": 52870595.0,
      "step": 30475
    },
    {
      "entropy": 5.512401580810547,
      "epoch": 2.3714452441159306,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004432261887330099,
      "loss": 5.173,
      "mean_token_accuracy": 0.1887656942009926,
      "num_tokens": 52879995.0,
      "step": 30480
    },
    {
      "entropy": 5.402894783020019,
      "epoch": 2.3718342734876483,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004432077824146898,
      "loss": 4.9765,
      "mean_token_accuracy": 0.20357853025197983,
      "num_tokens": 52888341.0,
      "step": 30485
    },
    {
      "entropy": 5.345517349243164,
      "epoch": 2.372223302859366,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004431893735440843,
      "loss": 4.9294,
      "mean_token_accuracy": 0.20372711420059203,
      "num_tokens": 52896617.0,
      "step": 30490
    },
    {
      "entropy": 5.350107288360595,
      "epoch": 2.3726123322310833,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004431709621214727,
      "loss": 4.9685,
      "mean_token_accuracy": 0.19610204100608825,
      "num_tokens": 52905071.0,
      "step": 30495
    },
    {
      "entropy": 5.3858555316925045,
      "epoch": 2.373001361602801,
      "grad_norm": 1.25,
      "learning_rate": 0.00044315254814713455,
      "loss": 4.9835,
      "mean_token_accuracy": 0.19958054572343825,
      "num_tokens": 52913756.0,
      "step": 30500
    },
    {
      "entropy": 5.374010181427002,
      "epoch": 2.3733903909745186,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044313413162134894,
      "loss": 4.9766,
      "mean_token_accuracy": 0.20740654766559602,
      "num_tokens": 52922895.0,
      "step": 30505
    },
    {
      "entropy": 5.422125720977784,
      "epoch": 2.373779420346236,
      "grad_norm": 1.25,
      "learning_rate": 0.0004431157125443958,
      "loss": 5.1075,
      "mean_token_accuracy": 0.18850268870592118,
      "num_tokens": 52931467.0,
      "step": 30510
    },
    {
      "entropy": 5.4414475440979,
      "epoch": 2.3741684497179536,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000443097290916554,
      "loss": 5.1719,
      "mean_token_accuracy": 0.18432148844003676,
      "num_tokens": 52941122.0,
      "step": 30515
    },
    {
      "entropy": 5.4889003276824955,
      "epoch": 2.3745574790896713,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004430788667381035,
      "loss": 5.1226,
      "mean_token_accuracy": 0.18984979689121245,
      "num_tokens": 52949881.0,
      "step": 30520
    },
    {
      "entropy": 5.379704141616822,
      "epoch": 2.374946508461389,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044306044000932374,
      "loss": 4.957,
      "mean_token_accuracy": 0.19538653790950775,
      "num_tokens": 52957836.0,
      "step": 30525
    },
    {
      "entropy": 5.384369039535523,
      "epoch": 2.3753355378331062,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004430420107304943,
      "loss": 5.0102,
      "mean_token_accuracy": 0.19991744607686995,
      "num_tokens": 52965948.0,
      "step": 30530
    },
    {
      "entropy": 5.417372131347657,
      "epoch": 2.375724567204824,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044302357890189475,
      "loss": 5.0166,
      "mean_token_accuracy": 0.19412731975317002,
      "num_tokens": 52974561.0,
      "step": 30535
    },
    {
      "entropy": 5.355967426300049,
      "epoch": 2.3761135965765416,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004430051445238049,
      "loss": 4.9443,
      "mean_token_accuracy": 0.20216686129570008,
      "num_tokens": 52982918.0,
      "step": 30540
    },
    {
      "entropy": 5.4079913139343265,
      "epoch": 2.376502625948259,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044298670759650426,
      "loss": 5.1231,
      "mean_token_accuracy": 0.18710657209157944,
      "num_tokens": 52992119.0,
      "step": 30545
    },
    {
      "entropy": 5.35042142868042,
      "epoch": 2.3768916553199766,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004429682681202728,
      "loss": 4.9246,
      "mean_token_accuracy": 0.199246609210968,
      "num_tokens": 53000427.0,
      "step": 30550
    },
    {
      "entropy": 5.412006473541259,
      "epoch": 2.3772806846916943,
      "grad_norm": 1.125,
      "learning_rate": 0.00044294982609539027,
      "loss": 5.1053,
      "mean_token_accuracy": 0.1859336107969284,
      "num_tokens": 53009615.0,
      "step": 30555
    },
    {
      "entropy": 5.4395664691925045,
      "epoch": 2.377669714063412,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004429313815221363,
      "loss": 5.0137,
      "mean_token_accuracy": 0.1973179742693901,
      "num_tokens": 53018357.0,
      "step": 30560
    },
    {
      "entropy": 5.375616121292114,
      "epoch": 2.3780587434351292,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044291293440079107,
      "loss": 5.0312,
      "mean_token_accuracy": 0.19325622618198396,
      "num_tokens": 53027209.0,
      "step": 30565
    },
    {
      "entropy": 5.340203332901001,
      "epoch": 2.378447772806847,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004428944847316342,
      "loss": 4.9967,
      "mean_token_accuracy": 0.19173188805580138,
      "num_tokens": 53035621.0,
      "step": 30570
    },
    {
      "entropy": 5.461109685897827,
      "epoch": 2.3788368021785646,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004428760325149458,
      "loss": 5.0571,
      "mean_token_accuracy": 0.19204505383968354,
      "num_tokens": 53044249.0,
      "step": 30575
    },
    {
      "entropy": 5.417042350769043,
      "epoch": 2.379225831550282,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044285757775100584,
      "loss": 5.0801,
      "mean_token_accuracy": 0.19229482859373093,
      "num_tokens": 53052977.0,
      "step": 30580
    },
    {
      "entropy": 5.394508981704712,
      "epoch": 2.3796148609219996,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044283912044009436,
      "loss": 5.0696,
      "mean_token_accuracy": 0.19335583001375198,
      "num_tokens": 53061071.0,
      "step": 30585
    },
    {
      "entropy": 5.487398386001587,
      "epoch": 2.3800038902937173,
      "grad_norm": 1.125,
      "learning_rate": 0.00044282066058249123,
      "loss": 5.0991,
      "mean_token_accuracy": 0.19450257420539857,
      "num_tokens": 53069237.0,
      "step": 30590
    },
    {
      "entropy": 5.483643817901611,
      "epoch": 2.380392919665435,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004428021981784768,
      "loss": 5.0971,
      "mean_token_accuracy": 0.1901790052652359,
      "num_tokens": 53077991.0,
      "step": 30595
    },
    {
      "entropy": 5.363138484954834,
      "epoch": 2.3807819490371522,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044278373322833106,
      "loss": 5.0033,
      "mean_token_accuracy": 0.19513794034719467,
      "num_tokens": 53086263.0,
      "step": 30600
    },
    {
      "entropy": 5.407528114318848,
      "epoch": 2.38117097840887,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044276526573233416,
      "loss": 5.0524,
      "mean_token_accuracy": 0.19346240013837815,
      "num_tokens": 53095048.0,
      "step": 30605
    },
    {
      "entropy": 5.507652950286865,
      "epoch": 2.381560007780587,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004427467956907664,
      "loss": 5.1044,
      "mean_token_accuracy": 0.1922258749604225,
      "num_tokens": 53103199.0,
      "step": 30610
    },
    {
      "entropy": 5.484123182296753,
      "epoch": 2.381949037152305,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044272832310390814,
      "loss": 5.1039,
      "mean_token_accuracy": 0.19115294963121415,
      "num_tokens": 53111677.0,
      "step": 30615
    },
    {
      "entropy": 5.40900297164917,
      "epoch": 2.3823380665240226,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004427098479720394,
      "loss": 5.0273,
      "mean_token_accuracy": 0.19584785401821136,
      "num_tokens": 53120136.0,
      "step": 30620
    },
    {
      "entropy": 5.420268964767456,
      "epoch": 2.3827270958957403,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044269137029544073,
      "loss": 5.0468,
      "mean_token_accuracy": 0.19238786399364471,
      "num_tokens": 53128847.0,
      "step": 30625
    },
    {
      "entropy": 5.484248256683349,
      "epoch": 2.3831161252674575,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004426728900743924,
      "loss": 5.1276,
      "mean_token_accuracy": 0.18973336070775987,
      "num_tokens": 53137307.0,
      "step": 30630
    },
    {
      "entropy": 5.4402038097381595,
      "epoch": 2.3835051546391752,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044265440730917484,
      "loss": 5.002,
      "mean_token_accuracy": 0.19154655933380127,
      "num_tokens": 53145854.0,
      "step": 30635
    },
    {
      "entropy": 5.372362041473389,
      "epoch": 2.383894184010893,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044263592200006845,
      "loss": 5.0836,
      "mean_token_accuracy": 0.18762978613376619,
      "num_tokens": 53154779.0,
      "step": 30640
    },
    {
      "entropy": 5.410637855529785,
      "epoch": 2.38428321338261,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044261743414735383,
      "loss": 5.0344,
      "mean_token_accuracy": 0.18690712600946427,
      "num_tokens": 53163344.0,
      "step": 30645
    },
    {
      "entropy": 5.376573371887207,
      "epoch": 2.384672242754328,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004425989437513114,
      "loss": 4.9408,
      "mean_token_accuracy": 0.18951437175273894,
      "num_tokens": 53171989.0,
      "step": 30650
    },
    {
      "entropy": 5.426846027374268,
      "epoch": 2.3850612721260456,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004425804508122218,
      "loss": 5.0714,
      "mean_token_accuracy": 0.19352902472019196,
      "num_tokens": 53180612.0,
      "step": 30655
    },
    {
      "entropy": 5.413616085052491,
      "epoch": 2.3854503014977633,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044256195533036566,
      "loss": 4.9875,
      "mean_token_accuracy": 0.19385881274938582,
      "num_tokens": 53188741.0,
      "step": 30660
    },
    {
      "entropy": 5.3680384159088135,
      "epoch": 2.3858393308694805,
      "grad_norm": 1.3125,
      "learning_rate": 0.00044254345730602345,
      "loss": 5.0034,
      "mean_token_accuracy": 0.19878605604171753,
      "num_tokens": 53196857.0,
      "step": 30665
    },
    {
      "entropy": 5.3369770526885985,
      "epoch": 2.386228360241198,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004425249567394759,
      "loss": 5.065,
      "mean_token_accuracy": 0.19140659272670746,
      "num_tokens": 53205264.0,
      "step": 30670
    },
    {
      "entropy": 5.4433478832244875,
      "epoch": 2.386617389612916,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044250645363100386,
      "loss": 5.0061,
      "mean_token_accuracy": 0.2015659123659134,
      "num_tokens": 53213459.0,
      "step": 30675
    },
    {
      "entropy": 5.465548086166382,
      "epoch": 2.387006418984633,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044248794798088805,
      "loss": 5.0896,
      "mean_token_accuracy": 0.19106371253728865,
      "num_tokens": 53221740.0,
      "step": 30680
    },
    {
      "entropy": 5.415329313278198,
      "epoch": 2.387395448356351,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004424694397894091,
      "loss": 5.0786,
      "mean_token_accuracy": 0.19390833377838135,
      "num_tokens": 53230177.0,
      "step": 30685
    },
    {
      "entropy": 5.437094020843506,
      "epoch": 2.3877844777280686,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044245092905684794,
      "loss": 4.9969,
      "mean_token_accuracy": 0.19128857403993607,
      "num_tokens": 53238238.0,
      "step": 30690
    },
    {
      "entropy": 5.420016288757324,
      "epoch": 2.3881735070997863,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044243241578348553,
      "loss": 5.0408,
      "mean_token_accuracy": 0.19420124292373658,
      "num_tokens": 53246519.0,
      "step": 30695
    },
    {
      "entropy": 5.418330192565918,
      "epoch": 2.3885625364715035,
      "grad_norm": 1.125,
      "learning_rate": 0.0004424138999696027,
      "loss": 5.1041,
      "mean_token_accuracy": 0.18944531232118605,
      "num_tokens": 53255522.0,
      "step": 30700
    },
    {
      "entropy": 5.330473470687866,
      "epoch": 2.388951565843221,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004423953816154804,
      "loss": 4.9813,
      "mean_token_accuracy": 0.20078271329402925,
      "num_tokens": 53264414.0,
      "step": 30705
    },
    {
      "entropy": 5.405040550231933,
      "epoch": 2.389340595214939,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044237686072139967,
      "loss": 5.0681,
      "mean_token_accuracy": 0.19158532321453095,
      "num_tokens": 53273403.0,
      "step": 30710
    },
    {
      "entropy": 5.4649817943573,
      "epoch": 2.389729624586656,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004423583372876414,
      "loss": 5.0959,
      "mean_token_accuracy": 0.19103813022375107,
      "num_tokens": 53282361.0,
      "step": 30715
    },
    {
      "entropy": 5.403137731552124,
      "epoch": 2.390118653958374,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044233981131448677,
      "loss": 5.0083,
      "mean_token_accuracy": 0.19460959434509278,
      "num_tokens": 53290890.0,
      "step": 30720
    },
    {
      "entropy": 5.301274585723877,
      "epoch": 2.3905076833300916,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044232128280221683,
      "loss": 4.9549,
      "mean_token_accuracy": 0.19965063333511351,
      "num_tokens": 53299781.0,
      "step": 30725
    },
    {
      "entropy": 5.3833301067352295,
      "epoch": 2.390896712701809,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004423027517511128,
      "loss": 5.062,
      "mean_token_accuracy": 0.19672009944915772,
      "num_tokens": 53308568.0,
      "step": 30730
    },
    {
      "entropy": 5.537405967712402,
      "epoch": 2.3912857420735265,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044228421816145573,
      "loss": 5.1174,
      "mean_token_accuracy": 0.19058251976966858,
      "num_tokens": 53317310.0,
      "step": 30735
    },
    {
      "entropy": 5.423778963088989,
      "epoch": 2.391674771445244,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044226568203352694,
      "loss": 5.0098,
      "mean_token_accuracy": 0.19284526109695435,
      "num_tokens": 53326022.0,
      "step": 30740
    },
    {
      "entropy": 5.411894702911377,
      "epoch": 2.3920638008169615,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044224714336760766,
      "loss": 5.0478,
      "mean_token_accuracy": 0.19370235055685042,
      "num_tokens": 53334090.0,
      "step": 30745
    },
    {
      "entropy": 5.418841600418091,
      "epoch": 2.392452830188679,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004422286021639792,
      "loss": 4.9766,
      "mean_token_accuracy": 0.18813541829586028,
      "num_tokens": 53342682.0,
      "step": 30750
    },
    {
      "entropy": 5.482299757003784,
      "epoch": 2.392841859560397,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004422100584229228,
      "loss": 5.1009,
      "mean_token_accuracy": 0.18999819457530975,
      "num_tokens": 53350652.0,
      "step": 30755
    },
    {
      "entropy": 5.387788534164429,
      "epoch": 2.3932308889321146,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044219151214472,
      "loss": 5.0179,
      "mean_token_accuracy": 0.19769892394542693,
      "num_tokens": 53359339.0,
      "step": 30760
    },
    {
      "entropy": 5.415327692031861,
      "epoch": 2.393619918303832,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004421729633296521,
      "loss": 5.0412,
      "mean_token_accuracy": 0.19402508735656737,
      "num_tokens": 53367617.0,
      "step": 30765
    },
    {
      "entropy": 5.470396089553833,
      "epoch": 2.3940089476755495,
      "grad_norm": 1.125,
      "learning_rate": 0.00044215441197800054,
      "loss": 5.1768,
      "mean_token_accuracy": 0.1790403351187706,
      "num_tokens": 53376725.0,
      "step": 30770
    },
    {
      "entropy": 5.437073707580566,
      "epoch": 2.394397977047267,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044213585809004685,
      "loss": 5.0664,
      "mean_token_accuracy": 0.19609026312828065,
      "num_tokens": 53384670.0,
      "step": 30775
    },
    {
      "entropy": 5.375916242599487,
      "epoch": 2.3947870064189845,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004421173016660725,
      "loss": 5.0124,
      "mean_token_accuracy": 0.197798390686512,
      "num_tokens": 53392944.0,
      "step": 30780
    },
    {
      "entropy": 5.433443593978882,
      "epoch": 2.395176035790702,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004420987427063592,
      "loss": 5.0988,
      "mean_token_accuracy": 0.18637786507606507,
      "num_tokens": 53401301.0,
      "step": 30785
    },
    {
      "entropy": 5.486806440353393,
      "epoch": 2.39556506516242,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044208018121118833,
      "loss": 5.1098,
      "mean_token_accuracy": 0.18687624782323836,
      "num_tokens": 53409490.0,
      "step": 30790
    },
    {
      "entropy": 5.480011415481568,
      "epoch": 2.3959540945341375,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044206161718084164,
      "loss": 5.1706,
      "mean_token_accuracy": 0.1847678914666176,
      "num_tokens": 53418595.0,
      "step": 30795
    },
    {
      "entropy": 5.3955549716949465,
      "epoch": 2.396343123905855,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004420430506156009,
      "loss": 5.0112,
      "mean_token_accuracy": 0.19414662718772888,
      "num_tokens": 53427217.0,
      "step": 30800
    },
    {
      "entropy": 5.38054347038269,
      "epoch": 2.3967321532775725,
      "grad_norm": 1.359375,
      "learning_rate": 0.00044202448151574764,
      "loss": 4.9818,
      "mean_token_accuracy": 0.20417545586824418,
      "num_tokens": 53435875.0,
      "step": 30805
    },
    {
      "entropy": 5.4141130447387695,
      "epoch": 2.39712118264929,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004420059098815638,
      "loss": 5.0706,
      "mean_token_accuracy": 0.19665136337280273,
      "num_tokens": 53445466.0,
      "step": 30810
    },
    {
      "entropy": 5.336964464187622,
      "epoch": 2.3975102120210074,
      "grad_norm": 1.25,
      "learning_rate": 0.0004419873357133311,
      "loss": 4.973,
      "mean_token_accuracy": 0.19582356959581376,
      "num_tokens": 53453813.0,
      "step": 30815
    },
    {
      "entropy": 5.41704626083374,
      "epoch": 2.397899241392725,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004419687590113313,
      "loss": 5.0577,
      "mean_token_accuracy": 0.19585483968257905,
      "num_tokens": 53462819.0,
      "step": 30820
    },
    {
      "entropy": 5.471425914764405,
      "epoch": 2.398288270764443,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044195017977584637,
      "loss": 5.0561,
      "mean_token_accuracy": 0.19361427128314973,
      "num_tokens": 53471297.0,
      "step": 30825
    },
    {
      "entropy": 5.479360055923462,
      "epoch": 2.39867730013616,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044193159800715823,
      "loss": 5.0579,
      "mean_token_accuracy": 0.19361462891101838,
      "num_tokens": 53479473.0,
      "step": 30830
    },
    {
      "entropy": 5.395021772384643,
      "epoch": 2.399066329507878,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044191301370554874,
      "loss": 5.0859,
      "mean_token_accuracy": 0.19065673649311066,
      "num_tokens": 53488542.0,
      "step": 30835
    },
    {
      "entropy": 5.425838613510132,
      "epoch": 2.3994553588795955,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044189442687129994,
      "loss": 4.9874,
      "mean_token_accuracy": 0.19537625163793565,
      "num_tokens": 53497586.0,
      "step": 30840
    },
    {
      "entropy": 5.48591046333313,
      "epoch": 2.3998443882513127,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004418758375046939,
      "loss": 5.0601,
      "mean_token_accuracy": 0.19503633081912994,
      "num_tokens": 53506368.0,
      "step": 30845
    },
    {
      "entropy": 5.39187068939209,
      "epoch": 2.4002334176230304,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044185724560601267,
      "loss": 5.0372,
      "mean_token_accuracy": 0.19493753761053084,
      "num_tokens": 53515735.0,
      "step": 30850
    },
    {
      "entropy": 5.452441883087158,
      "epoch": 2.400622446994748,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004418386511755382,
      "loss": 5.1159,
      "mean_token_accuracy": 0.1864190176129341,
      "num_tokens": 53523780.0,
      "step": 30855
    },
    {
      "entropy": 5.459221076965332,
      "epoch": 2.401011476366466,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004418200542135528,
      "loss": 5.0114,
      "mean_token_accuracy": 0.19394990503787995,
      "num_tokens": 53532633.0,
      "step": 30860
    },
    {
      "entropy": 5.461785221099854,
      "epoch": 2.401400505738183,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004418014547203386,
      "loss": 5.1406,
      "mean_token_accuracy": 0.1962772786617279,
      "num_tokens": 53540767.0,
      "step": 30865
    },
    {
      "entropy": 5.39872031211853,
      "epoch": 2.401789535109901,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004417828526961778,
      "loss": 5.1148,
      "mean_token_accuracy": 0.1927902951836586,
      "num_tokens": 53548925.0,
      "step": 30870
    },
    {
      "entropy": 5.331558084487915,
      "epoch": 2.4021785644816185,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044176424814135266,
      "loss": 4.8922,
      "mean_token_accuracy": 0.20545164942741395,
      "num_tokens": 53557086.0,
      "step": 30875
    },
    {
      "entropy": 5.447721147537232,
      "epoch": 2.4025675938533357,
      "grad_norm": 1.25,
      "learning_rate": 0.0004417456410561455,
      "loss": 5.134,
      "mean_token_accuracy": 0.1820867344737053,
      "num_tokens": 53565898.0,
      "step": 30880
    },
    {
      "entropy": 5.418444347381592,
      "epoch": 2.4029566232250534,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004417270314408387,
      "loss": 5.0977,
      "mean_token_accuracy": 0.18474151641130448,
      "num_tokens": 53574695.0,
      "step": 30885
    },
    {
      "entropy": 5.502459907531739,
      "epoch": 2.403345652596771,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044170841929571454,
      "loss": 5.1025,
      "mean_token_accuracy": 0.18591193556785585,
      "num_tokens": 53583854.0,
      "step": 30890
    },
    {
      "entropy": 5.423509359359741,
      "epoch": 2.403734681968489,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044168980462105543,
      "loss": 5.0553,
      "mean_token_accuracy": 0.2013448491692543,
      "num_tokens": 53593008.0,
      "step": 30895
    },
    {
      "entropy": 5.45743317604065,
      "epoch": 2.404123711340206,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004416711874171439,
      "loss": 4.9919,
      "mean_token_accuracy": 0.19569877535104752,
      "num_tokens": 53601548.0,
      "step": 30900
    },
    {
      "entropy": 5.43909797668457,
      "epoch": 2.404512740711924,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004416525676842624,
      "loss": 5.1664,
      "mean_token_accuracy": 0.18866202086210251,
      "num_tokens": 53609807.0,
      "step": 30905
    },
    {
      "entropy": 5.365352392196655,
      "epoch": 2.4049017700836415,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004416339454226933,
      "loss": 4.9539,
      "mean_token_accuracy": 0.20582334101200103,
      "num_tokens": 53618879.0,
      "step": 30910
    },
    {
      "entropy": 5.399794960021973,
      "epoch": 2.4052907994553587,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004416153206327194,
      "loss": 4.9783,
      "mean_token_accuracy": 0.199844391644001,
      "num_tokens": 53627318.0,
      "step": 30915
    },
    {
      "entropy": 5.331952476501465,
      "epoch": 2.4056798288270764,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044159669331462326,
      "loss": 5.0261,
      "mean_token_accuracy": 0.1918699711561203,
      "num_tokens": 53635872.0,
      "step": 30920
    },
    {
      "entropy": 5.408002233505249,
      "epoch": 2.406068858198794,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044157806346868737,
      "loss": 5.0255,
      "mean_token_accuracy": 0.19299423396587373,
      "num_tokens": 53644434.0,
      "step": 30925
    },
    {
      "entropy": 5.393380165100098,
      "epoch": 2.406457887570512,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044155943109519454,
      "loss": 4.9929,
      "mean_token_accuracy": 0.19712525457143784,
      "num_tokens": 53652721.0,
      "step": 30930
    },
    {
      "entropy": 5.431900405883789,
      "epoch": 2.406846916942229,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004415407961944274,
      "loss": 5.0474,
      "mean_token_accuracy": 0.1922902509570122,
      "num_tokens": 53660831.0,
      "step": 30935
    },
    {
      "entropy": 5.479037046432495,
      "epoch": 2.4072359463139468,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044152215876666883,
      "loss": 5.0326,
      "mean_token_accuracy": 0.19873662292957306,
      "num_tokens": 53669335.0,
      "step": 30940
    },
    {
      "entropy": 5.413552904129029,
      "epoch": 2.407624975685664,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004415035188122016,
      "loss": 5.0521,
      "mean_token_accuracy": 0.19206199944019317,
      "num_tokens": 53678004.0,
      "step": 30945
    },
    {
      "entropy": 5.385868310928345,
      "epoch": 2.4080140050573817,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044148487633130837,
      "loss": 5.0444,
      "mean_token_accuracy": 0.1947902575135231,
      "num_tokens": 53687300.0,
      "step": 30950
    },
    {
      "entropy": 5.350548505783081,
      "epoch": 2.4084030344290994,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044146623132427213,
      "loss": 4.9444,
      "mean_token_accuracy": 0.1985419809818268,
      "num_tokens": 53695692.0,
      "step": 30955
    },
    {
      "entropy": 5.410242080688477,
      "epoch": 2.408792063800817,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044144758379137584,
      "loss": 5.1015,
      "mean_token_accuracy": 0.18972747921943664,
      "num_tokens": 53704164.0,
      "step": 30960
    },
    {
      "entropy": 5.33400182723999,
      "epoch": 2.4091810931725344,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044142893373290236,
      "loss": 4.9767,
      "mean_token_accuracy": 0.19933097213506698,
      "num_tokens": 53712681.0,
      "step": 30965
    },
    {
      "entropy": 5.415587520599365,
      "epoch": 2.409570122544252,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004414102811491348,
      "loss": 4.9494,
      "mean_token_accuracy": 0.19964448511600494,
      "num_tokens": 53721208.0,
      "step": 30970
    },
    {
      "entropy": 5.393646669387818,
      "epoch": 2.4099591519159698,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044139162604035597,
      "loss": 5.0596,
      "mean_token_accuracy": 0.18882953226566315,
      "num_tokens": 53729590.0,
      "step": 30975
    },
    {
      "entropy": 5.404494142532348,
      "epoch": 2.410348181287687,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044137296840684917,
      "loss": 5.0684,
      "mean_token_accuracy": 0.1813428670167923,
      "num_tokens": 53737886.0,
      "step": 30980
    },
    {
      "entropy": 5.4490374565124515,
      "epoch": 2.4107372106594047,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004413543082488973,
      "loss": 5.0801,
      "mean_token_accuracy": 0.19190299808979033,
      "num_tokens": 53746397.0,
      "step": 30985
    },
    {
      "entropy": 5.407914352416992,
      "epoch": 2.4111262400311224,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004413356455667836,
      "loss": 5.0263,
      "mean_token_accuracy": 0.18529820442199707,
      "num_tokens": 53755319.0,
      "step": 30990
    },
    {
      "entropy": 5.3294103145599365,
      "epoch": 2.41151526940284,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004413169803607913,
      "loss": 5.0178,
      "mean_token_accuracy": 0.19203171730041504,
      "num_tokens": 53763655.0,
      "step": 30995
    },
    {
      "entropy": 5.440981721878051,
      "epoch": 2.4119042987745574,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044129831263120344,
      "loss": 5.0385,
      "mean_token_accuracy": 0.19603734761476516,
      "num_tokens": 53772922.0,
      "step": 31000
    },
    {
      "entropy": 5.455390310287475,
      "epoch": 2.412293328146275,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004412796423783034,
      "loss": 4.9974,
      "mean_token_accuracy": 0.19312087893486024,
      "num_tokens": 53781328.0,
      "step": 31005
    },
    {
      "entropy": 5.365654134750367,
      "epoch": 2.4126823575179928,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004412609696023745,
      "loss": 5.0183,
      "mean_token_accuracy": 0.1916339784860611,
      "num_tokens": 53791025.0,
      "step": 31010
    },
    {
      "entropy": 5.4606303691864015,
      "epoch": 2.41307138688971,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004412422943037,
      "loss": 5.0595,
      "mean_token_accuracy": 0.19348575174808502,
      "num_tokens": 53799423.0,
      "step": 31015
    },
    {
      "entropy": 5.420235633850098,
      "epoch": 2.4134604162614277,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044122361648256327,
      "loss": 5.0899,
      "mean_token_accuracy": 0.19369147568941117,
      "num_tokens": 53808001.0,
      "step": 31020
    },
    {
      "entropy": 5.44484429359436,
      "epoch": 2.4138494456331454,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044120493613924766,
      "loss": 4.9865,
      "mean_token_accuracy": 0.19721100777387618,
      "num_tokens": 53816354.0,
      "step": 31025
    },
    {
      "entropy": 5.395365953445435,
      "epoch": 2.414238475004863,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044118625327403677,
      "loss": 4.9633,
      "mean_token_accuracy": 0.20180853456258774,
      "num_tokens": 53825430.0,
      "step": 31030
    },
    {
      "entropy": 5.414460039138794,
      "epoch": 2.4146275043765804,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000441167567887214,
      "loss": 5.0664,
      "mean_token_accuracy": 0.19544458240270615,
      "num_tokens": 53833420.0,
      "step": 31035
    },
    {
      "entropy": 5.4033591747283936,
      "epoch": 2.415016533748298,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004411488799790629,
      "loss": 5.1128,
      "mean_token_accuracy": 0.1854367136955261,
      "num_tokens": 53841771.0,
      "step": 31040
    },
    {
      "entropy": 5.460783958435059,
      "epoch": 2.4154055631200158,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044113018954986693,
      "loss": 5.0424,
      "mean_token_accuracy": 0.19895725697278976,
      "num_tokens": 53850485.0,
      "step": 31045
    },
    {
      "entropy": 5.512038135528565,
      "epoch": 2.415794592491733,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004411114965999098,
      "loss": 5.1423,
      "mean_token_accuracy": 0.18421804010868073,
      "num_tokens": 53859491.0,
      "step": 31050
    },
    {
      "entropy": 5.382263231277466,
      "epoch": 2.4161836218634507,
      "grad_norm": 1.171875,
      "learning_rate": 0.000441092801129475,
      "loss": 5.0107,
      "mean_token_accuracy": 0.20050285756587982,
      "num_tokens": 53868838.0,
      "step": 31055
    },
    {
      "entropy": 5.523654747009277,
      "epoch": 2.4165726512351684,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044107410313884643,
      "loss": 5.1306,
      "mean_token_accuracy": 0.19027836620807648,
      "num_tokens": 53877689.0,
      "step": 31060
    },
    {
      "entropy": 5.306901597976685,
      "epoch": 2.4169616806068857,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044105540262830763,
      "loss": 4.8741,
      "mean_token_accuracy": 0.210520701110363,
      "num_tokens": 53885769.0,
      "step": 31065
    },
    {
      "entropy": 5.269044303894043,
      "epoch": 2.4173507099786034,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004410366995981424,
      "loss": 4.9425,
      "mean_token_accuracy": 0.19725072383880615,
      "num_tokens": 53894745.0,
      "step": 31070
    },
    {
      "entropy": 5.364822196960449,
      "epoch": 2.417739739350321,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044101799404863457,
      "loss": 5.0501,
      "mean_token_accuracy": 0.19158648550510407,
      "num_tokens": 53903888.0,
      "step": 31075
    },
    {
      "entropy": 5.444873476028443,
      "epoch": 2.4181287687220383,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004409992859800679,
      "loss": 5.0349,
      "mean_token_accuracy": 0.19962391406297683,
      "num_tokens": 53912422.0,
      "step": 31080
    },
    {
      "entropy": 5.381535959243775,
      "epoch": 2.418517798093756,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004409805753927263,
      "loss": 5.0283,
      "mean_token_accuracy": 0.19223244488239288,
      "num_tokens": 53921181.0,
      "step": 31085
    },
    {
      "entropy": 5.463592052459717,
      "epoch": 2.4189068274654737,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004409618622868936,
      "loss": 5.1405,
      "mean_token_accuracy": 0.19051187485456467,
      "num_tokens": 53929777.0,
      "step": 31090
    },
    {
      "entropy": 5.473381805419922,
      "epoch": 2.4192958568371914,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044094314666285385,
      "loss": 5.0503,
      "mean_token_accuracy": 0.1818574607372284,
      "num_tokens": 53938678.0,
      "step": 31095
    },
    {
      "entropy": 5.412472772598266,
      "epoch": 2.4196848862089086,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044092442852089095,
      "loss": 4.9883,
      "mean_token_accuracy": 0.19464339315891266,
      "num_tokens": 53946158.0,
      "step": 31100
    },
    {
      "entropy": 5.381982946395874,
      "epoch": 2.4200739155806263,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004409057078612889,
      "loss": 5.0444,
      "mean_token_accuracy": 0.1972884863615036,
      "num_tokens": 53954875.0,
      "step": 31105
    },
    {
      "entropy": 5.399390506744385,
      "epoch": 2.420462944952344,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044088698468433194,
      "loss": 5.0529,
      "mean_token_accuracy": 0.18851447701454163,
      "num_tokens": 53963355.0,
      "step": 31110
    },
    {
      "entropy": 5.395762491226196,
      "epoch": 2.4208519743240613,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044086825899030383,
      "loss": 5.0663,
      "mean_token_accuracy": 0.18999584317207335,
      "num_tokens": 53971861.0,
      "step": 31115
    },
    {
      "entropy": 5.403166246414185,
      "epoch": 2.421241003695779,
      "grad_norm": 1.25,
      "learning_rate": 0.000440849530779489,
      "loss": 5.0356,
      "mean_token_accuracy": 0.18615398406982422,
      "num_tokens": 53980415.0,
      "step": 31120
    },
    {
      "entropy": 5.466426372528076,
      "epoch": 2.4216300330674967,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004408308000521715,
      "loss": 5.0282,
      "mean_token_accuracy": 0.18950492441654204,
      "num_tokens": 53989107.0,
      "step": 31125
    },
    {
      "entropy": 5.4030272483825685,
      "epoch": 2.4220190624392144,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044081206680863556,
      "loss": 5.0083,
      "mean_token_accuracy": 0.1948394149541855,
      "num_tokens": 53997938.0,
      "step": 31130
    },
    {
      "entropy": 5.420228958129883,
      "epoch": 2.4224080918109316,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044079333104916554,
      "loss": 5.0015,
      "mean_token_accuracy": 0.19876229166984558,
      "num_tokens": 54006221.0,
      "step": 31135
    },
    {
      "entropy": 5.420237874984741,
      "epoch": 2.4227971211826493,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004407745927740454,
      "loss": 5.0846,
      "mean_token_accuracy": 0.19395023584365845,
      "num_tokens": 54014579.0,
      "step": 31140
    },
    {
      "entropy": 5.442969894409179,
      "epoch": 2.423186150554367,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004407558519835598,
      "loss": 5.0167,
      "mean_token_accuracy": 0.19489992111921312,
      "num_tokens": 54024119.0,
      "step": 31145
    },
    {
      "entropy": 5.4802202701568605,
      "epoch": 2.4235751799260843,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044073710867799295,
      "loss": 5.1086,
      "mean_token_accuracy": 0.18980143517255782,
      "num_tokens": 54032548.0,
      "step": 31150
    },
    {
      "entropy": 5.406494426727295,
      "epoch": 2.423964209297802,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004407183628576293,
      "loss": 5.0512,
      "mean_token_accuracy": 0.19437119513750076,
      "num_tokens": 54041210.0,
      "step": 31155
    },
    {
      "entropy": 5.429860162734985,
      "epoch": 2.4243532386695197,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004406996145227532,
      "loss": 5.0422,
      "mean_token_accuracy": 0.19113251119852065,
      "num_tokens": 54050056.0,
      "step": 31160
    },
    {
      "entropy": 5.412154912948608,
      "epoch": 2.424742268041237,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004406808636736492,
      "loss": 5.0075,
      "mean_token_accuracy": 0.1904846340417862,
      "num_tokens": 54058752.0,
      "step": 31165
    },
    {
      "entropy": 5.349294805526734,
      "epoch": 2.4251312974129546,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044066211031060183,
      "loss": 4.9545,
      "mean_token_accuracy": 0.19757064431905746,
      "num_tokens": 54067496.0,
      "step": 31170
    },
    {
      "entropy": 5.331160640716552,
      "epoch": 2.4255203267846723,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004406433544338956,
      "loss": 4.9734,
      "mean_token_accuracy": 0.19501854926347734,
      "num_tokens": 54075832.0,
      "step": 31175
    },
    {
      "entropy": 5.500713586807251,
      "epoch": 2.4259093561563896,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004406245960438151,
      "loss": 5.1321,
      "mean_token_accuracy": 0.18465981185436248,
      "num_tokens": 54084642.0,
      "step": 31180
    },
    {
      "entropy": 5.359854316711425,
      "epoch": 2.4262983855281073,
      "grad_norm": 1.25,
      "learning_rate": 0.000440605835140645,
      "loss": 4.9663,
      "mean_token_accuracy": 0.19971468895673752,
      "num_tokens": 54092759.0,
      "step": 31185
    },
    {
      "entropy": 5.46321587562561,
      "epoch": 2.426687414899825,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004405870717246699,
      "loss": 5.087,
      "mean_token_accuracy": 0.19501444846391677,
      "num_tokens": 54101413.0,
      "step": 31190
    },
    {
      "entropy": 5.476794481277466,
      "epoch": 2.4270764442715427,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044056830579617464,
      "loss": 5.0869,
      "mean_token_accuracy": 0.1885801687836647,
      "num_tokens": 54110237.0,
      "step": 31195
    },
    {
      "entropy": 5.439658308029175,
      "epoch": 2.42746547364326,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044054953735544376,
      "loss": 5.1595,
      "mean_token_accuracy": 0.18893886357545853,
      "num_tokens": 54119103.0,
      "step": 31200
    },
    {
      "entropy": 5.444906425476074,
      "epoch": 2.4278545030149776,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004405307664027622,
      "loss": 5.0558,
      "mean_token_accuracy": 0.18948551416397094,
      "num_tokens": 54127696.0,
      "step": 31205
    },
    {
      "entropy": 5.421993589401245,
      "epoch": 2.4282435323866953,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004405119929384147,
      "loss": 4.9636,
      "mean_token_accuracy": 0.20198196023702622,
      "num_tokens": 54135607.0,
      "step": 31210
    },
    {
      "entropy": 5.387712001800537,
      "epoch": 2.4286325617584126,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004404932169626862,
      "loss": 5.0434,
      "mean_token_accuracy": 0.18951854705810547,
      "num_tokens": 54144147.0,
      "step": 31215
    },
    {
      "entropy": 5.325083589553833,
      "epoch": 2.4290215911301303,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004404744384758615,
      "loss": 5.0102,
      "mean_token_accuracy": 0.19901714771986007,
      "num_tokens": 54152010.0,
      "step": 31220
    },
    {
      "entropy": 5.411715745925903,
      "epoch": 2.429410620501848,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004404556574782256,
      "loss": 5.1223,
      "mean_token_accuracy": 0.19664178490638734,
      "num_tokens": 54160923.0,
      "step": 31225
    },
    {
      "entropy": 5.443888235092163,
      "epoch": 2.4297996498735657,
      "grad_norm": 1.5625,
      "learning_rate": 0.00044043687397006336,
      "loss": 5.0337,
      "mean_token_accuracy": 0.19563821703195572,
      "num_tokens": 54169688.0,
      "step": 31230
    },
    {
      "entropy": 5.407328128814697,
      "epoch": 2.430188679245283,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044041808795166,
      "loss": 5.0263,
      "mean_token_accuracy": 0.19582823812961578,
      "num_tokens": 54178423.0,
      "step": 31235
    },
    {
      "entropy": 5.368948364257813,
      "epoch": 2.4305777086170006,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004403992994233004,
      "loss": 5.046,
      "mean_token_accuracy": 0.1967637926340103,
      "num_tokens": 54186939.0,
      "step": 31240
    },
    {
      "entropy": 5.411690425872803,
      "epoch": 2.4309667379887183,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044038050838526974,
      "loss": 5.0596,
      "mean_token_accuracy": 0.19335115253925322,
      "num_tokens": 54195758.0,
      "step": 31245
    },
    {
      "entropy": 5.387757205963135,
      "epoch": 2.4313557673604356,
      "grad_norm": 1.125,
      "learning_rate": 0.000440361714837853,
      "loss": 4.922,
      "mean_token_accuracy": 0.19931000024080275,
      "num_tokens": 54205139.0,
      "step": 31250
    },
    {
      "entropy": 5.342758750915527,
      "epoch": 2.4317447967321533,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004403429187813355,
      "loss": 4.976,
      "mean_token_accuracy": 0.19883002936840058,
      "num_tokens": 54213346.0,
      "step": 31255
    },
    {
      "entropy": 5.349114084243775,
      "epoch": 2.432133826103871,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004403241202160024,
      "loss": 4.9198,
      "mean_token_accuracy": 0.20366810858249665,
      "num_tokens": 54221796.0,
      "step": 31260
    },
    {
      "entropy": 5.484842157363891,
      "epoch": 2.432522855475588,
      "grad_norm": 1.125,
      "learning_rate": 0.0004403053191421388,
      "loss": 5.1349,
      "mean_token_accuracy": 0.18344210237264633,
      "num_tokens": 54230166.0,
      "step": 31265
    },
    {
      "entropy": 5.3375242233276365,
      "epoch": 2.432911884847306,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004402865155600302,
      "loss": 4.9842,
      "mean_token_accuracy": 0.19480786174535752,
      "num_tokens": 54238986.0,
      "step": 31270
    },
    {
      "entropy": 5.432356643676758,
      "epoch": 2.4333009142190236,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044026770946996183,
      "loss": 5.0122,
      "mean_token_accuracy": 0.19712936133146286,
      "num_tokens": 54248178.0,
      "step": 31275
    },
    {
      "entropy": 5.3971000671386715,
      "epoch": 2.433689943590741,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044024890087221896,
      "loss": 5.007,
      "mean_token_accuracy": 0.19304340034723283,
      "num_tokens": 54256320.0,
      "step": 31280
    },
    {
      "entropy": 5.384117984771729,
      "epoch": 2.4340789729624586,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044023008976708705,
      "loss": 4.9895,
      "mean_token_accuracy": 0.20007553398609162,
      "num_tokens": 54264209.0,
      "step": 31285
    },
    {
      "entropy": 5.335179376602173,
      "epoch": 2.4344680023341763,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004402112761548515,
      "loss": 4.9074,
      "mean_token_accuracy": 0.20254423916339875,
      "num_tokens": 54271943.0,
      "step": 31290
    },
    {
      "entropy": 5.374290227890015,
      "epoch": 2.434857031705894,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044019246003579776,
      "loss": 5.0655,
      "mean_token_accuracy": 0.18923926800489427,
      "num_tokens": 54280524.0,
      "step": 31295
    },
    {
      "entropy": 5.400473356246948,
      "epoch": 2.435246061077611,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004401736414102115,
      "loss": 5.0425,
      "mean_token_accuracy": 0.18969919234514238,
      "num_tokens": 54289479.0,
      "step": 31300
    },
    {
      "entropy": 5.434623289108276,
      "epoch": 2.435635090449329,
      "grad_norm": 1.265625,
      "learning_rate": 0.000440154820278378,
      "loss": 5.0427,
      "mean_token_accuracy": 0.19564686119556426,
      "num_tokens": 54297502.0,
      "step": 31305
    },
    {
      "entropy": 5.43638653755188,
      "epoch": 2.4360241198210466,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004401359966405831,
      "loss": 5.0592,
      "mean_token_accuracy": 0.18865828067064286,
      "num_tokens": 54305691.0,
      "step": 31310
    },
    {
      "entropy": 5.508402585983276,
      "epoch": 2.436413149192764,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044011717049711216,
      "loss": 5.1115,
      "mean_token_accuracy": 0.18017209619283675,
      "num_tokens": 54315582.0,
      "step": 31315
    },
    {
      "entropy": 5.432874870300293,
      "epoch": 2.4368021785644816,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044009834184825103,
      "loss": 5.0351,
      "mean_token_accuracy": 0.19515862911939622,
      "num_tokens": 54324102.0,
      "step": 31320
    },
    {
      "entropy": 5.472498083114624,
      "epoch": 2.4371912079361993,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004400795106942853,
      "loss": 5.0819,
      "mean_token_accuracy": 0.19353646039962769,
      "num_tokens": 54332618.0,
      "step": 31325
    },
    {
      "entropy": 5.438711500167846,
      "epoch": 2.437580237307917,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044006067703550083,
      "loss": 5.0425,
      "mean_token_accuracy": 0.19505256563425064,
      "num_tokens": 54341236.0,
      "step": 31330
    },
    {
      "entropy": 5.312912178039551,
      "epoch": 2.437969266679634,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004400418408721833,
      "loss": 4.9629,
      "mean_token_accuracy": 0.19714979976415634,
      "num_tokens": 54349817.0,
      "step": 31335
    },
    {
      "entropy": 5.463261651992798,
      "epoch": 2.438358296051352,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044002300220461843,
      "loss": 5.0819,
      "mean_token_accuracy": 0.18972368091344832,
      "num_tokens": 54358767.0,
      "step": 31340
    },
    {
      "entropy": 5.336778926849365,
      "epoch": 2.4387473254230696,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004400041610330922,
      "loss": 4.9287,
      "mean_token_accuracy": 0.20068173259496688,
      "num_tokens": 54367208.0,
      "step": 31345
    },
    {
      "entropy": 5.387504148483276,
      "epoch": 2.439136354794787,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004399853173578905,
      "loss": 5.0549,
      "mean_token_accuracy": 0.19442448914051055,
      "num_tokens": 54376234.0,
      "step": 31350
    },
    {
      "entropy": 5.425720977783203,
      "epoch": 2.4395253841665046,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043996647117929916,
      "loss": 5.0925,
      "mean_token_accuracy": 0.1931433469057083,
      "num_tokens": 54384828.0,
      "step": 31355
    },
    {
      "entropy": 5.446771764755249,
      "epoch": 2.4399144135382222,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004399476224976043,
      "loss": 5.0705,
      "mean_token_accuracy": 0.19233834743499756,
      "num_tokens": 54394068.0,
      "step": 31360
    },
    {
      "entropy": 5.412508821487426,
      "epoch": 2.44030344290994,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043992877131309165,
      "loss": 5.0375,
      "mean_token_accuracy": 0.19527519643306732,
      "num_tokens": 54403667.0,
      "step": 31365
    },
    {
      "entropy": 5.363367557525635,
      "epoch": 2.440692472281657,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004399099176260475,
      "loss": 4.955,
      "mean_token_accuracy": 0.20301275849342346,
      "num_tokens": 54411614.0,
      "step": 31370
    },
    {
      "entropy": 5.323710107803345,
      "epoch": 2.441081501653375,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004398910614367579,
      "loss": 4.9536,
      "mean_token_accuracy": 0.19913744330406188,
      "num_tokens": 54420084.0,
      "step": 31375
    },
    {
      "entropy": 5.427762699127197,
      "epoch": 2.441470531025092,
      "grad_norm": 1.125,
      "learning_rate": 0.00043987220274550876,
      "loss": 5.0765,
      "mean_token_accuracy": 0.19277286827564238,
      "num_tokens": 54429100.0,
      "step": 31380
    },
    {
      "entropy": 5.479410171508789,
      "epoch": 2.44185956039681,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043985334155258643,
      "loss": 5.1231,
      "mean_token_accuracy": 0.19430676847696304,
      "num_tokens": 54437900.0,
      "step": 31385
    },
    {
      "entropy": 5.455195474624634,
      "epoch": 2.4422485897685275,
      "grad_norm": 1.15625,
      "learning_rate": 0.000439834477858277,
      "loss": 5.0195,
      "mean_token_accuracy": 0.19736503064632416,
      "num_tokens": 54445840.0,
      "step": 31390
    },
    {
      "entropy": 5.378389596939087,
      "epoch": 2.4426376191402452,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043981561166286676,
      "loss": 5.0598,
      "mean_token_accuracy": 0.1921093463897705,
      "num_tokens": 54454116.0,
      "step": 31395
    },
    {
      "entropy": 5.394708585739136,
      "epoch": 2.4430266485119625,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000439796742966642,
      "loss": 5.0124,
      "mean_token_accuracy": 0.19211949557065963,
      "num_tokens": 54462787.0,
      "step": 31400
    },
    {
      "entropy": 5.411596393585205,
      "epoch": 2.44341567788368,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043977787176988884,
      "loss": 5.0871,
      "mean_token_accuracy": 0.1917489007115364,
      "num_tokens": 54472706.0,
      "step": 31405
    },
    {
      "entropy": 5.4799384593963625,
      "epoch": 2.443804707255398,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004397589980728938,
      "loss": 5.0443,
      "mean_token_accuracy": 0.19480502009391784,
      "num_tokens": 54482016.0,
      "step": 31410
    },
    {
      "entropy": 5.3771449565887455,
      "epoch": 2.444193736627115,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043974012187594324,
      "loss": 4.9582,
      "mean_token_accuracy": 0.20187467634677886,
      "num_tokens": 54489943.0,
      "step": 31415
    },
    {
      "entropy": 5.404350805282593,
      "epoch": 2.444582765998833,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004397212431793235,
      "loss": 5.1075,
      "mean_token_accuracy": 0.18785278797149657,
      "num_tokens": 54498251.0,
      "step": 31420
    },
    {
      "entropy": 5.335601472854615,
      "epoch": 2.4449717953705505,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043970236198332107,
      "loss": 4.938,
      "mean_token_accuracy": 0.2030004784464836,
      "num_tokens": 54506862.0,
      "step": 31425
    },
    {
      "entropy": 5.353122520446777,
      "epoch": 2.4453608247422682,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043968347828822246,
      "loss": 4.9378,
      "mean_token_accuracy": 0.20174659490585328,
      "num_tokens": 54515207.0,
      "step": 31430
    },
    {
      "entropy": 5.47895565032959,
      "epoch": 2.4457498541139855,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004396645920943142,
      "loss": 5.0788,
      "mean_token_accuracy": 0.19523707181215286,
      "num_tokens": 54523643.0,
      "step": 31435
    },
    {
      "entropy": 5.367811775207519,
      "epoch": 2.446138883485703,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004396457034018828,
      "loss": 4.9397,
      "mean_token_accuracy": 0.21111776679754257,
      "num_tokens": 54532805.0,
      "step": 31440
    },
    {
      "entropy": 5.4254895687103275,
      "epoch": 2.446527912857421,
      "grad_norm": 1.3125,
      "learning_rate": 0.000439626812211215,
      "loss": 5.0798,
      "mean_token_accuracy": 0.19395567774772643,
      "num_tokens": 54541525.0,
      "step": 31445
    },
    {
      "entropy": 5.411166477203369,
      "epoch": 2.446916942229138,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043960791852259735,
      "loss": 5.1128,
      "mean_token_accuracy": 0.19040479511022568,
      "num_tokens": 54550034.0,
      "step": 31450
    },
    {
      "entropy": 5.477800750732422,
      "epoch": 2.447305971600856,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004395890223363165,
      "loss": 5.0072,
      "mean_token_accuracy": 0.19512297660112382,
      "num_tokens": 54559498.0,
      "step": 31455
    },
    {
      "entropy": 5.431726169586182,
      "epoch": 2.4476950009725735,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00043957012365265915,
      "loss": 5.0564,
      "mean_token_accuracy": 0.19653663635253907,
      "num_tokens": 54568397.0,
      "step": 31460
    },
    {
      "entropy": 5.430154037475586,
      "epoch": 2.4480840303442912,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043955122247191214,
      "loss": 5.1262,
      "mean_token_accuracy": 0.1851029083132744,
      "num_tokens": 54576735.0,
      "step": 31465
    },
    {
      "entropy": 5.480307340621948,
      "epoch": 2.4484730597160085,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004395323187943623,
      "loss": 5.1143,
      "mean_token_accuracy": 0.19029008746147155,
      "num_tokens": 54586265.0,
      "step": 31470
    },
    {
      "entropy": 5.451926946640015,
      "epoch": 2.448862089087726,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004395134126202964,
      "loss": 5.0451,
      "mean_token_accuracy": 0.19518058598041535,
      "num_tokens": 54595307.0,
      "step": 31475
    },
    {
      "entropy": 5.42092752456665,
      "epoch": 2.449251118459444,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004394945039500012,
      "loss": 5.0362,
      "mean_token_accuracy": 0.19310922920703888,
      "num_tokens": 54603886.0,
      "step": 31480
    },
    {
      "entropy": 5.4572642803192135,
      "epoch": 2.449640147831161,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043947559278376385,
      "loss": 5.0685,
      "mean_token_accuracy": 0.19258410930633546,
      "num_tokens": 54613165.0,
      "step": 31485
    },
    {
      "entropy": 5.4899272441864015,
      "epoch": 2.450029177202879,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043945667912187117,
      "loss": 5.0122,
      "mean_token_accuracy": 0.19402024149894714,
      "num_tokens": 54622226.0,
      "step": 31490
    },
    {
      "entropy": 5.368931913375855,
      "epoch": 2.4504182065745965,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004394377629646101,
      "loss": 4.9946,
      "mean_token_accuracy": 0.19558170735836028,
      "num_tokens": 54631389.0,
      "step": 31495
    },
    {
      "entropy": 5.351899766921997,
      "epoch": 2.4508072359463138,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043941884431226763,
      "loss": 5.02,
      "mean_token_accuracy": 0.19144335240125657,
      "num_tokens": 54640033.0,
      "step": 31500
    },
    {
      "entropy": 5.416379785537719,
      "epoch": 2.4511962653180315,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000439399923165131,
      "loss": 4.9331,
      "mean_token_accuracy": 0.19484292417764665,
      "num_tokens": 54648828.0,
      "step": 31505
    },
    {
      "entropy": 5.410915422439575,
      "epoch": 2.451585294689749,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004393809995234872,
      "loss": 5.0029,
      "mean_token_accuracy": 0.19869796484708785,
      "num_tokens": 54657132.0,
      "step": 31510
    },
    {
      "entropy": 5.3974353790283205,
      "epoch": 2.4519743240614664,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004393620733876233,
      "loss": 5.0026,
      "mean_token_accuracy": 0.19140104651451112,
      "num_tokens": 54666261.0,
      "step": 31515
    },
    {
      "entropy": 5.348959493637085,
      "epoch": 2.452363353433184,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004393431447578267,
      "loss": 5.0009,
      "mean_token_accuracy": 0.19933670163154601,
      "num_tokens": 54675200.0,
      "step": 31520
    },
    {
      "entropy": 5.376762580871582,
      "epoch": 2.452752382804902,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043932421363438427,
      "loss": 4.9724,
      "mean_token_accuracy": 0.19923997074365615,
      "num_tokens": 54683659.0,
      "step": 31525
    },
    {
      "entropy": 5.285510110855102,
      "epoch": 2.4531414121766195,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004393052800175835,
      "loss": 4.9131,
      "mean_token_accuracy": 0.20845529884099961,
      "num_tokens": 54691903.0,
      "step": 31530
    },
    {
      "entropy": 5.291647863388062,
      "epoch": 2.4535304415483368,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043928634390771166,
      "loss": 4.9459,
      "mean_token_accuracy": 0.19873571246862412,
      "num_tokens": 54700512.0,
      "step": 31535
    },
    {
      "entropy": 5.392148065567016,
      "epoch": 2.4539194709200545,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043926740530505603,
      "loss": 5.0097,
      "mean_token_accuracy": 0.19435751140117646,
      "num_tokens": 54709144.0,
      "step": 31540
    },
    {
      "entropy": 5.419566488265991,
      "epoch": 2.454308500291772,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000439248464209904,
      "loss": 5.0469,
      "mean_token_accuracy": 0.19954792261123658,
      "num_tokens": 54717733.0,
      "step": 31545
    },
    {
      "entropy": 5.364417219161988,
      "epoch": 2.4546975296634894,
      "grad_norm": 1.1875,
      "learning_rate": 0.000439229520622543,
      "loss": 4.9954,
      "mean_token_accuracy": 0.19962529987096786,
      "num_tokens": 54726120.0,
      "step": 31550
    },
    {
      "entropy": 5.343695831298828,
      "epoch": 2.455086559035207,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004392105745432603,
      "loss": 4.9979,
      "mean_token_accuracy": 0.20072450637817382,
      "num_tokens": 54735639.0,
      "step": 31555
    },
    {
      "entropy": 5.401023054122925,
      "epoch": 2.455475588406925,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004391916259723436,
      "loss": 4.9504,
      "mean_token_accuracy": 0.1936843290925026,
      "num_tokens": 54743928.0,
      "step": 31560
    },
    {
      "entropy": 5.451465749740601,
      "epoch": 2.4558646177786425,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043917267491008023,
      "loss": 5.0317,
      "mean_token_accuracy": 0.19650884121656417,
      "num_tokens": 54752409.0,
      "step": 31565
    },
    {
      "entropy": 5.388997411727905,
      "epoch": 2.4562536471503598,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004391537213567579,
      "loss": 5.0707,
      "mean_token_accuracy": 0.19219435900449752,
      "num_tokens": 54761018.0,
      "step": 31570
    },
    {
      "entropy": 5.376125431060791,
      "epoch": 2.4566426765220775,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004391347653126641,
      "loss": 4.9895,
      "mean_token_accuracy": 0.19735456854104996,
      "num_tokens": 54769479.0,
      "step": 31575
    },
    {
      "entropy": 5.354591751098633,
      "epoch": 2.457031705893795,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043911580677808646,
      "loss": 4.927,
      "mean_token_accuracy": 0.19936426877975463,
      "num_tokens": 54778310.0,
      "step": 31580
    },
    {
      "entropy": 5.384905099868774,
      "epoch": 2.4574207352655124,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043909684575331264,
      "loss": 4.9999,
      "mean_token_accuracy": 0.19872778803110122,
      "num_tokens": 54788237.0,
      "step": 31585
    },
    {
      "entropy": 5.37748007774353,
      "epoch": 2.45780976463723,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004390778822386304,
      "loss": 5.0354,
      "mean_token_accuracy": 0.20008478611707686,
      "num_tokens": 54796746.0,
      "step": 31590
    },
    {
      "entropy": 5.3662457942962645,
      "epoch": 2.458198794008948,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043905891623432754,
      "loss": 4.9947,
      "mean_token_accuracy": 0.1929931476712227,
      "num_tokens": 54805572.0,
      "step": 31595
    },
    {
      "entropy": 5.3702186107635494,
      "epoch": 2.458587823380665,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043903994774069165,
      "loss": 4.941,
      "mean_token_accuracy": 0.2024587407708168,
      "num_tokens": 54814631.0,
      "step": 31600
    },
    {
      "entropy": 5.452630615234375,
      "epoch": 2.4589768527523828,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004390209767580107,
      "loss": 4.9784,
      "mean_token_accuracy": 0.1928611233830452,
      "num_tokens": 54822753.0,
      "step": 31605
    },
    {
      "entropy": 5.422944116592407,
      "epoch": 2.4593658821241005,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004390020032865725,
      "loss": 5.0143,
      "mean_token_accuracy": 0.19588242918252946,
      "num_tokens": 54830969.0,
      "step": 31610
    },
    {
      "entropy": 5.336540699005127,
      "epoch": 2.4597549114958177,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004389830273266649,
      "loss": 5.0377,
      "mean_token_accuracy": 0.1993071749806404,
      "num_tokens": 54840295.0,
      "step": 31615
    },
    {
      "entropy": 5.525561952590943,
      "epoch": 2.4601439408675354,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043896404887857596,
      "loss": 5.121,
      "mean_token_accuracy": 0.19116182625293732,
      "num_tokens": 54848583.0,
      "step": 31620
    },
    {
      "entropy": 5.417760372161865,
      "epoch": 2.460532970239253,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004389450679425935,
      "loss": 4.9547,
      "mean_token_accuracy": 0.19704277217388153,
      "num_tokens": 54857450.0,
      "step": 31625
    },
    {
      "entropy": 5.401632070541382,
      "epoch": 2.460921999610971,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004389260845190055,
      "loss": 5.0747,
      "mean_token_accuracy": 0.18853639364242553,
      "num_tokens": 54865928.0,
      "step": 31630
    },
    {
      "entropy": 5.457083034515381,
      "epoch": 2.461311028982688,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043890709860810014,
      "loss": 5.1018,
      "mean_token_accuracy": 0.19262714087963104,
      "num_tokens": 54874983.0,
      "step": 31635
    },
    {
      "entropy": 5.444676065444947,
      "epoch": 2.4617000583544058,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043888811021016545,
      "loss": 5.0272,
      "mean_token_accuracy": 0.19169287979602814,
      "num_tokens": 54883488.0,
      "step": 31640
    },
    {
      "entropy": 5.373010778427124,
      "epoch": 2.4620890877261234,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043886911932548953,
      "loss": 5.0027,
      "mean_token_accuracy": 0.19667731672525407,
      "num_tokens": 54892865.0,
      "step": 31645
    },
    {
      "entropy": 5.464204788208008,
      "epoch": 2.4624781170978407,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004388501259543605,
      "loss": 5.1063,
      "mean_token_accuracy": 0.18693741410970688,
      "num_tokens": 54901714.0,
      "step": 31650
    },
    {
      "entropy": 5.355846166610718,
      "epoch": 2.4628671464695584,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004388311300970667,
      "loss": 5.0776,
      "mean_token_accuracy": 0.19031663089990616,
      "num_tokens": 54910912.0,
      "step": 31655
    },
    {
      "entropy": 5.48328742980957,
      "epoch": 2.463256175841276,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004388121317538962,
      "loss": 5.1093,
      "mean_token_accuracy": 0.18409449309110643,
      "num_tokens": 54919613.0,
      "step": 31660
    },
    {
      "entropy": 5.453497505187988,
      "epoch": 2.463645205212994,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004387931309251374,
      "loss": 5.0038,
      "mean_token_accuracy": 0.19350139498710633,
      "num_tokens": 54928794.0,
      "step": 31665
    },
    {
      "entropy": 5.310459232330322,
      "epoch": 2.464034234584711,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004387741276110785,
      "loss": 4.9462,
      "mean_token_accuracy": 0.19767017513513566,
      "num_tokens": 54937826.0,
      "step": 31670
    },
    {
      "entropy": 5.399421453475952,
      "epoch": 2.4644232639564287,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043875512181200784,
      "loss": 5.057,
      "mean_token_accuracy": 0.1864125356078148,
      "num_tokens": 54946541.0,
      "step": 31675
    },
    {
      "entropy": 5.443987560272217,
      "epoch": 2.4648122933281464,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004387361135282139,
      "loss": 5.0476,
      "mean_token_accuracy": 0.19305471032857896,
      "num_tokens": 54954540.0,
      "step": 31680
    },
    {
      "entropy": 5.337622213363647,
      "epoch": 2.4652013226998637,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043871710275998507,
      "loss": 4.996,
      "mean_token_accuracy": 0.19901976436376573,
      "num_tokens": 54962784.0,
      "step": 31685
    },
    {
      "entropy": 5.32838921546936,
      "epoch": 2.4655903520715814,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004386980895076097,
      "loss": 4.9586,
      "mean_token_accuracy": 0.20501023381948472,
      "num_tokens": 54971572.0,
      "step": 31690
    },
    {
      "entropy": 5.442321014404297,
      "epoch": 2.465979381443299,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043867907377137646,
      "loss": 4.9854,
      "mean_token_accuracy": 0.19663703292608262,
      "num_tokens": 54980161.0,
      "step": 31695
    },
    {
      "entropy": 5.419276237487793,
      "epoch": 2.4663684108150163,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043866005555157376,
      "loss": 5.0199,
      "mean_token_accuracy": 0.19419140070676805,
      "num_tokens": 54988219.0,
      "step": 31700
    },
    {
      "entropy": 5.350657320022583,
      "epoch": 2.466757440186734,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004386410348484903,
      "loss": 4.9603,
      "mean_token_accuracy": 0.19908930361270905,
      "num_tokens": 54996550.0,
      "step": 31705
    },
    {
      "entropy": 5.382793045043945,
      "epoch": 2.4671464695584517,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004386220116624145,
      "loss": 5.1198,
      "mean_token_accuracy": 0.18964531868696213,
      "num_tokens": 55005269.0,
      "step": 31710
    },
    {
      "entropy": 5.39800238609314,
      "epoch": 2.467535498930169,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043860298599363506,
      "loss": 4.9104,
      "mean_token_accuracy": 0.20839183628559113,
      "num_tokens": 55014264.0,
      "step": 31715
    },
    {
      "entropy": 5.345554161071777,
      "epoch": 2.4679245283018867,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004385839578424408,
      "loss": 4.9938,
      "mean_token_accuracy": 0.19470368176698685,
      "num_tokens": 55023597.0,
      "step": 31720
    },
    {
      "entropy": 5.383526611328125,
      "epoch": 2.4683135576736044,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043856492720912037,
      "loss": 5.0121,
      "mean_token_accuracy": 0.19908594340085983,
      "num_tokens": 55032860.0,
      "step": 31725
    },
    {
      "entropy": 5.4351921558380125,
      "epoch": 2.468702587045322,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004385458940939624,
      "loss": 4.9907,
      "mean_token_accuracy": 0.20143966972827912,
      "num_tokens": 55041275.0,
      "step": 31730
    },
    {
      "entropy": 5.4054240703582765,
      "epoch": 2.4690916164170393,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004385268584972559,
      "loss": 5.0827,
      "mean_token_accuracy": 0.19058404713869095,
      "num_tokens": 55050871.0,
      "step": 31735
    },
    {
      "entropy": 5.406329345703125,
      "epoch": 2.469480645788757,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004385078204192896,
      "loss": 5.0106,
      "mean_token_accuracy": 0.19181863516569136,
      "num_tokens": 55060353.0,
      "step": 31740
    },
    {
      "entropy": 5.41583137512207,
      "epoch": 2.4698696751604747,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004384887798603523,
      "loss": 5.0161,
      "mean_token_accuracy": 0.19084177613258363,
      "num_tokens": 55069002.0,
      "step": 31745
    },
    {
      "entropy": 5.421838569641113,
      "epoch": 2.470258704532192,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000438469736820733,
      "loss": 5.0618,
      "mean_token_accuracy": 0.18500406444072723,
      "num_tokens": 55078369.0,
      "step": 31750
    },
    {
      "entropy": 5.523258447647095,
      "epoch": 2.4706477339039097,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043845069130072067,
      "loss": 5.1466,
      "mean_token_accuracy": 0.18580360859632492,
      "num_tokens": 55087597.0,
      "step": 31755
    },
    {
      "entropy": 5.383196592330933,
      "epoch": 2.4710367632756274,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004384316433006043,
      "loss": 4.9703,
      "mean_token_accuracy": 0.20030679553747177,
      "num_tokens": 55095906.0,
      "step": 31760
    },
    {
      "entropy": 5.34703803062439,
      "epoch": 2.471425792647345,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043841259282067276,
      "loss": 4.9869,
      "mean_token_accuracy": 0.19591302424669266,
      "num_tokens": 55104401.0,
      "step": 31765
    },
    {
      "entropy": 5.3689820766448975,
      "epoch": 2.4718148220190623,
      "grad_norm": 1.125,
      "learning_rate": 0.00043839353986121533,
      "loss": 5.05,
      "mean_token_accuracy": 0.193874728679657,
      "num_tokens": 55113075.0,
      "step": 31770
    },
    {
      "entropy": 5.38634762763977,
      "epoch": 2.47220385139078,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004383744844225209,
      "loss": 4.9375,
      "mean_token_accuracy": 0.1977776139974594,
      "num_tokens": 55122173.0,
      "step": 31775
    },
    {
      "entropy": 5.443068647384644,
      "epoch": 2.4725928807624977,
      "grad_norm": 1.796875,
      "learning_rate": 0.00043835542650487875,
      "loss": 5.0706,
      "mean_token_accuracy": 0.19355318546295167,
      "num_tokens": 55130356.0,
      "step": 31780
    },
    {
      "entropy": 5.382021379470825,
      "epoch": 2.472981910134215,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004383363661085781,
      "loss": 5.0201,
      "mean_token_accuracy": 0.19568648636341096,
      "num_tokens": 55138985.0,
      "step": 31785
    },
    {
      "entropy": 5.379597568511963,
      "epoch": 2.4733709395059327,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004383173032339079,
      "loss": 5.0331,
      "mean_token_accuracy": 0.20241547077894212,
      "num_tokens": 55148053.0,
      "step": 31790
    },
    {
      "entropy": 5.482140779495239,
      "epoch": 2.4737599688776504,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004382982378811577,
      "loss": 5.1237,
      "mean_token_accuracy": 0.18864275962114335,
      "num_tokens": 55156815.0,
      "step": 31795
    },
    {
      "entropy": 5.331408071517944,
      "epoch": 2.474148998249368,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004382791700506166,
      "loss": 4.8793,
      "mean_token_accuracy": 0.2092390313744545,
      "num_tokens": 55165007.0,
      "step": 31800
    },
    {
      "entropy": 5.392003202438355,
      "epoch": 2.4745380276210853,
      "grad_norm": 1.140625,
      "learning_rate": 0.000438260099742574,
      "loss": 5.0205,
      "mean_token_accuracy": 0.19403792470693587,
      "num_tokens": 55173784.0,
      "step": 31805
    },
    {
      "entropy": 5.388980293273926,
      "epoch": 2.474927056992803,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004382410269573192,
      "loss": 5.0101,
      "mean_token_accuracy": 0.1952759474515915,
      "num_tokens": 55182436.0,
      "step": 31810
    },
    {
      "entropy": 5.475616550445556,
      "epoch": 2.4753160863645203,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043822195169514167,
      "loss": 5.0568,
      "mean_token_accuracy": 0.19712310135364533,
      "num_tokens": 55190083.0,
      "step": 31815
    },
    {
      "entropy": 5.4534073829650875,
      "epoch": 2.475705115736238,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043820287395633086,
      "loss": 5.1154,
      "mean_token_accuracy": 0.1858028158545494,
      "num_tokens": 55200154.0,
      "step": 31820
    },
    {
      "entropy": 5.453097343444824,
      "epoch": 2.4760941451079557,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004381837937411761,
      "loss": 5.0405,
      "mean_token_accuracy": 0.19423019289970397,
      "num_tokens": 55209672.0,
      "step": 31825
    },
    {
      "entropy": 5.45835313796997,
      "epoch": 2.4764831744796734,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043816471104996706,
      "loss": 5.1038,
      "mean_token_accuracy": 0.1911303699016571,
      "num_tokens": 55218362.0,
      "step": 31830
    },
    {
      "entropy": 5.389842319488525,
      "epoch": 2.4768722038513906,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004381456258829933,
      "loss": 4.9486,
      "mean_token_accuracy": 0.20077385306358336,
      "num_tokens": 55226834.0,
      "step": 31835
    },
    {
      "entropy": 5.308271074295044,
      "epoch": 2.4772612332231083,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043812653824054426,
      "loss": 5.0639,
      "mean_token_accuracy": 0.18398825079202652,
      "num_tokens": 55236325.0,
      "step": 31840
    },
    {
      "entropy": 5.393371152877807,
      "epoch": 2.477650262594826,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004381074481229097,
      "loss": 4.9844,
      "mean_token_accuracy": 0.19987611174583436,
      "num_tokens": 55245007.0,
      "step": 31845
    },
    {
      "entropy": 5.4103227138519285,
      "epoch": 2.4780392919665433,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004380883555303791,
      "loss": 4.957,
      "mean_token_accuracy": 0.19747809022665025,
      "num_tokens": 55253967.0,
      "step": 31850
    },
    {
      "entropy": 5.421306848526001,
      "epoch": 2.478428321338261,
      "grad_norm": 1.5,
      "learning_rate": 0.0004380692604632424,
      "loss": 5.0255,
      "mean_token_accuracy": 0.191767817735672,
      "num_tokens": 55262226.0,
      "step": 31855
    },
    {
      "entropy": 5.490383434295654,
      "epoch": 2.4788173507099787,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043805016292178923,
      "loss": 5.1205,
      "mean_token_accuracy": 0.1917422741651535,
      "num_tokens": 55271523.0,
      "step": 31860
    },
    {
      "entropy": 5.439175081253052,
      "epoch": 2.4792063800816964,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004380310629063094,
      "loss": 5.0286,
      "mean_token_accuracy": 0.2009475991129875,
      "num_tokens": 55280604.0,
      "step": 31865
    },
    {
      "entropy": 5.322654151916504,
      "epoch": 2.4795954094534136,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004380119604170925,
      "loss": 4.9775,
      "mean_token_accuracy": 0.19952320158481598,
      "num_tokens": 55289521.0,
      "step": 31870
    },
    {
      "entropy": 5.408681106567383,
      "epoch": 2.4799844388251313,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043799285545442876,
      "loss": 4.9856,
      "mean_token_accuracy": 0.19501469731330873,
      "num_tokens": 55298211.0,
      "step": 31875
    },
    {
      "entropy": 5.352804708480835,
      "epoch": 2.480373468196849,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004379737480186078,
      "loss": 4.9256,
      "mean_token_accuracy": 0.2030505046248436,
      "num_tokens": 55306984.0,
      "step": 31880
    },
    {
      "entropy": 5.422357892990112,
      "epoch": 2.4807624975685663,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043795463810991956,
      "loss": 5.0797,
      "mean_token_accuracy": 0.18498249500989913,
      "num_tokens": 55316266.0,
      "step": 31885
    },
    {
      "entropy": 5.360296964645386,
      "epoch": 2.481151526940284,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004379355257286541,
      "loss": 4.9267,
      "mean_token_accuracy": 0.19890151172876358,
      "num_tokens": 55324698.0,
      "step": 31890
    },
    {
      "entropy": 5.411339998245239,
      "epoch": 2.4815405563120017,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043791641087510136,
      "loss": 4.9604,
      "mean_token_accuracy": 0.2020510032773018,
      "num_tokens": 55333033.0,
      "step": 31895
    },
    {
      "entropy": 5.427025318145752,
      "epoch": 2.4819295856837194,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004378972935495514,
      "loss": 5.08,
      "mean_token_accuracy": 0.1950684204697609,
      "num_tokens": 55341110.0,
      "step": 31900
    },
    {
      "entropy": 5.407880449295044,
      "epoch": 2.4823186150554366,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004378781737522943,
      "loss": 4.9981,
      "mean_token_accuracy": 0.19607090055942536,
      "num_tokens": 55350156.0,
      "step": 31905
    },
    {
      "entropy": 5.3996399402618405,
      "epoch": 2.4827076444271543,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043785905148362007,
      "loss": 5.0666,
      "mean_token_accuracy": 0.19022821336984636,
      "num_tokens": 55358737.0,
      "step": 31910
    },
    {
      "entropy": 5.365509796142578,
      "epoch": 2.483096673798872,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043783992674381903,
      "loss": 4.9482,
      "mean_token_accuracy": 0.20402962416410447,
      "num_tokens": 55366872.0,
      "step": 31915
    },
    {
      "entropy": 5.289077615737915,
      "epoch": 2.4834857031705893,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043782079953318126,
      "loss": 4.9408,
      "mean_token_accuracy": 0.19867077022790908,
      "num_tokens": 55375675.0,
      "step": 31920
    },
    {
      "entropy": 5.383708810806274,
      "epoch": 2.483874732542307,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004378016698519971,
      "loss": 4.9572,
      "mean_token_accuracy": 0.20084453821182252,
      "num_tokens": 55382750.0,
      "step": 31925
    },
    {
      "entropy": 5.348335599899292,
      "epoch": 2.4842637619140246,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043778253770055666,
      "loss": 5.0249,
      "mean_token_accuracy": 0.1913883313536644,
      "num_tokens": 55390852.0,
      "step": 31930
    },
    {
      "entropy": 5.363742876052856,
      "epoch": 2.484652791285742,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043776340307915033,
      "loss": 5.0025,
      "mean_token_accuracy": 0.19987063407897948,
      "num_tokens": 55398789.0,
      "step": 31935
    },
    {
      "entropy": 5.390597486495972,
      "epoch": 2.4850418206574596,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004377442659880684,
      "loss": 5.0623,
      "mean_token_accuracy": 0.19170828461647033,
      "num_tokens": 55408100.0,
      "step": 31940
    },
    {
      "entropy": 5.38057951927185,
      "epoch": 2.4854308500291773,
      "grad_norm": 1.125,
      "learning_rate": 0.0004377251264276013,
      "loss": 4.9823,
      "mean_token_accuracy": 0.19836508333683014,
      "num_tokens": 55416718.0,
      "step": 31945
    },
    {
      "entropy": 5.467417860031128,
      "epoch": 2.4858198794008945,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043770598439803936,
      "loss": 5.0812,
      "mean_token_accuracy": 0.19347298443317412,
      "num_tokens": 55426149.0,
      "step": 31950
    },
    {
      "entropy": 5.428702640533447,
      "epoch": 2.4862089087726122,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043768683989967304,
      "loss": 4.9969,
      "mean_token_accuracy": 0.1951591193675995,
      "num_tokens": 55434379.0,
      "step": 31955
    },
    {
      "entropy": 5.368360614776611,
      "epoch": 2.48659793814433,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043766769293279294,
      "loss": 5.0436,
      "mean_token_accuracy": 0.19439576268196107,
      "num_tokens": 55442880.0,
      "step": 31960
    },
    {
      "entropy": 5.394635534286499,
      "epoch": 2.4869869675160476,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043764854349768957,
      "loss": 5.0242,
      "mean_token_accuracy": 0.19372686445713044,
      "num_tokens": 55452090.0,
      "step": 31965
    },
    {
      "entropy": 5.448425054550171,
      "epoch": 2.487375996887765,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043762939159465333,
      "loss": 5.0513,
      "mean_token_accuracy": 0.19565532207489014,
      "num_tokens": 55460271.0,
      "step": 31970
    },
    {
      "entropy": 5.3439233779907225,
      "epoch": 2.4877650262594826,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043761023722397495,
      "loss": 4.9164,
      "mean_token_accuracy": 0.1992787942290306,
      "num_tokens": 55468414.0,
      "step": 31975
    },
    {
      "entropy": 5.411414957046508,
      "epoch": 2.4881540556312003,
      "grad_norm": 1.1875,
      "learning_rate": 0.000437591080385945,
      "loss": 5.0172,
      "mean_token_accuracy": 0.19629014283418655,
      "num_tokens": 55477658.0,
      "step": 31980
    },
    {
      "entropy": 5.4045580387115475,
      "epoch": 2.4885430850029175,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004375719210808543,
      "loss": 4.9432,
      "mean_token_accuracy": 0.19389043152332305,
      "num_tokens": 55486536.0,
      "step": 31985
    },
    {
      "entropy": 5.377507257461548,
      "epoch": 2.4889321143746352,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004375527593089934,
      "loss": 5.0212,
      "mean_token_accuracy": 0.19533667713403702,
      "num_tokens": 55494995.0,
      "step": 31990
    },
    {
      "entropy": 5.402934265136719,
      "epoch": 2.489321143746353,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004375335950706531,
      "loss": 5.0242,
      "mean_token_accuracy": 0.18993353843688965,
      "num_tokens": 55503392.0,
      "step": 31995
    },
    {
      "entropy": 5.363742923736572,
      "epoch": 2.4897101731180706,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004375144283661242,
      "loss": 4.985,
      "mean_token_accuracy": 0.20166463255882264,
      "num_tokens": 55512211.0,
      "step": 32000
    },
    {
      "entropy": 5.286008834838867,
      "epoch": 2.490099202489788,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004374952591956975,
      "loss": 4.9453,
      "mean_token_accuracy": 0.1990015059709549,
      "num_tokens": 55520412.0,
      "step": 32005
    },
    {
      "entropy": 5.3209484100341795,
      "epoch": 2.4904882318615056,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043747608755966396,
      "loss": 4.9794,
      "mean_token_accuracy": 0.19808338433504105,
      "num_tokens": 55529964.0,
      "step": 32010
    },
    {
      "entropy": 5.406563472747803,
      "epoch": 2.4908772612332233,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004374569134583143,
      "loss": 4.9805,
      "mean_token_accuracy": 0.19372447878122329,
      "num_tokens": 55539111.0,
      "step": 32015
    },
    {
      "entropy": 5.387434482574463,
      "epoch": 2.4912662906049405,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004374377368919396,
      "loss": 4.9411,
      "mean_token_accuracy": 0.2043435826897621,
      "num_tokens": 55547520.0,
      "step": 32020
    },
    {
      "entropy": 5.419068956375122,
      "epoch": 2.4916553199766582,
      "grad_norm": 1.125,
      "learning_rate": 0.00043741855786083085,
      "loss": 5.0649,
      "mean_token_accuracy": 0.18893383890390397,
      "num_tokens": 55557111.0,
      "step": 32025
    },
    {
      "entropy": 5.402033948898316,
      "epoch": 2.492044349348376,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043739937636527894,
      "loss": 5.0354,
      "mean_token_accuracy": 0.19404893815517427,
      "num_tokens": 55565383.0,
      "step": 32030
    },
    {
      "entropy": 5.427169561386108,
      "epoch": 2.492433378720093,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043738019240557494,
      "loss": 5.0017,
      "mean_token_accuracy": 0.1955266773700714,
      "num_tokens": 55573668.0,
      "step": 32035
    },
    {
      "entropy": 5.41478271484375,
      "epoch": 2.492822408091811,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043736100598201,
      "loss": 5.0499,
      "mean_token_accuracy": 0.1939924344420433,
      "num_tokens": 55582716.0,
      "step": 32040
    },
    {
      "entropy": 5.352471017837525,
      "epoch": 2.4932114374635286,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004373418170948752,
      "loss": 4.978,
      "mean_token_accuracy": 0.19209808111190796,
      "num_tokens": 55591472.0,
      "step": 32045
    },
    {
      "entropy": 5.488294363021851,
      "epoch": 2.493600466835246,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004373226257444617,
      "loss": 5.0378,
      "mean_token_accuracy": 0.19654901027679444,
      "num_tokens": 55600022.0,
      "step": 32050
    },
    {
      "entropy": 5.47860541343689,
      "epoch": 2.4939894962069635,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004373034319310608,
      "loss": 5.0589,
      "mean_token_accuracy": 0.19150239378213882,
      "num_tokens": 55608854.0,
      "step": 32055
    },
    {
      "entropy": 5.30726809501648,
      "epoch": 2.4943785255786812,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043728423565496367,
      "loss": 5.0182,
      "mean_token_accuracy": 0.1950073629617691,
      "num_tokens": 55618246.0,
      "step": 32060
    },
    {
      "entropy": 5.479041004180909,
      "epoch": 2.494767554950399,
      "grad_norm": 1.25,
      "learning_rate": 0.0004372650369164615,
      "loss": 5.0715,
      "mean_token_accuracy": 0.1917803019285202,
      "num_tokens": 55626706.0,
      "step": 32065
    },
    {
      "entropy": 5.379968786239624,
      "epoch": 2.495156584322116,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043724583571584565,
      "loss": 4.9922,
      "mean_token_accuracy": 0.199569234251976,
      "num_tokens": 55634810.0,
      "step": 32070
    },
    {
      "entropy": 5.384361982345581,
      "epoch": 2.495545613693834,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043722663205340753,
      "loss": 5.117,
      "mean_token_accuracy": 0.18615053296089173,
      "num_tokens": 55643801.0,
      "step": 32075
    },
    {
      "entropy": 5.397097253799439,
      "epoch": 2.4959346430655516,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004372074259294384,
      "loss": 5.0288,
      "mean_token_accuracy": 0.20048280656337739,
      "num_tokens": 55652236.0,
      "step": 32080
    },
    {
      "entropy": 5.419412422180176,
      "epoch": 2.496323672437269,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004371882173442299,
      "loss": 5.0651,
      "mean_token_accuracy": 0.20186380296945572,
      "num_tokens": 55660996.0,
      "step": 32085
    },
    {
      "entropy": 5.402106666564942,
      "epoch": 2.4967127018089865,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043716900629807323,
      "loss": 5.0639,
      "mean_token_accuracy": 0.19183885008096696,
      "num_tokens": 55669766.0,
      "step": 32090
    },
    {
      "entropy": 5.387622833251953,
      "epoch": 2.497101731180704,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043714979279126006,
      "loss": 5.0581,
      "mean_token_accuracy": 0.19885485023260116,
      "num_tokens": 55679179.0,
      "step": 32095
    },
    {
      "entropy": 5.4496582508087155,
      "epoch": 2.497490760552422,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004371305768240818,
      "loss": 4.9973,
      "mean_token_accuracy": 0.1970147654414177,
      "num_tokens": 55687554.0,
      "step": 32100
    },
    {
      "entropy": 5.3224235534667965,
      "epoch": 2.497879789924139,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004371113583968302,
      "loss": 4.886,
      "mean_token_accuracy": 0.20506118088960648,
      "num_tokens": 55695537.0,
      "step": 32105
    },
    {
      "entropy": 5.408735752105713,
      "epoch": 2.498268819295857,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004370921375097967,
      "loss": 5.0959,
      "mean_token_accuracy": 0.19134239703416825,
      "num_tokens": 55703432.0,
      "step": 32110
    },
    {
      "entropy": 5.411474752426147,
      "epoch": 2.4986578486675746,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043707291416327307,
      "loss": 5.0748,
      "mean_token_accuracy": 0.19670649915933608,
      "num_tokens": 55712169.0,
      "step": 32115
    },
    {
      "entropy": 5.474319219589233,
      "epoch": 2.499046878039292,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004370536883575509,
      "loss": 5.0769,
      "mean_token_accuracy": 0.19291894584894181,
      "num_tokens": 55720471.0,
      "step": 32120
    },
    {
      "entropy": 5.315148830413818,
      "epoch": 2.4994359074110095,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004370344600929219,
      "loss": 4.9496,
      "mean_token_accuracy": 0.19630141854286193,
      "num_tokens": 55729188.0,
      "step": 32125
    },
    {
      "entropy": 5.380610752105713,
      "epoch": 2.499824936782727,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004370152293696779,
      "loss": 5.0068,
      "mean_token_accuracy": 0.19659818708896637,
      "num_tokens": 55738313.0,
      "step": 32130
    },
    {
      "entropy": 5.381129407882691,
      "epoch": 2.500213966154445,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004369959961881107,
      "loss": 4.9711,
      "mean_token_accuracy": 0.19231171756982804,
      "num_tokens": 55746849.0,
      "step": 32135
    },
    {
      "entropy": 5.379983043670654,
      "epoch": 2.500602995526162,
      "grad_norm": 1.140625,
      "learning_rate": 0.000436976760548512,
      "loss": 5.0402,
      "mean_token_accuracy": 0.19847173392772674,
      "num_tokens": 55755151.0,
      "step": 32140
    },
    {
      "entropy": 5.309518766403198,
      "epoch": 2.50099202489788,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043695752245117393,
      "loss": 4.928,
      "mean_token_accuracy": 0.20493041276931762,
      "num_tokens": 55763716.0,
      "step": 32145
    },
    {
      "entropy": 5.4080119132995605,
      "epoch": 2.501381054269597,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004369382818963882,
      "loss": 5.0395,
      "mean_token_accuracy": 0.1925261065363884,
      "num_tokens": 55772424.0,
      "step": 32150
    },
    {
      "entropy": 5.496874141693115,
      "epoch": 2.501770083641315,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004369190388844467,
      "loss": 5.1057,
      "mean_token_accuracy": 0.1981392502784729,
      "num_tokens": 55780084.0,
      "step": 32155
    },
    {
      "entropy": 5.373549747467041,
      "epoch": 2.5021591130130325,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043689979341564154,
      "loss": 4.9365,
      "mean_token_accuracy": 0.20000428557395936,
      "num_tokens": 55788717.0,
      "step": 32160
    },
    {
      "entropy": 5.375211477279663,
      "epoch": 2.50254814238475,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004368805454902648,
      "loss": 4.9509,
      "mean_token_accuracy": 0.1992054283618927,
      "num_tokens": 55797166.0,
      "step": 32165
    },
    {
      "entropy": 5.4372090816497805,
      "epoch": 2.5029371717564675,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043686129510860826,
      "loss": 5.0585,
      "mean_token_accuracy": 0.1977046698331833,
      "num_tokens": 55805702.0,
      "step": 32170
    },
    {
      "entropy": 5.333179998397827,
      "epoch": 2.503326201128185,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043684204227096435,
      "loss": 4.9909,
      "mean_token_accuracy": 0.197385273873806,
      "num_tokens": 55814938.0,
      "step": 32175
    },
    {
      "entropy": 5.361462163925171,
      "epoch": 2.503715230499903,
      "grad_norm": 1.125,
      "learning_rate": 0.000436822786977625,
      "loss": 4.9946,
      "mean_token_accuracy": 0.19971792846918107,
      "num_tokens": 55823527.0,
      "step": 32180
    },
    {
      "entropy": 5.4863138675689695,
      "epoch": 2.50410425987162,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043680352922888244,
      "loss": 5.0611,
      "mean_token_accuracy": 0.19063250571489335,
      "num_tokens": 55831924.0,
      "step": 32185
    },
    {
      "entropy": 5.367499971389771,
      "epoch": 2.504493289243338,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004367842690250288,
      "loss": 4.9626,
      "mean_token_accuracy": 0.20445212870836257,
      "num_tokens": 55840799.0,
      "step": 32190
    },
    {
      "entropy": 5.2977509021759035,
      "epoch": 2.5048823186150555,
      "grad_norm": 1.25,
      "learning_rate": 0.0004367650063663565,
      "loss": 4.9316,
      "mean_token_accuracy": 0.20463903844356537,
      "num_tokens": 55848950.0,
      "step": 32195
    },
    {
      "entropy": 5.329190969467163,
      "epoch": 2.505271347986773,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004367457412531576,
      "loss": 4.9736,
      "mean_token_accuracy": 0.2054935410618782,
      "num_tokens": 55857140.0,
      "step": 32200
    },
    {
      "entropy": 5.4136405944824215,
      "epoch": 2.5056603773584905,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004367264736857245,
      "loss": 4.9846,
      "mean_token_accuracy": 0.19516116976737977,
      "num_tokens": 55865790.0,
      "step": 32205
    },
    {
      "entropy": 5.350301361083984,
      "epoch": 2.506049406730208,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004367072036643497,
      "loss": 4.9351,
      "mean_token_accuracy": 0.19605758637189866,
      "num_tokens": 55873598.0,
      "step": 32210
    },
    {
      "entropy": 5.324070405960083,
      "epoch": 2.506438436101926,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043668793118932536,
      "loss": 5.0801,
      "mean_token_accuracy": 0.19455282986164094,
      "num_tokens": 55882172.0,
      "step": 32215
    },
    {
      "entropy": 5.35164442062378,
      "epoch": 2.506827465473643,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043666865626094405,
      "loss": 4.9553,
      "mean_token_accuracy": 0.1988823965191841,
      "num_tokens": 55890701.0,
      "step": 32220
    },
    {
      "entropy": 5.479476690292358,
      "epoch": 2.507216494845361,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043664937887949825,
      "loss": 5.0716,
      "mean_token_accuracy": 0.1848250225186348,
      "num_tokens": 55899470.0,
      "step": 32225
    },
    {
      "entropy": 5.378156614303589,
      "epoch": 2.5076055242170785,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004366300990452803,
      "loss": 4.9875,
      "mean_token_accuracy": 0.1949670895934105,
      "num_tokens": 55907442.0,
      "step": 32230
    },
    {
      "entropy": 5.369243001937866,
      "epoch": 2.507994553588796,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043661081675858295,
      "loss": 5.0058,
      "mean_token_accuracy": 0.19026459604501725,
      "num_tokens": 55915761.0,
      "step": 32235
    },
    {
      "entropy": 5.365401411056519,
      "epoch": 2.5083835829605134,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043659153201969865,
      "loss": 4.9861,
      "mean_token_accuracy": 0.20049808770418168,
      "num_tokens": 55924101.0,
      "step": 32240
    },
    {
      "entropy": 5.475956821441651,
      "epoch": 2.508772612332231,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004365722448289201,
      "loss": 5.1217,
      "mean_token_accuracy": 0.18209327459335328,
      "num_tokens": 55933472.0,
      "step": 32245
    },
    {
      "entropy": 5.374709606170654,
      "epoch": 2.5091616417039484,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004365529551865399,
      "loss": 4.9875,
      "mean_token_accuracy": 0.19523807466030121,
      "num_tokens": 55941962.0,
      "step": 32250
    },
    {
      "entropy": 5.3876111030578615,
      "epoch": 2.509550671075666,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043653366309285066,
      "loss": 5.0236,
      "mean_token_accuracy": 0.19339352250099182,
      "num_tokens": 55951392.0,
      "step": 32255
    },
    {
      "entropy": 5.405840539932251,
      "epoch": 2.509939700447384,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004365143685481452,
      "loss": 4.9861,
      "mean_token_accuracy": 0.19456800520420076,
      "num_tokens": 55959995.0,
      "step": 32260
    },
    {
      "entropy": 5.353262615203858,
      "epoch": 2.5103287298191015,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004364950715527164,
      "loss": 4.9478,
      "mean_token_accuracy": 0.19543107748031616,
      "num_tokens": 55967960.0,
      "step": 32265
    },
    {
      "entropy": 5.41112871170044,
      "epoch": 2.5107177591908187,
      "grad_norm": 1.125,
      "learning_rate": 0.0004364757721068568,
      "loss": 5.0239,
      "mean_token_accuracy": 0.19811365455389024,
      "num_tokens": 55976620.0,
      "step": 32270
    },
    {
      "entropy": 5.363190317153931,
      "epoch": 2.5111067885625364,
      "grad_norm": 1.125,
      "learning_rate": 0.00043645647021085945,
      "loss": 4.9825,
      "mean_token_accuracy": 0.19481093883514405,
      "num_tokens": 55985478.0,
      "step": 32275
    },
    {
      "entropy": 5.39055061340332,
      "epoch": 2.511495817934254,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043643716586501706,
      "loss": 4.9679,
      "mean_token_accuracy": 0.19641793072223662,
      "num_tokens": 55994123.0,
      "step": 32280
    },
    {
      "entropy": 5.374627923965454,
      "epoch": 2.5118848473059714,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043641785906962276,
      "loss": 5.0407,
      "mean_token_accuracy": 0.19278645068407058,
      "num_tokens": 56003088.0,
      "step": 32285
    },
    {
      "entropy": 5.314338684082031,
      "epoch": 2.512273876677689,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004363985498249693,
      "loss": 4.9346,
      "mean_token_accuracy": 0.20161727219820022,
      "num_tokens": 56011349.0,
      "step": 32290
    },
    {
      "entropy": 5.378583097457886,
      "epoch": 2.512662906049407,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043637923813134974,
      "loss": 4.9037,
      "mean_token_accuracy": 0.2030582159757614,
      "num_tokens": 56020225.0,
      "step": 32295
    },
    {
      "entropy": 5.459575176239014,
      "epoch": 2.5130519354211245,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043635992398905713,
      "loss": 5.0323,
      "mean_token_accuracy": 0.19823600649833678,
      "num_tokens": 56028729.0,
      "step": 32300
    },
    {
      "entropy": 5.353182315826416,
      "epoch": 2.5134409647928417,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004363406073983844,
      "loss": 5.003,
      "mean_token_accuracy": 0.19163133203983307,
      "num_tokens": 56037173.0,
      "step": 32305
    },
    {
      "entropy": 5.364936685562133,
      "epoch": 2.5138299941645594,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043632128835962485,
      "loss": 5.0637,
      "mean_token_accuracy": 0.19085622429847718,
      "num_tokens": 56046013.0,
      "step": 32310
    },
    {
      "entropy": 5.4796045303344725,
      "epoch": 2.514219023536277,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043630196687307156,
      "loss": 5.1,
      "mean_token_accuracy": 0.19091798663139342,
      "num_tokens": 56054942.0,
      "step": 32315
    },
    {
      "entropy": 5.482224082946777,
      "epoch": 2.5146080529079944,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004362826429390176,
      "loss": 5.1137,
      "mean_token_accuracy": 0.18429044783115386,
      "num_tokens": 56064033.0,
      "step": 32320
    },
    {
      "entropy": 5.388293027877808,
      "epoch": 2.514997082279712,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043626331655775624,
      "loss": 5.0537,
      "mean_token_accuracy": 0.19777873903512955,
      "num_tokens": 56072534.0,
      "step": 32325
    },
    {
      "entropy": 5.363391733169555,
      "epoch": 2.51538611165143,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043624398772958075,
      "loss": 4.9514,
      "mean_token_accuracy": 0.19831036925315856,
      "num_tokens": 56081742.0,
      "step": 32330
    },
    {
      "entropy": 5.383007955551148,
      "epoch": 2.5157751410231475,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004362246564547844,
      "loss": 4.9473,
      "mean_token_accuracy": 0.20103092044591903,
      "num_tokens": 56090659.0,
      "step": 32335
    },
    {
      "entropy": 5.368412685394287,
      "epoch": 2.5161641703948647,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043620532273366044,
      "loss": 4.9989,
      "mean_token_accuracy": 0.19000377058982848,
      "num_tokens": 56099795.0,
      "step": 32340
    },
    {
      "entropy": 5.32804012298584,
      "epoch": 2.5165531997665824,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004361859865665024,
      "loss": 5.0158,
      "mean_token_accuracy": 0.19625585228204728,
      "num_tokens": 56107265.0,
      "step": 32345
    },
    {
      "entropy": 5.363792037963867,
      "epoch": 2.5169422291382997,
      "grad_norm": 1.125,
      "learning_rate": 0.00043616664795360346,
      "loss": 4.975,
      "mean_token_accuracy": 0.20111559927463532,
      "num_tokens": 56116434.0,
      "step": 32350
    },
    {
      "entropy": 5.417178106307984,
      "epoch": 2.5173312585100174,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043614730689525716,
      "loss": 5.1034,
      "mean_token_accuracy": 0.18328833431005478,
      "num_tokens": 56125356.0,
      "step": 32355
    },
    {
      "entropy": 5.500271654129028,
      "epoch": 2.517720287881735,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043612796339175707,
      "loss": 5.1374,
      "mean_token_accuracy": 0.18579833209514618,
      "num_tokens": 56133732.0,
      "step": 32360
    },
    {
      "entropy": 5.434199523925781,
      "epoch": 2.5181093172534528,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043610861744339653,
      "loss": 4.991,
      "mean_token_accuracy": 0.19995840340852739,
      "num_tokens": 56142037.0,
      "step": 32365
    },
    {
      "entropy": 5.463851022720337,
      "epoch": 2.5184983466251705,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004360892690504692,
      "loss": 5.0214,
      "mean_token_accuracy": 0.1966513529419899,
      "num_tokens": 56151064.0,
      "step": 32370
    },
    {
      "entropy": 5.403643846511841,
      "epoch": 2.5188873759968877,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004360699182132685,
      "loss": 4.978,
      "mean_token_accuracy": 0.1986454263329506,
      "num_tokens": 56159617.0,
      "step": 32375
    },
    {
      "entropy": 5.381380319595337,
      "epoch": 2.5192764053686054,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043605056493208825,
      "loss": 4.9986,
      "mean_token_accuracy": 0.19302793592214584,
      "num_tokens": 56168941.0,
      "step": 32380
    },
    {
      "entropy": 5.426887512207031,
      "epoch": 2.5196654347403227,
      "grad_norm": 1.15625,
      "learning_rate": 0.000436031209207222,
      "loss": 4.9638,
      "mean_token_accuracy": 0.2005198448896408,
      "num_tokens": 56177321.0,
      "step": 32385
    },
    {
      "entropy": 5.407274866104126,
      "epoch": 2.5200544641120404,
      "grad_norm": 1.125,
      "learning_rate": 0.00043601185103896346,
      "loss": 5.0363,
      "mean_token_accuracy": 0.191109699010849,
      "num_tokens": 56185775.0,
      "step": 32390
    },
    {
      "entropy": 5.337054586410522,
      "epoch": 2.520443493483758,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043599249042760636,
      "loss": 5.0499,
      "mean_token_accuracy": 0.19688768088817596,
      "num_tokens": 56194949.0,
      "step": 32395
    },
    {
      "entropy": 5.3785394668579105,
      "epoch": 2.5208325228554758,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043597312737344446,
      "loss": 4.9498,
      "mean_token_accuracy": 0.19493609964847564,
      "num_tokens": 56203314.0,
      "step": 32400
    },
    {
      "entropy": 5.336468696594238,
      "epoch": 2.521221552227193,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043595376187677157,
      "loss": 5.0059,
      "mean_token_accuracy": 0.19823409765958785,
      "num_tokens": 56211857.0,
      "step": 32405
    },
    {
      "entropy": 5.396268224716186,
      "epoch": 2.5216105815989107,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043593439393788147,
      "loss": 5.0012,
      "mean_token_accuracy": 0.19800431430339813,
      "num_tokens": 56220331.0,
      "step": 32410
    },
    {
      "entropy": 5.389956903457642,
      "epoch": 2.5219996109706284,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004359150235570682,
      "loss": 5.0055,
      "mean_token_accuracy": 0.19728570878505708,
      "num_tokens": 56228623.0,
      "step": 32415
    },
    {
      "entropy": 5.310672092437744,
      "epoch": 2.5223886403423457,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004358956507346255,
      "loss": 4.9028,
      "mean_token_accuracy": 0.2062796860933304,
      "num_tokens": 56237028.0,
      "step": 32420
    },
    {
      "entropy": 5.364142417907715,
      "epoch": 2.5227776697140634,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004358762754708474,
      "loss": 4.9567,
      "mean_token_accuracy": 0.19958708584308624,
      "num_tokens": 56244955.0,
      "step": 32425
    },
    {
      "entropy": 5.335987329483032,
      "epoch": 2.523166699085781,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043585689776602795,
      "loss": 4.9844,
      "mean_token_accuracy": 0.19856718331575393,
      "num_tokens": 56254020.0,
      "step": 32430
    },
    {
      "entropy": 5.39691309928894,
      "epoch": 2.5235557284574988,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043583751762046104,
      "loss": 5.0372,
      "mean_token_accuracy": 0.19404468238353728,
      "num_tokens": 56262829.0,
      "step": 32435
    },
    {
      "entropy": 5.426384305953979,
      "epoch": 2.523944757829216,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004358181350344408,
      "loss": 5.0093,
      "mean_token_accuracy": 0.20217810869216918,
      "num_tokens": 56271289.0,
      "step": 32440
    },
    {
      "entropy": 5.306826591491699,
      "epoch": 2.5243337872009337,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004357987500082613,
      "loss": 4.9332,
      "mean_token_accuracy": 0.20542809665203093,
      "num_tokens": 56279701.0,
      "step": 32445
    },
    {
      "entropy": 5.4275970458984375,
      "epoch": 2.524722816572651,
      "grad_norm": 1.125,
      "learning_rate": 0.00043577936254221675,
      "loss": 4.9548,
      "mean_token_accuracy": 0.2007301613688469,
      "num_tokens": 56288373.0,
      "step": 32450
    },
    {
      "entropy": 5.40239372253418,
      "epoch": 2.5251118459443687,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004357599726366013,
      "loss": 4.998,
      "mean_token_accuracy": 0.19390948861837387,
      "num_tokens": 56297853.0,
      "step": 32455
    },
    {
      "entropy": 5.447829151153565,
      "epoch": 2.5255008753160864,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004357405802917091,
      "loss": 5.1447,
      "mean_token_accuracy": 0.18836611658334732,
      "num_tokens": 56306456.0,
      "step": 32460
    },
    {
      "entropy": 5.366394472122193,
      "epoch": 2.525889904687804,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004357211855078345,
      "loss": 4.8926,
      "mean_token_accuracy": 0.19851909726858138,
      "num_tokens": 56314420.0,
      "step": 32465
    },
    {
      "entropy": 5.393977642059326,
      "epoch": 2.5262789340595218,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043570178828527157,
      "loss": 5.0811,
      "mean_token_accuracy": 0.18705134838819504,
      "num_tokens": 56324011.0,
      "step": 32470
    },
    {
      "entropy": 5.342649936676025,
      "epoch": 2.526667963431239,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004356823886243149,
      "loss": 4.958,
      "mean_token_accuracy": 0.20071180760860444,
      "num_tokens": 56332830.0,
      "step": 32475
    },
    {
      "entropy": 5.440405035018921,
      "epoch": 2.5270569928029567,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043566298652525875,
      "loss": 5.0282,
      "mean_token_accuracy": 0.19704444110393524,
      "num_tokens": 56341486.0,
      "step": 32480
    },
    {
      "entropy": 5.409482479095459,
      "epoch": 2.527446022174674,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043564358198839746,
      "loss": 5.0196,
      "mean_token_accuracy": 0.19949520379304886,
      "num_tokens": 56349382.0,
      "step": 32485
    },
    {
      "entropy": 5.407753324508667,
      "epoch": 2.5278350515463917,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004356241750140256,
      "loss": 5.0704,
      "mean_token_accuracy": 0.19491202682256697,
      "num_tokens": 56358335.0,
      "step": 32490
    },
    {
      "entropy": 5.463453960418701,
      "epoch": 2.5282240809181094,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043560476560243743,
      "loss": 5.0704,
      "mean_token_accuracy": 0.19193231910467148,
      "num_tokens": 56366777.0,
      "step": 32495
    },
    {
      "entropy": 5.464003419876098,
      "epoch": 2.528613110289827,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004355853537539276,
      "loss": 5.0714,
      "mean_token_accuracy": 0.19671618938446045,
      "num_tokens": 56375175.0,
      "step": 32500
    },
    {
      "entropy": 5.302193117141724,
      "epoch": 2.5290021396615443,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004355659394687906,
      "loss": 4.9836,
      "mean_token_accuracy": 0.19429001063108445,
      "num_tokens": 56384442.0,
      "step": 32505
    },
    {
      "entropy": 5.422477340698242,
      "epoch": 2.529391169033262,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004355465227473211,
      "loss": 5.0555,
      "mean_token_accuracy": 0.18986300230026246,
      "num_tokens": 56392420.0,
      "step": 32510
    },
    {
      "entropy": 5.41943941116333,
      "epoch": 2.5297801984049797,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004355271035898136,
      "loss": 5.0809,
      "mean_token_accuracy": 0.19782855808734895,
      "num_tokens": 56400633.0,
      "step": 32515
    },
    {
      "entropy": 5.38399305343628,
      "epoch": 2.530169227776697,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004355076819965628,
      "loss": 4.9798,
      "mean_token_accuracy": 0.19667167216539383,
      "num_tokens": 56409606.0,
      "step": 32520
    },
    {
      "entropy": 5.406248712539673,
      "epoch": 2.5305582571484146,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043548825796786353,
      "loss": 5.0245,
      "mean_token_accuracy": 0.20279947072267532,
      "num_tokens": 56417724.0,
      "step": 32525
    },
    {
      "entropy": 5.396423625946045,
      "epoch": 2.5309472865201323,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043546883150401023,
      "loss": 5.0015,
      "mean_token_accuracy": 0.19793458729982377,
      "num_tokens": 56426632.0,
      "step": 32530
    },
    {
      "entropy": 5.348414611816406,
      "epoch": 2.53133631589185,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043544940260529787,
      "loss": 4.9335,
      "mean_token_accuracy": 0.19994964003562926,
      "num_tokens": 56435266.0,
      "step": 32535
    },
    {
      "entropy": 5.452577590942383,
      "epoch": 2.5317253452635673,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004354299712720213,
      "loss": 5.0483,
      "mean_token_accuracy": 0.1944741725921631,
      "num_tokens": 56443335.0,
      "step": 32540
    },
    {
      "entropy": 5.415429401397705,
      "epoch": 2.532114374635285,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004354105375044752,
      "loss": 5.0079,
      "mean_token_accuracy": 0.2014174073934555,
      "num_tokens": 56452657.0,
      "step": 32545
    },
    {
      "entropy": 5.414242315292358,
      "epoch": 2.5325034040070027,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043539110130295445,
      "loss": 5.0888,
      "mean_token_accuracy": 0.19432550221681594,
      "num_tokens": 56461925.0,
      "step": 32550
    },
    {
      "entropy": 5.39635534286499,
      "epoch": 2.53289243337872,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043537166266775405,
      "loss": 4.9974,
      "mean_token_accuracy": 0.20099618285894394,
      "num_tokens": 56470356.0,
      "step": 32555
    },
    {
      "entropy": 5.385881185531616,
      "epoch": 2.5332814627504376,
      "grad_norm": 1.234375,
      "learning_rate": 0.000435352221599169,
      "loss": 5.0535,
      "mean_token_accuracy": 0.19082618355751038,
      "num_tokens": 56478754.0,
      "step": 32560
    },
    {
      "entropy": 5.408781290054321,
      "epoch": 2.5336704921221553,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004353327780974941,
      "loss": 4.9916,
      "mean_token_accuracy": 0.1979315087199211,
      "num_tokens": 56487089.0,
      "step": 32565
    },
    {
      "entropy": 5.419483232498169,
      "epoch": 2.534059521493873,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004353133321630246,
      "loss": 5.0163,
      "mean_token_accuracy": 0.1932085156440735,
      "num_tokens": 56496416.0,
      "step": 32570
    },
    {
      "entropy": 5.31560788154602,
      "epoch": 2.5344485508655903,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004352938837960554,
      "loss": 5.0124,
      "mean_token_accuracy": 0.19460261315107347,
      "num_tokens": 56505675.0,
      "step": 32575
    },
    {
      "entropy": 5.415780258178711,
      "epoch": 2.534837580237308,
      "grad_norm": 1.125,
      "learning_rate": 0.00043527443299688166,
      "loss": 5.0224,
      "mean_token_accuracy": 0.19545642882585526,
      "num_tokens": 56514855.0,
      "step": 32580
    },
    {
      "entropy": 5.514229440689087,
      "epoch": 2.5352266096090252,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043525497976579843,
      "loss": 5.1121,
      "mean_token_accuracy": 0.18686698228120804,
      "num_tokens": 56524073.0,
      "step": 32585
    },
    {
      "entropy": 5.436513948440552,
      "epoch": 2.535615638980743,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000435235524103101,
      "loss": 5.0518,
      "mean_token_accuracy": 0.195856511592865,
      "num_tokens": 56534127.0,
      "step": 32590
    },
    {
      "entropy": 5.522740650177002,
      "epoch": 2.5360046683524606,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004352160660090846,
      "loss": 5.1311,
      "mean_token_accuracy": 0.18736134320497513,
      "num_tokens": 56543219.0,
      "step": 32595
    },
    {
      "entropy": 5.448683834075927,
      "epoch": 2.5363936977241783,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004351966054840443,
      "loss": 5.0834,
      "mean_token_accuracy": 0.19782322347164155,
      "num_tokens": 56552146.0,
      "step": 32600
    },
    {
      "entropy": 5.457414627075195,
      "epoch": 2.5367827270958956,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043517714252827556,
      "loss": 5.1384,
      "mean_token_accuracy": 0.18486394435167314,
      "num_tokens": 56561780.0,
      "step": 32605
    },
    {
      "entropy": 5.37010498046875,
      "epoch": 2.5371717564676133,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004351576771420737,
      "loss": 4.944,
      "mean_token_accuracy": 0.19819338321685792,
      "num_tokens": 56569864.0,
      "step": 32610
    },
    {
      "entropy": 5.479020404815674,
      "epoch": 2.537560785839331,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004351382093257339,
      "loss": 5.124,
      "mean_token_accuracy": 0.18459818959236146,
      "num_tokens": 56578506.0,
      "step": 32615
    },
    {
      "entropy": 5.457563018798828,
      "epoch": 2.5379498152110482,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004351187390795517,
      "loss": 5.0448,
      "mean_token_accuracy": 0.19855828434228898,
      "num_tokens": 56587264.0,
      "step": 32620
    },
    {
      "entropy": 5.396497774124145,
      "epoch": 2.538338844582766,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043509926640382244,
      "loss": 4.9256,
      "mean_token_accuracy": 0.20001353621482848,
      "num_tokens": 56595779.0,
      "step": 32625
    },
    {
      "entropy": 5.33933391571045,
      "epoch": 2.5387278739544836,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004350797912988417,
      "loss": 4.9563,
      "mean_token_accuracy": 0.19855376332998276,
      "num_tokens": 56604251.0,
      "step": 32630
    },
    {
      "entropy": 5.318504428863525,
      "epoch": 2.5391169033262013,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004350603137649049,
      "loss": 4.9722,
      "mean_token_accuracy": 0.19659413546323776,
      "num_tokens": 56612527.0,
      "step": 32635
    },
    {
      "entropy": 5.423489713668824,
      "epoch": 2.5395059326979186,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004350408338023075,
      "loss": 5.014,
      "mean_token_accuracy": 0.19397053867578506,
      "num_tokens": 56621139.0,
      "step": 32640
    },
    {
      "entropy": 5.401403522491455,
      "epoch": 2.5398949620696363,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004350213514113453,
      "loss": 4.9743,
      "mean_token_accuracy": 0.19930481016635895,
      "num_tokens": 56629757.0,
      "step": 32645
    },
    {
      "entropy": 5.312555932998658,
      "epoch": 2.540283991441354,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043500186659231377,
      "loss": 4.9156,
      "mean_token_accuracy": 0.20199532210826873,
      "num_tokens": 56638518.0,
      "step": 32650
    },
    {
      "entropy": 5.336606359481811,
      "epoch": 2.5406730208130712,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004349823793455086,
      "loss": 5.0118,
      "mean_token_accuracy": 0.18905456513166427,
      "num_tokens": 56647542.0,
      "step": 32655
    },
    {
      "entropy": 5.42414813041687,
      "epoch": 2.541062050184789,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043496288967122547,
      "loss": 4.9969,
      "mean_token_accuracy": 0.1935635358095169,
      "num_tokens": 56655771.0,
      "step": 32660
    },
    {
      "entropy": 5.309195566177368,
      "epoch": 2.5414510795565066,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043494339756976007,
      "loss": 4.9366,
      "mean_token_accuracy": 0.2066170781850815,
      "num_tokens": 56663895.0,
      "step": 32665
    },
    {
      "entropy": 5.333622121810913,
      "epoch": 2.5418401089282243,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004349239030414082,
      "loss": 5.0061,
      "mean_token_accuracy": 0.19302317202091218,
      "num_tokens": 56672557.0,
      "step": 32670
    },
    {
      "entropy": 5.4179261207580565,
      "epoch": 2.5422291382999416,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043490440608646557,
      "loss": 5.0766,
      "mean_token_accuracy": 0.19722854495048522,
      "num_tokens": 56681095.0,
      "step": 32675
    },
    {
      "entropy": 5.4278827667236325,
      "epoch": 2.5426181676716593,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043488490670522817,
      "loss": 5.0723,
      "mean_token_accuracy": 0.1944474071264267,
      "num_tokens": 56689347.0,
      "step": 32680
    },
    {
      "entropy": 5.348727130889893,
      "epoch": 2.5430071970433765,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004348654048979918,
      "loss": 4.9644,
      "mean_token_accuracy": 0.19775613397359848,
      "num_tokens": 56698348.0,
      "step": 32685
    },
    {
      "entropy": 5.398197317123413,
      "epoch": 2.543396226415094,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004348459006650523,
      "loss": 5.0506,
      "mean_token_accuracy": 0.19802492558956147,
      "num_tokens": 56707215.0,
      "step": 32690
    },
    {
      "entropy": 5.407253074645996,
      "epoch": 2.543785255786812,
      "grad_norm": 1.125,
      "learning_rate": 0.00043482639400670576,
      "loss": 5.0713,
      "mean_token_accuracy": 0.1932977095246315,
      "num_tokens": 56716190.0,
      "step": 32695
    },
    {
      "entropy": 5.407024383544922,
      "epoch": 2.5441742851585296,
      "grad_norm": 1.140625,
      "learning_rate": 0.000434806884923248,
      "loss": 4.9545,
      "mean_token_accuracy": 0.20242834240198135,
      "num_tokens": 56724494.0,
      "step": 32700
    },
    {
      "entropy": 5.4086089611053465,
      "epoch": 2.5445633145302473,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004347873734149752,
      "loss": 4.9587,
      "mean_token_accuracy": 0.2055629700422287,
      "num_tokens": 56732648.0,
      "step": 32705
    },
    {
      "entropy": 5.498168420791626,
      "epoch": 2.5449523439019646,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043476785948218325,
      "loss": 5.2152,
      "mean_token_accuracy": 0.18798192143440245,
      "num_tokens": 56742270.0,
      "step": 32710
    },
    {
      "entropy": 5.420835733413696,
      "epoch": 2.5453413732736823,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043474834312516835,
      "loss": 4.9927,
      "mean_token_accuracy": 0.19400714486837387,
      "num_tokens": 56750982.0,
      "step": 32715
    },
    {
      "entropy": 5.33693528175354,
      "epoch": 2.5457304026453995,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004347288243442266,
      "loss": 4.9251,
      "mean_token_accuracy": 0.20153277218341828,
      "num_tokens": 56759443.0,
      "step": 32720
    },
    {
      "entropy": 5.45634708404541,
      "epoch": 2.546119432017117,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004347093031396543,
      "loss": 5.029,
      "mean_token_accuracy": 0.19521800577640533,
      "num_tokens": 56768704.0,
      "step": 32725
    },
    {
      "entropy": 5.404246711730957,
      "epoch": 2.546508461388835,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004346897795117474,
      "loss": 5.0473,
      "mean_token_accuracy": 0.1997896119952202,
      "num_tokens": 56777274.0,
      "step": 32730
    },
    {
      "entropy": 5.434302425384521,
      "epoch": 2.5468974907605526,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004346702534608023,
      "loss": 5.0352,
      "mean_token_accuracy": 0.1949635699391365,
      "num_tokens": 56786184.0,
      "step": 32735
    },
    {
      "entropy": 5.350253343582153,
      "epoch": 2.54728652013227,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004346507249871153,
      "loss": 5.0433,
      "mean_token_accuracy": 0.19283165484666825,
      "num_tokens": 56795156.0,
      "step": 32740
    },
    {
      "entropy": 5.444284915924072,
      "epoch": 2.5476755495039876,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004346311940909826,
      "loss": 5.0543,
      "mean_token_accuracy": 0.19348152130842208,
      "num_tokens": 56803829.0,
      "step": 32745
    },
    {
      "entropy": 5.408627128601074,
      "epoch": 2.5480645788757053,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043461166077270056,
      "loss": 5.0102,
      "mean_token_accuracy": 0.20693886131048203,
      "num_tokens": 56813191.0,
      "step": 32750
    },
    {
      "entropy": 5.405750036239624,
      "epoch": 2.5484536082474225,
      "grad_norm": 1.125,
      "learning_rate": 0.00043459212503256566,
      "loss": 5.0598,
      "mean_token_accuracy": 0.19557723999023438,
      "num_tokens": 56822338.0,
      "step": 32755
    },
    {
      "entropy": 5.428199291229248,
      "epoch": 2.54884263761914,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004345725868708743,
      "loss": 5.0052,
      "mean_token_accuracy": 0.19502134174108504,
      "num_tokens": 56830873.0,
      "step": 32760
    },
    {
      "entropy": 5.382936096191406,
      "epoch": 2.549231666990858,
      "grad_norm": 1.125,
      "learning_rate": 0.00043455304628792294,
      "loss": 4.9579,
      "mean_token_accuracy": 0.19868134111166,
      "num_tokens": 56839462.0,
      "step": 32765
    },
    {
      "entropy": 5.517848968505859,
      "epoch": 2.5496206963625756,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000434533503284008,
      "loss": 5.0456,
      "mean_token_accuracy": 0.1939234256744385,
      "num_tokens": 56847713.0,
      "step": 32770
    },
    {
      "entropy": 5.450197696685791,
      "epoch": 2.550009725734293,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004345139578594261,
      "loss": 4.9474,
      "mean_token_accuracy": 0.20427315235137938,
      "num_tokens": 56856507.0,
      "step": 32775
    },
    {
      "entropy": 5.358585357666016,
      "epoch": 2.5503987551060106,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004344944100144738,
      "loss": 5.0707,
      "mean_token_accuracy": 0.19834039956331254,
      "num_tokens": 56864772.0,
      "step": 32780
    },
    {
      "entropy": 5.434321546554566,
      "epoch": 2.550787784477728,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043447485974944765,
      "loss": 5.0417,
      "mean_token_accuracy": 0.1862793430685997,
      "num_tokens": 56872886.0,
      "step": 32785
    },
    {
      "entropy": 5.396774673461914,
      "epoch": 2.5511768138494455,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043445530706464437,
      "loss": 5.0074,
      "mean_token_accuracy": 0.1932521790266037,
      "num_tokens": 56882079.0,
      "step": 32790
    },
    {
      "entropy": 5.376486110687256,
      "epoch": 2.551565843221163,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004344357519603606,
      "loss": 4.9631,
      "mean_token_accuracy": 0.1986935555934906,
      "num_tokens": 56890237.0,
      "step": 32795
    },
    {
      "entropy": 5.492238998413086,
      "epoch": 2.551954872592881,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004344161944368931,
      "loss": 5.0525,
      "mean_token_accuracy": 0.18952853828668595,
      "num_tokens": 56898703.0,
      "step": 32800
    },
    {
      "entropy": 5.4621562480926515,
      "epoch": 2.5523439019645986,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004343966344945385,
      "loss": 5.1228,
      "mean_token_accuracy": 0.19284837543964387,
      "num_tokens": 56907954.0,
      "step": 32805
    },
    {
      "entropy": 5.2630205154418945,
      "epoch": 2.552732931336316,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043437707213359375,
      "loss": 4.8672,
      "mean_token_accuracy": 0.20343452841043472,
      "num_tokens": 56916316.0,
      "step": 32810
    },
    {
      "entropy": 5.339161109924317,
      "epoch": 2.5531219607080335,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043435750735435554,
      "loss": 4.9479,
      "mean_token_accuracy": 0.20337121188640594,
      "num_tokens": 56924386.0,
      "step": 32815
    },
    {
      "entropy": 5.450483703613282,
      "epoch": 2.553510990079751,
      "grad_norm": 1.125,
      "learning_rate": 0.00043433794015712085,
      "loss": 5.0234,
      "mean_token_accuracy": 0.20420075803995133,
      "num_tokens": 56932782.0,
      "step": 32820
    },
    {
      "entropy": 5.365355920791626,
      "epoch": 2.5539000194514685,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004343183705421865,
      "loss": 4.9564,
      "mean_token_accuracy": 0.20117889940738679,
      "num_tokens": 56941030.0,
      "step": 32825
    },
    {
      "entropy": 5.54454402923584,
      "epoch": 2.554289048823186,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004342987985098495,
      "loss": 5.2103,
      "mean_token_accuracy": 0.18563813269138335,
      "num_tokens": 56949525.0,
      "step": 32830
    },
    {
      "entropy": 5.361660051345825,
      "epoch": 2.554678078194904,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043427922406040676,
      "loss": 4.9559,
      "mean_token_accuracy": 0.19782355427742004,
      "num_tokens": 56958317.0,
      "step": 32835
    },
    {
      "entropy": 5.36912431716919,
      "epoch": 2.555067107566621,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043425964719415525,
      "loss": 5.0263,
      "mean_token_accuracy": 0.20390138328075408,
      "num_tokens": 56966848.0,
      "step": 32840
    },
    {
      "entropy": 5.482934093475341,
      "epoch": 2.555456136938339,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043424006791139213,
      "loss": 5.0955,
      "mean_token_accuracy": 0.18911695927381517,
      "num_tokens": 56975448.0,
      "step": 32845
    },
    {
      "entropy": 5.442656755447388,
      "epoch": 2.5558451663100565,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043422048621241443,
      "loss": 5.1118,
      "mean_token_accuracy": 0.19205926209688187,
      "num_tokens": 56983800.0,
      "step": 32850
    },
    {
      "entropy": 5.452989196777343,
      "epoch": 2.556234195681774,
      "grad_norm": 1.125,
      "learning_rate": 0.00043420090209751933,
      "loss": 5.0402,
      "mean_token_accuracy": 0.19671180695295334,
      "num_tokens": 56992505.0,
      "step": 32855
    },
    {
      "entropy": 5.42435998916626,
      "epoch": 2.5566232250534915,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004341813155670039,
      "loss": 4.9349,
      "mean_token_accuracy": 0.20290036350488663,
      "num_tokens": 57000471.0,
      "step": 32860
    },
    {
      "entropy": 5.4749256610870365,
      "epoch": 2.557012254425209,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004341617266211654,
      "loss": 5.0124,
      "mean_token_accuracy": 0.19197063893079758,
      "num_tokens": 57008708.0,
      "step": 32865
    },
    {
      "entropy": 5.464524507522583,
      "epoch": 2.557401283796927,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000434142135260301,
      "loss": 5.116,
      "mean_token_accuracy": 0.19214038103818892,
      "num_tokens": 57017395.0,
      "step": 32870
    },
    {
      "entropy": 5.459582424163818,
      "epoch": 2.557790313168644,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000434122541484708,
      "loss": 4.9621,
      "mean_token_accuracy": 0.19185912758111953,
      "num_tokens": 57025535.0,
      "step": 32875
    },
    {
      "entropy": 5.401855564117431,
      "epoch": 2.558179342540362,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004341029452946837,
      "loss": 4.9664,
      "mean_token_accuracy": 0.19888245910406113,
      "num_tokens": 57034029.0,
      "step": 32880
    },
    {
      "entropy": 5.385588788986206,
      "epoch": 2.558568371912079,
      "grad_norm": 1.125,
      "learning_rate": 0.0004340833466905254,
      "loss": 5.0896,
      "mean_token_accuracy": 0.19344945400953292,
      "num_tokens": 57042779.0,
      "step": 32885
    },
    {
      "entropy": 5.443098163604736,
      "epoch": 2.558957401283797,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043406374567253055,
      "loss": 5.053,
      "mean_token_accuracy": 0.19101290851831437,
      "num_tokens": 57052506.0,
      "step": 32890
    },
    {
      "entropy": 5.492218160629273,
      "epoch": 2.5593464306555145,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004340441422409965,
      "loss": 5.0858,
      "mean_token_accuracy": 0.1901238054037094,
      "num_tokens": 57061307.0,
      "step": 32895
    },
    {
      "entropy": 5.430581760406494,
      "epoch": 2.559735460027232,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004340245363962209,
      "loss": 5.0931,
      "mean_token_accuracy": 0.197430220246315,
      "num_tokens": 57069991.0,
      "step": 32900
    },
    {
      "entropy": 5.442956018447876,
      "epoch": 2.56012448939895,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043400492813850083,
      "loss": 4.9924,
      "mean_token_accuracy": 0.19772007167339326,
      "num_tokens": 57077768.0,
      "step": 32905
    },
    {
      "entropy": 5.337126350402832,
      "epoch": 2.560513518770667,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043398531746813426,
      "loss": 4.981,
      "mean_token_accuracy": 0.1988024741411209,
      "num_tokens": 57086666.0,
      "step": 32910
    },
    {
      "entropy": 5.342815780639649,
      "epoch": 2.560902548142385,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043396570438541845,
      "loss": 5.0053,
      "mean_token_accuracy": 0.20065800100564957,
      "num_tokens": 57095352.0,
      "step": 32915
    },
    {
      "entropy": 5.367441272735595,
      "epoch": 2.561291577514102,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004339460888906512,
      "loss": 5.0099,
      "mean_token_accuracy": 0.20147421360015869,
      "num_tokens": 57104535.0,
      "step": 32920
    },
    {
      "entropy": 5.463930463790893,
      "epoch": 2.5616806068858198,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004339264709841299,
      "loss": 4.9982,
      "mean_token_accuracy": 0.1912730574607849,
      "num_tokens": 57112523.0,
      "step": 32925
    },
    {
      "entropy": 5.317163133621216,
      "epoch": 2.5620696362575375,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043390685066615244,
      "loss": 4.8897,
      "mean_token_accuracy": 0.19990423917770386,
      "num_tokens": 57120921.0,
      "step": 32930
    },
    {
      "entropy": 5.4058739185333256,
      "epoch": 2.562458665629255,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004338872279370164,
      "loss": 5.0456,
      "mean_token_accuracy": 0.19243020862340926,
      "num_tokens": 57128398.0,
      "step": 32935
    },
    {
      "entropy": 5.4275659084320065,
      "epoch": 2.5628476950009724,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004338676027970196,
      "loss": 5.0415,
      "mean_token_accuracy": 0.19471668303012848,
      "num_tokens": 57136888.0,
      "step": 32940
    },
    {
      "entropy": 5.398860216140747,
      "epoch": 2.56323672437269,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043384797524645977,
      "loss": 4.9933,
      "mean_token_accuracy": 0.19868036806583406,
      "num_tokens": 57144795.0,
      "step": 32945
    },
    {
      "entropy": 5.448951768875122,
      "epoch": 2.563625753744408,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004338283452856348,
      "loss": 5.127,
      "mean_token_accuracy": 0.19479160904884338,
      "num_tokens": 57153073.0,
      "step": 32950
    },
    {
      "entropy": 5.449031019210816,
      "epoch": 2.564014783116125,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004338087129148425,
      "loss": 5.0846,
      "mean_token_accuracy": 0.19696416705846786,
      "num_tokens": 57161536.0,
      "step": 32955
    },
    {
      "entropy": 5.436656475067139,
      "epoch": 2.5644038124878428,
      "grad_norm": 1.34375,
      "learning_rate": 0.00043378907813438066,
      "loss": 5.06,
      "mean_token_accuracy": 0.19759908318519592,
      "num_tokens": 57170065.0,
      "step": 32960
    },
    {
      "entropy": 5.337686061859131,
      "epoch": 2.5647928418595605,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043376944094454734,
      "loss": 4.9291,
      "mean_token_accuracy": 0.1999871015548706,
      "num_tokens": 57179063.0,
      "step": 32965
    },
    {
      "entropy": 5.353264236450196,
      "epoch": 2.565181871231278,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043374980134564044,
      "loss": 4.8604,
      "mean_token_accuracy": 0.21347493678331375,
      "num_tokens": 57187803.0,
      "step": 32970
    },
    {
      "entropy": 5.475432920455932,
      "epoch": 2.5655709006029954,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043373015933795804,
      "loss": 5.1087,
      "mean_token_accuracy": 0.18489850014448167,
      "num_tokens": 57196428.0,
      "step": 32975
    },
    {
      "entropy": 5.4286253452301025,
      "epoch": 2.565959929974713,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004337105149217981,
      "loss": 5.0758,
      "mean_token_accuracy": 0.19324132800102234,
      "num_tokens": 57205541.0,
      "step": 32980
    },
    {
      "entropy": 5.393906211853027,
      "epoch": 2.566348959346431,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043369086809745875,
      "loss": 4.9919,
      "mean_token_accuracy": 0.20310668796300888,
      "num_tokens": 57214436.0,
      "step": 32985
    },
    {
      "entropy": 5.415428686141968,
      "epoch": 2.566737988718148,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043367121886523795,
      "loss": 5.0157,
      "mean_token_accuracy": 0.19501292407512666,
      "num_tokens": 57222359.0,
      "step": 32990
    },
    {
      "entropy": 5.361168718338012,
      "epoch": 2.5671270180898658,
      "grad_norm": 1.15625,
      "learning_rate": 0.000433651567225434,
      "loss": 4.9214,
      "mean_token_accuracy": 0.1974118486046791,
      "num_tokens": 57230226.0,
      "step": 32995
    },
    {
      "entropy": 5.38568606376648,
      "epoch": 2.5675160474615835,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004336319131783451,
      "loss": 5.0762,
      "mean_token_accuracy": 0.19096975922584533,
      "num_tokens": 57239585.0,
      "step": 33000
    },
    {
      "epoch": 2.5675160474615835,
      "eval_entropy": 5.215213312106825,
      "eval_loss": 5.131958484649658,
      "eval_mean_token_accuracy": 0.1991191391880634,
      "eval_num_tokens": 57239585.0,
      "eval_runtime": 28.8503,
      "eval_samples_per_second": 1440.468,
      "eval_steps_per_second": 180.067,
      "step": 33000
    },
    {
      "entropy": 5.376195192337036,
      "epoch": 2.567905076833301,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043361225672426933,
      "loss": 5.0413,
      "mean_token_accuracy": 0.20064694434404373,
      "num_tokens": 57248108.0,
      "step": 33005
    },
    {
      "entropy": 5.42994270324707,
      "epoch": 2.5682941062050184,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004335925978635051,
      "loss": 4.9929,
      "mean_token_accuracy": 0.19682959765195845,
      "num_tokens": 57257758.0,
      "step": 33010
    },
    {
      "entropy": 5.441091680526734,
      "epoch": 2.568683135576736,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043357293659635057,
      "loss": 5.1014,
      "mean_token_accuracy": 0.18894279152154922,
      "num_tokens": 57267180.0,
      "step": 33015
    },
    {
      "entropy": 5.381822347640991,
      "epoch": 2.5690721649484534,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004335532729231041,
      "loss": 4.9959,
      "mean_token_accuracy": 0.20056770890951156,
      "num_tokens": 57275756.0,
      "step": 33020
    },
    {
      "entropy": 5.388438367843628,
      "epoch": 2.569461194320171,
      "grad_norm": 1.328125,
      "learning_rate": 0.00043353360684406415,
      "loss": 5.0348,
      "mean_token_accuracy": 0.19274474680423737,
      "num_tokens": 57284593.0,
      "step": 33025
    },
    {
      "entropy": 5.395043230056762,
      "epoch": 2.5698502236918888,
      "grad_norm": 1.203125,
      "learning_rate": 0.000433513938359529,
      "loss": 5.0681,
      "mean_token_accuracy": 0.19115755409002305,
      "num_tokens": 57293304.0,
      "step": 33030
    },
    {
      "entropy": 5.407206392288208,
      "epoch": 2.5702392530636065,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004334942674697971,
      "loss": 5.0183,
      "mean_token_accuracy": 0.20076577961444855,
      "num_tokens": 57302057.0,
      "step": 33035
    },
    {
      "entropy": 5.456658983230591,
      "epoch": 2.5706282824353237,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043347459417516696,
      "loss": 5.0438,
      "mean_token_accuracy": 0.20029698610305785,
      "num_tokens": 57310576.0,
      "step": 33040
    },
    {
      "entropy": 5.452741622924805,
      "epoch": 2.5710173118070414,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004334549184759371,
      "loss": 5.0142,
      "mean_token_accuracy": 0.196185202896595,
      "num_tokens": 57319103.0,
      "step": 33045
    },
    {
      "entropy": 5.472675848007202,
      "epoch": 2.571406341178759,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004334352403724062,
      "loss": 5.1049,
      "mean_token_accuracy": 0.18930304050445557,
      "num_tokens": 57329075.0,
      "step": 33050
    },
    {
      "entropy": 5.429934644699097,
      "epoch": 2.5717953705504764,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00043341555986487254,
      "loss": 4.9844,
      "mean_token_accuracy": 0.1963612347841263,
      "num_tokens": 57337986.0,
      "step": 33055
    },
    {
      "entropy": 5.405981779098511,
      "epoch": 2.572184399922194,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004333958769536349,
      "loss": 5.0926,
      "mean_token_accuracy": 0.18935710191726685,
      "num_tokens": 57346440.0,
      "step": 33060
    },
    {
      "entropy": 5.461093378067017,
      "epoch": 2.5725734292939118,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004333761916389921,
      "loss": 5.091,
      "mean_token_accuracy": 0.1845938578248024,
      "num_tokens": 57354791.0,
      "step": 33065
    },
    {
      "entropy": 5.417821168899536,
      "epoch": 2.5729624586656294,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004333565039212425,
      "loss": 5.0061,
      "mean_token_accuracy": 0.19279139190912248,
      "num_tokens": 57363200.0,
      "step": 33070
    },
    {
      "entropy": 5.364356899261475,
      "epoch": 2.5733514880373467,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004333368138006851,
      "loss": 4.9916,
      "mean_token_accuracy": 0.19710914343595504,
      "num_tokens": 57371830.0,
      "step": 33075
    },
    {
      "entropy": 5.4708263874053955,
      "epoch": 2.5737405174090644,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004333171212776185,
      "loss": 5.1243,
      "mean_token_accuracy": 0.18771326839923858,
      "num_tokens": 57380123.0,
      "step": 33080
    },
    {
      "entropy": 5.442349529266357,
      "epoch": 2.574129546780782,
      "grad_norm": 1.125,
      "learning_rate": 0.0004332974263523416,
      "loss": 5.0978,
      "mean_token_accuracy": 0.19402647614479065,
      "num_tokens": 57389093.0,
      "step": 33085
    },
    {
      "entropy": 5.213416290283203,
      "epoch": 2.5745185761524993,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043327772902515327,
      "loss": 4.8117,
      "mean_token_accuracy": 0.2112378090620041,
      "num_tokens": 57397165.0,
      "step": 33090
    },
    {
      "entropy": 5.367502880096436,
      "epoch": 2.574907605524217,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004332580292963523,
      "loss": 5.105,
      "mean_token_accuracy": 0.19111934751272203,
      "num_tokens": 57406308.0,
      "step": 33095
    },
    {
      "entropy": 5.390132331848145,
      "epoch": 2.5752966348959347,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004332383271662376,
      "loss": 4.9116,
      "mean_token_accuracy": 0.1990741416811943,
      "num_tokens": 57414964.0,
      "step": 33100
    },
    {
      "entropy": 5.315077543258667,
      "epoch": 2.5756856642676524,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043321862263510816,
      "loss": 4.8523,
      "mean_token_accuracy": 0.2151111349463463,
      "num_tokens": 57422684.0,
      "step": 33105
    },
    {
      "entropy": 5.313579177856445,
      "epoch": 2.5760746936393697,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004331989157032629,
      "loss": 4.9788,
      "mean_token_accuracy": 0.19050878137350083,
      "num_tokens": 57431784.0,
      "step": 33110
    },
    {
      "entropy": 5.410853672027588,
      "epoch": 2.5764637230110874,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043317920637100097,
      "loss": 5.0575,
      "mean_token_accuracy": 0.1955583870410919,
      "num_tokens": 57440467.0,
      "step": 33115
    },
    {
      "entropy": 5.391645669937134,
      "epoch": 2.5768527523828046,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004331594946386213,
      "loss": 4.9589,
      "mean_token_accuracy": 0.1942771404981613,
      "num_tokens": 57449263.0,
      "step": 33120
    },
    {
      "entropy": 5.39660234451294,
      "epoch": 2.5772417817545223,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043313978050642303,
      "loss": 4.9615,
      "mean_token_accuracy": 0.20407168865203856,
      "num_tokens": 57457662.0,
      "step": 33125
    },
    {
      "entropy": 5.382514429092407,
      "epoch": 2.57763081112624,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004331200639747053,
      "loss": 4.9571,
      "mean_token_accuracy": 0.20231837034225464,
      "num_tokens": 57466049.0,
      "step": 33130
    },
    {
      "entropy": 5.35534987449646,
      "epoch": 2.5780198404979577,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004331003450437674,
      "loss": 5.0532,
      "mean_token_accuracy": 0.19660445153713227,
      "num_tokens": 57474923.0,
      "step": 33135
    },
    {
      "entropy": 5.323561954498291,
      "epoch": 2.5784088698696754,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004330806237139083,
      "loss": 4.9389,
      "mean_token_accuracy": 0.19867438375949859,
      "num_tokens": 57483292.0,
      "step": 33140
    },
    {
      "entropy": 5.316736936569214,
      "epoch": 2.5787978992413927,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004330608999854273,
      "loss": 4.9832,
      "mean_token_accuracy": 0.1959037736058235,
      "num_tokens": 57491894.0,
      "step": 33145
    },
    {
      "entropy": 5.372809743881225,
      "epoch": 2.5791869286131104,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004330411738586238,
      "loss": 5.0248,
      "mean_token_accuracy": 0.18888262808322906,
      "num_tokens": 57500999.0,
      "step": 33150
    },
    {
      "entropy": 5.521779632568359,
      "epoch": 2.5795759579848276,
      "grad_norm": 1.15625,
      "learning_rate": 0.000433021445333797,
      "loss": 5.0869,
      "mean_token_accuracy": 0.19092299938201904,
      "num_tokens": 57508964.0,
      "step": 33155
    },
    {
      "entropy": 5.495110511779785,
      "epoch": 2.5799649873565453,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043300171441124633,
      "loss": 5.1062,
      "mean_token_accuracy": 0.19211519956588746,
      "num_tokens": 57517605.0,
      "step": 33160
    },
    {
      "entropy": 5.480223083496094,
      "epoch": 2.580354016728263,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004329819810912711,
      "loss": 5.2056,
      "mean_token_accuracy": 0.18542506098747252,
      "num_tokens": 57527043.0,
      "step": 33165
    },
    {
      "entropy": 5.411193943023681,
      "epoch": 2.5807430460999807,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043296224537417075,
      "loss": 5.0231,
      "mean_token_accuracy": 0.19838768988847733,
      "num_tokens": 57535768.0,
      "step": 33170
    },
    {
      "entropy": 5.428645658493042,
      "epoch": 2.581132075471698,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043294250726024473,
      "loss": 5.0295,
      "mean_token_accuracy": 0.1945171132683754,
      "num_tokens": 57544060.0,
      "step": 33175
    },
    {
      "entropy": 5.340937709808349,
      "epoch": 2.5815211048434157,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004329227667497926,
      "loss": 4.9048,
      "mean_token_accuracy": 0.20319671779870987,
      "num_tokens": 57553139.0,
      "step": 33180
    },
    {
      "entropy": 5.30918517112732,
      "epoch": 2.5819101342151334,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043290302384311373,
      "loss": 4.945,
      "mean_token_accuracy": 0.1966753214597702,
      "num_tokens": 57561388.0,
      "step": 33185
    },
    {
      "entropy": 5.359861564636231,
      "epoch": 2.5822991635868506,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043288327854050794,
      "loss": 4.9665,
      "mean_token_accuracy": 0.19830524325370788,
      "num_tokens": 57570431.0,
      "step": 33190
    },
    {
      "entropy": 5.405405473709107,
      "epoch": 2.5826881929585683,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043286353084227467,
      "loss": 5.0097,
      "mean_token_accuracy": 0.19255893677473068,
      "num_tokens": 57579047.0,
      "step": 33195
    },
    {
      "entropy": 5.4494822978973385,
      "epoch": 2.583077222330286,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043284378074871354,
      "loss": 5.0415,
      "mean_token_accuracy": 0.1943253755569458,
      "num_tokens": 57587878.0,
      "step": 33200
    },
    {
      "entropy": 5.407710409164428,
      "epoch": 2.5834662517020037,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004328240282601243,
      "loss": 5.0408,
      "mean_token_accuracy": 0.20136262625455856,
      "num_tokens": 57596823.0,
      "step": 33205
    },
    {
      "entropy": 5.409446048736572,
      "epoch": 2.583855281073721,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004328042733768066,
      "loss": 5.1352,
      "mean_token_accuracy": 0.18554815798997878,
      "num_tokens": 57605983.0,
      "step": 33210
    },
    {
      "entropy": 5.396353673934937,
      "epoch": 2.5842443104454387,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043278451609906027,
      "loss": 4.9775,
      "mean_token_accuracy": 0.19769710451364517,
      "num_tokens": 57613987.0,
      "step": 33215
    },
    {
      "entropy": 5.405464315414429,
      "epoch": 2.584633339817156,
      "grad_norm": 1.25,
      "learning_rate": 0.00043276475642718503,
      "loss": 5.011,
      "mean_token_accuracy": 0.19645921438932418,
      "num_tokens": 57622409.0,
      "step": 33220
    },
    {
      "entropy": 5.404412698745728,
      "epoch": 2.5850223691888736,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004327449943614808,
      "loss": 4.9824,
      "mean_token_accuracy": 0.20155560672283174,
      "num_tokens": 57630664.0,
      "step": 33225
    },
    {
      "entropy": 5.3939738273620605,
      "epoch": 2.5854113985605913,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043272522990224727,
      "loss": 5.0483,
      "mean_token_accuracy": 0.19237309396266938,
      "num_tokens": 57639745.0,
      "step": 33230
    },
    {
      "entropy": 5.329434251785278,
      "epoch": 2.585800427932309,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004327054630497844,
      "loss": 4.9858,
      "mean_token_accuracy": 0.20279276072978974,
      "num_tokens": 57648628.0,
      "step": 33235
    },
    {
      "entropy": 5.407618713378906,
      "epoch": 2.5861894573040267,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043268569380439223,
      "loss": 4.9413,
      "mean_token_accuracy": 0.21495347023010253,
      "num_tokens": 57656322.0,
      "step": 33240
    },
    {
      "entropy": 5.375839519500732,
      "epoch": 2.586578486675744,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004326659221663705,
      "loss": 5.0054,
      "mean_token_accuracy": 0.19711500853300096,
      "num_tokens": 57664807.0,
      "step": 33245
    },
    {
      "entropy": 5.341571712493897,
      "epoch": 2.5869675160474617,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004326461481360194,
      "loss": 5.0045,
      "mean_token_accuracy": 0.20176856815814972,
      "num_tokens": 57673216.0,
      "step": 33250
    },
    {
      "entropy": 5.420721340179443,
      "epoch": 2.587356545419179,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043262637171363906,
      "loss": 5.0519,
      "mean_token_accuracy": 0.18821411430835724,
      "num_tokens": 57681615.0,
      "step": 33255
    },
    {
      "entropy": 5.553034543991089,
      "epoch": 2.5877455747908966,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043260659289952926,
      "loss": 5.1182,
      "mean_token_accuracy": 0.1940648838877678,
      "num_tokens": 57690951.0,
      "step": 33260
    },
    {
      "entropy": 5.443303346633911,
      "epoch": 2.5881346041626143,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043258681169399034,
      "loss": 5.0221,
      "mean_token_accuracy": 0.19450205862522124,
      "num_tokens": 57699723.0,
      "step": 33265
    },
    {
      "entropy": 5.3849842071533205,
      "epoch": 2.588523633534332,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043256702809732234,
      "loss": 4.9766,
      "mean_token_accuracy": 0.19638959914445878,
      "num_tokens": 57708380.0,
      "step": 33270
    },
    {
      "entropy": 5.4190967082977295,
      "epoch": 2.5889126629060493,
      "grad_norm": 1.125,
      "learning_rate": 0.0004325472421098255,
      "loss": 5.0327,
      "mean_token_accuracy": 0.19513487964868545,
      "num_tokens": 57717183.0,
      "step": 33275
    },
    {
      "entropy": 5.398415565490723,
      "epoch": 2.589301692277767,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043252745373180006,
      "loss": 4.9634,
      "mean_token_accuracy": 0.19339249283075333,
      "num_tokens": 57726369.0,
      "step": 33280
    },
    {
      "entropy": 5.4137535572052,
      "epoch": 2.5896907216494847,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004325076629635462,
      "loss": 4.9597,
      "mean_token_accuracy": 0.20786890238523484,
      "num_tokens": 57734863.0,
      "step": 33285
    },
    {
      "entropy": 5.403210639953613,
      "epoch": 2.590079751021202,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043248786980536424,
      "loss": 4.9969,
      "mean_token_accuracy": 0.20014824271202086,
      "num_tokens": 57743478.0,
      "step": 33290
    },
    {
      "entropy": 5.369860744476318,
      "epoch": 2.5904687803929196,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004324680742575545,
      "loss": 5.0718,
      "mean_token_accuracy": 0.20055337697267533,
      "num_tokens": 57752300.0,
      "step": 33295
    },
    {
      "entropy": 5.431903457641601,
      "epoch": 2.5908578097646373,
      "grad_norm": 1.125,
      "learning_rate": 0.00043244827632041744,
      "loss": 4.9999,
      "mean_token_accuracy": 0.1969797804951668,
      "num_tokens": 57760377.0,
      "step": 33300
    },
    {
      "entropy": 5.535237073898315,
      "epoch": 2.591246839136355,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004324284759942534,
      "loss": 5.1868,
      "mean_token_accuracy": 0.18544953167438508,
      "num_tokens": 57768665.0,
      "step": 33305
    },
    {
      "entropy": 5.352888774871826,
      "epoch": 2.5916358685080723,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004324086732793627,
      "loss": 5.0088,
      "mean_token_accuracy": 0.1928769215941429,
      "num_tokens": 57777387.0,
      "step": 33310
    },
    {
      "entropy": 5.522916221618653,
      "epoch": 2.59202489787979,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043238886817604605,
      "loss": 5.0946,
      "mean_token_accuracy": 0.1931093618273735,
      "num_tokens": 57785406.0,
      "step": 33315
    },
    {
      "entropy": 5.397273206710816,
      "epoch": 2.5924139272515077,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004323690606846039,
      "loss": 5.0054,
      "mean_token_accuracy": 0.1989477828145027,
      "num_tokens": 57794412.0,
      "step": 33320
    },
    {
      "entropy": 5.38226318359375,
      "epoch": 2.592802956623225,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043234925080533657,
      "loss": 4.936,
      "mean_token_accuracy": 0.197903174161911,
      "num_tokens": 57803310.0,
      "step": 33325
    },
    {
      "entropy": 5.3964029312133786,
      "epoch": 2.5931919859949426,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004323294385385448,
      "loss": 5.0174,
      "mean_token_accuracy": 0.1983078971505165,
      "num_tokens": 57812968.0,
      "step": 33330
    },
    {
      "entropy": 5.422370529174804,
      "epoch": 2.5935810153666603,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004323096238845293,
      "loss": 5.1586,
      "mean_token_accuracy": 0.1875622496008873,
      "num_tokens": 57821230.0,
      "step": 33335
    },
    {
      "entropy": 5.47182445526123,
      "epoch": 2.593970044738378,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004322898068435906,
      "loss": 5.0829,
      "mean_token_accuracy": 0.1912751853466034,
      "num_tokens": 57829643.0,
      "step": 33340
    },
    {
      "entropy": 5.451434803009033,
      "epoch": 2.5943590741100953,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004322699874160294,
      "loss": 4.9581,
      "mean_token_accuracy": 0.20072317719459534,
      "num_tokens": 57837994.0,
      "step": 33345
    },
    {
      "entropy": 5.421475887298584,
      "epoch": 2.594748103481813,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043225016560214654,
      "loss": 5.0678,
      "mean_token_accuracy": 0.19361352771520615,
      "num_tokens": 57846764.0,
      "step": 33350
    },
    {
      "entropy": 5.408279705047607,
      "epoch": 2.59513713285353,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043223034140224266,
      "loss": 4.9682,
      "mean_token_accuracy": 0.1983417108654976,
      "num_tokens": 57854887.0,
      "step": 33355
    },
    {
      "entropy": 5.424863338470459,
      "epoch": 2.595526162225248,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004322105148166187,
      "loss": 4.9872,
      "mean_token_accuracy": 0.19804560989141465,
      "num_tokens": 57863639.0,
      "step": 33360
    },
    {
      "entropy": 5.360683345794678,
      "epoch": 2.5959151915969656,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043219068584557526,
      "loss": 4.9182,
      "mean_token_accuracy": 0.2050188660621643,
      "num_tokens": 57872037.0,
      "step": 33365
    },
    {
      "entropy": 5.421139907836914,
      "epoch": 2.5963042209686833,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004321708544894134,
      "loss": 5.026,
      "mean_token_accuracy": 0.19619489461183548,
      "num_tokens": 57880809.0,
      "step": 33370
    },
    {
      "entropy": 5.464409446716308,
      "epoch": 2.5966932503404005,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043215102074843403,
      "loss": 5.1067,
      "mean_token_accuracy": 0.19468592554330827,
      "num_tokens": 57889184.0,
      "step": 33375
    },
    {
      "entropy": 5.383119916915893,
      "epoch": 2.5970822797121182,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043213118462293796,
      "loss": 4.9547,
      "mean_token_accuracy": 0.19801298826932906,
      "num_tokens": 57897695.0,
      "step": 33380
    },
    {
      "entropy": 5.458133506774902,
      "epoch": 2.597471309083836,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004321113461132264,
      "loss": 5.0487,
      "mean_token_accuracy": 0.1948784440755844,
      "num_tokens": 57906280.0,
      "step": 33385
    },
    {
      "entropy": 5.456954765319824,
      "epoch": 2.597860338455553,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043209150521960016,
      "loss": 5.0456,
      "mean_token_accuracy": 0.19694745242595674,
      "num_tokens": 57915112.0,
      "step": 33390
    },
    {
      "entropy": 5.439999008178711,
      "epoch": 2.598249367827271,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043207166194236037,
      "loss": 5.0196,
      "mean_token_accuracy": 0.19698575437068938,
      "num_tokens": 57924210.0,
      "step": 33395
    },
    {
      "entropy": 5.421382713317871,
      "epoch": 2.5986383971989886,
      "grad_norm": 1.125,
      "learning_rate": 0.0004320518162818082,
      "loss": 4.9831,
      "mean_token_accuracy": 0.2011178895831108,
      "num_tokens": 57933030.0,
      "step": 33400
    },
    {
      "entropy": 5.346721363067627,
      "epoch": 2.5990274265707063,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004320319682382445,
      "loss": 4.9212,
      "mean_token_accuracy": 0.19615161269903184,
      "num_tokens": 57941490.0,
      "step": 33405
    },
    {
      "entropy": 5.36622633934021,
      "epoch": 2.5994164559424235,
      "grad_norm": 1.125,
      "learning_rate": 0.00043201211781197075,
      "loss": 5.0086,
      "mean_token_accuracy": 0.19557689726352692,
      "num_tokens": 57950666.0,
      "step": 33410
    },
    {
      "entropy": 5.395963716506958,
      "epoch": 2.5998054853141412,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043199226500328805,
      "loss": 5.0711,
      "mean_token_accuracy": 0.19219135940074922,
      "num_tokens": 57959740.0,
      "step": 33415
    },
    {
      "entropy": 5.4872652053833,
      "epoch": 2.600194514685859,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004319724098124975,
      "loss": 5.0199,
      "mean_token_accuracy": 0.19232258647680284,
      "num_tokens": 57968391.0,
      "step": 33420
    },
    {
      "entropy": 5.424953126907349,
      "epoch": 2.600583544057576,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043195255223990055,
      "loss": 5.0473,
      "mean_token_accuracy": 0.19112542569637297,
      "num_tokens": 57976873.0,
      "step": 33425
    },
    {
      "entropy": 5.378559494018555,
      "epoch": 2.600972573429294,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004319326922857984,
      "loss": 5.0243,
      "mean_token_accuracy": 0.19390802830457687,
      "num_tokens": 57985085.0,
      "step": 33430
    },
    {
      "entropy": 5.417693758010865,
      "epoch": 2.6013616028010116,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004319128299504925,
      "loss": 5.0056,
      "mean_token_accuracy": 0.19598936587572097,
      "num_tokens": 57993686.0,
      "step": 33435
    },
    {
      "entropy": 5.435611963272095,
      "epoch": 2.6017506321727293,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043189296523428407,
      "loss": 5.113,
      "mean_token_accuracy": 0.18981040120124817,
      "num_tokens": 58001877.0,
      "step": 33440
    },
    {
      "entropy": 5.391393041610717,
      "epoch": 2.6021396615444465,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043187309813747464,
      "loss": 5.0084,
      "mean_token_accuracy": 0.1983030378818512,
      "num_tokens": 58010688.0,
      "step": 33445
    },
    {
      "entropy": 5.419939088821411,
      "epoch": 2.6025286909161642,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043185322866036565,
      "loss": 4.9442,
      "mean_token_accuracy": 0.20129207670688629,
      "num_tokens": 58019470.0,
      "step": 33450
    },
    {
      "entropy": 5.445365047454834,
      "epoch": 2.6029177202878815,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043183335680325866,
      "loss": 5.0671,
      "mean_token_accuracy": 0.18568008691072463,
      "num_tokens": 58028297.0,
      "step": 33455
    },
    {
      "entropy": 5.416705465316772,
      "epoch": 2.603306749659599,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043181348256645497,
      "loss": 5.1217,
      "mean_token_accuracy": 0.18392168134450912,
      "num_tokens": 58036815.0,
      "step": 33460
    },
    {
      "entropy": 5.5014698028564455,
      "epoch": 2.603695779031317,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043179360595025637,
      "loss": 5.1133,
      "mean_token_accuracy": 0.1936869889497757,
      "num_tokens": 58045682.0,
      "step": 33465
    },
    {
      "entropy": 5.432379198074341,
      "epoch": 2.6040848084030346,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004317737269549643,
      "loss": 5.073,
      "mean_token_accuracy": 0.18725499361753464,
      "num_tokens": 58054530.0,
      "step": 33470
    },
    {
      "entropy": 5.4447393894195555,
      "epoch": 2.604473837774752,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004317538455808805,
      "loss": 5.0851,
      "mean_token_accuracy": 0.19117185473442078,
      "num_tokens": 58063360.0,
      "step": 33475
    },
    {
      "entropy": 5.4212500095367435,
      "epoch": 2.6048628671464695,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004317339618283065,
      "loss": 5.054,
      "mean_token_accuracy": 0.20231692045927047,
      "num_tokens": 58071198.0,
      "step": 33480
    },
    {
      "entropy": 5.485487461090088,
      "epoch": 2.6052518965181872,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004317140756975442,
      "loss": 5.1156,
      "mean_token_accuracy": 0.18765053749084473,
      "num_tokens": 58080138.0,
      "step": 33485
    },
    {
      "entropy": 5.363854455947876,
      "epoch": 2.6056409258899045,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004316941871888951,
      "loss": 4.9696,
      "mean_token_accuracy": 0.2042372077703476,
      "num_tokens": 58088768.0,
      "step": 33490
    },
    {
      "entropy": 5.333820152282715,
      "epoch": 2.606029955261622,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043167429630266134,
      "loss": 5.0093,
      "mean_token_accuracy": 0.1946773946285248,
      "num_tokens": 58098309.0,
      "step": 33495
    },
    {
      "entropy": 5.444535589218139,
      "epoch": 2.60641898463334,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004316544030391444,
      "loss": 5.0351,
      "mean_token_accuracy": 0.1943730592727661,
      "num_tokens": 58106987.0,
      "step": 33500
    },
    {
      "entropy": 5.422795343399048,
      "epoch": 2.6068080140050576,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004316345073986461,
      "loss": 4.9967,
      "mean_token_accuracy": 0.1992761343717575,
      "num_tokens": 58115065.0,
      "step": 33505
    },
    {
      "entropy": 5.404860544204712,
      "epoch": 2.607197043376775,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004316146093814686,
      "loss": 5.1084,
      "mean_token_accuracy": 0.1904493451118469,
      "num_tokens": 58123995.0,
      "step": 33510
    },
    {
      "entropy": 5.361037492752075,
      "epoch": 2.6075860727484925,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004315947089879137,
      "loss": 4.9153,
      "mean_token_accuracy": 0.20740990936756135,
      "num_tokens": 58132080.0,
      "step": 33515
    },
    {
      "entropy": 5.458073711395263,
      "epoch": 2.60797510212021,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043157480621828327,
      "loss": 5.0771,
      "mean_token_accuracy": 0.194709937274456,
      "num_tokens": 58140199.0,
      "step": 33520
    },
    {
      "entropy": 5.3023748874664305,
      "epoch": 2.6083641314919275,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043155490107287935,
      "loss": 4.9215,
      "mean_token_accuracy": 0.20740467607975005,
      "num_tokens": 58148852.0,
      "step": 33525
    },
    {
      "entropy": 5.390206718444825,
      "epoch": 2.608753160863645,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004315349935520041,
      "loss": 5.0765,
      "mean_token_accuracy": 0.1993390366435051,
      "num_tokens": 58157290.0,
      "step": 33530
    },
    {
      "entropy": 5.478251314163208,
      "epoch": 2.609142190235363,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004315150836559594,
      "loss": 5.0337,
      "mean_token_accuracy": 0.18843597322702407,
      "num_tokens": 58166086.0,
      "step": 33535
    },
    {
      "entropy": 5.435611629486084,
      "epoch": 2.6095312196070806,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004314951713850474,
      "loss": 4.9864,
      "mean_token_accuracy": 0.19927272647619249,
      "num_tokens": 58174865.0,
      "step": 33540
    },
    {
      "entropy": 5.360013484954834,
      "epoch": 2.609920248978798,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043147525673957026,
      "loss": 5.0103,
      "mean_token_accuracy": 0.20238753259181977,
      "num_tokens": 58182313.0,
      "step": 33545
    },
    {
      "entropy": 5.382126235961914,
      "epoch": 2.6103092783505155,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043145533971983025,
      "loss": 5.0407,
      "mean_token_accuracy": 0.1875401943922043,
      "num_tokens": 58190758.0,
      "step": 33550
    },
    {
      "entropy": 5.410359954833984,
      "epoch": 2.6106983077222328,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043143542032612935,
      "loss": 5.0505,
      "mean_token_accuracy": 0.19337413012981414,
      "num_tokens": 58199308.0,
      "step": 33555
    },
    {
      "entropy": 5.421206426620484,
      "epoch": 2.6110873370939505,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004314154985587701,
      "loss": 4.9693,
      "mean_token_accuracy": 0.19400418251752855,
      "num_tokens": 58206995.0,
      "step": 33560
    },
    {
      "entropy": 5.350102472305298,
      "epoch": 2.611476366465668,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043139557441805457,
      "loss": 4.9874,
      "mean_token_accuracy": 0.2004566088318825,
      "num_tokens": 58215186.0,
      "step": 33565
    },
    {
      "entropy": 5.455255842208862,
      "epoch": 2.611865395837386,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043137564790428503,
      "loss": 5.023,
      "mean_token_accuracy": 0.19396957159042358,
      "num_tokens": 58224347.0,
      "step": 33570
    },
    {
      "entropy": 5.324423837661743,
      "epoch": 2.6122544252091036,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043135571901776397,
      "loss": 4.9245,
      "mean_token_accuracy": 0.20258399695158005,
      "num_tokens": 58232477.0,
      "step": 33575
    },
    {
      "entropy": 5.313129615783692,
      "epoch": 2.612643454580821,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004313357877587937,
      "loss": 4.9633,
      "mean_token_accuracy": 0.20570655465126036,
      "num_tokens": 58241186.0,
      "step": 33580
    },
    {
      "entropy": 5.399720001220703,
      "epoch": 2.6130324839525385,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004313158541276768,
      "loss": 5.0261,
      "mean_token_accuracy": 0.2012824073433876,
      "num_tokens": 58249140.0,
      "step": 33585
    },
    {
      "entropy": 5.400412845611572,
      "epoch": 2.6134215133242558,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004312959181247156,
      "loss": 5.0115,
      "mean_token_accuracy": 0.19189297258853913,
      "num_tokens": 58258313.0,
      "step": 33590
    },
    {
      "entropy": 5.381920337677002,
      "epoch": 2.6138105426959735,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004312759797502126,
      "loss": 4.9868,
      "mean_token_accuracy": 0.19745534062385559,
      "num_tokens": 58267013.0,
      "step": 33595
    },
    {
      "entropy": 5.413403606414795,
      "epoch": 2.614199572067691,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043125603900447026,
      "loss": 5.0723,
      "mean_token_accuracy": 0.18996374905109406,
      "num_tokens": 58275844.0,
      "step": 33600
    },
    {
      "entropy": 5.411362028121948,
      "epoch": 2.614588601439409,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004312360958877913,
      "loss": 5.0342,
      "mean_token_accuracy": 0.19286447614431382,
      "num_tokens": 58284416.0,
      "step": 33605
    },
    {
      "entropy": 5.412449073791504,
      "epoch": 2.614977630811126,
      "grad_norm": 1.125,
      "learning_rate": 0.0004312161504004782,
      "loss": 5.0201,
      "mean_token_accuracy": 0.19352453500032424,
      "num_tokens": 58293564.0,
      "step": 33610
    },
    {
      "entropy": 5.381396770477295,
      "epoch": 2.615366660182844,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043119620254283375,
      "loss": 4.9776,
      "mean_token_accuracy": 0.1954403430223465,
      "num_tokens": 58302230.0,
      "step": 33615
    },
    {
      "entropy": 5.421072196960449,
      "epoch": 2.6157556895545615,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043117625231516054,
      "loss": 5.0113,
      "mean_token_accuracy": 0.19653632938861848,
      "num_tokens": 58309540.0,
      "step": 33620
    },
    {
      "entropy": 5.355747032165527,
      "epoch": 2.6161447189262788,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004311562997177612,
      "loss": 4.9359,
      "mean_token_accuracy": 0.20713720023632048,
      "num_tokens": 58317885.0,
      "step": 33625
    },
    {
      "entropy": 5.412367820739746,
      "epoch": 2.6165337482979965,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004311363447509386,
      "loss": 5.0226,
      "mean_token_accuracy": 0.18666871786117553,
      "num_tokens": 58326133.0,
      "step": 33630
    },
    {
      "entropy": 5.421963214874268,
      "epoch": 2.616922777669714,
      "grad_norm": 1.25,
      "learning_rate": 0.0004311163874149955,
      "loss": 5.0323,
      "mean_token_accuracy": 0.19386831820011138,
      "num_tokens": 58335085.0,
      "step": 33635
    },
    {
      "entropy": 5.4929666996002195,
      "epoch": 2.617311807041432,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043109642771023464,
      "loss": 5.0475,
      "mean_token_accuracy": 0.19512614160776137,
      "num_tokens": 58344222.0,
      "step": 33640
    },
    {
      "entropy": 5.38620138168335,
      "epoch": 2.617700836413149,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004310764656369591,
      "loss": 4.9379,
      "mean_token_accuracy": 0.19684532433748245,
      "num_tokens": 58353256.0,
      "step": 33645
    },
    {
      "entropy": 5.451818370819092,
      "epoch": 2.618089865784867,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004310565011954715,
      "loss": 5.0364,
      "mean_token_accuracy": 0.19929524809122084,
      "num_tokens": 58361197.0,
      "step": 33650
    },
    {
      "entropy": 5.522153663635254,
      "epoch": 2.618478895156584,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004310365343860748,
      "loss": 5.1084,
      "mean_token_accuracy": 0.19607727229595184,
      "num_tokens": 58370653.0,
      "step": 33655
    },
    {
      "entropy": 5.422746849060059,
      "epoch": 2.6188679245283017,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043101656520907225,
      "loss": 5.0787,
      "mean_token_accuracy": 0.1901749327778816,
      "num_tokens": 58379804.0,
      "step": 33660
    },
    {
      "entropy": 5.380604839324951,
      "epoch": 2.6192569539000194,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004309965936647665,
      "loss": 5.01,
      "mean_token_accuracy": 0.1916694387793541,
      "num_tokens": 58388443.0,
      "step": 33665
    },
    {
      "entropy": 5.459115505218506,
      "epoch": 2.619645983271737,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004309766197534608,
      "loss": 5.1225,
      "mean_token_accuracy": 0.19072893112897873,
      "num_tokens": 58397165.0,
      "step": 33670
    },
    {
      "entropy": 5.415544891357422,
      "epoch": 2.620035012643455,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043095664347545816,
      "loss": 5.0387,
      "mean_token_accuracy": 0.1935420125722885,
      "num_tokens": 58405679.0,
      "step": 33675
    },
    {
      "entropy": 5.368338918685913,
      "epoch": 2.620424042015172,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004309366648310616,
      "loss": 4.9378,
      "mean_token_accuracy": 0.2014945402741432,
      "num_tokens": 58414998.0,
      "step": 33680
    },
    {
      "entropy": 5.441775989532471,
      "epoch": 2.62081307138689,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043091668382057443,
      "loss": 4.9695,
      "mean_token_accuracy": 0.19644484668970108,
      "num_tokens": 58423535.0,
      "step": 33685
    },
    {
      "entropy": 5.367545509338379,
      "epoch": 2.621202100758607,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043089670044429973,
      "loss": 4.9467,
      "mean_token_accuracy": 0.1996975526213646,
      "num_tokens": 58432379.0,
      "step": 33690
    },
    {
      "entropy": 5.3892780303955075,
      "epoch": 2.6215911301303247,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043087671470254076,
      "loss": 5.0527,
      "mean_token_accuracy": 0.19272695928812028,
      "num_tokens": 58441013.0,
      "step": 33695
    },
    {
      "entropy": 5.4753190040588375,
      "epoch": 2.6219801595020424,
      "grad_norm": 1.125,
      "learning_rate": 0.00043085672659560077,
      "loss": 5.0742,
      "mean_token_accuracy": 0.20055238753557206,
      "num_tokens": 58450353.0,
      "step": 33700
    },
    {
      "entropy": 5.462785148620606,
      "epoch": 2.62236918887376,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043083673612378295,
      "loss": 4.9499,
      "mean_token_accuracy": 0.19458895176649094,
      "num_tokens": 58458833.0,
      "step": 33705
    },
    {
      "entropy": 5.3792500495910645,
      "epoch": 2.6227582182454774,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043081674328739066,
      "loss": 5.0711,
      "mean_token_accuracy": 0.1904326021671295,
      "num_tokens": 58468325.0,
      "step": 33710
    },
    {
      "entropy": 5.390144348144531,
      "epoch": 2.623147247617195,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043079674808672744,
      "loss": 4.9481,
      "mean_token_accuracy": 0.19750515669584273,
      "num_tokens": 58476333.0,
      "step": 33715
    },
    {
      "entropy": 5.368268442153931,
      "epoch": 2.623536276988913,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004307767505220964,
      "loss": 5.0069,
      "mean_token_accuracy": 0.20425021648406982,
      "num_tokens": 58484435.0,
      "step": 33720
    },
    {
      "entropy": 5.386517286300659,
      "epoch": 2.62392530636063,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004307567505938012,
      "loss": 5.1126,
      "mean_token_accuracy": 0.1851356491446495,
      "num_tokens": 58493709.0,
      "step": 33725
    },
    {
      "entropy": 5.449285411834717,
      "epoch": 2.6243143357323477,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004307367483021452,
      "loss": 5.0006,
      "mean_token_accuracy": 0.19844175577163697,
      "num_tokens": 58502522.0,
      "step": 33730
    },
    {
      "entropy": 5.438796043395996,
      "epoch": 2.6247033651040654,
      "grad_norm": 1.15625,
      "learning_rate": 0.000430716743647432,
      "loss": 5.0396,
      "mean_token_accuracy": 0.1899079442024231,
      "num_tokens": 58511459.0,
      "step": 33735
    },
    {
      "entropy": 5.410864019393921,
      "epoch": 2.625092394475783,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000430696736629965,
      "loss": 5.0676,
      "mean_token_accuracy": 0.19566015303134918,
      "num_tokens": 58520197.0,
      "step": 33740
    },
    {
      "entropy": 5.435902690887451,
      "epoch": 2.6254814238475004,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004306767272500478,
      "loss": 5.0808,
      "mean_token_accuracy": 0.1920143887400627,
      "num_tokens": 58528587.0,
      "step": 33745
    },
    {
      "entropy": 5.440750455856323,
      "epoch": 2.625870453219218,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043065671550798415,
      "loss": 4.992,
      "mean_token_accuracy": 0.19872858226299286,
      "num_tokens": 58537348.0,
      "step": 33750
    },
    {
      "entropy": 5.341591310501099,
      "epoch": 2.626259482590936,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043063670140407753,
      "loss": 4.9105,
      "mean_token_accuracy": 0.21187319606542587,
      "num_tokens": 58545637.0,
      "step": 33755
    },
    {
      "entropy": 5.3664209842681885,
      "epoch": 2.626648511962653,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004306166849386317,
      "loss": 4.996,
      "mean_token_accuracy": 0.19240718483924865,
      "num_tokens": 58554575.0,
      "step": 33760
    },
    {
      "entropy": 5.328170299530029,
      "epoch": 2.6270375413343707,
      "grad_norm": 1.125,
      "learning_rate": 0.00043059666611195037,
      "loss": 4.8427,
      "mean_token_accuracy": 0.20593435168266297,
      "num_tokens": 58563150.0,
      "step": 33765
    },
    {
      "entropy": 5.3378674507141115,
      "epoch": 2.6274265707060884,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004305766449243372,
      "loss": 4.9961,
      "mean_token_accuracy": 0.19201348572969437,
      "num_tokens": 58571790.0,
      "step": 33770
    },
    {
      "entropy": 5.3259368419647215,
      "epoch": 2.627815600077806,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004305566213760962,
      "loss": 4.9377,
      "mean_token_accuracy": 0.20279304534196854,
      "num_tokens": 58581094.0,
      "step": 33775
    },
    {
      "entropy": 5.430250215530395,
      "epoch": 2.6282046294495234,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043053659546753094,
      "loss": 5.0094,
      "mean_token_accuracy": 0.19294605404138565,
      "num_tokens": 58590836.0,
      "step": 33780
    },
    {
      "entropy": 5.403907108306885,
      "epoch": 2.628593658821241,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004305165671989455,
      "loss": 4.9914,
      "mean_token_accuracy": 0.19500521421432496,
      "num_tokens": 58599315.0,
      "step": 33785
    },
    {
      "entropy": 5.375732135772705,
      "epoch": 2.6289826881929583,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004304965365706437,
      "loss": 4.9652,
      "mean_token_accuracy": 0.2018195942044258,
      "num_tokens": 58607663.0,
      "step": 33790
    },
    {
      "entropy": 5.303044080734253,
      "epoch": 2.629371717564676,
      "grad_norm": 1.125,
      "learning_rate": 0.0004304765035829294,
      "loss": 4.9781,
      "mean_token_accuracy": 0.20152345597743987,
      "num_tokens": 58616513.0,
      "step": 33795
    },
    {
      "entropy": 5.45792145729065,
      "epoch": 2.6297607469363937,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043045646823610664,
      "loss": 5.1101,
      "mean_token_accuracy": 0.1933145195245743,
      "num_tokens": 58625483.0,
      "step": 33800
    },
    {
      "entropy": 5.407931327819824,
      "epoch": 2.6301497763081114,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043043643053047935,
      "loss": 5.0133,
      "mean_token_accuracy": 0.19704361408948898,
      "num_tokens": 58633880.0,
      "step": 33805
    },
    {
      "entropy": 5.43102879524231,
      "epoch": 2.6305388056798287,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004304163904663517,
      "loss": 5.0039,
      "mean_token_accuracy": 0.19708012491464616,
      "num_tokens": 58642940.0,
      "step": 33810
    },
    {
      "entropy": 5.348496198654175,
      "epoch": 2.6309278350515464,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043039634804402767,
      "loss": 4.9676,
      "mean_token_accuracy": 0.2047894313931465,
      "num_tokens": 58651736.0,
      "step": 33815
    },
    {
      "entropy": 5.43419041633606,
      "epoch": 2.631316864423264,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004303763032638114,
      "loss": 5.0002,
      "mean_token_accuracy": 0.19973311424255372,
      "num_tokens": 58660494.0,
      "step": 33820
    },
    {
      "entropy": 5.3585161685943605,
      "epoch": 2.6317058937949813,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004303562561260071,
      "loss": 4.9971,
      "mean_token_accuracy": 0.19196023494005204,
      "num_tokens": 58669150.0,
      "step": 33825
    },
    {
      "entropy": 5.3444366455078125,
      "epoch": 2.632094923166699,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043033620663091883,
      "loss": 4.9482,
      "mean_token_accuracy": 0.21073953956365585,
      "num_tokens": 58677806.0,
      "step": 33830
    },
    {
      "entropy": 5.34049072265625,
      "epoch": 2.6324839525384167,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004303161547788509,
      "loss": 4.9465,
      "mean_token_accuracy": 0.20084349662065507,
      "num_tokens": 58686406.0,
      "step": 33835
    },
    {
      "entropy": 5.454685163497925,
      "epoch": 2.6328729819101344,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004302961005701074,
      "loss": 5.071,
      "mean_token_accuracy": 0.19278971552848817,
      "num_tokens": 58696074.0,
      "step": 33840
    },
    {
      "entropy": 5.433259057998657,
      "epoch": 2.6332620112818517,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043027604400499295,
      "loss": 5.0322,
      "mean_token_accuracy": 0.20097335129976274,
      "num_tokens": 58704860.0,
      "step": 33845
    },
    {
      "entropy": 5.439332675933838,
      "epoch": 2.6336510406535694,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043025598508381155,
      "loss": 5.013,
      "mean_token_accuracy": 0.19542540460824967,
      "num_tokens": 58713455.0,
      "step": 33850
    },
    {
      "entropy": 5.291840124130249,
      "epoch": 2.634040070025287,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004302359238068677,
      "loss": 4.9266,
      "mean_token_accuracy": 0.19805506914854049,
      "num_tokens": 58722382.0,
      "step": 33855
    },
    {
      "entropy": 5.43573226928711,
      "epoch": 2.6344290993970043,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043021586017446585,
      "loss": 4.9524,
      "mean_token_accuracy": 0.20068986117839813,
      "num_tokens": 58730816.0,
      "step": 33860
    },
    {
      "entropy": 5.382688522338867,
      "epoch": 2.634818128768722,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043019579418691027,
      "loss": 4.9636,
      "mean_token_accuracy": 0.2000237688422203,
      "num_tokens": 58740143.0,
      "step": 33865
    },
    {
      "entropy": 5.30120964050293,
      "epoch": 2.6352071581404397,
      "grad_norm": 1.125,
      "learning_rate": 0.0004301757258445056,
      "loss": 4.9107,
      "mean_token_accuracy": 0.20167735517024993,
      "num_tokens": 58748347.0,
      "step": 33870
    },
    {
      "entropy": 5.287467050552368,
      "epoch": 2.6355961875121574,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004301556551475563,
      "loss": 4.9432,
      "mean_token_accuracy": 0.19536378532648085,
      "num_tokens": 58756274.0,
      "step": 33875
    },
    {
      "entropy": 5.358163356781006,
      "epoch": 2.6359852168838747,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004301355820963669,
      "loss": 4.9607,
      "mean_token_accuracy": 0.20452135801315308,
      "num_tokens": 58764564.0,
      "step": 33880
    },
    {
      "entropy": 5.321017456054688,
      "epoch": 2.6363742462555924,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004301155066912419,
      "loss": 5.0225,
      "mean_token_accuracy": 0.19594679027795792,
      "num_tokens": 58772914.0,
      "step": 33885
    },
    {
      "entropy": 5.405888986587525,
      "epoch": 2.6367632756273096,
      "grad_norm": 1.25,
      "learning_rate": 0.000430095428932486,
      "loss": 5.0139,
      "mean_token_accuracy": 0.19712991416454315,
      "num_tokens": 58781572.0,
      "step": 33890
    },
    {
      "entropy": 5.40854663848877,
      "epoch": 2.6371523049990273,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043007534882040385,
      "loss": 4.9885,
      "mean_token_accuracy": 0.1997486710548401,
      "num_tokens": 58789259.0,
      "step": 33895
    },
    {
      "entropy": 5.43879246711731,
      "epoch": 2.637541334370745,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004300552663553001,
      "loss": 5.0266,
      "mean_token_accuracy": 0.20097511857748032,
      "num_tokens": 58797984.0,
      "step": 33900
    },
    {
      "entropy": 5.399646186828614,
      "epoch": 2.6379303637424627,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004300351815374795,
      "loss": 4.9023,
      "mean_token_accuracy": 0.20719758570194244,
      "num_tokens": 58806474.0,
      "step": 33905
    },
    {
      "entropy": 5.387241697311401,
      "epoch": 2.63831939311418,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004300150943672467,
      "loss": 5.0846,
      "mean_token_accuracy": 0.19039632081985475,
      "num_tokens": 58816223.0,
      "step": 33910
    },
    {
      "entropy": 5.415988874435425,
      "epoch": 2.6387084224858977,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004299950048449067,
      "loss": 5.0007,
      "mean_token_accuracy": 0.19084305316209793,
      "num_tokens": 58824950.0,
      "step": 33915
    },
    {
      "entropy": 5.421756172180176,
      "epoch": 2.6390974518576154,
      "grad_norm": 1.25,
      "learning_rate": 0.0004299749129707641,
      "loss": 4.9947,
      "mean_token_accuracy": 0.19469838440418244,
      "num_tokens": 58833201.0,
      "step": 33920
    },
    {
      "entropy": 5.33601942062378,
      "epoch": 2.6394864812293326,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004299548187451239,
      "loss": 4.9458,
      "mean_token_accuracy": 0.20292457491159438,
      "num_tokens": 58841621.0,
      "step": 33925
    },
    {
      "entropy": 5.321476554870605,
      "epoch": 2.6398755106010503,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042993472216829097,
      "loss": 4.9519,
      "mean_token_accuracy": 0.19912744164466858,
      "num_tokens": 58849831.0,
      "step": 33930
    },
    {
      "entropy": 5.339437007904053,
      "epoch": 2.640264539972768,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042991462324057025,
      "loss": 5.0393,
      "mean_token_accuracy": 0.19632008373737336,
      "num_tokens": 58858530.0,
      "step": 33935
    },
    {
      "entropy": 5.420591068267822,
      "epoch": 2.6406535693444857,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004298945219622667,
      "loss": 5.1068,
      "mean_token_accuracy": 0.1883932739496231,
      "num_tokens": 58868127.0,
      "step": 33940
    },
    {
      "entropy": 5.406493902206421,
      "epoch": 2.641042598716203,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042987441833368525,
      "loss": 4.9675,
      "mean_token_accuracy": 0.19873608648777008,
      "num_tokens": 58876959.0,
      "step": 33945
    },
    {
      "entropy": 5.367155504226685,
      "epoch": 2.6414316280879206,
      "grad_norm": 1.125,
      "learning_rate": 0.00042985431235513104,
      "loss": 4.9915,
      "mean_token_accuracy": 0.19830975085496902,
      "num_tokens": 58885534.0,
      "step": 33950
    },
    {
      "entropy": 5.4311511516571045,
      "epoch": 2.6418206574596383,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042983420402690917,
      "loss": 5.0398,
      "mean_token_accuracy": 0.20027358829975128,
      "num_tokens": 58894600.0,
      "step": 33955
    },
    {
      "entropy": 5.41059455871582,
      "epoch": 2.6422096868313556,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042981409334932457,
      "loss": 5.038,
      "mean_token_accuracy": 0.19373781383037567,
      "num_tokens": 58903243.0,
      "step": 33960
    },
    {
      "entropy": 5.400194835662842,
      "epoch": 2.6425987162030733,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004297939803226826,
      "loss": 5.053,
      "mean_token_accuracy": 0.19630462527275086,
      "num_tokens": 58911058.0,
      "step": 33965
    },
    {
      "entropy": 5.432136535644531,
      "epoch": 2.642987745574791,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004297738649472884,
      "loss": 5.0508,
      "mean_token_accuracy": 0.19819913655519486,
      "num_tokens": 58919698.0,
      "step": 33970
    },
    {
      "entropy": 5.4930871486663815,
      "epoch": 2.6433767749465087,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042975374722344713,
      "loss": 5.1941,
      "mean_token_accuracy": 0.18508311957120896,
      "num_tokens": 58929276.0,
      "step": 33975
    },
    {
      "entropy": 5.529504680633545,
      "epoch": 2.643765804318226,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042973362715146397,
      "loss": 5.1106,
      "mean_token_accuracy": 0.1930304065346718,
      "num_tokens": 58938148.0,
      "step": 33980
    },
    {
      "entropy": 5.413777685165405,
      "epoch": 2.6441548336899436,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004297135047316444,
      "loss": 4.9456,
      "mean_token_accuracy": 0.19399710595607758,
      "num_tokens": 58946406.0,
      "step": 33985
    },
    {
      "entropy": 5.423583221435547,
      "epoch": 2.644543863061661,
      "grad_norm": 1.125,
      "learning_rate": 0.0004296933799642936,
      "loss": 5.0627,
      "mean_token_accuracy": 0.18798491805791856,
      "num_tokens": 58954914.0,
      "step": 33990
    },
    {
      "entropy": 5.42130036354065,
      "epoch": 2.6449328924333786,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000429673252849717,
      "loss": 4.9597,
      "mean_token_accuracy": 0.20014220774173735,
      "num_tokens": 58963236.0,
      "step": 33995
    },
    {
      "entropy": 5.3958714485168455,
      "epoch": 2.6453219218050963,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042965312338822005,
      "loss": 5.0339,
      "mean_token_accuracy": 0.19641783982515335,
      "num_tokens": 58972141.0,
      "step": 34000
    },
    {
      "entropy": 5.392393398284912,
      "epoch": 2.645710951176814,
      "grad_norm": 1.078125,
      "learning_rate": 0.000429632991580108,
      "loss": 5.0868,
      "mean_token_accuracy": 0.1937030404806137,
      "num_tokens": 58982348.0,
      "step": 34005
    },
    {
      "entropy": 5.448680448532104,
      "epoch": 2.6460999805485317,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004296128574256864,
      "loss": 5.0166,
      "mean_token_accuracy": 0.19493293911218643,
      "num_tokens": 58990858.0,
      "step": 34010
    },
    {
      "entropy": 5.299312353134155,
      "epoch": 2.646489009920249,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042959272092526086,
      "loss": 4.8856,
      "mean_token_accuracy": 0.20283929109573365,
      "num_tokens": 58999841.0,
      "step": 34015
    },
    {
      "entropy": 5.306032752990722,
      "epoch": 2.6468780392919666,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042957258207913687,
      "loss": 4.9094,
      "mean_token_accuracy": 0.20322803109884263,
      "num_tokens": 59008538.0,
      "step": 34020
    },
    {
      "entropy": 5.318527460098267,
      "epoch": 2.647267068663684,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042955244088761985,
      "loss": 4.9155,
      "mean_token_accuracy": 0.19806606322526932,
      "num_tokens": 59016656.0,
      "step": 34025
    },
    {
      "entropy": 5.4330912113189695,
      "epoch": 2.6476560980354016,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004295322973510156,
      "loss": 4.9965,
      "mean_token_accuracy": 0.20220135748386384,
      "num_tokens": 59025682.0,
      "step": 34030
    },
    {
      "entropy": 5.41906852722168,
      "epoch": 2.6480451274071193,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004295121514696297,
      "loss": 5.0676,
      "mean_token_accuracy": 0.19688004702329637,
      "num_tokens": 59033448.0,
      "step": 34035
    },
    {
      "entropy": 5.383317661285401,
      "epoch": 2.648434156778837,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042949200324376784,
      "loss": 4.9809,
      "mean_token_accuracy": 0.19861141145229338,
      "num_tokens": 59042220.0,
      "step": 34040
    },
    {
      "entropy": 5.398624849319458,
      "epoch": 2.6488231861505542,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004294718526737357,
      "loss": 5.0261,
      "mean_token_accuracy": 0.19952138215303422,
      "num_tokens": 59050337.0,
      "step": 34045
    },
    {
      "entropy": 5.361309766769409,
      "epoch": 2.649212215522272,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004294516997598391,
      "loss": 4.998,
      "mean_token_accuracy": 0.19630155116319656,
      "num_tokens": 59059186.0,
      "step": 34050
    },
    {
      "entropy": 5.395581007003784,
      "epoch": 2.6496012448939896,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004294315445023838,
      "loss": 5.047,
      "mean_token_accuracy": 0.19041151255369188,
      "num_tokens": 59068714.0,
      "step": 34055
    },
    {
      "entropy": 5.485059595108032,
      "epoch": 2.649990274265707,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042941138690167556,
      "loss": 5.1509,
      "mean_token_accuracy": 0.18468718379735946,
      "num_tokens": 59077015.0,
      "step": 34060
    },
    {
      "entropy": 5.418365430831909,
      "epoch": 2.6503793036374246,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004293912269580204,
      "loss": 5.0105,
      "mean_token_accuracy": 0.1970279261469841,
      "num_tokens": 59085766.0,
      "step": 34065
    },
    {
      "entropy": 5.467449903488159,
      "epoch": 2.6507683330091423,
      "grad_norm": 1.125,
      "learning_rate": 0.000429371064671724,
      "loss": 5.0556,
      "mean_token_accuracy": 0.1883290097117424,
      "num_tokens": 59094390.0,
      "step": 34070
    },
    {
      "entropy": 5.48091950416565,
      "epoch": 2.65115736238086,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042935090004309247,
      "loss": 5.0319,
      "mean_token_accuracy": 0.19576746523380278,
      "num_tokens": 59102310.0,
      "step": 34075
    },
    {
      "entropy": 5.456078004837036,
      "epoch": 2.6515463917525772,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042933073307243165,
      "loss": 5.001,
      "mean_token_accuracy": 0.19584300369024277,
      "num_tokens": 59111385.0,
      "step": 34080
    },
    {
      "entropy": 5.404288625717163,
      "epoch": 2.651935421124295,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042931056376004767,
      "loss": 5.0713,
      "mean_token_accuracy": 0.19850219488143922,
      "num_tokens": 59120152.0,
      "step": 34085
    },
    {
      "entropy": 5.381229114532471,
      "epoch": 2.652324450496012,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004292903921062465,
      "loss": 4.8746,
      "mean_token_accuracy": 0.21270565390586854,
      "num_tokens": 59128782.0,
      "step": 34090
    },
    {
      "entropy": 5.441132831573486,
      "epoch": 2.65271347986773,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042927021811133416,
      "loss": 5.02,
      "mean_token_accuracy": 0.19743793159723283,
      "num_tokens": 59136946.0,
      "step": 34095
    },
    {
      "entropy": 5.355271530151367,
      "epoch": 2.6531025092394476,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004292500417756167,
      "loss": 4.9521,
      "mean_token_accuracy": 0.20467088371515274,
      "num_tokens": 59145845.0,
      "step": 34100
    },
    {
      "entropy": 5.480632495880127,
      "epoch": 2.6534915386111653,
      "grad_norm": 1.3125,
      "learning_rate": 0.00042922986309940054,
      "loss": 5.0926,
      "mean_token_accuracy": 0.18817767202854158,
      "num_tokens": 59154607.0,
      "step": 34105
    },
    {
      "entropy": 5.345821142196655,
      "epoch": 2.653880567982883,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042920968208299165,
      "loss": 4.9979,
      "mean_token_accuracy": 0.1961749315261841,
      "num_tokens": 59164160.0,
      "step": 34110
    },
    {
      "entropy": 5.413113260269165,
      "epoch": 2.6542695973546,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004291894987266963,
      "loss": 4.975,
      "mean_token_accuracy": 0.2036278709769249,
      "num_tokens": 59171976.0,
      "step": 34115
    },
    {
      "entropy": 5.366276836395263,
      "epoch": 2.654658626726318,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042916931303082064,
      "loss": 5.0274,
      "mean_token_accuracy": 0.19248372912406922,
      "num_tokens": 59181168.0,
      "step": 34120
    },
    {
      "entropy": 5.501958894729614,
      "epoch": 2.655047656098035,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042914912499567113,
      "loss": 5.1644,
      "mean_token_accuracy": 0.18492101430892943,
      "num_tokens": 59190047.0,
      "step": 34125
    },
    {
      "entropy": 5.442483282089233,
      "epoch": 2.655436685469753,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042912893462155395,
      "loss": 5.0152,
      "mean_token_accuracy": 0.1948254033923149,
      "num_tokens": 59199065.0,
      "step": 34130
    },
    {
      "entropy": 5.347232627868652,
      "epoch": 2.6558257148414706,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042910874190877545,
      "loss": 4.9185,
      "mean_token_accuracy": 0.20756661593914033,
      "num_tokens": 59207827.0,
      "step": 34135
    },
    {
      "entropy": 5.431786108016968,
      "epoch": 2.6562147442131883,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004290885468576422,
      "loss": 5.1286,
      "mean_token_accuracy": 0.18795362561941148,
      "num_tokens": 59216429.0,
      "step": 34140
    },
    {
      "entropy": 5.361321020126343,
      "epoch": 2.6566037735849055,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004290683494684604,
      "loss": 4.97,
      "mean_token_accuracy": 0.189467254281044,
      "num_tokens": 59224447.0,
      "step": 34145
    },
    {
      "entropy": 5.4632689476013185,
      "epoch": 2.656992802956623,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004290481497415367,
      "loss": 5.066,
      "mean_token_accuracy": 0.18852092027664186,
      "num_tokens": 59233124.0,
      "step": 34150
    },
    {
      "entropy": 5.327345323562622,
      "epoch": 2.657381832328341,
      "grad_norm": 1.125,
      "learning_rate": 0.0004290279476771775,
      "loss": 4.9047,
      "mean_token_accuracy": 0.20259241312742232,
      "num_tokens": 59241450.0,
      "step": 34155
    },
    {
      "entropy": 5.403840112686157,
      "epoch": 2.657770861700058,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004290077432756894,
      "loss": 5.1408,
      "mean_token_accuracy": 0.18461692333221436,
      "num_tokens": 59250797.0,
      "step": 34160
    },
    {
      "entropy": 5.393539953231811,
      "epoch": 2.658159891071776,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004289875365373788,
      "loss": 4.9075,
      "mean_token_accuracy": 0.20092692077159882,
      "num_tokens": 59258854.0,
      "step": 34165
    },
    {
      "entropy": 5.438014078140259,
      "epoch": 2.6585489204434936,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042896732746255256,
      "loss": 5.0034,
      "mean_token_accuracy": 0.19441579282283783,
      "num_tokens": 59267797.0,
      "step": 34170
    },
    {
      "entropy": 5.361431407928467,
      "epoch": 2.6589379498152113,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042894711605151714,
      "loss": 4.9037,
      "mean_token_accuracy": 0.2024035394191742,
      "num_tokens": 59276370.0,
      "step": 34175
    },
    {
      "entropy": 5.352745389938354,
      "epoch": 2.6593269791869285,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042892690230457924,
      "loss": 4.9805,
      "mean_token_accuracy": 0.19391236305236817,
      "num_tokens": 59285641.0,
      "step": 34180
    },
    {
      "entropy": 5.367085409164429,
      "epoch": 2.659716008558646,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042890668622204566,
      "loss": 4.9927,
      "mean_token_accuracy": 0.1983859658241272,
      "num_tokens": 59293816.0,
      "step": 34185
    },
    {
      "entropy": 5.387168025970459,
      "epoch": 2.660105037930364,
      "grad_norm": 1.125,
      "learning_rate": 0.000428886467804223,
      "loss": 4.919,
      "mean_token_accuracy": 0.20364300012588502,
      "num_tokens": 59301854.0,
      "step": 34190
    },
    {
      "entropy": 5.289494228363037,
      "epoch": 2.660494067302081,
      "grad_norm": 1.125,
      "learning_rate": 0.00042886624705141825,
      "loss": 4.9387,
      "mean_token_accuracy": 0.1990928828716278,
      "num_tokens": 59310798.0,
      "step": 34195
    },
    {
      "entropy": 5.347465562820434,
      "epoch": 2.660883096673799,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042884602396393796,
      "loss": 4.9863,
      "mean_token_accuracy": 0.19715733379125594,
      "num_tokens": 59319085.0,
      "step": 34200
    },
    {
      "entropy": 5.381764078140259,
      "epoch": 2.6612721260455166,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004288257985420892,
      "loss": 5.0505,
      "mean_token_accuracy": 0.19189455956220627,
      "num_tokens": 59327805.0,
      "step": 34205
    },
    {
      "entropy": 5.465325927734375,
      "epoch": 2.6616611554172342,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004288055707861788,
      "loss": 4.9766,
      "mean_token_accuracy": 0.20451024919748306,
      "num_tokens": 59336520.0,
      "step": 34210
    },
    {
      "entropy": 5.375922393798828,
      "epoch": 2.6620501847889515,
      "grad_norm": 1.125,
      "learning_rate": 0.00042878534069651364,
      "loss": 5.0458,
      "mean_token_accuracy": 0.19396799355745314,
      "num_tokens": 59345827.0,
      "step": 34215
    },
    {
      "entropy": 5.384785985946655,
      "epoch": 2.662439214160669,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004287651082734007,
      "loss": 5.047,
      "mean_token_accuracy": 0.2000302717089653,
      "num_tokens": 59355024.0,
      "step": 34220
    },
    {
      "entropy": 5.360146951675415,
      "epoch": 2.6628282435323865,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000428744873517147,
      "loss": 4.9317,
      "mean_token_accuracy": 0.19957145899534226,
      "num_tokens": 59364461.0,
      "step": 34225
    },
    {
      "entropy": 5.392741584777832,
      "epoch": 2.663217272904104,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004287246364280596,
      "loss": 5.0904,
      "mean_token_accuracy": 0.1864723950624466,
      "num_tokens": 59372807.0,
      "step": 34230
    },
    {
      "entropy": 5.464121723175049,
      "epoch": 2.663606302275822,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004287043970064455,
      "loss": 5.0407,
      "mean_token_accuracy": 0.19846223890781403,
      "num_tokens": 59381824.0,
      "step": 34235
    },
    {
      "entropy": 5.371962642669677,
      "epoch": 2.6639953316475395,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004286841552526118,
      "loss": 4.9417,
      "mean_token_accuracy": 0.19834139347076415,
      "num_tokens": 59391174.0,
      "step": 34240
    },
    {
      "entropy": 5.349871683120727,
      "epoch": 2.664384361019257,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042866391116686567,
      "loss": 5.0421,
      "mean_token_accuracy": 0.19739351272583008,
      "num_tokens": 59399927.0,
      "step": 34245
    },
    {
      "entropy": 5.431089878082275,
      "epoch": 2.6647733903909745,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004286436647495142,
      "loss": 5.0332,
      "mean_token_accuracy": 0.1994730606675148,
      "num_tokens": 59408016.0,
      "step": 34250
    },
    {
      "entropy": 5.401451826095581,
      "epoch": 2.665162419762692,
      "grad_norm": 1.375,
      "learning_rate": 0.00042862341600086477,
      "loss": 5.067,
      "mean_token_accuracy": 0.19018494635820388,
      "num_tokens": 59416702.0,
      "step": 34255
    },
    {
      "entropy": 5.391842317581177,
      "epoch": 2.6655514491344094,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042860316492122447,
      "loss": 4.9491,
      "mean_token_accuracy": 0.1969820663332939,
      "num_tokens": 59424766.0,
      "step": 34260
    },
    {
      "entropy": 5.400101804733277,
      "epoch": 2.665940478506127,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004285829115109006,
      "loss": 5.0564,
      "mean_token_accuracy": 0.19883065670728683,
      "num_tokens": 59433691.0,
      "step": 34265
    },
    {
      "entropy": 5.43451714515686,
      "epoch": 2.666329507877845,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004285626557702005,
      "loss": 5.0533,
      "mean_token_accuracy": 0.19789710640907288,
      "num_tokens": 59442495.0,
      "step": 34270
    },
    {
      "entropy": 5.367371845245361,
      "epoch": 2.6667185372495625,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004285423976994316,
      "loss": 4.9517,
      "mean_token_accuracy": 0.20110870003700257,
      "num_tokens": 59451038.0,
      "step": 34275
    },
    {
      "entropy": 5.298448514938355,
      "epoch": 2.66710756662128,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042852213729890117,
      "loss": 4.9649,
      "mean_token_accuracy": 0.19888628274202347,
      "num_tokens": 59459119.0,
      "step": 34280
    },
    {
      "entropy": 5.378402948379517,
      "epoch": 2.6674965959929975,
      "grad_norm": 1.125,
      "learning_rate": 0.0004285018745689166,
      "loss": 4.974,
      "mean_token_accuracy": 0.20053848773241043,
      "num_tokens": 59467725.0,
      "step": 34285
    },
    {
      "entropy": 5.42381534576416,
      "epoch": 2.667885625364715,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004284816095097855,
      "loss": 4.9761,
      "mean_token_accuracy": 0.1975323513150215,
      "num_tokens": 59476231.0,
      "step": 34290
    },
    {
      "entropy": 5.352923011779785,
      "epoch": 2.6682746547364324,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004284613421218152,
      "loss": 5.0035,
      "mean_token_accuracy": 0.19331594705581664,
      "num_tokens": 59484898.0,
      "step": 34295
    },
    {
      "entropy": 5.359823894500733,
      "epoch": 2.66866368410815,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042844107240531324,
      "loss": 4.9263,
      "mean_token_accuracy": 0.200864040851593,
      "num_tokens": 59493546.0,
      "step": 34300
    },
    {
      "entropy": 5.3814184188842775,
      "epoch": 2.669052713479868,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004284208003605873,
      "loss": 5.0033,
      "mean_token_accuracy": 0.19196025282144547,
      "num_tokens": 59502729.0,
      "step": 34305
    },
    {
      "entropy": 5.3861700057983395,
      "epoch": 2.6694417428515855,
      "grad_norm": 1.109375,
      "learning_rate": 0.000428400525987945,
      "loss": 5.0584,
      "mean_token_accuracy": 0.18843467980623246,
      "num_tokens": 59512069.0,
      "step": 34310
    },
    {
      "entropy": 5.415060997009277,
      "epoch": 2.669830772223303,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042838024928769375,
      "loss": 5.0131,
      "mean_token_accuracy": 0.19946872293949128,
      "num_tokens": 59520639.0,
      "step": 34315
    },
    {
      "entropy": 5.400388383865357,
      "epoch": 2.6702198015950205,
      "grad_norm": 1.125,
      "learning_rate": 0.00042835997026014145,
      "loss": 4.9278,
      "mean_token_accuracy": 0.20618724972009658,
      "num_tokens": 59528549.0,
      "step": 34320
    },
    {
      "entropy": 5.354205560684204,
      "epoch": 2.6706088309667377,
      "grad_norm": 1.125,
      "learning_rate": 0.0004283396889055957,
      "loss": 4.9586,
      "mean_token_accuracy": 0.20648440420627595,
      "num_tokens": 59537345.0,
      "step": 34325
    },
    {
      "entropy": 5.358095026016235,
      "epoch": 2.6709978603384554,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004283194052243641,
      "loss": 5.0439,
      "mean_token_accuracy": 0.1978743180632591,
      "num_tokens": 59546866.0,
      "step": 34330
    },
    {
      "entropy": 5.457716226577759,
      "epoch": 2.671386889710173,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042829911921675456,
      "loss": 4.9965,
      "mean_token_accuracy": 0.2010033667087555,
      "num_tokens": 59555363.0,
      "step": 34335
    },
    {
      "entropy": 5.331245756149292,
      "epoch": 2.671775919081891,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000428278830883075,
      "loss": 4.8783,
      "mean_token_accuracy": 0.2030148908495903,
      "num_tokens": 59564417.0,
      "step": 34340
    },
    {
      "entropy": 5.502124977111817,
      "epoch": 2.6721649484536085,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042825854022363307,
      "loss": 4.9891,
      "mean_token_accuracy": 0.19220607876777648,
      "num_tokens": 59572948.0,
      "step": 34345
    },
    {
      "entropy": 5.322700786590576,
      "epoch": 2.6725539778253258,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042823824723873676,
      "loss": 4.9359,
      "mean_token_accuracy": 0.19934329837560655,
      "num_tokens": 59581764.0,
      "step": 34350
    },
    {
      "entropy": 5.314413452148438,
      "epoch": 2.6729430071970435,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042821795192869407,
      "loss": 4.9643,
      "mean_token_accuracy": 0.20451496988534928,
      "num_tokens": 59590281.0,
      "step": 34355
    },
    {
      "entropy": 5.489963531494141,
      "epoch": 2.6733320365687607,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004281976542938127,
      "loss": 5.1313,
      "mean_token_accuracy": 0.1904164284467697,
      "num_tokens": 59598497.0,
      "step": 34360
    },
    {
      "entropy": 5.411137866973877,
      "epoch": 2.6737210659404784,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042817735433440076,
      "loss": 4.9635,
      "mean_token_accuracy": 0.2024264469742775,
      "num_tokens": 59606901.0,
      "step": 34365
    },
    {
      "entropy": 5.494289302825928,
      "epoch": 2.674110095312196,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004281570520507663,
      "loss": 5.1121,
      "mean_token_accuracy": 0.19492388814687728,
      "num_tokens": 59616045.0,
      "step": 34370
    },
    {
      "entropy": 5.472842264175415,
      "epoch": 2.674499124683914,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042813674744321737,
      "loss": 5.0377,
      "mean_token_accuracy": 0.20080640465021132,
      "num_tokens": 59624840.0,
      "step": 34375
    },
    {
      "entropy": 5.411509370803833,
      "epoch": 2.674888154055631,
      "grad_norm": 1.21875,
      "learning_rate": 0.000428116440512062,
      "loss": 5.0087,
      "mean_token_accuracy": 0.19539084285497665,
      "num_tokens": 59632887.0,
      "step": 34380
    },
    {
      "entropy": 5.33553729057312,
      "epoch": 2.6752771834273488,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004280961312576084,
      "loss": 4.948,
      "mean_token_accuracy": 0.20470093041658402,
      "num_tokens": 59641671.0,
      "step": 34385
    },
    {
      "entropy": 5.269271183013916,
      "epoch": 2.6756662127990665,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004280758196801646,
      "loss": 4.8923,
      "mean_token_accuracy": 0.2071111962199211,
      "num_tokens": 59649406.0,
      "step": 34390
    },
    {
      "entropy": 5.416578245162964,
      "epoch": 2.6760552421707837,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042805550578003894,
      "loss": 5.0778,
      "mean_token_accuracy": 0.19426416754722595,
      "num_tokens": 59658114.0,
      "step": 34395
    },
    {
      "entropy": 5.437091732025147,
      "epoch": 2.6764442715425014,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004280351895575396,
      "loss": 5.0175,
      "mean_token_accuracy": 0.1957835778594017,
      "num_tokens": 59667174.0,
      "step": 34400
    },
    {
      "entropy": 5.342769956588745,
      "epoch": 2.676833300914219,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004280148710129748,
      "loss": 4.9987,
      "mean_token_accuracy": 0.2011121168732643,
      "num_tokens": 59675740.0,
      "step": 34405
    },
    {
      "entropy": 5.474993705749512,
      "epoch": 2.677222330285937,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042799455014665296,
      "loss": 5.1539,
      "mean_token_accuracy": 0.18720866590738297,
      "num_tokens": 59684550.0,
      "step": 34410
    },
    {
      "entropy": 5.523981428146362,
      "epoch": 2.677611359657654,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042797422695888226,
      "loss": 5.0701,
      "mean_token_accuracy": 0.19565058797597884,
      "num_tokens": 59692810.0,
      "step": 34415
    },
    {
      "entropy": 5.40265679359436,
      "epoch": 2.6780003890293718,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004279539014499712,
      "loss": 4.991,
      "mean_token_accuracy": 0.19077727496623992,
      "num_tokens": 59701742.0,
      "step": 34420
    },
    {
      "entropy": 5.413672828674317,
      "epoch": 2.678389418401089,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004279335736202281,
      "loss": 5.0169,
      "mean_token_accuracy": 0.19683320671319962,
      "num_tokens": 59710440.0,
      "step": 34425
    },
    {
      "entropy": 5.496515560150146,
      "epoch": 2.6787784477728067,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004279132434699615,
      "loss": 5.0835,
      "mean_token_accuracy": 0.19308008402585983,
      "num_tokens": 59720430.0,
      "step": 34430
    },
    {
      "entropy": 5.342096757888794,
      "epoch": 2.6791674771445244,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004278929109994798,
      "loss": 4.95,
      "mean_token_accuracy": 0.19330374002456666,
      "num_tokens": 59730249.0,
      "step": 34435
    },
    {
      "entropy": 5.362269926071167,
      "epoch": 2.679556506516242,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042787257620909144,
      "loss": 4.8993,
      "mean_token_accuracy": 0.20127840340137482,
      "num_tokens": 59738815.0,
      "step": 34440
    },
    {
      "entropy": 5.289106607437134,
      "epoch": 2.67994553588796,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042785223909910515,
      "loss": 4.8953,
      "mean_token_accuracy": 0.19872929900884628,
      "num_tokens": 59747092.0,
      "step": 34445
    },
    {
      "entropy": 5.4077647686004635,
      "epoch": 2.680334565259677,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004278318996698294,
      "loss": 5.0701,
      "mean_token_accuracy": 0.19181589186191558,
      "num_tokens": 59756832.0,
      "step": 34450
    },
    {
      "entropy": 5.50298318862915,
      "epoch": 2.6807235946313948,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004278115579215728,
      "loss": 5.0688,
      "mean_token_accuracy": 0.19121240228414535,
      "num_tokens": 59766052.0,
      "step": 34455
    },
    {
      "entropy": 5.407856464385986,
      "epoch": 2.681112624003112,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004277912138546441,
      "loss": 4.9782,
      "mean_token_accuracy": 0.19869401752948762,
      "num_tokens": 59774369.0,
      "step": 34460
    },
    {
      "entropy": 5.313616418838501,
      "epoch": 2.6815016533748297,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004277708674693519,
      "loss": 4.9266,
      "mean_token_accuracy": 0.20080707520246505,
      "num_tokens": 59782509.0,
      "step": 34465
    },
    {
      "entropy": 5.312063217163086,
      "epoch": 2.6818906827465474,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042775051876600483,
      "loss": 4.886,
      "mean_token_accuracy": 0.20883483588695526,
      "num_tokens": 59790808.0,
      "step": 34470
    },
    {
      "entropy": 5.364171504974365,
      "epoch": 2.682279712118265,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004277301677449119,
      "loss": 4.8725,
      "mean_token_accuracy": 0.20297887325286865,
      "num_tokens": 59799260.0,
      "step": 34475
    },
    {
      "entropy": 5.420628786087036,
      "epoch": 2.6826687414899824,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004277098144063818,
      "loss": 5.0339,
      "mean_token_accuracy": 0.19999630451202394,
      "num_tokens": 59807350.0,
      "step": 34480
    },
    {
      "entropy": 5.384455585479737,
      "epoch": 2.6830577708617,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042768945875072326,
      "loss": 5.0383,
      "mean_token_accuracy": 0.19103001356124877,
      "num_tokens": 59816071.0,
      "step": 34485
    },
    {
      "entropy": 5.463081359863281,
      "epoch": 2.6834468002334178,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004276691007782451,
      "loss": 5.0159,
      "mean_token_accuracy": 0.2002073734998703,
      "num_tokens": 59825332.0,
      "step": 34490
    },
    {
      "entropy": 5.400717067718506,
      "epoch": 2.683835829605135,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004276487404892565,
      "loss": 5.0158,
      "mean_token_accuracy": 0.19342595636844634,
      "num_tokens": 59834082.0,
      "step": 34495
    },
    {
      "entropy": 5.44166841506958,
      "epoch": 2.6842248589768527,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042762837788406615,
      "loss": 5.0308,
      "mean_token_accuracy": 0.19211960285902024,
      "num_tokens": 59842633.0,
      "step": 34500
    },
    {
      "entropy": 5.441576242446899,
      "epoch": 2.6846138883485704,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004276080129629831,
      "loss": 4.9891,
      "mean_token_accuracy": 0.19263278543949128,
      "num_tokens": 59850517.0,
      "step": 34505
    },
    {
      "entropy": 5.427461194992065,
      "epoch": 2.685002917720288,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00042758764572631644,
      "loss": 5.0889,
      "mean_token_accuracy": 0.1949665367603302,
      "num_tokens": 59859112.0,
      "step": 34510
    },
    {
      "entropy": 5.371607875823974,
      "epoch": 2.6853919470920053,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042756727617437504,
      "loss": 4.9154,
      "mean_token_accuracy": 0.20073309987783433,
      "num_tokens": 59868413.0,
      "step": 34515
    },
    {
      "entropy": 5.493725347518921,
      "epoch": 2.685780976463723,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042754690430746805,
      "loss": 5.1118,
      "mean_token_accuracy": 0.18919164538383484,
      "num_tokens": 59877300.0,
      "step": 34520
    },
    {
      "entropy": 5.385222578048706,
      "epoch": 2.6861700058354403,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004275265301259046,
      "loss": 5.0244,
      "mean_token_accuracy": 0.20145601779222488,
      "num_tokens": 59885440.0,
      "step": 34525
    },
    {
      "entropy": 5.488229560852051,
      "epoch": 2.686559035207158,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042750615362999395,
      "loss": 5.1289,
      "mean_token_accuracy": 0.18318473249673844,
      "num_tokens": 59893954.0,
      "step": 34530
    },
    {
      "entropy": 5.422585344314575,
      "epoch": 2.6869480645788757,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042748577482004495,
      "loss": 5.0393,
      "mean_token_accuracy": 0.19546912014484405,
      "num_tokens": 59902336.0,
      "step": 34535
    },
    {
      "entropy": 5.309809732437134,
      "epoch": 2.6873370939505934,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042746539369636715,
      "loss": 4.8922,
      "mean_token_accuracy": 0.20755211263895035,
      "num_tokens": 59911134.0,
      "step": 34540
    },
    {
      "entropy": 5.295792007446289,
      "epoch": 2.687726123322311,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00042744501025926966,
      "loss": 4.8894,
      "mean_token_accuracy": 0.20303991585969924,
      "num_tokens": 59920450.0,
      "step": 34545
    },
    {
      "entropy": 5.46213846206665,
      "epoch": 2.6881151526940283,
      "grad_norm": 1.125,
      "learning_rate": 0.0004274246245090618,
      "loss": 4.9616,
      "mean_token_accuracy": 0.20705340802669525,
      "num_tokens": 59928919.0,
      "step": 34550
    },
    {
      "entropy": 5.389980745315552,
      "epoch": 2.688504182065746,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004274042364460529,
      "loss": 4.9909,
      "mean_token_accuracy": 0.19752189218997956,
      "num_tokens": 59937973.0,
      "step": 34555
    },
    {
      "entropy": 5.34364070892334,
      "epoch": 2.6888932114374633,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042738384607055214,
      "loss": 5.0151,
      "mean_token_accuracy": 0.19624349027872084,
      "num_tokens": 59946177.0,
      "step": 34560
    },
    {
      "entropy": 5.387221240997315,
      "epoch": 2.689282240809181,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042736345338286917,
      "loss": 4.9937,
      "mean_token_accuracy": 0.20159117877483368,
      "num_tokens": 59955351.0,
      "step": 34565
    },
    {
      "entropy": 5.474174213409424,
      "epoch": 2.6896712701808987,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004273430583833133,
      "loss": 5.0686,
      "mean_token_accuracy": 0.1945489227771759,
      "num_tokens": 59963839.0,
      "step": 34570
    },
    {
      "entropy": 5.3866644382476805,
      "epoch": 2.6900602995526164,
      "grad_norm": 1.15625,
      "learning_rate": 0.000427322661072194,
      "loss": 4.9966,
      "mean_token_accuracy": 0.19049329310655594,
      "num_tokens": 59971779.0,
      "step": 34575
    },
    {
      "entropy": 5.392128753662109,
      "epoch": 2.6904493289243336,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004273022614498208,
      "loss": 4.9626,
      "mean_token_accuracy": 0.2020797312259674,
      "num_tokens": 59980932.0,
      "step": 34580
    },
    {
      "entropy": 5.382230615615844,
      "epoch": 2.6908383582960513,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004272818595165031,
      "loss": 4.9731,
      "mean_token_accuracy": 0.20196616351604463,
      "num_tokens": 59990281.0,
      "step": 34585
    },
    {
      "entropy": 5.297256755828857,
      "epoch": 2.691227387667769,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004272614552725506,
      "loss": 4.8847,
      "mean_token_accuracy": 0.21121360063552858,
      "num_tokens": 59998642.0,
      "step": 34590
    },
    {
      "entropy": 5.405024909973145,
      "epoch": 2.6916164170394863,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042724104871827293,
      "loss": 5.0735,
      "mean_token_accuracy": 0.1906057357788086,
      "num_tokens": 60006998.0,
      "step": 34595
    },
    {
      "entropy": 5.3436497211456295,
      "epoch": 2.692005446411204,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004272206398539795,
      "loss": 4.9472,
      "mean_token_accuracy": 0.20375546663999558,
      "num_tokens": 60015279.0,
      "step": 34600
    },
    {
      "entropy": 5.365096139907837,
      "epoch": 2.6923944757829217,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042720022867998026,
      "loss": 5.0204,
      "mean_token_accuracy": 0.20281146764755248,
      "num_tokens": 60024648.0,
      "step": 34605
    },
    {
      "entropy": 5.503239297866822,
      "epoch": 2.6927835051546394,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004271798151965848,
      "loss": 5.0943,
      "mean_token_accuracy": 0.18786617666482924,
      "num_tokens": 60033821.0,
      "step": 34610
    },
    {
      "entropy": 5.390892791748047,
      "epoch": 2.6931725345263566,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004271593994041029,
      "loss": 5.0021,
      "mean_token_accuracy": 0.19315481036901475,
      "num_tokens": 60041745.0,
      "step": 34615
    },
    {
      "entropy": 5.400434875488282,
      "epoch": 2.6935615638980743,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004271389813028443,
      "loss": 5.0292,
      "mean_token_accuracy": 0.19832102805376053,
      "num_tokens": 60051659.0,
      "step": 34620
    },
    {
      "entropy": 5.462645769119263,
      "epoch": 2.693950593269792,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004271185608931187,
      "loss": 5.0768,
      "mean_token_accuracy": 0.19329693168401718,
      "num_tokens": 60060498.0,
      "step": 34625
    },
    {
      "entropy": 5.359845685958862,
      "epoch": 2.6943396226415093,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042709813817523617,
      "loss": 4.9926,
      "mean_token_accuracy": 0.19633451998233795,
      "num_tokens": 60068678.0,
      "step": 34630
    },
    {
      "entropy": 5.323875570297242,
      "epoch": 2.694728652013227,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042707771314950645,
      "loss": 4.9081,
      "mean_token_accuracy": 0.19849919229745866,
      "num_tokens": 60077889.0,
      "step": 34635
    },
    {
      "entropy": 5.39540662765503,
      "epoch": 2.6951176813849447,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042705728581623944,
      "loss": 4.9748,
      "mean_token_accuracy": 0.1911034792661667,
      "num_tokens": 60085989.0,
      "step": 34640
    },
    {
      "entropy": 5.412735319137573,
      "epoch": 2.6955067107566624,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004270368561757452,
      "loss": 4.9707,
      "mean_token_accuracy": 0.204974664747715,
      "num_tokens": 60095012.0,
      "step": 34645
    },
    {
      "entropy": 5.436860370635986,
      "epoch": 2.6958957401283796,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004270164242283336,
      "loss": 5.013,
      "mean_token_accuracy": 0.19279265999794007,
      "num_tokens": 60103894.0,
      "step": 34650
    },
    {
      "entropy": 5.33751277923584,
      "epoch": 2.6962847695000973,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004269959899743148,
      "loss": 4.9008,
      "mean_token_accuracy": 0.20609208643436433,
      "num_tokens": 60112456.0,
      "step": 34655
    },
    {
      "entropy": 5.39109845161438,
      "epoch": 2.6966737988718146,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004269755534139987,
      "loss": 4.9825,
      "mean_token_accuracy": 0.20097634792327881,
      "num_tokens": 60120378.0,
      "step": 34660
    },
    {
      "entropy": 5.446466064453125,
      "epoch": 2.6970628282435323,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042695511454769557,
      "loss": 5.0586,
      "mean_token_accuracy": 0.19718556404113768,
      "num_tokens": 60128091.0,
      "step": 34665
    },
    {
      "entropy": 5.418723487854004,
      "epoch": 2.69745185761525,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042693467337571534,
      "loss": 5.0478,
      "mean_token_accuracy": 0.20279632210731507,
      "num_tokens": 60136298.0,
      "step": 34670
    },
    {
      "entropy": 5.442491340637207,
      "epoch": 2.6978408869869677,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004269142298983684,
      "loss": 4.9901,
      "mean_token_accuracy": 0.2012015625834465,
      "num_tokens": 60145556.0,
      "step": 34675
    },
    {
      "entropy": 5.412729978561401,
      "epoch": 2.698229916358685,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042689378411596467,
      "loss": 4.9602,
      "mean_token_accuracy": 0.1987512767314911,
      "num_tokens": 60154301.0,
      "step": 34680
    },
    {
      "entropy": 5.2845447063446045,
      "epoch": 2.6986189457304026,
      "grad_norm": 1.125,
      "learning_rate": 0.0004268733360288146,
      "loss": 4.9098,
      "mean_token_accuracy": 0.2080233082175255,
      "num_tokens": 60162646.0,
      "step": 34685
    },
    {
      "entropy": 5.364597988128662,
      "epoch": 2.6990079751021203,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042685288563722835,
      "loss": 4.9622,
      "mean_token_accuracy": 0.1982418715953827,
      "num_tokens": 60170749.0,
      "step": 34690
    },
    {
      "entropy": 5.3218918800354,
      "epoch": 2.6993970044738376,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004268324329415163,
      "loss": 4.8866,
      "mean_token_accuracy": 0.20682495087385178,
      "num_tokens": 60178798.0,
      "step": 34695
    },
    {
      "entropy": 5.352897453308105,
      "epoch": 2.6997860338455553,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004268119779419887,
      "loss": 4.9297,
      "mean_token_accuracy": 0.20208481699228287,
      "num_tokens": 60186974.0,
      "step": 34700
    },
    {
      "entropy": 5.391630601882935,
      "epoch": 2.700175063217273,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042679152063895603,
      "loss": 4.9153,
      "mean_token_accuracy": 0.20124580711126328,
      "num_tokens": 60195271.0,
      "step": 34705
    },
    {
      "entropy": 5.404238700866699,
      "epoch": 2.7005640925889907,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042677106103272867,
      "loss": 4.9507,
      "mean_token_accuracy": 0.20059803873300552,
      "num_tokens": 60204476.0,
      "step": 34710
    },
    {
      "entropy": 5.415479421615601,
      "epoch": 2.700953121960708,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000426750599123617,
      "loss": 5.0623,
      "mean_token_accuracy": 0.19499545842409133,
      "num_tokens": 60213285.0,
      "step": 34715
    },
    {
      "entropy": 5.379653644561768,
      "epoch": 2.7013421513324256,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042673013491193145,
      "loss": 4.9693,
      "mean_token_accuracy": 0.20879622250795365,
      "num_tokens": 60222321.0,
      "step": 34720
    },
    {
      "entropy": 5.3582813262939455,
      "epoch": 2.7017311807041433,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042670966839798263,
      "loss": 4.9474,
      "mean_token_accuracy": 0.21028258204460143,
      "num_tokens": 60230759.0,
      "step": 34725
    },
    {
      "entropy": 5.415563917160034,
      "epoch": 2.7021202100758606,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004266891995820811,
      "loss": 4.9794,
      "mean_token_accuracy": 0.20133963078260422,
      "num_tokens": 60238570.0,
      "step": 34730
    },
    {
      "entropy": 5.359295320510864,
      "epoch": 2.7025092394475783,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042666872846453734,
      "loss": 5.0461,
      "mean_token_accuracy": 0.19327437728643418,
      "num_tokens": 60247445.0,
      "step": 34735
    },
    {
      "entropy": 5.363235855102539,
      "epoch": 2.702898268819296,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00042664825504566206,
      "loss": 4.9988,
      "mean_token_accuracy": 0.1964689165353775,
      "num_tokens": 60256358.0,
      "step": 34740
    },
    {
      "entropy": 5.389095687866211,
      "epoch": 2.7032872981910137,
      "grad_norm": 1.203125,
      "learning_rate": 0.000426627779325766,
      "loss": 4.9935,
      "mean_token_accuracy": 0.20379738956689836,
      "num_tokens": 60264347.0,
      "step": 34745
    },
    {
      "entropy": 5.389885330200196,
      "epoch": 2.703676327562731,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004266073013051595,
      "loss": 4.9476,
      "mean_token_accuracy": 0.196377931535244,
      "num_tokens": 60273858.0,
      "step": 34750
    },
    {
      "entropy": 5.383081769943237,
      "epoch": 2.7040653569344486,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004265868209841537,
      "loss": 4.9471,
      "mean_token_accuracy": 0.20187449753284453,
      "num_tokens": 60282089.0,
      "step": 34755
    },
    {
      "entropy": 5.4112996578216555,
      "epoch": 2.704454386306166,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004265663383630591,
      "loss": 5.0469,
      "mean_token_accuracy": 0.19911534935235978,
      "num_tokens": 60291253.0,
      "step": 34760
    },
    {
      "entropy": 5.504092645645142,
      "epoch": 2.7048434156778836,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042654585344218657,
      "loss": 5.0416,
      "mean_token_accuracy": 0.1932744041085243,
      "num_tokens": 60300584.0,
      "step": 34765
    },
    {
      "entropy": 5.399906015396118,
      "epoch": 2.7052324450496013,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004265253662218469,
      "loss": 5.0002,
      "mean_token_accuracy": 0.20235911756753922,
      "num_tokens": 60308822.0,
      "step": 34770
    },
    {
      "entropy": 5.375939226150512,
      "epoch": 2.705621474421319,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042650487670235093,
      "loss": 5.0596,
      "mean_token_accuracy": 0.19862616062164307,
      "num_tokens": 60317466.0,
      "step": 34775
    },
    {
      "entropy": 5.455323505401611,
      "epoch": 2.7060105037930366,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004264843848840097,
      "loss": 5.0386,
      "mean_token_accuracy": 0.19244395494461058,
      "num_tokens": 60325307.0,
      "step": 34780
    },
    {
      "entropy": 5.439924526214599,
      "epoch": 2.706399533164754,
      "grad_norm": 1.203125,
      "learning_rate": 0.000426463890767134,
      "loss": 5.1075,
      "mean_token_accuracy": 0.1866210252046585,
      "num_tokens": 60334300.0,
      "step": 34785
    },
    {
      "entropy": 5.372380638122559,
      "epoch": 2.7067885625364716,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004264433943520349,
      "loss": 5.0249,
      "mean_token_accuracy": 0.19149001687765121,
      "num_tokens": 60343481.0,
      "step": 34790
    },
    {
      "entropy": 5.407073640823365,
      "epoch": 2.707177591908189,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042642289563902326,
      "loss": 4.9988,
      "mean_token_accuracy": 0.19364880621433259,
      "num_tokens": 60353134.0,
      "step": 34795
    },
    {
      "entropy": 5.3637144565582275,
      "epoch": 2.7075666212799065,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042640239462841023,
      "loss": 4.9335,
      "mean_token_accuracy": 0.20837823450565338,
      "num_tokens": 60362389.0,
      "step": 34800
    },
    {
      "entropy": 5.305854320526123,
      "epoch": 2.7079556506516242,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042638189132050675,
      "loss": 4.9029,
      "mean_token_accuracy": 0.20706166028976442,
      "num_tokens": 60370963.0,
      "step": 34805
    },
    {
      "entropy": 5.354073333740234,
      "epoch": 2.708344680023342,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042636138571562415,
      "loss": 5.0374,
      "mean_token_accuracy": 0.18988647758960725,
      "num_tokens": 60379534.0,
      "step": 34810
    },
    {
      "entropy": 5.43061146736145,
      "epoch": 2.708733709395059,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042634087781407337,
      "loss": 5.0269,
      "mean_token_accuracy": 0.19171185344457625,
      "num_tokens": 60389234.0,
      "step": 34815
    },
    {
      "entropy": 5.415381574630738,
      "epoch": 2.709122738766777,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042632036761616564,
      "loss": 4.9603,
      "mean_token_accuracy": 0.1974749132990837,
      "num_tokens": 60398571.0,
      "step": 34820
    },
    {
      "entropy": 5.407800340652466,
      "epoch": 2.7095117681384946,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042629985512221224,
      "loss": 5.0637,
      "mean_token_accuracy": 0.19485462903976442,
      "num_tokens": 60407356.0,
      "step": 34825
    },
    {
      "entropy": 5.437600183486938,
      "epoch": 2.709900797510212,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004262793403325243,
      "loss": 5.0982,
      "mean_token_accuracy": 0.2004051223397255,
      "num_tokens": 60415617.0,
      "step": 34830
    },
    {
      "entropy": 5.417517614364624,
      "epoch": 2.7102898268819295,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042625882324741316,
      "loss": 5.0425,
      "mean_token_accuracy": 0.19766895473003387,
      "num_tokens": 60424453.0,
      "step": 34835
    },
    {
      "entropy": 5.374994611740112,
      "epoch": 2.7106788562536472,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004262383038671902,
      "loss": 4.9562,
      "mean_token_accuracy": 0.19841166585683823,
      "num_tokens": 60433358.0,
      "step": 34840
    },
    {
      "entropy": 5.341975212097168,
      "epoch": 2.711067885625365,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004262177821921666,
      "loss": 4.9821,
      "mean_token_accuracy": 0.19429868906736375,
      "num_tokens": 60441793.0,
      "step": 34845
    },
    {
      "entropy": 5.405882501602173,
      "epoch": 2.711456914997082,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042619725822265394,
      "loss": 4.9934,
      "mean_token_accuracy": 0.20677744299173356,
      "num_tokens": 60449981.0,
      "step": 34850
    },
    {
      "entropy": 5.389493465423584,
      "epoch": 2.7118459443688,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042617673195896345,
      "loss": 5.0528,
      "mean_token_accuracy": 0.19665377140045165,
      "num_tokens": 60459050.0,
      "step": 34855
    },
    {
      "entropy": 5.297450304031372,
      "epoch": 2.712234973740517,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042615620340140674,
      "loss": 4.8937,
      "mean_token_accuracy": 0.2078096389770508,
      "num_tokens": 60467300.0,
      "step": 34860
    },
    {
      "entropy": 5.301272630691528,
      "epoch": 2.712624003112235,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042613567255029525,
      "loss": 4.98,
      "mean_token_accuracy": 0.19621365368366242,
      "num_tokens": 60475969.0,
      "step": 34865
    },
    {
      "entropy": 5.374357223510742,
      "epoch": 2.7130130324839525,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042611513940594037,
      "loss": 4.9986,
      "mean_token_accuracy": 0.2007204368710518,
      "num_tokens": 60484886.0,
      "step": 34870
    },
    {
      "entropy": 5.4231274127960205,
      "epoch": 2.7134020618556702,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004260946039686539,
      "loss": 5.0504,
      "mean_token_accuracy": 0.1943580761551857,
      "num_tokens": 60494172.0,
      "step": 34875
    },
    {
      "entropy": 5.402447319030761,
      "epoch": 2.713791091227388,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042607406623874726,
      "loss": 5.0427,
      "mean_token_accuracy": 0.19055717885494233,
      "num_tokens": 60501722.0,
      "step": 34880
    },
    {
      "entropy": 5.349048852920532,
      "epoch": 2.714180120599105,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004260535262165322,
      "loss": 4.8908,
      "mean_token_accuracy": 0.20973397344350814,
      "num_tokens": 60510613.0,
      "step": 34885
    },
    {
      "entropy": 5.403182554244995,
      "epoch": 2.714569149970823,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042603298390232015,
      "loss": 5.0567,
      "mean_token_accuracy": 0.19642326682806016,
      "num_tokens": 60519868.0,
      "step": 34890
    },
    {
      "entropy": 5.404860782623291,
      "epoch": 2.71495817934254,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042601243929642305,
      "loss": 5.0245,
      "mean_token_accuracy": 0.19419406950473786,
      "num_tokens": 60528832.0,
      "step": 34895
    },
    {
      "entropy": 5.360406303405762,
      "epoch": 2.715347208714258,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042599189239915253,
      "loss": 4.9529,
      "mean_token_accuracy": 0.1954007089138031,
      "num_tokens": 60537870.0,
      "step": 34900
    },
    {
      "entropy": 5.398197650909424,
      "epoch": 2.7157362380859755,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004259713432108204,
      "loss": 4.9855,
      "mean_token_accuracy": 0.1972329393029213,
      "num_tokens": 60546157.0,
      "step": 34905
    },
    {
      "entropy": 5.331812191009521,
      "epoch": 2.7161252674576932,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004259507917317384,
      "loss": 4.9713,
      "mean_token_accuracy": 0.19816379994153976,
      "num_tokens": 60554411.0,
      "step": 34910
    },
    {
      "entropy": 5.458487510681152,
      "epoch": 2.7165142968294105,
      "grad_norm": 1.125,
      "learning_rate": 0.00042593023796221843,
      "loss": 5.0129,
      "mean_token_accuracy": 0.19824106991291046,
      "num_tokens": 60562748.0,
      "step": 34915
    },
    {
      "entropy": 5.4000410556793215,
      "epoch": 2.716903326201128,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004259096819025723,
      "loss": 5.0141,
      "mean_token_accuracy": 0.19799044877290725,
      "num_tokens": 60572004.0,
      "step": 34920
    },
    {
      "entropy": 5.431330728530884,
      "epoch": 2.717292355572846,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042588912355311196,
      "loss": 5.0043,
      "mean_token_accuracy": 0.20066782236099243,
      "num_tokens": 60580596.0,
      "step": 34925
    },
    {
      "entropy": 5.336384868621826,
      "epoch": 2.717681384944563,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042586856291414925,
      "loss": 4.9949,
      "mean_token_accuracy": 0.1967260479927063,
      "num_tokens": 60588734.0,
      "step": 34930
    },
    {
      "entropy": 5.405439376831055,
      "epoch": 2.718070414316281,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042584799998599633,
      "loss": 5.0027,
      "mean_token_accuracy": 0.19739606529474257,
      "num_tokens": 60597302.0,
      "step": 34935
    },
    {
      "entropy": 5.484164619445801,
      "epoch": 2.7184594436879985,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004258274347689651,
      "loss": 5.0481,
      "mean_token_accuracy": 0.1951994627714157,
      "num_tokens": 60606928.0,
      "step": 34940
    },
    {
      "entropy": 5.366205835342408,
      "epoch": 2.718848473059716,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042580686726336766,
      "loss": 5.0181,
      "mean_token_accuracy": 0.19663025289773942,
      "num_tokens": 60615970.0,
      "step": 34945
    },
    {
      "entropy": 5.3720861911773685,
      "epoch": 2.7192375024314335,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042578629746951597,
      "loss": 5.041,
      "mean_token_accuracy": 0.19405727833509445,
      "num_tokens": 60625453.0,
      "step": 34950
    },
    {
      "entropy": 5.456085968017578,
      "epoch": 2.719626531803151,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004257657253877222,
      "loss": 5.0198,
      "mean_token_accuracy": 0.19812843799591065,
      "num_tokens": 60634622.0,
      "step": 34955
    },
    {
      "entropy": 5.4255115509033205,
      "epoch": 2.720015561174869,
      "grad_norm": 1.125,
      "learning_rate": 0.00042574515101829856,
      "loss": 5.0046,
      "mean_token_accuracy": 0.1989461973309517,
      "num_tokens": 60643463.0,
      "step": 34960
    },
    {
      "entropy": 5.336925363540649,
      "epoch": 2.720404590546586,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042572457436155717,
      "loss": 4.9638,
      "mean_token_accuracy": 0.205049367249012,
      "num_tokens": 60652192.0,
      "step": 34965
    },
    {
      "entropy": 5.469555282592774,
      "epoch": 2.720793619918304,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042570399541781023,
      "loss": 5.0815,
      "mean_token_accuracy": 0.19384343028068543,
      "num_tokens": 60661071.0,
      "step": 34970
    },
    {
      "entropy": 5.526416969299317,
      "epoch": 2.7211826492900215,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042568341418737016,
      "loss": 5.1505,
      "mean_token_accuracy": 0.18243410289287568,
      "num_tokens": 60670165.0,
      "step": 34975
    },
    {
      "entropy": 5.370257902145386,
      "epoch": 2.721571678661739,
      "grad_norm": 1.15625,
      "learning_rate": 0.000425662830670549,
      "loss": 4.9527,
      "mean_token_accuracy": 0.1994159683585167,
      "num_tokens": 60678260.0,
      "step": 34980
    },
    {
      "entropy": 5.385682678222656,
      "epoch": 2.7219607080334565,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042564224486765925,
      "loss": 4.9335,
      "mean_token_accuracy": 0.19755038619041443,
      "num_tokens": 60686919.0,
      "step": 34985
    },
    {
      "entropy": 5.281902599334717,
      "epoch": 2.722349737405174,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004256216567790131,
      "loss": 4.9134,
      "mean_token_accuracy": 0.20358132123947142,
      "num_tokens": 60695549.0,
      "step": 34990
    },
    {
      "entropy": 5.311908292770386,
      "epoch": 2.7227387667768914,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004256010664049232,
      "loss": 4.9416,
      "mean_token_accuracy": 0.19887105673551558,
      "num_tokens": 60704258.0,
      "step": 34995
    },
    {
      "entropy": 5.413995981216431,
      "epoch": 2.723127796148609,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004255804737457017,
      "loss": 5.0102,
      "mean_token_accuracy": 0.1923614487051964,
      "num_tokens": 60712860.0,
      "step": 35000
    },
    {
      "entropy": 5.390444612503051,
      "epoch": 2.723516825520327,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042555987880166127,
      "loss": 4.9799,
      "mean_token_accuracy": 0.20344469994306563,
      "num_tokens": 60721861.0,
      "step": 35005
    },
    {
      "entropy": 5.386259603500366,
      "epoch": 2.7239058548920445,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004255392815731142,
      "loss": 5.0002,
      "mean_token_accuracy": 0.19885495454072952,
      "num_tokens": 60730760.0,
      "step": 35010
    },
    {
      "entropy": 5.402735900878906,
      "epoch": 2.7242948842637618,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042551868206037324,
      "loss": 5.0293,
      "mean_token_accuracy": 0.19492748826742173,
      "num_tokens": 60739192.0,
      "step": 35015
    },
    {
      "entropy": 5.363769102096557,
      "epoch": 2.7246839136354795,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042549808026375084,
      "loss": 4.9711,
      "mean_token_accuracy": 0.20069366842508315,
      "num_tokens": 60748009.0,
      "step": 35020
    },
    {
      "entropy": 5.370988368988037,
      "epoch": 2.725072943007197,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042547747618355966,
      "loss": 4.9991,
      "mean_token_accuracy": 0.19621071517467498,
      "num_tokens": 60756663.0,
      "step": 35025
    },
    {
      "entropy": 5.4907842636108395,
      "epoch": 2.7254619723789144,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004254568698201122,
      "loss": 5.1419,
      "mean_token_accuracy": 0.19140998870134354,
      "num_tokens": 60765055.0,
      "step": 35030
    },
    {
      "entropy": 5.451573896408081,
      "epoch": 2.725851001750632,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042543626117372133,
      "loss": 5.0592,
      "mean_token_accuracy": 0.1909284323453903,
      "num_tokens": 60773745.0,
      "step": 35035
    },
    {
      "entropy": 5.424426651000976,
      "epoch": 2.72624003112235,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042541565024469945,
      "loss": 5.0374,
      "mean_token_accuracy": 0.19808966666460037,
      "num_tokens": 60782244.0,
      "step": 35040
    },
    {
      "entropy": 5.324416589736939,
      "epoch": 2.7266290604940675,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042539503703335963,
      "loss": 4.8298,
      "mean_token_accuracy": 0.20676670521497725,
      "num_tokens": 60791156.0,
      "step": 35045
    },
    {
      "entropy": 5.353069019317627,
      "epoch": 2.7270180898657848,
      "grad_norm": 1.125,
      "learning_rate": 0.00042537442154001455,
      "loss": 4.8808,
      "mean_token_accuracy": 0.19783241003751756,
      "num_tokens": 60799498.0,
      "step": 35050
    },
    {
      "entropy": 5.388478136062622,
      "epoch": 2.7274071192375025,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042535380376497684,
      "loss": 5.115,
      "mean_token_accuracy": 0.18409560173749923,
      "num_tokens": 60807991.0,
      "step": 35055
    },
    {
      "entropy": 5.480797672271729,
      "epoch": 2.72779614860922,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004253331837085595,
      "loss": 4.9644,
      "mean_token_accuracy": 0.19217465519905091,
      "num_tokens": 60815674.0,
      "step": 35060
    },
    {
      "entropy": 5.3376611232757565,
      "epoch": 2.7281851779809374,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042531256137107545,
      "loss": 4.943,
      "mean_token_accuracy": 0.19427911341190338,
      "num_tokens": 60823490.0,
      "step": 35065
    },
    {
      "entropy": 5.384861087799072,
      "epoch": 2.728574207352655,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004252919367528375,
      "loss": 5.0255,
      "mean_token_accuracy": 0.19740456193685532,
      "num_tokens": 60832437.0,
      "step": 35070
    },
    {
      "entropy": 5.415588140487671,
      "epoch": 2.728963236724373,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042527130985415857,
      "loss": 5.0185,
      "mean_token_accuracy": 0.19756570011377333,
      "num_tokens": 60841450.0,
      "step": 35075
    },
    {
      "entropy": 5.419904136657715,
      "epoch": 2.7293522660960905,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004252506806753517,
      "loss": 4.9367,
      "mean_token_accuracy": 0.20415988564491272,
      "num_tokens": 60849848.0,
      "step": 35080
    },
    {
      "entropy": 5.267904949188233,
      "epoch": 2.7297412954678077,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042523004921673,
      "loss": 4.8579,
      "mean_token_accuracy": 0.19961386024951935,
      "num_tokens": 60858912.0,
      "step": 35085
    },
    {
      "entropy": 5.3423693656921385,
      "epoch": 2.7301303248395254,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004252094154786063,
      "loss": 4.9633,
      "mean_token_accuracy": 0.19486998915672302,
      "num_tokens": 60867204.0,
      "step": 35090
    },
    {
      "entropy": 5.416849279403687,
      "epoch": 2.7305193542112427,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004251887794612939,
      "loss": 5.0127,
      "mean_token_accuracy": 0.19130531549453736,
      "num_tokens": 60875921.0,
      "step": 35095
    },
    {
      "entropy": 5.3920722007751465,
      "epoch": 2.7309083835829604,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004251681411651057,
      "loss": 4.9638,
      "mean_token_accuracy": 0.19613010734319686,
      "num_tokens": 60884639.0,
      "step": 35100
    },
    {
      "entropy": 5.379074192047119,
      "epoch": 2.731297412954678,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042514750059035504,
      "loss": 5.1116,
      "mean_token_accuracy": 0.19532786011695863,
      "num_tokens": 60893517.0,
      "step": 35105
    },
    {
      "entropy": 5.3309181213378904,
      "epoch": 2.731686442326396,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004251268577373551,
      "loss": 4.9496,
      "mean_token_accuracy": 0.20360612571239473,
      "num_tokens": 60902515.0,
      "step": 35110
    },
    {
      "entropy": 5.401984405517578,
      "epoch": 2.732075471698113,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000425106212606419,
      "loss": 4.9937,
      "mean_token_accuracy": 0.1958647385239601,
      "num_tokens": 60911453.0,
      "step": 35115
    },
    {
      "entropy": 5.398989295959472,
      "epoch": 2.7324645010698307,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042508556519786004,
      "loss": 5.0038,
      "mean_token_accuracy": 0.19824443012475967,
      "num_tokens": 60920549.0,
      "step": 35120
    },
    {
      "entropy": 5.454034280776978,
      "epoch": 2.7328535304415484,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004250649155119915,
      "loss": 5.006,
      "mean_token_accuracy": 0.19725531637668609,
      "num_tokens": 60929271.0,
      "step": 35125
    },
    {
      "entropy": 5.420857620239258,
      "epoch": 2.7332425598132657,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042504426354912666,
      "loss": 4.9805,
      "mean_token_accuracy": 0.19455235302448273,
      "num_tokens": 60938228.0,
      "step": 35130
    },
    {
      "entropy": 5.423171186447144,
      "epoch": 2.7336315891849834,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042502360930957904,
      "loss": 5.0379,
      "mean_token_accuracy": 0.19683623313903809,
      "num_tokens": 60945777.0,
      "step": 35135
    },
    {
      "entropy": 5.351422882080078,
      "epoch": 2.734020618556701,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042500295279366196,
      "loss": 5.0418,
      "mean_token_accuracy": 0.19246007651090621,
      "num_tokens": 60954500.0,
      "step": 35140
    },
    {
      "entropy": 5.397111225128174,
      "epoch": 2.734409647928419,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042498229400168874,
      "loss": 5.0052,
      "mean_token_accuracy": 0.19866187870502472,
      "num_tokens": 60962918.0,
      "step": 35145
    },
    {
      "entropy": 5.377806663513184,
      "epoch": 2.734798677300136,
      "grad_norm": 1.125,
      "learning_rate": 0.000424961632933973,
      "loss": 4.9512,
      "mean_token_accuracy": 0.20498688668012618,
      "num_tokens": 60971544.0,
      "step": 35150
    },
    {
      "entropy": 5.377746152877807,
      "epoch": 2.7351877066718537,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004249409695908281,
      "loss": 4.9728,
      "mean_token_accuracy": 0.2047186017036438,
      "num_tokens": 60979637.0,
      "step": 35155
    },
    {
      "entropy": 5.329731082916259,
      "epoch": 2.7355767360435714,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004249203039725677,
      "loss": 4.9885,
      "mean_token_accuracy": 0.1947261095046997,
      "num_tokens": 60988417.0,
      "step": 35160
    },
    {
      "entropy": 5.45766863822937,
      "epoch": 2.7359657654152887,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042489963607950527,
      "loss": 5.0025,
      "mean_token_accuracy": 0.19932710528373718,
      "num_tokens": 60997282.0,
      "step": 35165
    },
    {
      "entropy": 5.402943849563599,
      "epoch": 2.7363547947870064,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042487896591195454,
      "loss": 4.9889,
      "mean_token_accuracy": 0.19214026629924774,
      "num_tokens": 61006191.0,
      "step": 35170
    },
    {
      "entropy": 5.3461854457855225,
      "epoch": 2.736743824158724,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042485829347022893,
      "loss": 4.9639,
      "mean_token_accuracy": 0.2022184759378433,
      "num_tokens": 61014780.0,
      "step": 35175
    },
    {
      "entropy": 5.424778795242309,
      "epoch": 2.737132853530442,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042483761875464226,
      "loss": 5.0155,
      "mean_token_accuracy": 0.19151908159255981,
      "num_tokens": 61024328.0,
      "step": 35180
    },
    {
      "entropy": 5.406956434249878,
      "epoch": 2.737521882902159,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042481694176550825,
      "loss": 4.9776,
      "mean_token_accuracy": 0.1990225702524185,
      "num_tokens": 61032716.0,
      "step": 35185
    },
    {
      "entropy": 5.44702525138855,
      "epoch": 2.7379109122738767,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042479626250314066,
      "loss": 5.0351,
      "mean_token_accuracy": 0.19466724395751953,
      "num_tokens": 61041495.0,
      "step": 35190
    },
    {
      "entropy": 5.323679780960083,
      "epoch": 2.738299941645594,
      "grad_norm": 1.125,
      "learning_rate": 0.00042477558096785316,
      "loss": 4.9859,
      "mean_token_accuracy": 0.2002753958106041,
      "num_tokens": 61050275.0,
      "step": 35195
    },
    {
      "entropy": 5.3710469722747805,
      "epoch": 2.7386889710173117,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042475489715995956,
      "loss": 4.8835,
      "mean_token_accuracy": 0.2082016348838806,
      "num_tokens": 61058774.0,
      "step": 35200
    },
    {
      "entropy": 5.420380067825318,
      "epoch": 2.7390780003890294,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042473421107977385,
      "loss": 4.9519,
      "mean_token_accuracy": 0.20523722171783448,
      "num_tokens": 61067475.0,
      "step": 35205
    },
    {
      "entropy": 5.289706897735596,
      "epoch": 2.739467029760747,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042471352272760974,
      "loss": 4.9244,
      "mean_token_accuracy": 0.20416066497564317,
      "num_tokens": 61075943.0,
      "step": 35210
    },
    {
      "entropy": 5.388561248779297,
      "epoch": 2.7398560591324648,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004246928321037813,
      "loss": 5.0067,
      "mean_token_accuracy": 0.19975289106369018,
      "num_tokens": 61084460.0,
      "step": 35215
    },
    {
      "entropy": 5.478441047668457,
      "epoch": 2.740245088504182,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004246721392086023,
      "loss": 5.0753,
      "mean_token_accuracy": 0.1918073445558548,
      "num_tokens": 61093423.0,
      "step": 35220
    },
    {
      "entropy": 5.39250545501709,
      "epoch": 2.7406341178758997,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042465144404238686,
      "loss": 5.0223,
      "mean_token_accuracy": 0.1852139040827751,
      "num_tokens": 61102953.0,
      "step": 35225
    },
    {
      "entropy": 5.398007869720459,
      "epoch": 2.741023147247617,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042463074660544897,
      "loss": 4.9154,
      "mean_token_accuracy": 0.20541236847639083,
      "num_tokens": 61112114.0,
      "step": 35230
    },
    {
      "entropy": 5.428777885437012,
      "epoch": 2.7414121766193347,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004246100468981026,
      "loss": 4.9947,
      "mean_token_accuracy": 0.20100147128105164,
      "num_tokens": 61120603.0,
      "step": 35235
    },
    {
      "entropy": 5.370364999771118,
      "epoch": 2.7418012059910524,
      "grad_norm": 1.203125,
      "learning_rate": 0.000424589344920662,
      "loss": 4.9721,
      "mean_token_accuracy": 0.1999620243906975,
      "num_tokens": 61129238.0,
      "step": 35240
    },
    {
      "entropy": 5.348665809631347,
      "epoch": 2.74219023536277,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004245686406734411,
      "loss": 4.8992,
      "mean_token_accuracy": 0.20454926192760467,
      "num_tokens": 61138082.0,
      "step": 35245
    },
    {
      "entropy": 5.4401849746704105,
      "epoch": 2.7425792647344873,
      "grad_norm": 1.125,
      "learning_rate": 0.0004245479341567541,
      "loss": 5.0123,
      "mean_token_accuracy": 0.1971248522400856,
      "num_tokens": 61146779.0,
      "step": 35250
    },
    {
      "entropy": 5.2980536937713625,
      "epoch": 2.742968294106205,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004245272253709154,
      "loss": 4.8964,
      "mean_token_accuracy": 0.20928891152143478,
      "num_tokens": 61155492.0,
      "step": 35255
    },
    {
      "entropy": 5.387368679046631,
      "epoch": 2.7433573234779227,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004245065143162389,
      "loss": 4.8747,
      "mean_token_accuracy": 0.20668141096830367,
      "num_tokens": 61164235.0,
      "step": 35260
    },
    {
      "entropy": 5.380212450027466,
      "epoch": 2.74374635284964,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042448580099303904,
      "loss": 4.9548,
      "mean_token_accuracy": 0.20073190629482268,
      "num_tokens": 61172203.0,
      "step": 35265
    },
    {
      "entropy": 5.355871725082397,
      "epoch": 2.7441353822213577,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042446508540163017,
      "loss": 4.9588,
      "mean_token_accuracy": 0.1987817496061325,
      "num_tokens": 61180827.0,
      "step": 35270
    },
    {
      "entropy": 5.39039454460144,
      "epoch": 2.7445244115930754,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004244443675423264,
      "loss": 5.0118,
      "mean_token_accuracy": 0.19967636317014695,
      "num_tokens": 61189899.0,
      "step": 35275
    },
    {
      "entropy": 5.437172698974609,
      "epoch": 2.744913440964793,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042442364741544227,
      "loss": 5.0225,
      "mean_token_accuracy": 0.19061192870140076,
      "num_tokens": 61197981.0,
      "step": 35280
    },
    {
      "entropy": 5.426348304748535,
      "epoch": 2.7453024703365103,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042440292502129214,
      "loss": 5.0096,
      "mean_token_accuracy": 0.2003627225756645,
      "num_tokens": 61206085.0,
      "step": 35285
    },
    {
      "entropy": 5.368020677566529,
      "epoch": 2.745691499708228,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004243822003601904,
      "loss": 4.9733,
      "mean_token_accuracy": 0.19767421185970308,
      "num_tokens": 61215124.0,
      "step": 35290
    },
    {
      "entropy": 5.463925075531006,
      "epoch": 2.7460805290799453,
      "grad_norm": 1.125,
      "learning_rate": 0.0004243614734324515,
      "loss": 5.1337,
      "mean_token_accuracy": 0.19371851682662963,
      "num_tokens": 61223926.0,
      "step": 35295
    },
    {
      "entropy": 5.372512006759644,
      "epoch": 2.746469558451663,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004243407442383901,
      "loss": 4.9673,
      "mean_token_accuracy": 0.2006288006901741,
      "num_tokens": 61232526.0,
      "step": 35300
    },
    {
      "entropy": 5.296121549606323,
      "epoch": 2.7468585878233807,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042432001277832045,
      "loss": 4.8986,
      "mean_token_accuracy": 0.20792423337697982,
      "num_tokens": 61241591.0,
      "step": 35305
    },
    {
      "entropy": 5.385372543334961,
      "epoch": 2.7472476171950984,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004242992790525573,
      "loss": 4.9722,
      "mean_token_accuracy": 0.202564936876297,
      "num_tokens": 61250851.0,
      "step": 35310
    },
    {
      "entropy": 5.365449476242065,
      "epoch": 2.747636646566816,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004242785430614154,
      "loss": 4.8962,
      "mean_token_accuracy": 0.20249408781528472,
      "num_tokens": 61259600.0,
      "step": 35315
    },
    {
      "entropy": 5.337045288085937,
      "epoch": 2.7480256759385333,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042425780480520906,
      "loss": 5.0056,
      "mean_token_accuracy": 0.19437334090471267,
      "num_tokens": 61268458.0,
      "step": 35320
    },
    {
      "entropy": 5.330884122848511,
      "epoch": 2.748414705310251,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042423706428425305,
      "loss": 4.9466,
      "mean_token_accuracy": 0.20158275365829467,
      "num_tokens": 61276974.0,
      "step": 35325
    },
    {
      "entropy": 5.489245748519897,
      "epoch": 2.7488037346819683,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004242163214988622,
      "loss": 5.0897,
      "mean_token_accuracy": 0.19465400725603105,
      "num_tokens": 61285165.0,
      "step": 35330
    },
    {
      "entropy": 5.415948009490966,
      "epoch": 2.749192764053686,
      "grad_norm": 1.125,
      "learning_rate": 0.0004241955764493512,
      "loss": 4.9966,
      "mean_token_accuracy": 0.20466747730970383,
      "num_tokens": 61294061.0,
      "step": 35335
    },
    {
      "entropy": 5.366667699813843,
      "epoch": 2.7495817934254037,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042417482913603477,
      "loss": 4.9195,
      "mean_token_accuracy": 0.2015632137656212,
      "num_tokens": 61302435.0,
      "step": 35340
    },
    {
      "entropy": 5.3102428913116455,
      "epoch": 2.7499708227971214,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042415407955922774,
      "loss": 4.9752,
      "mean_token_accuracy": 0.20758775472640992,
      "num_tokens": 61310462.0,
      "step": 35345
    },
    {
      "entropy": 5.460335683822632,
      "epoch": 2.7503598521688386,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042413332771924496,
      "loss": 5.1222,
      "mean_token_accuracy": 0.17978277653455735,
      "num_tokens": 61319490.0,
      "step": 35350
    },
    {
      "entropy": 5.517118740081787,
      "epoch": 2.7507488815405563,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004241125736164013,
      "loss": 5.0024,
      "mean_token_accuracy": 0.1991927981376648,
      "num_tokens": 61328070.0,
      "step": 35355
    },
    {
      "entropy": 5.364162302017212,
      "epoch": 2.751137910912274,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004240918172510117,
      "loss": 4.8925,
      "mean_token_accuracy": 0.204372838139534,
      "num_tokens": 61336319.0,
      "step": 35360
    },
    {
      "entropy": 5.322314119338989,
      "epoch": 2.7515269402839913,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000424071058623391,
      "loss": 5.0552,
      "mean_token_accuracy": 0.19193875342607497,
      "num_tokens": 61345564.0,
      "step": 35365
    },
    {
      "entropy": 5.393060350418091,
      "epoch": 2.751915969655709,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004240502977338543,
      "loss": 5.0264,
      "mean_token_accuracy": 0.19632163792848586,
      "num_tokens": 61354823.0,
      "step": 35370
    },
    {
      "entropy": 5.444690084457397,
      "epoch": 2.7523049990274266,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004240295345827166,
      "loss": 5.0567,
      "mean_token_accuracy": 0.19158421456813812,
      "num_tokens": 61363441.0,
      "step": 35375
    },
    {
      "entropy": 5.33777117729187,
      "epoch": 2.7526940283991443,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004240087691702929,
      "loss": 4.9364,
      "mean_token_accuracy": 0.19296832978725434,
      "num_tokens": 61371795.0,
      "step": 35380
    },
    {
      "entropy": 5.385474061965942,
      "epoch": 2.7530830577708616,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004239880014968983,
      "loss": 4.9737,
      "mean_token_accuracy": 0.19952446967363358,
      "num_tokens": 61379698.0,
      "step": 35385
    },
    {
      "entropy": 5.3688901424407955,
      "epoch": 2.7534720871425793,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042396723156284786,
      "loss": 4.9396,
      "mean_token_accuracy": 0.2054091900587082,
      "num_tokens": 61388693.0,
      "step": 35390
    },
    {
      "entropy": 5.4397115230560305,
      "epoch": 2.753861116514297,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004239464593684568,
      "loss": 5.0264,
      "mean_token_accuracy": 0.1969677522778511,
      "num_tokens": 61397995.0,
      "step": 35395
    },
    {
      "entropy": 5.486526250839233,
      "epoch": 2.7542501458860142,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042392568491404043,
      "loss": 5.0411,
      "mean_token_accuracy": 0.19704239070415497,
      "num_tokens": 61405567.0,
      "step": 35400
    },
    {
      "entropy": 5.360042333602905,
      "epoch": 2.754639175257732,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042390490819991367,
      "loss": 4.955,
      "mean_token_accuracy": 0.20568713843822478,
      "num_tokens": 61413458.0,
      "step": 35405
    },
    {
      "entropy": 5.410577487945557,
      "epoch": 2.7550282046294496,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000423884129226392,
      "loss": 4.9691,
      "mean_token_accuracy": 0.20628975033760072,
      "num_tokens": 61422026.0,
      "step": 35410
    },
    {
      "entropy": 5.449905586242676,
      "epoch": 2.7554172340011673,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004238633479937906,
      "loss": 4.9932,
      "mean_token_accuracy": 0.20050131231546403,
      "num_tokens": 61430843.0,
      "step": 35415
    },
    {
      "entropy": 5.346636056900024,
      "epoch": 2.7558062633728846,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004238425645024249,
      "loss": 4.9011,
      "mean_token_accuracy": 0.20080612897872924,
      "num_tokens": 61438340.0,
      "step": 35420
    },
    {
      "entropy": 5.333720779418945,
      "epoch": 2.7561952927446023,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004238217787526102,
      "loss": 4.9632,
      "mean_token_accuracy": 0.20215589553117752,
      "num_tokens": 61447302.0,
      "step": 35425
    },
    {
      "entropy": 5.294369077682495,
      "epoch": 2.7565843221163195,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004238009907446618,
      "loss": 4.896,
      "mean_token_accuracy": 0.20605918616056443,
      "num_tokens": 61456688.0,
      "step": 35430
    },
    {
      "entropy": 5.462442255020141,
      "epoch": 2.7569733514880372,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042378020047889527,
      "loss": 5.0037,
      "mean_token_accuracy": 0.19084039628505706,
      "num_tokens": 61465718.0,
      "step": 35435
    },
    {
      "entropy": 5.498924493789673,
      "epoch": 2.757362380859755,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042375940795562594,
      "loss": 5.0991,
      "mean_token_accuracy": 0.1949687719345093,
      "num_tokens": 61474113.0,
      "step": 35440
    },
    {
      "entropy": 5.311561727523804,
      "epoch": 2.7577514102314726,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042373861317516944,
      "loss": 4.9229,
      "mean_token_accuracy": 0.20258758068084717,
      "num_tokens": 61482876.0,
      "step": 35445
    },
    {
      "entropy": 5.36276478767395,
      "epoch": 2.75814043960319,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004237178161378412,
      "loss": 4.8787,
      "mean_token_accuracy": 0.2029843121767044,
      "num_tokens": 61491229.0,
      "step": 35450
    },
    {
      "entropy": 5.337359762191772,
      "epoch": 2.7585294689749076,
      "grad_norm": 1.125,
      "learning_rate": 0.0004236970168439568,
      "loss": 4.9238,
      "mean_token_accuracy": 0.19890360832214354,
      "num_tokens": 61499935.0,
      "step": 35455
    },
    {
      "entropy": 5.473676252365112,
      "epoch": 2.7589184983466253,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042367621529383186,
      "loss": 4.9929,
      "mean_token_accuracy": 0.198628668487072,
      "num_tokens": 61508249.0,
      "step": 35460
    },
    {
      "entropy": 5.3854443550109865,
      "epoch": 2.7593075277183425,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042365541148778205,
      "loss": 4.8932,
      "mean_token_accuracy": 0.20191843211650848,
      "num_tokens": 61516530.0,
      "step": 35465
    },
    {
      "entropy": 5.411546516418457,
      "epoch": 2.7596965570900602,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042363460542612285,
      "loss": 5.0619,
      "mean_token_accuracy": 0.19656858295202256,
      "num_tokens": 61524998.0,
      "step": 35470
    },
    {
      "entropy": 5.223538255691528,
      "epoch": 2.760085586461778,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004236137971091702,
      "loss": 4.8353,
      "mean_token_accuracy": 0.21118010729551315,
      "num_tokens": 61533661.0,
      "step": 35475
    },
    {
      "entropy": 5.367352294921875,
      "epoch": 2.7604746158334956,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042359298653723963,
      "loss": 4.9603,
      "mean_token_accuracy": 0.20268310308456422,
      "num_tokens": 61542387.0,
      "step": 35480
    },
    {
      "entropy": 5.424409294128418,
      "epoch": 2.760863645205213,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004235721737106471,
      "loss": 4.9282,
      "mean_token_accuracy": 0.20485030710697175,
      "num_tokens": 61550626.0,
      "step": 35485
    },
    {
      "entropy": 5.433951568603516,
      "epoch": 2.7612526745769306,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004235513586297082,
      "loss": 5.0734,
      "mean_token_accuracy": 0.20028118640184403,
      "num_tokens": 61559170.0,
      "step": 35490
    },
    {
      "entropy": 5.360174083709717,
      "epoch": 2.7616417039486483,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042353054129473894,
      "loss": 5.0718,
      "mean_token_accuracy": 0.19668426513671874,
      "num_tokens": 61567484.0,
      "step": 35495
    },
    {
      "entropy": 5.360730981826782,
      "epoch": 2.7620307333203655,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004235097217060552,
      "loss": 4.893,
      "mean_token_accuracy": 0.2114207550883293,
      "num_tokens": 61575542.0,
      "step": 35500
    },
    {
      "entropy": 5.393539476394653,
      "epoch": 2.7624197626920832,
      "grad_norm": 1.125,
      "learning_rate": 0.0004234888998639727,
      "loss": 5.0084,
      "mean_token_accuracy": 0.19615163058042526,
      "num_tokens": 61584273.0,
      "step": 35505
    },
    {
      "entropy": 5.341640758514404,
      "epoch": 2.762808792063801,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004234680757688076,
      "loss": 4.9751,
      "mean_token_accuracy": 0.2040863111615181,
      "num_tokens": 61593514.0,
      "step": 35510
    },
    {
      "entropy": 5.4047425270080565,
      "epoch": 2.7631978214355186,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042344724942087564,
      "loss": 5.0166,
      "mean_token_accuracy": 0.2002140074968338,
      "num_tokens": 61601146.0,
      "step": 35515
    },
    {
      "entropy": 5.422760438919068,
      "epoch": 2.763586850807236,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000423426420820493,
      "loss": 5.0686,
      "mean_token_accuracy": 0.19631607681512833,
      "num_tokens": 61609521.0,
      "step": 35520
    },
    {
      "entropy": 5.421096515655518,
      "epoch": 2.7639758801789536,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042340558996797564,
      "loss": 5.0332,
      "mean_token_accuracy": 0.1936247318983078,
      "num_tokens": 61618382.0,
      "step": 35525
    },
    {
      "entropy": 5.354607725143433,
      "epoch": 2.764364909550671,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042338475686363964,
      "loss": 4.9225,
      "mean_token_accuracy": 0.20196730196475982,
      "num_tokens": 61627098.0,
      "step": 35530
    },
    {
      "entropy": 5.312275171279907,
      "epoch": 2.7647539389223885,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004233639215078012,
      "loss": 4.9691,
      "mean_token_accuracy": 0.20299019813537597,
      "num_tokens": 61635747.0,
      "step": 35535
    },
    {
      "entropy": 5.393980216979981,
      "epoch": 2.765142968294106,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042334308390077637,
      "loss": 4.9938,
      "mean_token_accuracy": 0.19790416657924653,
      "num_tokens": 61645151.0,
      "step": 35540
    },
    {
      "entropy": 5.382141494750977,
      "epoch": 2.765531997665824,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004233222440428814,
      "loss": 5.0151,
      "mean_token_accuracy": 0.20035903602838517,
      "num_tokens": 61653216.0,
      "step": 35545
    },
    {
      "entropy": 5.396370124816895,
      "epoch": 2.765921027037541,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004233014019344324,
      "loss": 5.0053,
      "mean_token_accuracy": 0.2015206351876259,
      "num_tokens": 61661583.0,
      "step": 35550
    },
    {
      "entropy": 5.387295484542847,
      "epoch": 2.766310056409259,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004232805575757457,
      "loss": 4.9855,
      "mean_token_accuracy": 0.19570852667093278,
      "num_tokens": 61670465.0,
      "step": 35555
    },
    {
      "entropy": 5.465928554534912,
      "epoch": 2.7666990857809766,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004232597109671375,
      "loss": 4.9669,
      "mean_token_accuracy": 0.2031568169593811,
      "num_tokens": 61678670.0,
      "step": 35560
    },
    {
      "entropy": 5.450640058517456,
      "epoch": 2.767088115152694,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004232388621089242,
      "loss": 5.0399,
      "mean_token_accuracy": 0.1926266849040985,
      "num_tokens": 61687860.0,
      "step": 35565
    },
    {
      "entropy": 5.341442584991455,
      "epoch": 2.7674771445244115,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042321801100142216,
      "loss": 5.0045,
      "mean_token_accuracy": 0.20018403977155685,
      "num_tokens": 61696395.0,
      "step": 35570
    },
    {
      "entropy": 5.464934682846069,
      "epoch": 2.767866173896129,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042319715764494765,
      "loss": 5.0615,
      "mean_token_accuracy": 0.19958362132310867,
      "num_tokens": 61705029.0,
      "step": 35575
    },
    {
      "entropy": 5.416219472885132,
      "epoch": 2.768255203267847,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042317630203981714,
      "loss": 4.9695,
      "mean_token_accuracy": 0.20319342911243438,
      "num_tokens": 61713160.0,
      "step": 35580
    },
    {
      "entropy": 5.370386028289795,
      "epoch": 2.768644232639564,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042315544418634714,
      "loss": 4.9406,
      "mean_token_accuracy": 0.20982572734355925,
      "num_tokens": 61722094.0,
      "step": 35585
    },
    {
      "entropy": 5.457256174087524,
      "epoch": 2.769033262011282,
      "grad_norm": 1.125,
      "learning_rate": 0.00042313458408485414,
      "loss": 5.089,
      "mean_token_accuracy": 0.19261547774076462,
      "num_tokens": 61730771.0,
      "step": 35590
    },
    {
      "entropy": 5.383885335922241,
      "epoch": 2.7694222913829996,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042311372173565456,
      "loss": 4.9981,
      "mean_token_accuracy": 0.19807071685791017,
      "num_tokens": 61740059.0,
      "step": 35595
    },
    {
      "entropy": 5.4095909118652346,
      "epoch": 2.769811320754717,
      "grad_norm": 1.125,
      "learning_rate": 0.000423092857139065,
      "loss": 4.9568,
      "mean_token_accuracy": 0.2043505623936653,
      "num_tokens": 61748466.0,
      "step": 35600
    },
    {
      "entropy": 5.330094909667968,
      "epoch": 2.7702003501264345,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000423071990295402,
      "loss": 4.9007,
      "mean_token_accuracy": 0.20384887605905533,
      "num_tokens": 61756862.0,
      "step": 35605
    },
    {
      "entropy": 5.392449808120728,
      "epoch": 2.770589379498152,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004230511212049822,
      "loss": 4.9482,
      "mean_token_accuracy": 0.2052776634693146,
      "num_tokens": 61765867.0,
      "step": 35610
    },
    {
      "entropy": 5.427014446258545,
      "epoch": 2.77097840886987,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004230302498681224,
      "loss": 4.9788,
      "mean_token_accuracy": 0.19852515459060668,
      "num_tokens": 61774297.0,
      "step": 35615
    },
    {
      "entropy": 5.32357268333435,
      "epoch": 2.771367438241587,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004230093762851392,
      "loss": 4.9203,
      "mean_token_accuracy": 0.20734080374240876,
      "num_tokens": 61782841.0,
      "step": 35620
    },
    {
      "entropy": 5.391433906555176,
      "epoch": 2.771756467613305,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004229885004563491,
      "loss": 5.0967,
      "mean_token_accuracy": 0.19104552417993545,
      "num_tokens": 61791719.0,
      "step": 35625
    },
    {
      "entropy": 5.441840314865113,
      "epoch": 2.772145496985022,
      "grad_norm": 1.125,
      "learning_rate": 0.0004229676223820692,
      "loss": 5.0081,
      "mean_token_accuracy": 0.19271782636642457,
      "num_tokens": 61800267.0,
      "step": 35630
    },
    {
      "entropy": 5.403231430053711,
      "epoch": 2.77253452635674,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042294674206261616,
      "loss": 4.9906,
      "mean_token_accuracy": 0.2009415417909622,
      "num_tokens": 61808482.0,
      "step": 35635
    },
    {
      "entropy": 5.290844535827636,
      "epoch": 2.7729235557284575,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004229258594983067,
      "loss": 4.8583,
      "mean_token_accuracy": 0.2154325157403946,
      "num_tokens": 61816950.0,
      "step": 35640
    },
    {
      "entropy": 5.373343515396118,
      "epoch": 2.773312585100175,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004229049746894578,
      "loss": 4.9954,
      "mean_token_accuracy": 0.19665665030479432,
      "num_tokens": 61826011.0,
      "step": 35645
    },
    {
      "entropy": 5.455793619155884,
      "epoch": 2.773701614471893,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004228840876363864,
      "loss": 5.0253,
      "mean_token_accuracy": 0.1907666116952896,
      "num_tokens": 61834705.0,
      "step": 35650
    },
    {
      "entropy": 5.486746692657471,
      "epoch": 2.77409064384361,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004228631983394093,
      "loss": 4.9861,
      "mean_token_accuracy": 0.1986302465200424,
      "num_tokens": 61844166.0,
      "step": 35655
    },
    {
      "entropy": 5.396489095687866,
      "epoch": 2.774479673215328,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004228423067988434,
      "loss": 5.0007,
      "mean_token_accuracy": 0.2049995705485344,
      "num_tokens": 61852141.0,
      "step": 35660
    },
    {
      "entropy": 5.278471422195435,
      "epoch": 2.774868702587045,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042282141301500597,
      "loss": 4.9774,
      "mean_token_accuracy": 0.20007776468992233,
      "num_tokens": 61861566.0,
      "step": 35665
    },
    {
      "entropy": 5.399405479431152,
      "epoch": 2.775257731958763,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004228005169882138,
      "loss": 4.9639,
      "mean_token_accuracy": 0.2048620566725731,
      "num_tokens": 61869878.0,
      "step": 35670
    },
    {
      "entropy": 5.367794466018677,
      "epoch": 2.7756467613304805,
      "grad_norm": 1.125,
      "learning_rate": 0.00042277961871878413,
      "loss": 5.0251,
      "mean_token_accuracy": 0.18964139074087144,
      "num_tokens": 61879220.0,
      "step": 35675
    },
    {
      "entropy": 5.433588027954102,
      "epoch": 2.776035790702198,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004227587182070338,
      "loss": 4.9522,
      "mean_token_accuracy": 0.20772101283073424,
      "num_tokens": 61887534.0,
      "step": 35680
    },
    {
      "entropy": 5.42021107673645,
      "epoch": 2.7764248200739154,
      "grad_norm": 1.125,
      "learning_rate": 0.0004227378154532802,
      "loss": 5.1405,
      "mean_token_accuracy": 0.18695973008871078,
      "num_tokens": 61896420.0,
      "step": 35685
    },
    {
      "entropy": 5.4332122802734375,
      "epoch": 2.776813849445633,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004227169104578404,
      "loss": 5.076,
      "mean_token_accuracy": 0.19625899195671082,
      "num_tokens": 61905024.0,
      "step": 35690
    },
    {
      "entropy": 5.320305490493775,
      "epoch": 2.777202878817351,
      "grad_norm": 1.125,
      "learning_rate": 0.00042269600322103164,
      "loss": 4.9511,
      "mean_token_accuracy": 0.20457398146390915,
      "num_tokens": 61913272.0,
      "step": 35695
    },
    {
      "entropy": 5.38574538230896,
      "epoch": 2.777591908189068,
      "grad_norm": 1.09375,
      "learning_rate": 0.000422675093743171,
      "loss": 4.9458,
      "mean_token_accuracy": 0.20305030196905136,
      "num_tokens": 61922142.0,
      "step": 35700
    },
    {
      "entropy": 5.393837881088257,
      "epoch": 2.777980937560786,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042265418202457603,
      "loss": 4.9685,
      "mean_token_accuracy": 0.2021106079220772,
      "num_tokens": 61930140.0,
      "step": 35705
    },
    {
      "entropy": 5.392801141738891,
      "epoch": 2.7783699669325035,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004226332680655637,
      "loss": 5.0434,
      "mean_token_accuracy": 0.1945040225982666,
      "num_tokens": 61938993.0,
      "step": 35710
    },
    {
      "entropy": 5.361865663528443,
      "epoch": 2.778758996304221,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004226123518664516,
      "loss": 5.0282,
      "mean_token_accuracy": 0.19247539341449738,
      "num_tokens": 61948063.0,
      "step": 35715
    },
    {
      "entropy": 5.383347892761231,
      "epoch": 2.7791480256759384,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004225914334275569,
      "loss": 4.9489,
      "mean_token_accuracy": 0.19550112038850784,
      "num_tokens": 61956610.0,
      "step": 35720
    },
    {
      "entropy": 5.319733667373657,
      "epoch": 2.779537055047656,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004225705127491972,
      "loss": 4.8179,
      "mean_token_accuracy": 0.20196427702903746,
      "num_tokens": 61965359.0,
      "step": 35725
    },
    {
      "entropy": 5.34215202331543,
      "epoch": 2.7799260844193734,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004225495898316899,
      "loss": 4.9901,
      "mean_token_accuracy": 0.19938546568155288,
      "num_tokens": 61973523.0,
      "step": 35730
    },
    {
      "entropy": 5.322200965881348,
      "epoch": 2.780315113791091,
      "grad_norm": 1.125,
      "learning_rate": 0.00042252866467535233,
      "loss": 4.9192,
      "mean_token_accuracy": 0.2075161561369896,
      "num_tokens": 61981988.0,
      "step": 35735
    },
    {
      "entropy": 5.392977476119995,
      "epoch": 2.780704143162809,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042250773728050214,
      "loss": 4.9819,
      "mean_token_accuracy": 0.19958182126283647,
      "num_tokens": 61991400.0,
      "step": 35740
    },
    {
      "entropy": 5.3875421524047855,
      "epoch": 2.7810931725345265,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004224868076474568,
      "loss": 5.0229,
      "mean_token_accuracy": 0.20196807235479355,
      "num_tokens": 61999287.0,
      "step": 35745
    },
    {
      "entropy": 5.3896137237548825,
      "epoch": 2.781482201906244,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042246587577653394,
      "loss": 4.9967,
      "mean_token_accuracy": 0.19513302594423293,
      "num_tokens": 62008578.0,
      "step": 35750
    },
    {
      "entropy": 5.370587301254273,
      "epoch": 2.7818712312779614,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004224449416680511,
      "loss": 4.9229,
      "mean_token_accuracy": 0.2039752945303917,
      "num_tokens": 62017222.0,
      "step": 35755
    },
    {
      "entropy": 5.281582546234131,
      "epoch": 2.782260260649679,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042242400532232593,
      "loss": 4.8744,
      "mean_token_accuracy": 0.2103859394788742,
      "num_tokens": 62026224.0,
      "step": 35760
    },
    {
      "entropy": 5.4089947700500485,
      "epoch": 2.7826492900213964,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042240306673967614,
      "loss": 5.1603,
      "mean_token_accuracy": 0.19131201803684234,
      "num_tokens": 62036201.0,
      "step": 35765
    },
    {
      "entropy": 5.333034467697144,
      "epoch": 2.783038319393114,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042238212592041945,
      "loss": 4.9071,
      "mean_token_accuracy": 0.20250097066164016,
      "num_tokens": 62044780.0,
      "step": 35770
    },
    {
      "entropy": 5.357460880279541,
      "epoch": 2.7834273487648318,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042236118286487364,
      "loss": 4.9208,
      "mean_token_accuracy": 0.2043831929564476,
      "num_tokens": 62052720.0,
      "step": 35775
    },
    {
      "entropy": 5.3845460414886475,
      "epoch": 2.7838163781365495,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042234023757335633,
      "loss": 4.9644,
      "mean_token_accuracy": 0.19399232119321824,
      "num_tokens": 62061797.0,
      "step": 35780
    },
    {
      "entropy": 5.289451837539673,
      "epoch": 2.7842054075082667,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042231929004618547,
      "loss": 4.9158,
      "mean_token_accuracy": 0.20419395267963408,
      "num_tokens": 62070543.0,
      "step": 35785
    },
    {
      "entropy": 5.473853540420532,
      "epoch": 2.7845944368799844,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004222983402836789,
      "loss": 5.0829,
      "mean_token_accuracy": 0.19314807802438735,
      "num_tokens": 62079531.0,
      "step": 35790
    },
    {
      "entropy": 5.34035472869873,
      "epoch": 2.784983466251702,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042227738828615435,
      "loss": 4.9266,
      "mean_token_accuracy": 0.20471401810646056,
      "num_tokens": 62088429.0,
      "step": 35795
    },
    {
      "entropy": 5.281514358520508,
      "epoch": 2.7853724956234194,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042225643405392993,
      "loss": 4.9258,
      "mean_token_accuracy": 0.20528976768255233,
      "num_tokens": 62096926.0,
      "step": 35800
    },
    {
      "entropy": 5.383842277526855,
      "epoch": 2.785761524995137,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042223547758732356,
      "loss": 5.1224,
      "mean_token_accuracy": 0.18358878940343856,
      "num_tokens": 62106673.0,
      "step": 35805
    },
    {
      "entropy": 5.421551465988159,
      "epoch": 2.7861505543668548,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042221451888665307,
      "loss": 4.9458,
      "mean_token_accuracy": 0.19915224313735963,
      "num_tokens": 62115091.0,
      "step": 35810
    },
    {
      "entropy": 5.440556240081787,
      "epoch": 2.7865395837385725,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042219355795223657,
      "loss": 5.0159,
      "mean_token_accuracy": 0.19717124104499817,
      "num_tokens": 62124062.0,
      "step": 35815
    },
    {
      "entropy": 5.411426973342896,
      "epoch": 2.7869286131102897,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042217259478439216,
      "loss": 4.9442,
      "mean_token_accuracy": 0.2024976819753647,
      "num_tokens": 62133141.0,
      "step": 35820
    },
    {
      "entropy": 5.384665298461914,
      "epoch": 2.7873176424820074,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004221516293834379,
      "loss": 4.9771,
      "mean_token_accuracy": 0.20215948224067687,
      "num_tokens": 62142056.0,
      "step": 35825
    },
    {
      "entropy": 5.42564287185669,
      "epoch": 2.787706671853725,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042213066174969177,
      "loss": 4.9673,
      "mean_token_accuracy": 0.2007246881723404,
      "num_tokens": 62150064.0,
      "step": 35830
    },
    {
      "entropy": 5.307141351699829,
      "epoch": 2.7880957012254424,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042210969188347217,
      "loss": 4.9073,
      "mean_token_accuracy": 0.2046535938978195,
      "num_tokens": 62159300.0,
      "step": 35835
    },
    {
      "entropy": 5.322756385803222,
      "epoch": 2.78848473059716,
      "grad_norm": 1.125,
      "learning_rate": 0.0004220887197850971,
      "loss": 4.8874,
      "mean_token_accuracy": 0.20173195153474807,
      "num_tokens": 62167774.0,
      "step": 35840
    },
    {
      "entropy": 5.3925244331359865,
      "epoch": 2.7888737599688778,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042206774545488476,
      "loss": 4.9581,
      "mean_token_accuracy": 0.20503920316696167,
      "num_tokens": 62176301.0,
      "step": 35845
    },
    {
      "entropy": 5.335413408279419,
      "epoch": 2.7892627893405955,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004220467688931535,
      "loss": 4.9492,
      "mean_token_accuracy": 0.19678831696510315,
      "num_tokens": 62184804.0,
      "step": 35850
    },
    {
      "entropy": 5.371379423141479,
      "epoch": 2.7896518187123127,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004220257901002216,
      "loss": 4.9477,
      "mean_token_accuracy": 0.19621988236904145,
      "num_tokens": 62193864.0,
      "step": 35855
    },
    {
      "entropy": 5.379216814041138,
      "epoch": 2.7900408480840304,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004220048090764073,
      "loss": 4.9578,
      "mean_token_accuracy": 0.20145267695188523,
      "num_tokens": 62202906.0,
      "step": 35860
    },
    {
      "entropy": 5.28515853881836,
      "epoch": 2.7904298774557477,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042198382582202905,
      "loss": 4.8094,
      "mean_token_accuracy": 0.21349625438451766,
      "num_tokens": 62211292.0,
      "step": 35865
    },
    {
      "entropy": 5.362348461151123,
      "epoch": 2.7908189068274654,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004219628403374052,
      "loss": 5.0862,
      "mean_token_accuracy": 0.19391706436872483,
      "num_tokens": 62220830.0,
      "step": 35870
    },
    {
      "entropy": 5.383690261840821,
      "epoch": 2.791207936199183,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004219418526228541,
      "loss": 4.9777,
      "mean_token_accuracy": 0.19670870900154114,
      "num_tokens": 62228659.0,
      "step": 35875
    },
    {
      "entropy": 5.431873559951782,
      "epoch": 2.7915969655709008,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004219208626786943,
      "loss": 5.0718,
      "mean_token_accuracy": 0.18918053209781646,
      "num_tokens": 62238098.0,
      "step": 35880
    },
    {
      "entropy": 5.476975965499878,
      "epoch": 2.791985994942618,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004218998705052443,
      "loss": 5.0867,
      "mean_token_accuracy": 0.18456562906503676,
      "num_tokens": 62247523.0,
      "step": 35885
    },
    {
      "entropy": 5.4104971408844,
      "epoch": 2.7923750243143357,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042187887610282255,
      "loss": 4.9677,
      "mean_token_accuracy": 0.19273096472024917,
      "num_tokens": 62255601.0,
      "step": 35890
    },
    {
      "entropy": 5.398213291168213,
      "epoch": 2.7927640536860534,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004218578794717476,
      "loss": 4.9799,
      "mean_token_accuracy": 0.19614312350749968,
      "num_tokens": 62263650.0,
      "step": 35895
    },
    {
      "entropy": 5.3360267162323,
      "epoch": 2.7931530830577707,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042183688061233813,
      "loss": 5.0083,
      "mean_token_accuracy": 0.19483959525823594,
      "num_tokens": 62272474.0,
      "step": 35900
    },
    {
      "entropy": 5.350556468963623,
      "epoch": 2.7935421124294884,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042181587952491267,
      "loss": 4.8607,
      "mean_token_accuracy": 0.21262108087539672,
      "num_tokens": 62281364.0,
      "step": 35905
    },
    {
      "entropy": 5.40102219581604,
      "epoch": 2.793931141801206,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042179487620978997,
      "loss": 4.9479,
      "mean_token_accuracy": 0.19825297594070435,
      "num_tokens": 62289025.0,
      "step": 35910
    },
    {
      "entropy": 5.34331431388855,
      "epoch": 2.7943201711729238,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004217738706672886,
      "loss": 4.9872,
      "mean_token_accuracy": 0.19918045550584793,
      "num_tokens": 62298063.0,
      "step": 35915
    },
    {
      "entropy": 5.354018497467041,
      "epoch": 2.794709200544641,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00042175286289772746,
      "loss": 4.9628,
      "mean_token_accuracy": 0.19780047237873077,
      "num_tokens": 62306806.0,
      "step": 35920
    },
    {
      "entropy": 5.337471961975098,
      "epoch": 2.7950982299163587,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004217318529014251,
      "loss": 4.9636,
      "mean_token_accuracy": 0.19914084821939468,
      "num_tokens": 62315119.0,
      "step": 35925
    },
    {
      "entropy": 5.404162311553955,
      "epoch": 2.7954872592880764,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004217108406787005,
      "loss": 5.0551,
      "mean_token_accuracy": 0.19554538875818253,
      "num_tokens": 62323615.0,
      "step": 35930
    },
    {
      "entropy": 5.422103309631348,
      "epoch": 2.7958762886597937,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004216898262298724,
      "loss": 4.9811,
      "mean_token_accuracy": 0.20338421911001206,
      "num_tokens": 62332313.0,
      "step": 35935
    },
    {
      "entropy": 5.234785079956055,
      "epoch": 2.7962653180315113,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004216688095552596,
      "loss": 4.808,
      "mean_token_accuracy": 0.20752618610858917,
      "num_tokens": 62341060.0,
      "step": 35940
    },
    {
      "entropy": 5.397421216964721,
      "epoch": 2.796654347403229,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004216477906551811,
      "loss": 5.0463,
      "mean_token_accuracy": 0.1946691781282425,
      "num_tokens": 62349616.0,
      "step": 35945
    },
    {
      "entropy": 5.412206935882568,
      "epoch": 2.7970433767749467,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042162676952995585,
      "loss": 4.9952,
      "mean_token_accuracy": 0.20287972539663315,
      "num_tokens": 62358083.0,
      "step": 35950
    },
    {
      "entropy": 5.416095161437989,
      "epoch": 2.797432406146664,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042160574617990267,
      "loss": 4.9903,
      "mean_token_accuracy": 0.1979239284992218,
      "num_tokens": 62366494.0,
      "step": 35955
    },
    {
      "entropy": 5.397483539581299,
      "epoch": 2.7978214355183817,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042158472060534066,
      "loss": 4.9798,
      "mean_token_accuracy": 0.19367599338293076,
      "num_tokens": 62375590.0,
      "step": 35960
    },
    {
      "entropy": 5.478774452209473,
      "epoch": 2.798210464890099,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042156369280658886,
      "loss": 5.1304,
      "mean_token_accuracy": 0.19774450212717057,
      "num_tokens": 62384186.0,
      "step": 35965
    },
    {
      "entropy": 5.4219160079956055,
      "epoch": 2.7985994942618166,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004215426627839662,
      "loss": 4.9443,
      "mean_token_accuracy": 0.2010745882987976,
      "num_tokens": 62392789.0,
      "step": 35970
    },
    {
      "entropy": 5.322008562088013,
      "epoch": 2.7989885236335343,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042152163053779196,
      "loss": 4.9545,
      "mean_token_accuracy": 0.19932419806718826,
      "num_tokens": 62401808.0,
      "step": 35975
    },
    {
      "entropy": 5.430076313018799,
      "epoch": 2.799377553005252,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004215005960683852,
      "loss": 5.0252,
      "mean_token_accuracy": 0.1908600762486458,
      "num_tokens": 62410700.0,
      "step": 35980
    },
    {
      "entropy": 5.437100982666015,
      "epoch": 2.7997665823769697,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004214795593760651,
      "loss": 4.9484,
      "mean_token_accuracy": 0.2061524957418442,
      "num_tokens": 62419337.0,
      "step": 35985
    },
    {
      "entropy": 5.363708066940307,
      "epoch": 2.800155611748687,
      "grad_norm": 1.25,
      "learning_rate": 0.0004214585204611508,
      "loss": 5.0308,
      "mean_token_accuracy": 0.1986408516764641,
      "num_tokens": 62427189.0,
      "step": 35990
    },
    {
      "entropy": 5.388799667358398,
      "epoch": 2.8005446411204047,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042143747932396153,
      "loss": 5.0846,
      "mean_token_accuracy": 0.19581374078989028,
      "num_tokens": 62436531.0,
      "step": 35995
    },
    {
      "entropy": 5.432786846160889,
      "epoch": 2.800933670492122,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004214164359648166,
      "loss": 5.0456,
      "mean_token_accuracy": 0.19871318340301514,
      "num_tokens": 62445250.0,
      "step": 36000
    },
    {
      "epoch": 2.800933670492122,
      "eval_entropy": 5.240249944949403,
      "eval_loss": 5.0924835205078125,
      "eval_mean_token_accuracy": 0.2027466549063899,
      "eval_num_tokens": 62445250.0,
      "eval_runtime": 28.7364,
      "eval_samples_per_second": 1446.178,
      "eval_steps_per_second": 180.781,
      "step": 36000
    },
    {
      "entropy": 5.401690864562989,
      "epoch": 2.8013226998638396,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042139539038403535,
      "loss": 4.9324,
      "mean_token_accuracy": 0.20196448862552643,
      "num_tokens": 62454221.0,
      "step": 36005
    },
    {
      "entropy": 5.3202269077301025,
      "epoch": 2.8017117292355573,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000421374342581937,
      "loss": 4.8701,
      "mean_token_accuracy": 0.21134181767702104,
      "num_tokens": 62463237.0,
      "step": 36010
    },
    {
      "entropy": 5.384362125396729,
      "epoch": 2.802100758607275,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042135329255884105,
      "loss": 4.9941,
      "mean_token_accuracy": 0.1996095985174179,
      "num_tokens": 62472227.0,
      "step": 36015
    },
    {
      "entropy": 5.363177728652954,
      "epoch": 2.8024897879789923,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004213322403150668,
      "loss": 4.9058,
      "mean_token_accuracy": 0.20486340075731277,
      "num_tokens": 62480225.0,
      "step": 36020
    },
    {
      "entropy": 5.411304903030396,
      "epoch": 2.80287881735071,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004213111858509337,
      "loss": 4.9977,
      "mean_token_accuracy": 0.19507963806390763,
      "num_tokens": 62487858.0,
      "step": 36025
    },
    {
      "entropy": 5.432738733291626,
      "epoch": 2.8032678467224277,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004212901291667612,
      "loss": 5.1026,
      "mean_token_accuracy": 0.1950969859957695,
      "num_tokens": 62496004.0,
      "step": 36030
    },
    {
      "entropy": 5.387632131576538,
      "epoch": 2.803656876094145,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042126907026286887,
      "loss": 4.9944,
      "mean_token_accuracy": 0.19514856934547425,
      "num_tokens": 62504011.0,
      "step": 36035
    },
    {
      "entropy": 5.420150995254517,
      "epoch": 2.8040459054658626,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042124800913957624,
      "loss": 5.0021,
      "mean_token_accuracy": 0.19884813725948333,
      "num_tokens": 62512143.0,
      "step": 36040
    },
    {
      "entropy": 5.465098094940186,
      "epoch": 2.8044349348375803,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004212269457972028,
      "loss": 5.0069,
      "mean_token_accuracy": 0.19502864480018617,
      "num_tokens": 62520374.0,
      "step": 36045
    },
    {
      "entropy": 5.491916799545288,
      "epoch": 2.804823964209298,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004212058802360682,
      "loss": 5.0711,
      "mean_token_accuracy": 0.1921910747885704,
      "num_tokens": 62529345.0,
      "step": 36050
    },
    {
      "entropy": 5.326232767105102,
      "epoch": 2.8052129935810153,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004211848124564922,
      "loss": 4.9309,
      "mean_token_accuracy": 0.1997879147529602,
      "num_tokens": 62538622.0,
      "step": 36055
    },
    {
      "entropy": 5.365319871902466,
      "epoch": 2.805602022952733,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004211637424587942,
      "loss": 4.9133,
      "mean_token_accuracy": 0.2012769713997841,
      "num_tokens": 62547071.0,
      "step": 36060
    },
    {
      "entropy": 5.4547303199768065,
      "epoch": 2.8059910523244502,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004211426702432941,
      "loss": 5.0209,
      "mean_token_accuracy": 0.19625279754400254,
      "num_tokens": 62556321.0,
      "step": 36065
    },
    {
      "entropy": 5.445640611648559,
      "epoch": 2.806380081696168,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004211215958103116,
      "loss": 5.1363,
      "mean_token_accuracy": 0.18879806697368623,
      "num_tokens": 62565549.0,
      "step": 36070
    },
    {
      "entropy": 5.35904541015625,
      "epoch": 2.8067691110678856,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042110051916016645,
      "loss": 4.9267,
      "mean_token_accuracy": 0.20534685552120208,
      "num_tokens": 62573753.0,
      "step": 36075
    },
    {
      "entropy": 5.373360061645508,
      "epoch": 2.8071581404396033,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004210794402931785,
      "loss": 5.0186,
      "mean_token_accuracy": 0.19346022307872773,
      "num_tokens": 62583919.0,
      "step": 36080
    },
    {
      "entropy": 5.339261388778686,
      "epoch": 2.807547169811321,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004210583592096675,
      "loss": 4.9485,
      "mean_token_accuracy": 0.1957052767276764,
      "num_tokens": 62592136.0,
      "step": 36085
    },
    {
      "entropy": 5.394687938690185,
      "epoch": 2.8079361991830383,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004210372759099534,
      "loss": 4.9245,
      "mean_token_accuracy": 0.20260759443044662,
      "num_tokens": 62600742.0,
      "step": 36090
    },
    {
      "entropy": 5.388784265518188,
      "epoch": 2.808325228554756,
      "grad_norm": 1.234375,
      "learning_rate": 0.000421016190394356,
      "loss": 4.9213,
      "mean_token_accuracy": 0.202702035009861,
      "num_tokens": 62608239.0,
      "step": 36095
    },
    {
      "entropy": 5.3343925952911375,
      "epoch": 2.8087142579264732,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042099510266319545,
      "loss": 4.9612,
      "mean_token_accuracy": 0.19917184710502625,
      "num_tokens": 62617445.0,
      "step": 36100
    },
    {
      "entropy": 5.334157466888428,
      "epoch": 2.809103287298191,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004209740127167915,
      "loss": 4.9575,
      "mean_token_accuracy": 0.19956833124160767,
      "num_tokens": 62625578.0,
      "step": 36105
    },
    {
      "entropy": 5.352154874801636,
      "epoch": 2.8094923166699086,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004209529205554643,
      "loss": 4.9561,
      "mean_token_accuracy": 0.1968429207801819,
      "num_tokens": 62635028.0,
      "step": 36110
    },
    {
      "entropy": 5.4406829357147215,
      "epoch": 2.8098813460416263,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004209318261795339,
      "loss": 4.9796,
      "mean_token_accuracy": 0.2022951975464821,
      "num_tokens": 62643739.0,
      "step": 36115
    },
    {
      "entropy": 5.447633266448975,
      "epoch": 2.8102703754133436,
      "grad_norm": 1.125,
      "learning_rate": 0.0004209107295893202,
      "loss": 5.1049,
      "mean_token_accuracy": 0.19537079483270645,
      "num_tokens": 62652769.0,
      "step": 36120
    },
    {
      "entropy": 5.314122486114502,
      "epoch": 2.8106594047850613,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042088963078514344,
      "loss": 4.9541,
      "mean_token_accuracy": 0.19805045425891876,
      "num_tokens": 62661412.0,
      "step": 36125
    },
    {
      "entropy": 5.335757923126221,
      "epoch": 2.811048434156779,
      "grad_norm": 1.125,
      "learning_rate": 0.0004208685297673238,
      "loss": 4.9993,
      "mean_token_accuracy": 0.19910851269960403,
      "num_tokens": 62670048.0,
      "step": 36130
    },
    {
      "entropy": 5.403274822235107,
      "epoch": 2.811437463528496,
      "grad_norm": 1.125,
      "learning_rate": 0.0004208474265361813,
      "loss": 4.9761,
      "mean_token_accuracy": 0.19658947587013245,
      "num_tokens": 62678901.0,
      "step": 36135
    },
    {
      "entropy": 5.342027091979981,
      "epoch": 2.811826492900214,
      "grad_norm": 1.125,
      "learning_rate": 0.00042082632109203627,
      "loss": 4.8927,
      "mean_token_accuracy": 0.20444838851690292,
      "num_tokens": 62687124.0,
      "step": 36140
    },
    {
      "entropy": 5.390532684326172,
      "epoch": 2.8122155222719316,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000420805213435209,
      "loss": 5.0231,
      "mean_token_accuracy": 0.1904981777071953,
      "num_tokens": 62696029.0,
      "step": 36145
    },
    {
      "entropy": 5.372088289260864,
      "epoch": 2.8126045516436493,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004207841035660196,
      "loss": 4.903,
      "mean_token_accuracy": 0.20387211591005325,
      "num_tokens": 62704181.0,
      "step": 36150
    },
    {
      "entropy": 5.328259706497192,
      "epoch": 2.8129935810153666,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004207629914847885,
      "loss": 5.018,
      "mean_token_accuracy": 0.20008323043584825,
      "num_tokens": 62713440.0,
      "step": 36155
    },
    {
      "entropy": 5.330566644668579,
      "epoch": 2.8133826103870843,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000420741877191836,
      "loss": 4.979,
      "mean_token_accuracy": 0.20319633185863495,
      "num_tokens": 62723024.0,
      "step": 36160
    },
    {
      "entropy": 5.3710874080657955,
      "epoch": 2.8137716397588015,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004207207606874825,
      "loss": 4.9599,
      "mean_token_accuracy": 0.19447644650936127,
      "num_tokens": 62731441.0,
      "step": 36165
    },
    {
      "entropy": 5.427125644683838,
      "epoch": 2.814160669130519,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004206996419720484,
      "loss": 5.0529,
      "mean_token_accuracy": 0.1998141050338745,
      "num_tokens": 62740073.0,
      "step": 36170
    },
    {
      "entropy": 5.305306148529053,
      "epoch": 2.814549698502237,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004206785210458541,
      "loss": 4.8523,
      "mean_token_accuracy": 0.20848419070243834,
      "num_tokens": 62749192.0,
      "step": 36175
    },
    {
      "entropy": 5.362651205062866,
      "epoch": 2.8149387278739546,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004206573979092202,
      "loss": 4.9699,
      "mean_token_accuracy": 0.20324881225824357,
      "num_tokens": 62757408.0,
      "step": 36180
    },
    {
      "entropy": 5.377953958511353,
      "epoch": 2.8153277572456723,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042063627256246706,
      "loss": 4.9923,
      "mean_token_accuracy": 0.20035659074783324,
      "num_tokens": 62767475.0,
      "step": 36185
    },
    {
      "entropy": 5.308944416046143,
      "epoch": 2.8157167866173896,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004206151450059153,
      "loss": 4.8636,
      "mean_token_accuracy": 0.2066209852695465,
      "num_tokens": 62775802.0,
      "step": 36190
    },
    {
      "entropy": 5.26205587387085,
      "epoch": 2.8161058159891073,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042059401523988546,
      "loss": 4.9244,
      "mean_token_accuracy": 0.19986067712306976,
      "num_tokens": 62784718.0,
      "step": 36195
    },
    {
      "entropy": 5.358785200119018,
      "epoch": 2.8164948453608245,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004205728832646982,
      "loss": 5.0446,
      "mean_token_accuracy": 0.19115493595600128,
      "num_tokens": 62793692.0,
      "step": 36200
    },
    {
      "entropy": 5.389555263519287,
      "epoch": 2.816883874732542,
      "grad_norm": 1.125,
      "learning_rate": 0.00042055174908067414,
      "loss": 5.013,
      "mean_token_accuracy": 0.20038746744394303,
      "num_tokens": 62802583.0,
      "step": 36205
    },
    {
      "entropy": 5.402655458450317,
      "epoch": 2.81727290410426,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042053061268813397,
      "loss": 4.9674,
      "mean_token_accuracy": 0.2008891597390175,
      "num_tokens": 62812374.0,
      "step": 36210
    },
    {
      "entropy": 5.391308784484863,
      "epoch": 2.8176619334759776,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042050947408739836,
      "loss": 5.0456,
      "mean_token_accuracy": 0.19457284212112427,
      "num_tokens": 62821796.0,
      "step": 36215
    },
    {
      "entropy": 5.405890846252442,
      "epoch": 2.818050962847695,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004204883332787881,
      "loss": 4.9681,
      "mean_token_accuracy": 0.19584135562181473,
      "num_tokens": 62831349.0,
      "step": 36220
    },
    {
      "entropy": 5.362933683395386,
      "epoch": 2.8184399922194125,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042046719026262396,
      "loss": 4.9529,
      "mean_token_accuracy": 0.20298338681459427,
      "num_tokens": 62838842.0,
      "step": 36225
    },
    {
      "entropy": 5.329693222045899,
      "epoch": 2.8188290215911302,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042044604503922674,
      "loss": 4.9331,
      "mean_token_accuracy": 0.2032444953918457,
      "num_tokens": 62847067.0,
      "step": 36230
    },
    {
      "entropy": 5.301738548278808,
      "epoch": 2.8192180509628475,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004204248976089173,
      "loss": 4.8088,
      "mean_token_accuracy": 0.2089138373732567,
      "num_tokens": 62855374.0,
      "step": 36235
    },
    {
      "entropy": 5.345616674423217,
      "epoch": 2.819607080334565,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004204037479720165,
      "loss": 4.9283,
      "mean_token_accuracy": 0.20207460671663285,
      "num_tokens": 62863141.0,
      "step": 36240
    },
    {
      "entropy": 5.3851696968078615,
      "epoch": 2.819996109706283,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042038259612884514,
      "loss": 5.0487,
      "mean_token_accuracy": 0.19783153682947158,
      "num_tokens": 62872080.0,
      "step": 36245
    },
    {
      "entropy": 5.395944118499756,
      "epoch": 2.8203851390780006,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042036144207972445,
      "loss": 5.0123,
      "mean_token_accuracy": 0.20139943361282348,
      "num_tokens": 62880492.0,
      "step": 36250
    },
    {
      "entropy": 5.366403865814209,
      "epoch": 2.820774168449718,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004203402858249752,
      "loss": 4.9113,
      "mean_token_accuracy": 0.20440810173749924,
      "num_tokens": 62888961.0,
      "step": 36255
    },
    {
      "entropy": 5.383540868759155,
      "epoch": 2.8211631978214355,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004203191273649184,
      "loss": 4.9755,
      "mean_token_accuracy": 0.20121616274118423,
      "num_tokens": 62897224.0,
      "step": 36260
    },
    {
      "entropy": 5.377414655685425,
      "epoch": 2.8215522271931532,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042029796669987513,
      "loss": 4.9681,
      "mean_token_accuracy": 0.20221128165721894,
      "num_tokens": 62905873.0,
      "step": 36265
    },
    {
      "entropy": 5.408347034454346,
      "epoch": 2.8219412565648705,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004202768038301665,
      "loss": 5.0022,
      "mean_token_accuracy": 0.2004620224237442,
      "num_tokens": 62914438.0,
      "step": 36270
    },
    {
      "entropy": 5.418831157684326,
      "epoch": 2.822330285936588,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042025563875611367,
      "loss": 5.0142,
      "mean_token_accuracy": 0.19713575541973113,
      "num_tokens": 62923310.0,
      "step": 36275
    },
    {
      "entropy": 5.380051755905152,
      "epoch": 2.822719315308306,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004202344714780376,
      "loss": 5.0332,
      "mean_token_accuracy": 0.19681963473558425,
      "num_tokens": 62932031.0,
      "step": 36280
    },
    {
      "entropy": 5.381773042678833,
      "epoch": 2.8231083446800236,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042021330199625966,
      "loss": 4.9771,
      "mean_token_accuracy": 0.19716326594352723,
      "num_tokens": 62939695.0,
      "step": 36285
    },
    {
      "entropy": 5.3902459144592285,
      "epoch": 2.823497374051741,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042019213031110105,
      "loss": 4.9443,
      "mean_token_accuracy": 0.20004019886255264,
      "num_tokens": 62948251.0,
      "step": 36290
    },
    {
      "entropy": 5.414561414718628,
      "epoch": 2.8238864034234585,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004201709564228829,
      "loss": 4.9783,
      "mean_token_accuracy": 0.19726343154907228,
      "num_tokens": 62956881.0,
      "step": 36295
    },
    {
      "entropy": 5.520067548751831,
      "epoch": 2.824275432795176,
      "grad_norm": 1.125,
      "learning_rate": 0.00042014978033192655,
      "loss": 5.1438,
      "mean_token_accuracy": 0.18636136502027512,
      "num_tokens": 62966241.0,
      "step": 36300
    },
    {
      "entropy": 5.349583768844605,
      "epoch": 2.8246644621668935,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042012860203855337,
      "loss": 4.8555,
      "mean_token_accuracy": 0.20345052182674409,
      "num_tokens": 62974826.0,
      "step": 36305
    },
    {
      "entropy": 5.445868492126465,
      "epoch": 2.825053491538611,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042010742154308453,
      "loss": 5.0604,
      "mean_token_accuracy": 0.19412871301174164,
      "num_tokens": 62983800.0,
      "step": 36310
    },
    {
      "entropy": 5.399209499359131,
      "epoch": 2.825442520910329,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042008623884584167,
      "loss": 5.0534,
      "mean_token_accuracy": 0.1923176482319832,
      "num_tokens": 62993219.0,
      "step": 36315
    },
    {
      "entropy": 5.457874441146851,
      "epoch": 2.825831550282046,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042006505394714597,
      "loss": 5.0286,
      "mean_token_accuracy": 0.19659639596939088,
      "num_tokens": 63002455.0,
      "step": 36320
    },
    {
      "entropy": 5.479030132293701,
      "epoch": 2.826220579653764,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042004386684731896,
      "loss": 5.1029,
      "mean_token_accuracy": 0.18663970828056337,
      "num_tokens": 63012200.0,
      "step": 36325
    },
    {
      "entropy": 5.356777143478394,
      "epoch": 2.8266096090254815,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042002267754668216,
      "loss": 4.9914,
      "mean_token_accuracy": 0.20466650277376175,
      "num_tokens": 63021870.0,
      "step": 36330
    },
    {
      "entropy": 5.414930772781372,
      "epoch": 2.826998638397199,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000420001486045557,
      "loss": 4.9802,
      "mean_token_accuracy": 0.19850059300661088,
      "num_tokens": 63030941.0,
      "step": 36335
    },
    {
      "entropy": 5.465569257736206,
      "epoch": 2.8273876677689165,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004199802923442652,
      "loss": 5.0447,
      "mean_token_accuracy": 0.19225822389125824,
      "num_tokens": 63039264.0,
      "step": 36340
    },
    {
      "entropy": 5.4531008243560795,
      "epoch": 2.827776697140634,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004199590964431282,
      "loss": 5.0449,
      "mean_token_accuracy": 0.19052287936210632,
      "num_tokens": 63048723.0,
      "step": 36345
    },
    {
      "entropy": 5.362220096588135,
      "epoch": 2.828165726512352,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004199378983424676,
      "loss": 4.9565,
      "mean_token_accuracy": 0.2024417906999588,
      "num_tokens": 63058147.0,
      "step": 36350
    },
    {
      "entropy": 5.387423753738403,
      "epoch": 2.828554755884069,
      "grad_norm": 1.125,
      "learning_rate": 0.00041991669804260505,
      "loss": 4.9509,
      "mean_token_accuracy": 0.19898790270090103,
      "num_tokens": 63066614.0,
      "step": 36355
    },
    {
      "entropy": 5.364709663391113,
      "epoch": 2.828943785255787,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004198954955438623,
      "loss": 4.9241,
      "mean_token_accuracy": 0.20080677717924117,
      "num_tokens": 63075174.0,
      "step": 36360
    },
    {
      "entropy": 5.3992424488067625,
      "epoch": 2.8293328146275045,
      "grad_norm": 1.234375,
      "learning_rate": 0.000419874290846561,
      "loss": 4.9634,
      "mean_token_accuracy": 0.19825672805309297,
      "num_tokens": 63083375.0,
      "step": 36365
    },
    {
      "entropy": 5.3866822719573975,
      "epoch": 2.8297218439992218,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041985308395102303,
      "loss": 5.0446,
      "mean_token_accuracy": 0.20821838080883026,
      "num_tokens": 63091669.0,
      "step": 36370
    },
    {
      "entropy": 5.385388326644898,
      "epoch": 2.8301108733709395,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041983187485756997,
      "loss": 5.0218,
      "mean_token_accuracy": 0.19797692000865935,
      "num_tokens": 63100597.0,
      "step": 36375
    },
    {
      "entropy": 5.4726152420043945,
      "epoch": 2.830499902742657,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041981066356652373,
      "loss": 5.1166,
      "mean_token_accuracy": 0.19607223123311995,
      "num_tokens": 63109445.0,
      "step": 36380
    },
    {
      "entropy": 5.426414918899536,
      "epoch": 2.830888932114375,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041978945007820626,
      "loss": 5.0086,
      "mean_token_accuracy": 0.19027449935674667,
      "num_tokens": 63118485.0,
      "step": 36385
    },
    {
      "entropy": 5.365958738327026,
      "epoch": 2.831277961486092,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041976823439293925,
      "loss": 4.9807,
      "mean_token_accuracy": 0.20073189437389374,
      "num_tokens": 63127366.0,
      "step": 36390
    },
    {
      "entropy": 5.331878805160523,
      "epoch": 2.83166699085781,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004197470165110447,
      "loss": 5.0146,
      "mean_token_accuracy": 0.1915379211306572,
      "num_tokens": 63136192.0,
      "step": 36395
    },
    {
      "entropy": 5.476261854171753,
      "epoch": 2.832056020229527,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004197257964328446,
      "loss": 5.0828,
      "mean_token_accuracy": 0.19480252563953399,
      "num_tokens": 63144250.0,
      "step": 36400
    },
    {
      "entropy": 5.42934775352478,
      "epoch": 2.8324450496012448,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004197045741586609,
      "loss": 4.971,
      "mean_token_accuracy": 0.20400989949703216,
      "num_tokens": 63152685.0,
      "step": 36405
    },
    {
      "entropy": 5.389996814727783,
      "epoch": 2.8328340789729625,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004196833496888156,
      "loss": 4.9068,
      "mean_token_accuracy": 0.20530979633331298,
      "num_tokens": 63161853.0,
      "step": 36410
    },
    {
      "entropy": 5.3254420280456545,
      "epoch": 2.83322310834468,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004196621230236308,
      "loss": 4.925,
      "mean_token_accuracy": 0.20006163418293,
      "num_tokens": 63170626.0,
      "step": 36415
    },
    {
      "entropy": 5.336709070205688,
      "epoch": 2.833612137716398,
      "grad_norm": 1.125,
      "learning_rate": 0.00041964089416342845,
      "loss": 4.992,
      "mean_token_accuracy": 0.19946582168340682,
      "num_tokens": 63180183.0,
      "step": 36420
    },
    {
      "entropy": 5.435316038131714,
      "epoch": 2.834001167088115,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004196196631085308,
      "loss": 5.028,
      "mean_token_accuracy": 0.19180624485015868,
      "num_tokens": 63188993.0,
      "step": 36425
    },
    {
      "entropy": 5.39701886177063,
      "epoch": 2.834390196459833,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041959842985925987,
      "loss": 5.0259,
      "mean_token_accuracy": 0.19429241120815277,
      "num_tokens": 63197592.0,
      "step": 36430
    },
    {
      "entropy": 5.372807312011719,
      "epoch": 2.83477922583155,
      "grad_norm": 1.109375,
      "learning_rate": 0.000419577194415938,
      "loss": 5.0016,
      "mean_token_accuracy": 0.19639111757278443,
      "num_tokens": 63206216.0,
      "step": 36435
    },
    {
      "entropy": 5.3529191493988035,
      "epoch": 2.8351682552032678,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041955595677888735,
      "loss": 4.9347,
      "mean_token_accuracy": 0.20346073508262635,
      "num_tokens": 63214633.0,
      "step": 36440
    },
    {
      "entropy": 5.383954477310181,
      "epoch": 2.8355572845749855,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004195347169484301,
      "loss": 4.9662,
      "mean_token_accuracy": 0.19904121309518813,
      "num_tokens": 63223666.0,
      "step": 36445
    },
    {
      "entropy": 5.320629262924195,
      "epoch": 2.835946313946703,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004195134749248886,
      "loss": 4.8769,
      "mean_token_accuracy": 0.20694139003753662,
      "num_tokens": 63232192.0,
      "step": 36450
    },
    {
      "entropy": 5.292669868469238,
      "epoch": 2.8363353433184204,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004194922307085851,
      "loss": 4.9437,
      "mean_token_accuracy": 0.1955632135272026,
      "num_tokens": 63241235.0,
      "step": 36455
    },
    {
      "entropy": 5.4641725540161135,
      "epoch": 2.836724372690138,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004194709842998419,
      "loss": 5.0699,
      "mean_token_accuracy": 0.19470468610525132,
      "num_tokens": 63249742.0,
      "step": 36460
    },
    {
      "entropy": 5.365504121780395,
      "epoch": 2.837113402061856,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041944973569898157,
      "loss": 4.9409,
      "mean_token_accuracy": 0.20193239003419877,
      "num_tokens": 63258004.0,
      "step": 36465
    },
    {
      "entropy": 5.455755376815796,
      "epoch": 2.837502431433573,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004194284849063265,
      "loss": 5.0415,
      "mean_token_accuracy": 0.19460777044296265,
      "num_tokens": 63266478.0,
      "step": 36470
    },
    {
      "entropy": 5.37844295501709,
      "epoch": 2.8378914608052908,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004194072319221989,
      "loss": 4.9664,
      "mean_token_accuracy": 0.20050947964191437,
      "num_tokens": 63274868.0,
      "step": 36475
    },
    {
      "entropy": 5.38673963546753,
      "epoch": 2.8382804901770085,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004193859767469214,
      "loss": 5.0803,
      "mean_token_accuracy": 0.18968232423067094,
      "num_tokens": 63284207.0,
      "step": 36480
    },
    {
      "entropy": 5.431429052352906,
      "epoch": 2.838669519548726,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041936471938081667,
      "loss": 5.0033,
      "mean_token_accuracy": 0.19469614773988725,
      "num_tokens": 63292477.0,
      "step": 36485
    },
    {
      "entropy": 5.37626314163208,
      "epoch": 2.8390585489204434,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041934345982420695,
      "loss": 5.0008,
      "mean_token_accuracy": 0.19936856925487517,
      "num_tokens": 63301457.0,
      "step": 36490
    },
    {
      "entropy": 5.445898628234863,
      "epoch": 2.839447578292161,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000419322198077415,
      "loss": 5.0406,
      "mean_token_accuracy": 0.1945921376347542,
      "num_tokens": 63309967.0,
      "step": 36495
    },
    {
      "entropy": 5.381873750686646,
      "epoch": 2.8398366076638784,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041930093414076344,
      "loss": 4.8829,
      "mean_token_accuracy": 0.20467058718204498,
      "num_tokens": 63319329.0,
      "step": 36500
    },
    {
      "entropy": 5.324381256103516,
      "epoch": 2.840225637035596,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041927966801457486,
      "loss": 4.9846,
      "mean_token_accuracy": 0.19861678034067154,
      "num_tokens": 63328600.0,
      "step": 36505
    },
    {
      "entropy": 5.3899774074554445,
      "epoch": 2.8406146664073137,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000419258399699172,
      "loss": 4.9897,
      "mean_token_accuracy": 0.19936752766370774,
      "num_tokens": 63337297.0,
      "step": 36510
    },
    {
      "entropy": 5.37494387626648,
      "epoch": 2.8410036957790314,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004192371291948775,
      "loss": 4.9433,
      "mean_token_accuracy": 0.2073831394314766,
      "num_tokens": 63346634.0,
      "step": 36515
    },
    {
      "entropy": 5.438672924041748,
      "epoch": 2.841392725150749,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041921585650201413,
      "loss": 5.0457,
      "mean_token_accuracy": 0.19019139260053636,
      "num_tokens": 63354870.0,
      "step": 36520
    },
    {
      "entropy": 5.335669660568238,
      "epoch": 2.8417817545224664,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004191945816209047,
      "loss": 5.0411,
      "mean_token_accuracy": 0.1908004492521286,
      "num_tokens": 63363571.0,
      "step": 36525
    },
    {
      "entropy": 5.416158628463745,
      "epoch": 2.842170783894184,
      "grad_norm": 1.125,
      "learning_rate": 0.00041917330455187195,
      "loss": 4.9414,
      "mean_token_accuracy": 0.204941126704216,
      "num_tokens": 63371716.0,
      "step": 36530
    },
    {
      "entropy": 5.355167102813721,
      "epoch": 2.8425598132659013,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041915202529523894,
      "loss": 5.042,
      "mean_token_accuracy": 0.2016516759991646,
      "num_tokens": 63381399.0,
      "step": 36535
    },
    {
      "entropy": 5.418057537078857,
      "epoch": 2.842948842637619,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004191307438513283,
      "loss": 5.0376,
      "mean_token_accuracy": 0.20107784271240234,
      "num_tokens": 63390548.0,
      "step": 36540
    },
    {
      "entropy": 5.382865476608276,
      "epoch": 2.8433378720093367,
      "grad_norm": 1.171875,
      "learning_rate": 0.000419109460220463,
      "loss": 4.9827,
      "mean_token_accuracy": 0.19784849435091018,
      "num_tokens": 63400219.0,
      "step": 36545
    },
    {
      "entropy": 5.350049924850464,
      "epoch": 2.8437269013810544,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041908817440296605,
      "loss": 4.9057,
      "mean_token_accuracy": 0.20449953973293306,
      "num_tokens": 63409131.0,
      "step": 36550
    },
    {
      "entropy": 5.438841056823731,
      "epoch": 2.8441159307527717,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041906688639916035,
      "loss": 5.1277,
      "mean_token_accuracy": 0.18973740488290786,
      "num_tokens": 63418830.0,
      "step": 36555
    },
    {
      "entropy": 5.40243706703186,
      "epoch": 2.8445049601244894,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004190455962093689,
      "loss": 4.929,
      "mean_token_accuracy": 0.19739658683538436,
      "num_tokens": 63427467.0,
      "step": 36560
    },
    {
      "entropy": 5.386296987533569,
      "epoch": 2.844893989496207,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041902430383391494,
      "loss": 4.9958,
      "mean_token_accuracy": 0.1969866618514061,
      "num_tokens": 63436957.0,
      "step": 36565
    },
    {
      "entropy": 5.362722253799438,
      "epoch": 2.8452830188679243,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004190030092731214,
      "loss": 5.0419,
      "mean_token_accuracy": 0.19847180247306823,
      "num_tokens": 63446010.0,
      "step": 36570
    },
    {
      "entropy": 5.333335208892822,
      "epoch": 2.845672048239642,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004189817125273113,
      "loss": 4.9273,
      "mean_token_accuracy": 0.20335300862789155,
      "num_tokens": 63454666.0,
      "step": 36575
    },
    {
      "entropy": 5.355603742599487,
      "epoch": 2.8460610776113597,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041896041359680797,
      "loss": 4.9477,
      "mean_token_accuracy": 0.20321038514375686,
      "num_tokens": 63463076.0,
      "step": 36580
    },
    {
      "entropy": 5.429598474502564,
      "epoch": 2.8464501069830774,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041893911248193437,
      "loss": 4.9523,
      "mean_token_accuracy": 0.2025564655661583,
      "num_tokens": 63471547.0,
      "step": 36585
    },
    {
      "entropy": 5.373992919921875,
      "epoch": 2.8468391363547947,
      "grad_norm": 1.3125,
      "learning_rate": 0.000418917809183014,
      "loss": 4.9905,
      "mean_token_accuracy": 0.19798582643270493,
      "num_tokens": 63479710.0,
      "step": 36590
    },
    {
      "entropy": 5.311467599868775,
      "epoch": 2.8472281657265124,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041889650370036986,
      "loss": 4.9296,
      "mean_token_accuracy": 0.2042201951146126,
      "num_tokens": 63488639.0,
      "step": 36595
    },
    {
      "entropy": 5.417302942276001,
      "epoch": 2.84761719509823,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004188751960343253,
      "loss": 5.042,
      "mean_token_accuracy": 0.19292029738426208,
      "num_tokens": 63496956.0,
      "step": 36600
    },
    {
      "entropy": 5.366496562957764,
      "epoch": 2.8480062244699473,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004188538861852037,
      "loss": 4.9564,
      "mean_token_accuracy": 0.20153767466545106,
      "num_tokens": 63505538.0,
      "step": 36605
    },
    {
      "entropy": 5.269507837295532,
      "epoch": 2.848395253841665,
      "grad_norm": 1.125,
      "learning_rate": 0.0004188325741533283,
      "loss": 4.8585,
      "mean_token_accuracy": 0.20440756231546403,
      "num_tokens": 63513664.0,
      "step": 36610
    },
    {
      "entropy": 5.3249414443969725,
      "epoch": 2.8487842832133827,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004188112599390225,
      "loss": 4.8682,
      "mean_token_accuracy": 0.20675662010908127,
      "num_tokens": 63521910.0,
      "step": 36615
    },
    {
      "entropy": 5.3695649147033695,
      "epoch": 2.8491733125851004,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004187899435426098,
      "loss": 4.8832,
      "mean_token_accuracy": 0.20887090265750885,
      "num_tokens": 63530272.0,
      "step": 36620
    },
    {
      "entropy": 5.343402671813965,
      "epoch": 2.8495623419568177,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041876862496441347,
      "loss": 5.0378,
      "mean_token_accuracy": 0.19705433398485184,
      "num_tokens": 63538652.0,
      "step": 36625
    },
    {
      "entropy": 5.372191095352173,
      "epoch": 2.8499513713285354,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041874730420475716,
      "loss": 4.9426,
      "mean_token_accuracy": 0.20222785025835038,
      "num_tokens": 63547016.0,
      "step": 36630
    },
    {
      "entropy": 5.336967658996582,
      "epoch": 2.8503404007002526,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041872598126396434,
      "loss": 4.9392,
      "mean_token_accuracy": 0.20588555186986923,
      "num_tokens": 63555561.0,
      "step": 36635
    },
    {
      "entropy": 5.365342330932617,
      "epoch": 2.8507294300719703,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041870465614235843,
      "loss": 4.9973,
      "mean_token_accuracy": 0.20036884248256684,
      "num_tokens": 63563510.0,
      "step": 36640
    },
    {
      "entropy": 5.34108681678772,
      "epoch": 2.851118459443688,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041868332884026317,
      "loss": 4.9607,
      "mean_token_accuracy": 0.20552650541067125,
      "num_tokens": 63572495.0,
      "step": 36645
    },
    {
      "entropy": 5.411440229415893,
      "epoch": 2.8515074888154057,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004186619993580021,
      "loss": 5.0234,
      "mean_token_accuracy": 0.195866858959198,
      "num_tokens": 63581525.0,
      "step": 36650
    },
    {
      "entropy": 5.367895936965942,
      "epoch": 2.851896518187123,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041864066769589875,
      "loss": 4.9444,
      "mean_token_accuracy": 0.19944222569465636,
      "num_tokens": 63590136.0,
      "step": 36655
    },
    {
      "entropy": 5.4395287990570065,
      "epoch": 2.8522855475588407,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041861933385427697,
      "loss": 5.0611,
      "mean_token_accuracy": 0.2006852760910988,
      "num_tokens": 63599353.0,
      "step": 36660
    },
    {
      "entropy": 5.360436248779297,
      "epoch": 2.8526745769305584,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004185979978334604,
      "loss": 4.8891,
      "mean_token_accuracy": 0.20667565912008284,
      "num_tokens": 63607586.0,
      "step": 36665
    },
    {
      "entropy": 5.338217306137085,
      "epoch": 2.8530636063022756,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004185766596337727,
      "loss": 4.9159,
      "mean_token_accuracy": 0.20061647593975068,
      "num_tokens": 63616444.0,
      "step": 36670
    },
    {
      "entropy": 5.3502647399902346,
      "epoch": 2.8534526356739933,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041855531925553773,
      "loss": 5.0199,
      "mean_token_accuracy": 0.1964164823293686,
      "num_tokens": 63625591.0,
      "step": 36675
    },
    {
      "entropy": 5.50525918006897,
      "epoch": 2.853841665045711,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004185339766990794,
      "loss": 5.0974,
      "mean_token_accuracy": 0.19667548388242723,
      "num_tokens": 63633681.0,
      "step": 36680
    },
    {
      "entropy": 5.385427618026734,
      "epoch": 2.8542306944174287,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004185126319647213,
      "loss": 5.0248,
      "mean_token_accuracy": 0.20020008236169815,
      "num_tokens": 63642298.0,
      "step": 36685
    },
    {
      "entropy": 5.445777177810669,
      "epoch": 2.854619723789146,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004184912850527875,
      "loss": 5.0781,
      "mean_token_accuracy": 0.1971253678202629,
      "num_tokens": 63650969.0,
      "step": 36690
    },
    {
      "entropy": 5.390744590759278,
      "epoch": 2.8550087531608637,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004184699359636018,
      "loss": 4.9363,
      "mean_token_accuracy": 0.2120387777686119,
      "num_tokens": 63659477.0,
      "step": 36695
    },
    {
      "entropy": 5.373928785324097,
      "epoch": 2.8553977825325814,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004184485846974882,
      "loss": 4.9981,
      "mean_token_accuracy": 0.19902394860982894,
      "num_tokens": 63668575.0,
      "step": 36700
    },
    {
      "entropy": 5.4341874599456785,
      "epoch": 2.8557868119042986,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004184272312547706,
      "loss": 5.0452,
      "mean_token_accuracy": 0.19717740565538405,
      "num_tokens": 63677818.0,
      "step": 36705
    },
    {
      "entropy": 5.42542839050293,
      "epoch": 2.8561758412760163,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041840587563577315,
      "loss": 4.8926,
      "mean_token_accuracy": 0.20150510519742965,
      "num_tokens": 63686360.0,
      "step": 36710
    },
    {
      "entropy": 5.33306245803833,
      "epoch": 2.856564870647734,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004183845178408197,
      "loss": 5.0457,
      "mean_token_accuracy": 0.19386502355337143,
      "num_tokens": 63695602.0,
      "step": 36715
    },
    {
      "entropy": 5.333541011810302,
      "epoch": 2.8569539000194517,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041836315787023456,
      "loss": 4.9913,
      "mean_token_accuracy": 0.1991768643260002,
      "num_tokens": 63704225.0,
      "step": 36720
    },
    {
      "entropy": 5.415351629257202,
      "epoch": 2.857342929391169,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041834179572434153,
      "loss": 4.9529,
      "mean_token_accuracy": 0.1959248185157776,
      "num_tokens": 63712814.0,
      "step": 36725
    },
    {
      "entropy": 5.386776971817016,
      "epoch": 2.8577319587628867,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041832043140346495,
      "loss": 4.9536,
      "mean_token_accuracy": 0.19837044775485993,
      "num_tokens": 63721808.0,
      "step": 36730
    },
    {
      "entropy": 5.398809766769409,
      "epoch": 2.858120988134604,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004182990649079289,
      "loss": 5.0068,
      "mean_token_accuracy": 0.2005328431725502,
      "num_tokens": 63730226.0,
      "step": 36735
    },
    {
      "entropy": 5.384431266784668,
      "epoch": 2.8585100175063216,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004182776962380577,
      "loss": 4.9295,
      "mean_token_accuracy": 0.20164803862571717,
      "num_tokens": 63738400.0,
      "step": 36740
    },
    {
      "entropy": 5.3479115009307865,
      "epoch": 2.8588990468780393,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041825632539417546,
      "loss": 4.9073,
      "mean_token_accuracy": 0.20507562458515166,
      "num_tokens": 63746385.0,
      "step": 36745
    },
    {
      "entropy": 5.383835172653198,
      "epoch": 2.859288076249757,
      "grad_norm": 1.125,
      "learning_rate": 0.0004182349523766065,
      "loss": 5.0599,
      "mean_token_accuracy": 0.19847557544708253,
      "num_tokens": 63755354.0,
      "step": 36750
    },
    {
      "entropy": 5.473566055297852,
      "epoch": 2.8596771056214743,
      "grad_norm": 1.125,
      "learning_rate": 0.00041821357718567514,
      "loss": 5.0736,
      "mean_token_accuracy": 0.19454247057437896,
      "num_tokens": 63764262.0,
      "step": 36755
    },
    {
      "entropy": 5.377267169952392,
      "epoch": 2.860066134993192,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004181921998217057,
      "loss": 4.9099,
      "mean_token_accuracy": 0.20494703501462935,
      "num_tokens": 63772770.0,
      "step": 36760
    },
    {
      "entropy": 5.405041599273682,
      "epoch": 2.8604551643649097,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004181708202850225,
      "loss": 4.9686,
      "mean_token_accuracy": 0.20490303933620452,
      "num_tokens": 63781353.0,
      "step": 36765
    },
    {
      "entropy": 5.4674512386322025,
      "epoch": 2.860844193736627,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041814943857595,
      "loss": 5.0831,
      "mean_token_accuracy": 0.1940636232495308,
      "num_tokens": 63789262.0,
      "step": 36770
    },
    {
      "entropy": 5.4633262157440186,
      "epoch": 2.8612332231083446,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004181280546948127,
      "loss": 5.1344,
      "mean_token_accuracy": 0.19183164536952974,
      "num_tokens": 63797792.0,
      "step": 36775
    },
    {
      "entropy": 5.421915674209595,
      "epoch": 2.8616222524800623,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004181066686419349,
      "loss": 4.8997,
      "mean_token_accuracy": 0.2105755552649498,
      "num_tokens": 63805362.0,
      "step": 36780
    },
    {
      "entropy": 5.3843225002288815,
      "epoch": 2.86201128185178,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041808528041764115,
      "loss": 4.9683,
      "mean_token_accuracy": 0.20286853462457657,
      "num_tokens": 63813942.0,
      "step": 36785
    },
    {
      "entropy": 5.287630224227906,
      "epoch": 2.8624003112234973,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004180638900222561,
      "loss": 4.8737,
      "mean_token_accuracy": 0.21257918775081636,
      "num_tokens": 63823077.0,
      "step": 36790
    },
    {
      "entropy": 5.306641340255737,
      "epoch": 2.862789340595215,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004180424974561042,
      "loss": 4.874,
      "mean_token_accuracy": 0.20539965033531188,
      "num_tokens": 63831048.0,
      "step": 36795
    },
    {
      "entropy": 5.428457307815552,
      "epoch": 2.8631783699669326,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041802110271951,
      "loss": 5.0315,
      "mean_token_accuracy": 0.191106478869915,
      "num_tokens": 63839449.0,
      "step": 36800
    },
    {
      "entropy": 5.3788567066192625,
      "epoch": 2.86356739933865,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004179997058127983,
      "loss": 4.9828,
      "mean_token_accuracy": 0.2013300836086273,
      "num_tokens": 63847945.0,
      "step": 36805
    },
    {
      "entropy": 5.312235975265503,
      "epoch": 2.8639564287103676,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004179783067362936,
      "loss": 4.9076,
      "mean_token_accuracy": 0.21026684939861298,
      "num_tokens": 63857334.0,
      "step": 36810
    },
    {
      "entropy": 5.366594314575195,
      "epoch": 2.8643454580820853,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004179569054903207,
      "loss": 5.0197,
      "mean_token_accuracy": 0.19128962457180024,
      "num_tokens": 63866272.0,
      "step": 36815
    },
    {
      "entropy": 5.4136659622192385,
      "epoch": 2.864734487453803,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041793550207520436,
      "loss": 4.943,
      "mean_token_accuracy": 0.20066686123609542,
      "num_tokens": 63874798.0,
      "step": 36820
    },
    {
      "entropy": 5.425258827209473,
      "epoch": 2.8651235168255202,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004179140964912692,
      "loss": 4.9309,
      "mean_token_accuracy": 0.20420774221420288,
      "num_tokens": 63883106.0,
      "step": 36825
    },
    {
      "entropy": 5.3564714908599855,
      "epoch": 2.865512546197238,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004178926887388401,
      "loss": 4.9551,
      "mean_token_accuracy": 0.20449693500995636,
      "num_tokens": 63892172.0,
      "step": 36830
    },
    {
      "entropy": 5.411362171173096,
      "epoch": 2.865901575568955,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041787127881824193,
      "loss": 5.0768,
      "mean_token_accuracy": 0.19120672196149827,
      "num_tokens": 63900436.0,
      "step": 36835
    },
    {
      "entropy": 5.477596092224121,
      "epoch": 2.866290604940673,
      "grad_norm": 1.125,
      "learning_rate": 0.00041784986672979945,
      "loss": 5.0734,
      "mean_token_accuracy": 0.19527046382427216,
      "num_tokens": 63908937.0,
      "step": 36840
    },
    {
      "entropy": 5.392542839050293,
      "epoch": 2.8666796343123906,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041782845247383767,
      "loss": 4.992,
      "mean_token_accuracy": 0.20587743520736695,
      "num_tokens": 63916350.0,
      "step": 36845
    },
    {
      "entropy": 5.351164865493774,
      "epoch": 2.8670686636841083,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041780703605068145,
      "loss": 4.98,
      "mean_token_accuracy": 0.19533114433288573,
      "num_tokens": 63925227.0,
      "step": 36850
    },
    {
      "entropy": 5.39546046257019,
      "epoch": 2.867457693055826,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004177856174606558,
      "loss": 4.968,
      "mean_token_accuracy": 0.19777505844831467,
      "num_tokens": 63934075.0,
      "step": 36855
    },
    {
      "entropy": 5.398231506347656,
      "epoch": 2.8678467224275432,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004177641967040856,
      "loss": 5.0197,
      "mean_token_accuracy": 0.19691864401102066,
      "num_tokens": 63942554.0,
      "step": 36860
    },
    {
      "entropy": 5.332274866104126,
      "epoch": 2.868235751799261,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041774277378129595,
      "loss": 4.9251,
      "mean_token_accuracy": 0.20171054899692537,
      "num_tokens": 63951850.0,
      "step": 36865
    },
    {
      "entropy": 5.448313283920288,
      "epoch": 2.868624781170978,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041772134869261186,
      "loss": 4.9872,
      "mean_token_accuracy": 0.19829238057136536,
      "num_tokens": 63960709.0,
      "step": 36870
    },
    {
      "entropy": 5.361208486557007,
      "epoch": 2.869013810542696,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004176999214383585,
      "loss": 4.9634,
      "mean_token_accuracy": 0.2049262449145317,
      "num_tokens": 63969859.0,
      "step": 36875
    },
    {
      "entropy": 5.397562789916992,
      "epoch": 2.8694028399144136,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041767849201886103,
      "loss": 4.9465,
      "mean_token_accuracy": 0.1991907238960266,
      "num_tokens": 63977524.0,
      "step": 36880
    },
    {
      "entropy": 5.28074197769165,
      "epoch": 2.8697918692861313,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004176570604344445,
      "loss": 4.842,
      "mean_token_accuracy": 0.20927724391222,
      "num_tokens": 63985938.0,
      "step": 36885
    },
    {
      "entropy": 5.2829827785491945,
      "epoch": 2.8701808986578485,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041763562668543415,
      "loss": 4.9888,
      "mean_token_accuracy": 0.20340211391448976,
      "num_tokens": 63993864.0,
      "step": 36890
    },
    {
      "entropy": 5.466585874557495,
      "epoch": 2.8705699280295662,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004176141907721552,
      "loss": 5.0395,
      "mean_token_accuracy": 0.1997930407524109,
      "num_tokens": 64003479.0,
      "step": 36895
    },
    {
      "entropy": 5.46111536026001,
      "epoch": 2.870958957401284,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004175927526949329,
      "loss": 4.9907,
      "mean_token_accuracy": 0.20680253505706786,
      "num_tokens": 64012492.0,
      "step": 36900
    },
    {
      "entropy": 5.452385711669922,
      "epoch": 2.871347986773001,
      "grad_norm": 1.125,
      "learning_rate": 0.0004175713124540925,
      "loss": 5.0731,
      "mean_token_accuracy": 0.18616908937692642,
      "num_tokens": 64021756.0,
      "step": 36905
    },
    {
      "entropy": 5.422436904907227,
      "epoch": 2.871737016144719,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041754987004995935,
      "loss": 5.0395,
      "mean_token_accuracy": 0.19796013385057448,
      "num_tokens": 64030351.0,
      "step": 36910
    },
    {
      "entropy": 5.3681854724884035,
      "epoch": 2.8721260455164366,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041752842548285887,
      "loss": 4.884,
      "mean_token_accuracy": 0.2010513111948967,
      "num_tokens": 64039523.0,
      "step": 36915
    },
    {
      "entropy": 5.315818881988525,
      "epoch": 2.8725150748881543,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004175069787531163,
      "loss": 4.8977,
      "mean_token_accuracy": 0.20945239812135696,
      "num_tokens": 64048264.0,
      "step": 36920
    },
    {
      "entropy": 5.395058345794678,
      "epoch": 2.8729041042598715,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041748552986105726,
      "loss": 5.0342,
      "mean_token_accuracy": 0.1934139460325241,
      "num_tokens": 64057013.0,
      "step": 36925
    },
    {
      "entropy": 5.353413343429565,
      "epoch": 2.8732931336315892,
      "grad_norm": 1.125,
      "learning_rate": 0.000417464078807007,
      "loss": 4.9198,
      "mean_token_accuracy": 0.20274354368448258,
      "num_tokens": 64065962.0,
      "step": 36930
    },
    {
      "entropy": 5.358920669555664,
      "epoch": 2.8736821630033065,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004174426255912912,
      "loss": 4.9813,
      "mean_token_accuracy": 0.2022714853286743,
      "num_tokens": 64074493.0,
      "step": 36935
    },
    {
      "entropy": 5.422954511642456,
      "epoch": 2.874071192375024,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004174211702142352,
      "loss": 5.0682,
      "mean_token_accuracy": 0.1948063537478447,
      "num_tokens": 64083351.0,
      "step": 36940
    },
    {
      "entropy": 5.428135108947754,
      "epoch": 2.874460221746742,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041739971267616466,
      "loss": 4.9508,
      "mean_token_accuracy": 0.20139007717370988,
      "num_tokens": 64091775.0,
      "step": 36945
    },
    {
      "entropy": 5.276426696777344,
      "epoch": 2.8748492511184596,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004173782529774051,
      "loss": 4.9694,
      "mean_token_accuracy": 0.20655950605869294,
      "num_tokens": 64100684.0,
      "step": 36950
    },
    {
      "entropy": 5.314303827285767,
      "epoch": 2.8752382804901773,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041735679111828223,
      "loss": 4.9606,
      "mean_token_accuracy": 0.20499653965234757,
      "num_tokens": 64109102.0,
      "step": 36955
    },
    {
      "entropy": 5.3591780185699465,
      "epoch": 2.8756273098618945,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041733532709912157,
      "loss": 4.9411,
      "mean_token_accuracy": 0.2022224098443985,
      "num_tokens": 64118341.0,
      "step": 36960
    },
    {
      "entropy": 5.357262897491455,
      "epoch": 2.876016339233612,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041731386092024893,
      "loss": 4.8648,
      "mean_token_accuracy": 0.20841473937034607,
      "num_tokens": 64126647.0,
      "step": 36965
    },
    {
      "entropy": 5.2909399509429935,
      "epoch": 2.8764053686053295,
      "grad_norm": 1.125,
      "learning_rate": 0.00041729239258198996,
      "loss": 4.7731,
      "mean_token_accuracy": 0.20860371589660645,
      "num_tokens": 64135534.0,
      "step": 36970
    },
    {
      "entropy": 5.3558337688446045,
      "epoch": 2.876794397977047,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041727092208467036,
      "loss": 5.0212,
      "mean_token_accuracy": 0.1969813048839569,
      "num_tokens": 64144108.0,
      "step": 36975
    },
    {
      "entropy": 5.316230964660645,
      "epoch": 2.877183427348765,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041724944942861603,
      "loss": 4.8051,
      "mean_token_accuracy": 0.210329869389534,
      "num_tokens": 64152599.0,
      "step": 36980
    },
    {
      "entropy": 5.370316839218139,
      "epoch": 2.8775724567204826,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041722797461415267,
      "loss": 4.9447,
      "mean_token_accuracy": 0.19478170871734618,
      "num_tokens": 64161733.0,
      "step": 36985
    },
    {
      "entropy": 5.360448884963989,
      "epoch": 2.8779614860922,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004172064976416062,
      "loss": 4.9284,
      "mean_token_accuracy": 0.20668711215257646,
      "num_tokens": 64171045.0,
      "step": 36990
    },
    {
      "entropy": 5.38868727684021,
      "epoch": 2.8783505154639175,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004171850185113024,
      "loss": 4.9113,
      "mean_token_accuracy": 0.20555466562509536,
      "num_tokens": 64179725.0,
      "step": 36995
    },
    {
      "entropy": 5.32879228591919,
      "epoch": 2.878739544835635,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041716353722356735,
      "loss": 5.0013,
      "mean_token_accuracy": 0.19792623519897462,
      "num_tokens": 64188799.0,
      "step": 37000
    },
    {
      "entropy": 5.420260858535767,
      "epoch": 2.8791285742073525,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004171420537787269,
      "loss": 5.0129,
      "mean_token_accuracy": 0.19890602678060532,
      "num_tokens": 64197809.0,
      "step": 37005
    },
    {
      "entropy": 5.413884067535401,
      "epoch": 2.87951760357907,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041712056817710697,
      "loss": 5.0406,
      "mean_token_accuracy": 0.18851754516363145,
      "num_tokens": 64207113.0,
      "step": 37010
    },
    {
      "entropy": 5.432380294799804,
      "epoch": 2.879906632950788,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004170990804190337,
      "loss": 4.9854,
      "mean_token_accuracy": 0.20316566973924638,
      "num_tokens": 64216846.0,
      "step": 37015
    },
    {
      "entropy": 5.331421279907227,
      "epoch": 2.8802956623225056,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041707759050483304,
      "loss": 4.9355,
      "mean_token_accuracy": 0.2037827715277672,
      "num_tokens": 64225348.0,
      "step": 37020
    },
    {
      "entropy": 5.347870969772339,
      "epoch": 2.880684691694223,
      "grad_norm": 1.15625,
      "learning_rate": 0.000417056098434831,
      "loss": 4.9465,
      "mean_token_accuracy": 0.20148575752973558,
      "num_tokens": 64234065.0,
      "step": 37025
    },
    {
      "entropy": 5.399331045150757,
      "epoch": 2.8810737210659405,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004170346042093538,
      "loss": 4.9855,
      "mean_token_accuracy": 0.2000252291560173,
      "num_tokens": 64242484.0,
      "step": 37030
    },
    {
      "entropy": 5.27199387550354,
      "epoch": 2.881462750437658,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004170131078287276,
      "loss": 4.8371,
      "mean_token_accuracy": 0.21090682595968246,
      "num_tokens": 64251056.0,
      "step": 37035
    },
    {
      "entropy": 5.3463311195373535,
      "epoch": 2.8818517798093755,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004169916092932785,
      "loss": 4.934,
      "mean_token_accuracy": 0.20172853916883468,
      "num_tokens": 64258961.0,
      "step": 37040
    },
    {
      "entropy": 5.315022754669189,
      "epoch": 2.882240809181093,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004169701086033327,
      "loss": 4.9023,
      "mean_token_accuracy": 0.2012649491429329,
      "num_tokens": 64268133.0,
      "step": 37045
    },
    {
      "entropy": 5.422477102279663,
      "epoch": 2.882629838552811,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041694860575921654,
      "loss": 4.9993,
      "mean_token_accuracy": 0.1935962527990341,
      "num_tokens": 64276487.0,
      "step": 37050
    },
    {
      "entropy": 5.4019567489624025,
      "epoch": 2.8830188679245285,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041692710076125615,
      "loss": 4.9073,
      "mean_token_accuracy": 0.20424478650093078,
      "num_tokens": 64284983.0,
      "step": 37055
    },
    {
      "entropy": 5.393069553375244,
      "epoch": 2.883407897296246,
      "grad_norm": 1.078125,
      "learning_rate": 0.000416905593609778,
      "loss": 4.9842,
      "mean_token_accuracy": 0.20398012101650237,
      "num_tokens": 64293875.0,
      "step": 37060
    },
    {
      "entropy": 5.4037182331085205,
      "epoch": 2.8837969266679635,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041688408430510827,
      "loss": 4.9563,
      "mean_token_accuracy": 0.20789678692817687,
      "num_tokens": 64302628.0,
      "step": 37065
    },
    {
      "entropy": 5.424954032897949,
      "epoch": 2.8841859560396808,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004168625728475734,
      "loss": 5.0062,
      "mean_token_accuracy": 0.20004134774208068,
      "num_tokens": 64311024.0,
      "step": 37070
    },
    {
      "entropy": 5.325345420837403,
      "epoch": 2.8845749854113985,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041684105923749977,
      "loss": 4.9737,
      "mean_token_accuracy": 0.20403173118829726,
      "num_tokens": 64318836.0,
      "step": 37075
    },
    {
      "entropy": 5.3220258235931395,
      "epoch": 2.884964014783116,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004168195434752139,
      "loss": 5.0071,
      "mean_token_accuracy": 0.2015474632382393,
      "num_tokens": 64326983.0,
      "step": 37080
    },
    {
      "entropy": 5.396118021011352,
      "epoch": 2.885353044154834,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041679802556104207,
      "loss": 4.9975,
      "mean_token_accuracy": 0.19967755377292634,
      "num_tokens": 64335304.0,
      "step": 37085
    },
    {
      "entropy": 5.399141073226929,
      "epoch": 2.885742073526551,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041677650549531097,
      "loss": 4.8666,
      "mean_token_accuracy": 0.2068141371011734,
      "num_tokens": 64343515.0,
      "step": 37090
    },
    {
      "entropy": 5.34257435798645,
      "epoch": 2.886131102898269,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041675498327834706,
      "loss": 4.9584,
      "mean_token_accuracy": 0.19879851788282393,
      "num_tokens": 64351936.0,
      "step": 37095
    },
    {
      "entropy": 5.402570629119873,
      "epoch": 2.8865201322699865,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041673345891047686,
      "loss": 5.0234,
      "mean_token_accuracy": 0.2007183775305748,
      "num_tokens": 64360332.0,
      "step": 37100
    },
    {
      "entropy": 5.3923859119415285,
      "epoch": 2.8869091616417037,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004167119323920271,
      "loss": 4.9691,
      "mean_token_accuracy": 0.19890433102846145,
      "num_tokens": 64369615.0,
      "step": 37105
    },
    {
      "entropy": 5.368736124038696,
      "epoch": 2.8872981910134214,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004166904037233243,
      "loss": 4.8921,
      "mean_token_accuracy": 0.20254307687282563,
      "num_tokens": 64378355.0,
      "step": 37110
    },
    {
      "entropy": 5.264634513854981,
      "epoch": 2.887687220385139,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041666887290469506,
      "loss": 4.8627,
      "mean_token_accuracy": 0.20478395223617554,
      "num_tokens": 64386120.0,
      "step": 37115
    },
    {
      "entropy": 5.2881354808807375,
      "epoch": 2.888076249756857,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004166473399364663,
      "loss": 4.9679,
      "mean_token_accuracy": 0.20222768038511277,
      "num_tokens": 64395153.0,
      "step": 37120
    },
    {
      "entropy": 5.459886646270752,
      "epoch": 2.888465279128574,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041662580481896446,
      "loss": 4.9995,
      "mean_token_accuracy": 0.19711934477090837,
      "num_tokens": 64404266.0,
      "step": 37125
    },
    {
      "entropy": 5.361165332794189,
      "epoch": 2.888854308500292,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004166042675525165,
      "loss": 4.9632,
      "mean_token_accuracy": 0.19904866218566894,
      "num_tokens": 64412999.0,
      "step": 37130
    },
    {
      "entropy": 5.350707769393921,
      "epoch": 2.8892433378720095,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041658272813744924,
      "loss": 4.9473,
      "mean_token_accuracy": 0.20951907187700272,
      "num_tokens": 64421561.0,
      "step": 37135
    },
    {
      "entropy": 5.400559282302856,
      "epoch": 2.8896323672437267,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004165611865740894,
      "loss": 5.0228,
      "mean_token_accuracy": 0.19931823015213013,
      "num_tokens": 64430254.0,
      "step": 37140
    },
    {
      "entropy": 5.387171125411987,
      "epoch": 2.8900213966154444,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004165396428627638,
      "loss": 4.9157,
      "mean_token_accuracy": 0.20624625235795974,
      "num_tokens": 64438452.0,
      "step": 37145
    },
    {
      "entropy": 5.486778736114502,
      "epoch": 2.890410425987162,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041651809700379947,
      "loss": 5.0704,
      "mean_token_accuracy": 0.1996098443865776,
      "num_tokens": 64447442.0,
      "step": 37150
    },
    {
      "entropy": 5.399266576766967,
      "epoch": 2.89079945535888,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041649654899752326,
      "loss": 5.0458,
      "mean_token_accuracy": 0.19848165959119796,
      "num_tokens": 64456179.0,
      "step": 37155
    },
    {
      "entropy": 5.510623598098755,
      "epoch": 2.891188484730597,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004164749988442621,
      "loss": 5.0963,
      "mean_token_accuracy": 0.19883234053850174,
      "num_tokens": 64464498.0,
      "step": 37160
    },
    {
      "entropy": 5.405330562591553,
      "epoch": 2.891577514102315,
      "grad_norm": 1.15625,
      "learning_rate": 0.000416453446544343,
      "loss": 4.9214,
      "mean_token_accuracy": 0.19809215068817138,
      "num_tokens": 64473047.0,
      "step": 37165
    },
    {
      "entropy": 5.304905891418457,
      "epoch": 2.891966543474032,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004164318920980931,
      "loss": 4.8519,
      "mean_token_accuracy": 0.21838342845439912,
      "num_tokens": 64482338.0,
      "step": 37170
    },
    {
      "entropy": 5.267415761947632,
      "epoch": 2.8923555728457497,
      "grad_norm": 1.125,
      "learning_rate": 0.0004164103355058393,
      "loss": 4.868,
      "mean_token_accuracy": 0.20211563110351563,
      "num_tokens": 64490611.0,
      "step": 37175
    },
    {
      "entropy": 5.334226512908936,
      "epoch": 2.8927446022174674,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004163887767679087,
      "loss": 4.896,
      "mean_token_accuracy": 0.20980583429336547,
      "num_tokens": 64499180.0,
      "step": 37180
    },
    {
      "entropy": 5.402279281616211,
      "epoch": 2.893133631589185,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004163672158846285,
      "loss": 4.9983,
      "mean_token_accuracy": 0.19483935534954072,
      "num_tokens": 64508664.0,
      "step": 37185
    },
    {
      "entropy": 5.520532321929932,
      "epoch": 2.8935226609609024,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041634565285632573,
      "loss": 5.1212,
      "mean_token_accuracy": 0.1844704657793045,
      "num_tokens": 64518059.0,
      "step": 37190
    },
    {
      "entropy": 5.45718789100647,
      "epoch": 2.89391169033262,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041632408768332776,
      "loss": 5.0815,
      "mean_token_accuracy": 0.19127696007490158,
      "num_tokens": 64526616.0,
      "step": 37195
    },
    {
      "entropy": 5.467284393310547,
      "epoch": 2.8943007197043378,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041630252036596165,
      "loss": 5.0697,
      "mean_token_accuracy": 0.19320171028375627,
      "num_tokens": 64535008.0,
      "step": 37200
    },
    {
      "entropy": 5.358973836898803,
      "epoch": 2.894689749076055,
      "grad_norm": 1.125,
      "learning_rate": 0.0004162809509045547,
      "loss": 4.8741,
      "mean_token_accuracy": 0.2024504601955414,
      "num_tokens": 64543741.0,
      "step": 37205
    },
    {
      "entropy": 5.269455194473267,
      "epoch": 2.8950787784477727,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041625937929943424,
      "loss": 4.81,
      "mean_token_accuracy": 0.21030043661594391,
      "num_tokens": 64552409.0,
      "step": 37210
    },
    {
      "entropy": 5.376355838775635,
      "epoch": 2.8954678078194904,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00041623780555092747,
      "loss": 5.0671,
      "mean_token_accuracy": 0.18901106864213943,
      "num_tokens": 64561308.0,
      "step": 37215
    },
    {
      "entropy": 5.373002243041992,
      "epoch": 2.895856837191208,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004162162296593618,
      "loss": 4.8998,
      "mean_token_accuracy": 0.2083124950528145,
      "num_tokens": 64569671.0,
      "step": 37220
    },
    {
      "entropy": 5.454429388046265,
      "epoch": 2.8962458665629254,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041619465162506466,
      "loss": 5.0311,
      "mean_token_accuracy": 0.19633175432682037,
      "num_tokens": 64578186.0,
      "step": 37225
    },
    {
      "entropy": 5.350834655761719,
      "epoch": 2.896634895934643,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004161730714483633,
      "loss": 4.9484,
      "mean_token_accuracy": 0.20673643946647643,
      "num_tokens": 64586703.0,
      "step": 37230
    },
    {
      "entropy": 5.345397567749023,
      "epoch": 2.8970239253063608,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004161514891295854,
      "loss": 4.9826,
      "mean_token_accuracy": 0.19618725031614304,
      "num_tokens": 64595119.0,
      "step": 37235
    },
    {
      "entropy": 5.432996368408203,
      "epoch": 2.897412954678078,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041612990466905825,
      "loss": 5.0147,
      "mean_token_accuracy": 0.19914722740650176,
      "num_tokens": 64604312.0,
      "step": 37240
    },
    {
      "entropy": 5.409211730957031,
      "epoch": 2.8978019840497957,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041610831806710954,
      "loss": 4.9512,
      "mean_token_accuracy": 0.20491380840539933,
      "num_tokens": 64613088.0,
      "step": 37245
    },
    {
      "entropy": 5.46361289024353,
      "epoch": 2.8981910134215134,
      "grad_norm": 1.125,
      "learning_rate": 0.00041608672932406666,
      "loss": 5.1063,
      "mean_token_accuracy": 0.1960842028260231,
      "num_tokens": 64621659.0,
      "step": 37250
    },
    {
      "entropy": 5.405196619033814,
      "epoch": 2.898580042793231,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041606513844025716,
      "loss": 4.9216,
      "mean_token_accuracy": 0.20775201469659804,
      "num_tokens": 64630436.0,
      "step": 37255
    },
    {
      "entropy": 5.363085556030273,
      "epoch": 2.8989690721649484,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004160435454160087,
      "loss": 4.9635,
      "mean_token_accuracy": 0.20452069789171218,
      "num_tokens": 64639836.0,
      "step": 37260
    },
    {
      "entropy": 5.357938528060913,
      "epoch": 2.899358101536666,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000416021950251649,
      "loss": 4.9388,
      "mean_token_accuracy": 0.19422908127307892,
      "num_tokens": 64648941.0,
      "step": 37265
    },
    {
      "entropy": 5.331830406188965,
      "epoch": 2.8997471309083833,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004160003529475057,
      "loss": 5.007,
      "mean_token_accuracy": 0.19518893510103225,
      "num_tokens": 64657632.0,
      "step": 37270
    },
    {
      "entropy": 5.411461114883423,
      "epoch": 2.900136160280101,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004159787535039064,
      "loss": 5.0154,
      "mean_token_accuracy": 0.1925535410642624,
      "num_tokens": 64665733.0,
      "step": 37275
    },
    {
      "entropy": 5.328467178344726,
      "epoch": 2.9005251896518187,
      "grad_norm": 1.125,
      "learning_rate": 0.0004159571519211789,
      "loss": 4.8844,
      "mean_token_accuracy": 0.21159991323947908,
      "num_tokens": 64673873.0,
      "step": 37280
    },
    {
      "entropy": 5.364060115814209,
      "epoch": 2.9009142190235364,
      "grad_norm": 1.1875,
      "learning_rate": 0.000415935548199651,
      "loss": 4.9918,
      "mean_token_accuracy": 0.1946658417582512,
      "num_tokens": 64682282.0,
      "step": 37285
    },
    {
      "entropy": 5.306412220001221,
      "epoch": 2.901303248395254,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004159139423396504,
      "loss": 4.9168,
      "mean_token_accuracy": 0.20639852583408355,
      "num_tokens": 64690179.0,
      "step": 37290
    },
    {
      "entropy": 5.376539421081543,
      "epoch": 2.9016922777669714,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004158923343415051,
      "loss": 5.0582,
      "mean_token_accuracy": 0.2010296329855919,
      "num_tokens": 64698935.0,
      "step": 37295
    },
    {
      "entropy": 5.3281032085418705,
      "epoch": 2.902081307138689,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004158707242055429,
      "loss": 4.8948,
      "mean_token_accuracy": 0.2016885757446289,
      "num_tokens": 64708328.0,
      "step": 37300
    },
    {
      "entropy": 5.452783203125,
      "epoch": 2.9024703365104063,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041584911193209167,
      "loss": 5.0436,
      "mean_token_accuracy": 0.19289521276950836,
      "num_tokens": 64717003.0,
      "step": 37305
    },
    {
      "entropy": 5.343239879608154,
      "epoch": 2.902859365882124,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004158274975214794,
      "loss": 4.9347,
      "mean_token_accuracy": 0.20231957137584686,
      "num_tokens": 64725951.0,
      "step": 37310
    },
    {
      "entropy": 5.349320602416992,
      "epoch": 2.9032483952538417,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041580588097403384,
      "loss": 4.9551,
      "mean_token_accuracy": 0.19835673719644548,
      "num_tokens": 64733949.0,
      "step": 37315
    },
    {
      "entropy": 5.400386238098145,
      "epoch": 2.9036374246255594,
      "grad_norm": 1.125,
      "learning_rate": 0.00041578426229008325,
      "loss": 5.0232,
      "mean_token_accuracy": 0.2016855537891388,
      "num_tokens": 64742729.0,
      "step": 37320
    },
    {
      "entropy": 5.38262152671814,
      "epoch": 2.9040264539972767,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041576264146995557,
      "loss": 4.965,
      "mean_token_accuracy": 0.19412308633327485,
      "num_tokens": 64751655.0,
      "step": 37325
    },
    {
      "entropy": 5.411255264282227,
      "epoch": 2.9044154833689944,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041574101851397883,
      "loss": 4.957,
      "mean_token_accuracy": 0.20771207809448242,
      "num_tokens": 64760019.0,
      "step": 37330
    },
    {
      "entropy": 5.358091354370117,
      "epoch": 2.904804512740712,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004157193934224811,
      "loss": 5.0186,
      "mean_token_accuracy": 0.20424339920282364,
      "num_tokens": 64768528.0,
      "step": 37335
    },
    {
      "entropy": 5.413808584213257,
      "epoch": 2.9051935421124293,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004156977661957906,
      "loss": 4.9997,
      "mean_token_accuracy": 0.19535810947418214,
      "num_tokens": 64776903.0,
      "step": 37340
    },
    {
      "entropy": 5.457866764068603,
      "epoch": 2.905582571484147,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004156761368342355,
      "loss": 5.0795,
      "mean_token_accuracy": 0.1968183770775795,
      "num_tokens": 64785361.0,
      "step": 37345
    },
    {
      "entropy": 5.444344472885132,
      "epoch": 2.9059716008558647,
      "grad_norm": 1.125,
      "learning_rate": 0.00041565450533814384,
      "loss": 5.0061,
      "mean_token_accuracy": 0.2005004107952118,
      "num_tokens": 64793720.0,
      "step": 37350
    },
    {
      "entropy": 5.3388941287994385,
      "epoch": 2.9063606302275824,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041563287170784396,
      "loss": 4.8984,
      "mean_token_accuracy": 0.20896690040826799,
      "num_tokens": 64802049.0,
      "step": 37355
    },
    {
      "entropy": 5.381278848648071,
      "epoch": 2.9067496595992997,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004156112359436641,
      "loss": 4.9516,
      "mean_token_accuracy": 0.2014062836766243,
      "num_tokens": 64810754.0,
      "step": 37360
    },
    {
      "entropy": 5.414401483535767,
      "epoch": 2.9071386889710173,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041558959804593255,
      "loss": 5.0502,
      "mean_token_accuracy": 0.19641206413507462,
      "num_tokens": 64819466.0,
      "step": 37365
    },
    {
      "entropy": 5.393006086349487,
      "epoch": 2.9075277183427346,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041556795801497766,
      "loss": 4.9905,
      "mean_token_accuracy": 0.19391808211803435,
      "num_tokens": 64828135.0,
      "step": 37370
    },
    {
      "entropy": 5.413118362426758,
      "epoch": 2.9079167477144523,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004155463158511277,
      "loss": 5.0496,
      "mean_token_accuracy": 0.19979741871356965,
      "num_tokens": 64836542.0,
      "step": 37375
    },
    {
      "entropy": 5.364656162261963,
      "epoch": 2.90830577708617,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041552467155471105,
      "loss": 4.8949,
      "mean_token_accuracy": 0.2096607968211174,
      "num_tokens": 64844470.0,
      "step": 37380
    },
    {
      "entropy": 5.364316034317016,
      "epoch": 2.9086948064578877,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004155030251260563,
      "loss": 4.9695,
      "mean_token_accuracy": 0.19923162162303926,
      "num_tokens": 64854011.0,
      "step": 37385
    },
    {
      "entropy": 5.469547843933105,
      "epoch": 2.9090838358296054,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041548137656549167,
      "loss": 5.0101,
      "mean_token_accuracy": 0.20190962851047517,
      "num_tokens": 64862349.0,
      "step": 37390
    },
    {
      "entropy": 5.43422646522522,
      "epoch": 2.9094728652013226,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004154597258733457,
      "loss": 4.9709,
      "mean_token_accuracy": 0.19839250445365905,
      "num_tokens": 64870634.0,
      "step": 37395
    },
    {
      "entropy": 5.3955028533935545,
      "epoch": 2.9098618945730403,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004154380730499472,
      "loss": 5.0244,
      "mean_token_accuracy": 0.19922563284635544,
      "num_tokens": 64879870.0,
      "step": 37400
    },
    {
      "entropy": 5.352818775177002,
      "epoch": 2.9102509239447576,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041541641809562425,
      "loss": 4.92,
      "mean_token_accuracy": 0.20635703802108765,
      "num_tokens": 64888650.0,
      "step": 37405
    },
    {
      "entropy": 5.390845108032226,
      "epoch": 2.9106399533164753,
      "grad_norm": 1.125,
      "learning_rate": 0.0004153947610107057,
      "loss": 4.9366,
      "mean_token_accuracy": 0.2032993569970131,
      "num_tokens": 64896788.0,
      "step": 37410
    },
    {
      "entropy": 5.33636507987976,
      "epoch": 2.911028982688193,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041537310179552017,
      "loss": 4.8744,
      "mean_token_accuracy": 0.21089080423116685,
      "num_tokens": 64905134.0,
      "step": 37415
    },
    {
      "entropy": 5.398211431503296,
      "epoch": 2.9114180120599107,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004153514404503962,
      "loss": 5.0227,
      "mean_token_accuracy": 0.19696078449487686,
      "num_tokens": 64913662.0,
      "step": 37420
    },
    {
      "entropy": 5.419208908081055,
      "epoch": 2.911807041431628,
      "grad_norm": 1.125,
      "learning_rate": 0.0004153297769756625,
      "loss": 4.9301,
      "mean_token_accuracy": 0.20317360311746596,
      "num_tokens": 64922159.0,
      "step": 37425
    },
    {
      "entropy": 5.394416046142578,
      "epoch": 2.9121960708033456,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004153081113716478,
      "loss": 4.9205,
      "mean_token_accuracy": 0.20740756243467331,
      "num_tokens": 64930672.0,
      "step": 37430
    },
    {
      "entropy": 5.348020076751709,
      "epoch": 2.9125851001750633,
      "grad_norm": 1.125,
      "learning_rate": 0.00041528644363868087,
      "loss": 4.9662,
      "mean_token_accuracy": 0.19713263660669328,
      "num_tokens": 64940424.0,
      "step": 37435
    },
    {
      "entropy": 5.448840856552124,
      "epoch": 2.9129741295467806,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004152647737770904,
      "loss": 5.1014,
      "mean_token_accuracy": 0.19356604665517807,
      "num_tokens": 64949510.0,
      "step": 37440
    },
    {
      "entropy": 5.372235345840454,
      "epoch": 2.9133631589184983,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004152431017872053,
      "loss": 4.9177,
      "mean_token_accuracy": 0.20535041391849518,
      "num_tokens": 64958240.0,
      "step": 37445
    },
    {
      "entropy": 5.351169204711914,
      "epoch": 2.913752188290216,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041522142766935426,
      "loss": 4.9136,
      "mean_token_accuracy": 0.20330470353364943,
      "num_tokens": 64967071.0,
      "step": 37450
    },
    {
      "entropy": 5.239471054077148,
      "epoch": 2.9141412176619337,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041519975142386627,
      "loss": 4.9341,
      "mean_token_accuracy": 0.20740776509046555,
      "num_tokens": 64974949.0,
      "step": 37455
    },
    {
      "entropy": 5.40181303024292,
      "epoch": 2.914530247033651,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004151780730510702,
      "loss": 4.9634,
      "mean_token_accuracy": 0.19663360416889192,
      "num_tokens": 64983510.0,
      "step": 37460
    },
    {
      "entropy": 5.521481466293335,
      "epoch": 2.9149192764053686,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041515639255129504,
      "loss": 5.125,
      "mean_token_accuracy": 0.19464604556560516,
      "num_tokens": 64993055.0,
      "step": 37465
    },
    {
      "entropy": 5.414715576171875,
      "epoch": 2.9153083057770863,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004151347099248697,
      "loss": 4.9517,
      "mean_token_accuracy": 0.19747292399406433,
      "num_tokens": 65002406.0,
      "step": 37470
    },
    {
      "entropy": 5.344550848007202,
      "epoch": 2.9156973351488036,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004151130251721231,
      "loss": 5.0652,
      "mean_token_accuracy": 0.19346767365932466,
      "num_tokens": 65011949.0,
      "step": 37475
    },
    {
      "entropy": 5.5293275833129885,
      "epoch": 2.9160863645205213,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041509133829338454,
      "loss": 5.1143,
      "mean_token_accuracy": 0.19104818254709244,
      "num_tokens": 65020215.0,
      "step": 37480
    },
    {
      "entropy": 5.378177356719971,
      "epoch": 2.916475393892239,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004150696492889827,
      "loss": 4.9321,
      "mean_token_accuracy": 0.2071759045124054,
      "num_tokens": 65028423.0,
      "step": 37485
    },
    {
      "entropy": 5.286511945724487,
      "epoch": 2.9168644232639567,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041504795815924697,
      "loss": 4.9084,
      "mean_token_accuracy": 0.20513925999403,
      "num_tokens": 65037804.0,
      "step": 37490
    },
    {
      "entropy": 5.353257465362549,
      "epoch": 2.917253452635674,
      "grad_norm": 1.125,
      "learning_rate": 0.00041502626490450635,
      "loss": 4.887,
      "mean_token_accuracy": 0.20681657791137695,
      "num_tokens": 65046389.0,
      "step": 37495
    },
    {
      "entropy": 5.305055713653564,
      "epoch": 2.9176424820073916,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004150045695250901,
      "loss": 4.9352,
      "mean_token_accuracy": 0.20170525163412095,
      "num_tokens": 65055499.0,
      "step": 37500
    },
    {
      "entropy": 5.314117765426635,
      "epoch": 2.918031511379109,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004149828720213272,
      "loss": 4.9003,
      "mean_token_accuracy": 0.21101456731557847,
      "num_tokens": 65063170.0,
      "step": 37505
    },
    {
      "entropy": 5.4019958019256595,
      "epoch": 2.9184205407508266,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004149611723935472,
      "loss": 4.985,
      "mean_token_accuracy": 0.2028233602643013,
      "num_tokens": 65071046.0,
      "step": 37510
    },
    {
      "entropy": 5.35547366142273,
      "epoch": 2.9188095701225443,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004149394706420791,
      "loss": 4.9781,
      "mean_token_accuracy": 0.2030557408928871,
      "num_tokens": 65079135.0,
      "step": 37515
    },
    {
      "entropy": 5.345481014251709,
      "epoch": 2.919198599494262,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041491776676725223,
      "loss": 4.9795,
      "mean_token_accuracy": 0.2001962348818779,
      "num_tokens": 65087861.0,
      "step": 37520
    },
    {
      "entropy": 5.339496850967407,
      "epoch": 2.9195876288659792,
      "grad_norm": 1.125,
      "learning_rate": 0.0004148960607693961,
      "loss": 4.8978,
      "mean_token_accuracy": 0.20552366822957993,
      "num_tokens": 65096561.0,
      "step": 37525
    },
    {
      "entropy": 5.398003673553466,
      "epoch": 2.919976658237697,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041487435264883974,
      "loss": 4.9797,
      "mean_token_accuracy": 0.2019769787788391,
      "num_tokens": 65105383.0,
      "step": 37530
    },
    {
      "entropy": 5.358969831466675,
      "epoch": 2.9203656876094146,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004148526424059128,
      "loss": 4.9348,
      "mean_token_accuracy": 0.1994137316942215,
      "num_tokens": 65113551.0,
      "step": 37535
    },
    {
      "entropy": 5.329093027114868,
      "epoch": 2.920754716981132,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004148309300409446,
      "loss": 4.8745,
      "mean_token_accuracy": 0.21325115859508514,
      "num_tokens": 65121599.0,
      "step": 37540
    },
    {
      "entropy": 5.417471265792846,
      "epoch": 2.9211437463528496,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004148092155542646,
      "loss": 5.0413,
      "mean_token_accuracy": 0.1939526692032814,
      "num_tokens": 65130881.0,
      "step": 37545
    },
    {
      "entropy": 5.348291873931885,
      "epoch": 2.9215327757245673,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004147874989462023,
      "loss": 4.9987,
      "mean_token_accuracy": 0.20159822404384614,
      "num_tokens": 65139233.0,
      "step": 37550
    },
    {
      "entropy": 5.3407673835754395,
      "epoch": 2.921921805096285,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004147657802170871,
      "loss": 4.9224,
      "mean_token_accuracy": 0.20072155743837355,
      "num_tokens": 65147738.0,
      "step": 37555
    },
    {
      "entropy": 5.403707122802734,
      "epoch": 2.922310834468002,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041474405936724873,
      "loss": 4.9644,
      "mean_token_accuracy": 0.19597127437591552,
      "num_tokens": 65156413.0,
      "step": 37560
    },
    {
      "entropy": 5.428874731063843,
      "epoch": 2.92269986383972,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004147223363970166,
      "loss": 4.9569,
      "mean_token_accuracy": 0.20637882351875306,
      "num_tokens": 65164614.0,
      "step": 37565
    },
    {
      "entropy": 5.408313322067261,
      "epoch": 2.9230888932114376,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004147006113067205,
      "loss": 4.9934,
      "mean_token_accuracy": 0.19283372163772583,
      "num_tokens": 65173159.0,
      "step": 37570
    },
    {
      "entropy": 5.336662626266479,
      "epoch": 2.923477922583155,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004146788840966901,
      "loss": 4.9569,
      "mean_token_accuracy": 0.19978110641241073,
      "num_tokens": 65181999.0,
      "step": 37575
    },
    {
      "entropy": 5.480209064483643,
      "epoch": 2.9238669519548726,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004146571547672548,
      "loss": 5.1495,
      "mean_token_accuracy": 0.18244872689247132,
      "num_tokens": 65191291.0,
      "step": 37580
    },
    {
      "entropy": 5.407338333129883,
      "epoch": 2.9242559813265903,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041463542331874443,
      "loss": 4.8855,
      "mean_token_accuracy": 0.20636311173439026,
      "num_tokens": 65198891.0,
      "step": 37585
    },
    {
      "entropy": 5.354855537414551,
      "epoch": 2.924645010698308,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041461368975148876,
      "loss": 4.987,
      "mean_token_accuracy": 0.2017035037279129,
      "num_tokens": 65208206.0,
      "step": 37590
    },
    {
      "entropy": 5.36383285522461,
      "epoch": 2.925034040070025,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041459195406581763,
      "loss": 4.9213,
      "mean_token_accuracy": 0.20416442304849625,
      "num_tokens": 65216384.0,
      "step": 37595
    },
    {
      "entropy": 5.366391277313232,
      "epoch": 2.925423069441743,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004145702162620608,
      "loss": 4.8911,
      "mean_token_accuracy": 0.2038661539554596,
      "num_tokens": 65224612.0,
      "step": 37600
    },
    {
      "entropy": 5.416227102279663,
      "epoch": 2.92581209881346,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041454847634054805,
      "loss": 4.9761,
      "mean_token_accuracy": 0.20649711787700653,
      "num_tokens": 65233389.0,
      "step": 37605
    },
    {
      "entropy": 5.435948514938355,
      "epoch": 2.926201128185178,
      "grad_norm": 1.125,
      "learning_rate": 0.00041452673430160923,
      "loss": 5.0364,
      "mean_token_accuracy": 0.20066017359495164,
      "num_tokens": 65242432.0,
      "step": 37610
    },
    {
      "entropy": 5.301699495315551,
      "epoch": 2.9265901575568956,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004145049901455744,
      "loss": 4.845,
      "mean_token_accuracy": 0.20612765550613404,
      "num_tokens": 65250161.0,
      "step": 37615
    },
    {
      "entropy": 5.362723779678345,
      "epoch": 2.9269791869286133,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041448324387277337,
      "loss": 4.9939,
      "mean_token_accuracy": 0.1973414048552513,
      "num_tokens": 65258660.0,
      "step": 37620
    },
    {
      "entropy": 5.439974403381347,
      "epoch": 2.927368216300331,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041446149548353606,
      "loss": 4.9563,
      "mean_token_accuracy": 0.20136415511369704,
      "num_tokens": 65267297.0,
      "step": 37625
    },
    {
      "entropy": 5.436471557617187,
      "epoch": 2.927757245672048,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004144397449781926,
      "loss": 5.0415,
      "mean_token_accuracy": 0.19909771978855134,
      "num_tokens": 65276064.0,
      "step": 37630
    },
    {
      "entropy": 5.425273370742798,
      "epoch": 2.928146275043766,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004144179923570729,
      "loss": 4.9982,
      "mean_token_accuracy": 0.20273493081331254,
      "num_tokens": 65284624.0,
      "step": 37635
    },
    {
      "entropy": 5.335787773132324,
      "epoch": 2.928535304415483,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004143962376205071,
      "loss": 4.9383,
      "mean_token_accuracy": 0.203912653028965,
      "num_tokens": 65293318.0,
      "step": 37640
    },
    {
      "entropy": 5.4403656959533695,
      "epoch": 2.928924333787201,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041437448076882526,
      "loss": 5.0628,
      "mean_token_accuracy": 0.19714379608631133,
      "num_tokens": 65301379.0,
      "step": 37645
    },
    {
      "entropy": 5.404918622970581,
      "epoch": 2.9293133631589185,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041435272180235757,
      "loss": 5.0431,
      "mean_token_accuracy": 0.19528946429491043,
      "num_tokens": 65309722.0,
      "step": 37650
    },
    {
      "entropy": 5.3694384574890135,
      "epoch": 2.9297023925306362,
      "grad_norm": 1.140625,
      "learning_rate": 0.000414330960721434,
      "loss": 4.9711,
      "mean_token_accuracy": 0.20003531873226166,
      "num_tokens": 65318616.0,
      "step": 37655
    },
    {
      "entropy": 5.42270302772522,
      "epoch": 2.9300914219023535,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041430919752638497,
      "loss": 4.9595,
      "mean_token_accuracy": 0.20351346135139464,
      "num_tokens": 65327046.0,
      "step": 37660
    },
    {
      "entropy": 5.330277347564698,
      "epoch": 2.930480451274071,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004142874322175406,
      "loss": 4.9387,
      "mean_token_accuracy": 0.19928795099258423,
      "num_tokens": 65335032.0,
      "step": 37665
    },
    {
      "entropy": 5.358166933059692,
      "epoch": 2.930869480645789,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041426566479523116,
      "loss": 5.0047,
      "mean_token_accuracy": 0.1915602207183838,
      "num_tokens": 65343573.0,
      "step": 37670
    },
    {
      "entropy": 5.352674674987793,
      "epoch": 2.931258510017506,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041424389525978683,
      "loss": 4.8486,
      "mean_token_accuracy": 0.2078844279050827,
      "num_tokens": 65352660.0,
      "step": 37675
    },
    {
      "entropy": 5.396851587295532,
      "epoch": 2.931647539389224,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004142221236115381,
      "loss": 4.9945,
      "mean_token_accuracy": 0.19501526951789855,
      "num_tokens": 65362938.0,
      "step": 37680
    },
    {
      "entropy": 5.359987783432007,
      "epoch": 2.9320365687609415,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004142003498508152,
      "loss": 4.8996,
      "mean_token_accuracy": 0.20601724088191986,
      "num_tokens": 65372000.0,
      "step": 37685
    },
    {
      "entropy": 5.3622230052947994,
      "epoch": 2.9324255981326592,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004141785739779486,
      "loss": 4.9302,
      "mean_token_accuracy": 0.20755930989980698,
      "num_tokens": 65379389.0,
      "step": 37690
    },
    {
      "entropy": 5.333988952636719,
      "epoch": 2.9328146275043765,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041415679599326853,
      "loss": 4.9597,
      "mean_token_accuracy": 0.19340699911117554,
      "num_tokens": 65388335.0,
      "step": 37695
    },
    {
      "entropy": 5.405156135559082,
      "epoch": 2.933203656876094,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041413501589710567,
      "loss": 5.0662,
      "mean_token_accuracy": 0.19486627131700515,
      "num_tokens": 65397280.0,
      "step": 37700
    },
    {
      "entropy": 5.473224735260009,
      "epoch": 2.9335926862478114,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004141132336897904,
      "loss": 5.1037,
      "mean_token_accuracy": 0.19134899973869324,
      "num_tokens": 65406600.0,
      "step": 37705
    },
    {
      "entropy": 5.408292579650879,
      "epoch": 2.933981715619529,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004140914493716533,
      "loss": 4.9486,
      "mean_token_accuracy": 0.20452744960784913,
      "num_tokens": 65415642.0,
      "step": 37710
    },
    {
      "entropy": 5.403013801574707,
      "epoch": 2.934370744991247,
      "grad_norm": 1.125,
      "learning_rate": 0.0004140696629430248,
      "loss": 4.9582,
      "mean_token_accuracy": 0.1998807743191719,
      "num_tokens": 65423977.0,
      "step": 37715
    },
    {
      "entropy": 5.282024669647217,
      "epoch": 2.9347597743629645,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041404787440423534,
      "loss": 4.9648,
      "mean_token_accuracy": 0.20175246745347977,
      "num_tokens": 65432567.0,
      "step": 37720
    },
    {
      "entropy": 5.3721386909484865,
      "epoch": 2.9351488037346822,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041402608375561595,
      "loss": 4.9874,
      "mean_token_accuracy": 0.2007876768708229,
      "num_tokens": 65441341.0,
      "step": 37725
    },
    {
      "entropy": 5.34037446975708,
      "epoch": 2.9355378331063995,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041400429099749687,
      "loss": 4.88,
      "mean_token_accuracy": 0.21553498059511184,
      "num_tokens": 65449612.0,
      "step": 37730
    },
    {
      "entropy": 5.3398548603057865,
      "epoch": 2.935926862478117,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000413982496130209,
      "loss": 4.9399,
      "mean_token_accuracy": 0.1986027777194977,
      "num_tokens": 65457216.0,
      "step": 37735
    },
    {
      "entropy": 5.320475387573242,
      "epoch": 2.9363158918498344,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041396069915408294,
      "loss": 4.8778,
      "mean_token_accuracy": 0.20334037840366365,
      "num_tokens": 65465420.0,
      "step": 37740
    },
    {
      "entropy": 5.470016670227051,
      "epoch": 2.936704921221552,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041393890006944945,
      "loss": 4.9852,
      "mean_token_accuracy": 0.19464935809373857,
      "num_tokens": 65474608.0,
      "step": 37745
    },
    {
      "entropy": 5.394106674194336,
      "epoch": 2.93709395059327,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004139170988766393,
      "loss": 4.9326,
      "mean_token_accuracy": 0.20110880583524704,
      "num_tokens": 65483204.0,
      "step": 37750
    },
    {
      "entropy": 5.308521032333374,
      "epoch": 2.9374829799649875,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004138952955759833,
      "loss": 4.8866,
      "mean_token_accuracy": 0.2100418284535408,
      "num_tokens": 65491472.0,
      "step": 37755
    },
    {
      "entropy": 5.404114723205566,
      "epoch": 2.937872009336705,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041387349016781225,
      "loss": 4.9983,
      "mean_token_accuracy": 0.199154894053936,
      "num_tokens": 65500018.0,
      "step": 37760
    },
    {
      "entropy": 5.4392087936401365,
      "epoch": 2.9382610387084225,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000413851682652457,
      "loss": 5.0502,
      "mean_token_accuracy": 0.20067093670368194,
      "num_tokens": 65508845.0,
      "step": 37765
    },
    {
      "entropy": 5.4023490905761715,
      "epoch": 2.93865006808014,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004138298730302485,
      "loss": 4.9534,
      "mean_token_accuracy": 0.20764517188072204,
      "num_tokens": 65517696.0,
      "step": 37770
    },
    {
      "entropy": 5.304414844512939,
      "epoch": 2.9390390974518574,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004138080613015176,
      "loss": 4.9281,
      "mean_token_accuracy": 0.2026049330830574,
      "num_tokens": 65526563.0,
      "step": 37775
    },
    {
      "entropy": 5.408280849456787,
      "epoch": 2.939428126823575,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041378624746659536,
      "loss": 5.0128,
      "mean_token_accuracy": 0.19605775624513627,
      "num_tokens": 65535719.0,
      "step": 37780
    },
    {
      "entropy": 5.451245594024658,
      "epoch": 2.939817156195293,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041376443152581265,
      "loss": 5.0342,
      "mean_token_accuracy": 0.1945420280098915,
      "num_tokens": 65543823.0,
      "step": 37785
    },
    {
      "entropy": 5.387151718139648,
      "epoch": 2.9402061855670105,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041374261347950064,
      "loss": 4.913,
      "mean_token_accuracy": 0.20796197950839995,
      "num_tokens": 65551923.0,
      "step": 37790
    },
    {
      "entropy": 5.3681152820587155,
      "epoch": 2.9405952149387278,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004137207933279903,
      "loss": 4.9287,
      "mean_token_accuracy": 0.20895612835884095,
      "num_tokens": 65560552.0,
      "step": 37795
    },
    {
      "entropy": 5.371119832992553,
      "epoch": 2.9409842443104455,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041369897107161266,
      "loss": 4.9673,
      "mean_token_accuracy": 0.20324310809373855,
      "num_tokens": 65568897.0,
      "step": 37800
    },
    {
      "entropy": 5.4015179634094235,
      "epoch": 2.941373273682163,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041367714671069897,
      "loss": 4.9889,
      "mean_token_accuracy": 0.20055605173110963,
      "num_tokens": 65577611.0,
      "step": 37805
    },
    {
      "entropy": 5.38393087387085,
      "epoch": 2.9417623030538804,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041365532024558024,
      "loss": 4.9399,
      "mean_token_accuracy": 0.2006349191069603,
      "num_tokens": 65586284.0,
      "step": 37810
    },
    {
      "entropy": 5.420541715621948,
      "epoch": 2.942151332425598,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004136334916765877,
      "loss": 5.0047,
      "mean_token_accuracy": 0.20125560313463212,
      "num_tokens": 65595065.0,
      "step": 37815
    },
    {
      "entropy": 5.431440114974976,
      "epoch": 2.942540361797316,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041361166100405264,
      "loss": 5.0072,
      "mean_token_accuracy": 0.19855813533067704,
      "num_tokens": 65604478.0,
      "step": 37820
    },
    {
      "entropy": 5.366732978820801,
      "epoch": 2.9429293911690335,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004135898282283062,
      "loss": 5.0024,
      "mean_token_accuracy": 0.20280594676733016,
      "num_tokens": 65613473.0,
      "step": 37825
    },
    {
      "entropy": 5.369474029541015,
      "epoch": 2.9433184205407508,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004135679933496797,
      "loss": 4.9526,
      "mean_token_accuracy": 0.20257633924484253,
      "num_tokens": 65622498.0,
      "step": 37830
    },
    {
      "entropy": 5.382475900650024,
      "epoch": 2.9437074499124685,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004135461563685045,
      "loss": 5.0422,
      "mean_token_accuracy": 0.1956923261284828,
      "num_tokens": 65630939.0,
      "step": 37835
    },
    {
      "entropy": 5.414383506774902,
      "epoch": 2.9440964792841857,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041352431728511194,
      "loss": 4.9595,
      "mean_token_accuracy": 0.1996999442577362,
      "num_tokens": 65639667.0,
      "step": 37840
    },
    {
      "entropy": 5.424587392807007,
      "epoch": 2.9444855086559034,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004135024760998333,
      "loss": 4.9682,
      "mean_token_accuracy": 0.20248344987630845,
      "num_tokens": 65648413.0,
      "step": 37845
    },
    {
      "entropy": 5.37522702217102,
      "epoch": 2.944874538027621,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041348063281299995,
      "loss": 5.0522,
      "mean_token_accuracy": 0.19733929634094238,
      "num_tokens": 65657486.0,
      "step": 37850
    },
    {
      "entropy": 5.355013132095337,
      "epoch": 2.945263567399339,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041345878742494346,
      "loss": 4.9309,
      "mean_token_accuracy": 0.203348571062088,
      "num_tokens": 65666328.0,
      "step": 37855
    },
    {
      "entropy": 5.32621431350708,
      "epoch": 2.945652596771056,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041343693993599535,
      "loss": 4.9632,
      "mean_token_accuracy": 0.19722681641578674,
      "num_tokens": 65674107.0,
      "step": 37860
    },
    {
      "entropy": 5.300283288955688,
      "epoch": 2.9460416261427738,
      "grad_norm": 1.125,
      "learning_rate": 0.0004134150903464869,
      "loss": 4.88,
      "mean_token_accuracy": 0.20362711250782012,
      "num_tokens": 65682534.0,
      "step": 37865
    },
    {
      "entropy": 5.4423047542572025,
      "epoch": 2.9464306555144915,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004133932386567498,
      "loss": 5.0885,
      "mean_token_accuracy": 0.19161452054977418,
      "num_tokens": 65690781.0,
      "step": 37870
    },
    {
      "entropy": 5.37935881614685,
      "epoch": 2.9468196848862087,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004133713848671156,
      "loss": 4.9346,
      "mean_token_accuracy": 0.20045971274375915,
      "num_tokens": 65698668.0,
      "step": 37875
    },
    {
      "entropy": 5.330772113800049,
      "epoch": 2.9472087142579264,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004133495289779159,
      "loss": 4.925,
      "mean_token_accuracy": 0.2017100363969803,
      "num_tokens": 65706568.0,
      "step": 37880
    },
    {
      "entropy": 5.3536817073822025,
      "epoch": 2.947597743629644,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004133276709894823,
      "loss": 4.9539,
      "mean_token_accuracy": 0.20344362854957582,
      "num_tokens": 65714652.0,
      "step": 37885
    },
    {
      "entropy": 5.337714719772339,
      "epoch": 2.947986773001362,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004133058109021464,
      "loss": 4.9459,
      "mean_token_accuracy": 0.205729678273201,
      "num_tokens": 65723296.0,
      "step": 37890
    },
    {
      "entropy": 5.4146990299224855,
      "epoch": 2.948375802373079,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041328394871624003,
      "loss": 5.0331,
      "mean_token_accuracy": 0.1958726868033409,
      "num_tokens": 65732257.0,
      "step": 37895
    },
    {
      "entropy": 5.425130605697632,
      "epoch": 2.9487648317447968,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004132620844320948,
      "loss": 4.9345,
      "mean_token_accuracy": 0.20518406331539155,
      "num_tokens": 65740152.0,
      "step": 37900
    },
    {
      "entropy": 5.309505414962769,
      "epoch": 2.9491538611165145,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004132402180500425,
      "loss": 4.9483,
      "mean_token_accuracy": 0.19853665679693222,
      "num_tokens": 65748927.0,
      "step": 37905
    },
    {
      "entropy": 5.359899663925171,
      "epoch": 2.9495428904882317,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004132183495704149,
      "loss": 4.9594,
      "mean_token_accuracy": 0.20121003836393356,
      "num_tokens": 65756949.0,
      "step": 37910
    },
    {
      "entropy": 5.462262725830078,
      "epoch": 2.9499319198599494,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004131964789935439,
      "loss": 5.0823,
      "mean_token_accuracy": 0.19762555062770842,
      "num_tokens": 65765076.0,
      "step": 37915
    },
    {
      "entropy": 5.360346126556396,
      "epoch": 2.950320949231667,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004131746063197612,
      "loss": 4.9667,
      "mean_token_accuracy": 0.20599135756492615,
      "num_tokens": 65773581.0,
      "step": 37920
    },
    {
      "entropy": 5.485965585708618,
      "epoch": 2.950709978603385,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004131527315493989,
      "loss": 5.1192,
      "mean_token_accuracy": 0.19339931160211563,
      "num_tokens": 65782312.0,
      "step": 37925
    },
    {
      "entropy": 5.367971611022949,
      "epoch": 2.951099007975102,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041313085468278873,
      "loss": 4.9517,
      "mean_token_accuracy": 0.20491400212049485,
      "num_tokens": 65790601.0,
      "step": 37930
    },
    {
      "entropy": 5.395820999145508,
      "epoch": 2.9514880373468197,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004131089757202627,
      "loss": 4.9664,
      "mean_token_accuracy": 0.20425473004579545,
      "num_tokens": 65799393.0,
      "step": 37935
    },
    {
      "entropy": 5.398208093643189,
      "epoch": 2.951877066718537,
      "grad_norm": 1.0,
      "learning_rate": 0.00041308709466215276,
      "loss": 4.975,
      "mean_token_accuracy": 0.2023208722472191,
      "num_tokens": 65808871.0,
      "step": 37940
    },
    {
      "entropy": 5.274767446517944,
      "epoch": 2.9522660960902547,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000413065211508791,
      "loss": 4.8001,
      "mean_token_accuracy": 0.2135943055152893,
      "num_tokens": 65817075.0,
      "step": 37945
    },
    {
      "entropy": 5.347618198394775,
      "epoch": 2.9526551254619724,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041304332626050937,
      "loss": 4.8674,
      "mean_token_accuracy": 0.2061361625790596,
      "num_tokens": 65825081.0,
      "step": 37950
    },
    {
      "entropy": 5.323006534576416,
      "epoch": 2.95304415483369,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041302143891764,
      "loss": 4.9911,
      "mean_token_accuracy": 0.196773199737072,
      "num_tokens": 65833183.0,
      "step": 37955
    },
    {
      "entropy": 5.355693197250366,
      "epoch": 2.9534331842054073,
      "grad_norm": 1.109375,
      "learning_rate": 0.000412999549480515,
      "loss": 4.8782,
      "mean_token_accuracy": 0.20206838399171828,
      "num_tokens": 65841514.0,
      "step": 37960
    },
    {
      "entropy": 5.481353807449341,
      "epoch": 2.953822213577125,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004129776579494665,
      "loss": 5.025,
      "mean_token_accuracy": 0.20750745087862016,
      "num_tokens": 65850433.0,
      "step": 37965
    },
    {
      "entropy": 5.416760778427124,
      "epoch": 2.9542112429488427,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004129557643248265,
      "loss": 4.9999,
      "mean_token_accuracy": 0.19766815900802612,
      "num_tokens": 65859807.0,
      "step": 37970
    },
    {
      "entropy": 5.311580467224121,
      "epoch": 2.95460027232056,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004129338686069276,
      "loss": 4.9436,
      "mean_token_accuracy": 0.20746409595012666,
      "num_tokens": 65867901.0,
      "step": 37975
    },
    {
      "entropy": 5.370877265930176,
      "epoch": 2.9549893016922777,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004129119707961016,
      "loss": 4.9932,
      "mean_token_accuracy": 0.19823826104402542,
      "num_tokens": 65876660.0,
      "step": 37980
    },
    {
      "entropy": 5.462342309951782,
      "epoch": 2.9553783310639954,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000412890070892681,
      "loss": 5.0334,
      "mean_token_accuracy": 0.1943318650126457,
      "num_tokens": 65885258.0,
      "step": 37985
    },
    {
      "entropy": 5.443876314163208,
      "epoch": 2.955767360435713,
      "grad_norm": 1.171875,
      "learning_rate": 0.000412868168896998,
      "loss": 5.0672,
      "mean_token_accuracy": 0.1881442129611969,
      "num_tokens": 65894260.0,
      "step": 37990
    },
    {
      "entropy": 5.447087192535401,
      "epoch": 2.9561563898074303,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004128462648093851,
      "loss": 4.9889,
      "mean_token_accuracy": 0.1995323896408081,
      "num_tokens": 65903000.0,
      "step": 37995
    },
    {
      "entropy": 5.394429302215576,
      "epoch": 2.956545419179148,
      "grad_norm": 1.125,
      "learning_rate": 0.0004128243586301745,
      "loss": 4.9735,
      "mean_token_accuracy": 0.20282233506441116,
      "num_tokens": 65911550.0,
      "step": 38000
    },
    {
      "entropy": 5.3915001392364506,
      "epoch": 2.9569344485508657,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004128024503596986,
      "loss": 4.9414,
      "mean_token_accuracy": 0.20307868272066115,
      "num_tokens": 65920091.0,
      "step": 38005
    },
    {
      "entropy": 5.32342791557312,
      "epoch": 2.957323477922583,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041278053999828986,
      "loss": 4.956,
      "mean_token_accuracy": 0.19966159760951996,
      "num_tokens": 65929313.0,
      "step": 38010
    },
    {
      "entropy": 5.396011304855347,
      "epoch": 2.9577125072943007,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004127586275462806,
      "loss": 5.0269,
      "mean_token_accuracy": 0.1913701370358467,
      "num_tokens": 65938407.0,
      "step": 38015
    },
    {
      "entropy": 5.414344930648804,
      "epoch": 2.9581015366660184,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041273671300400363,
      "loss": 4.9412,
      "mean_token_accuracy": 0.2063109040260315,
      "num_tokens": 65946644.0,
      "step": 38020
    },
    {
      "entropy": 5.410267686843872,
      "epoch": 2.958490566037736,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041271479637179116,
      "loss": 5.0424,
      "mean_token_accuracy": 0.19689824432134628,
      "num_tokens": 65955695.0,
      "step": 38025
    },
    {
      "entropy": 5.445426893234253,
      "epoch": 2.9588795954094533,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004126928776499759,
      "loss": 5.0732,
      "mean_token_accuracy": 0.18842373937368392,
      "num_tokens": 65964326.0,
      "step": 38030
    },
    {
      "entropy": 5.357020854949951,
      "epoch": 2.959268624781171,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004126709568388903,
      "loss": 4.9057,
      "mean_token_accuracy": 0.2022950157523155,
      "num_tokens": 65972986.0,
      "step": 38035
    },
    {
      "entropy": 5.342325878143311,
      "epoch": 2.9596576541528883,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004126490339388672,
      "loss": 4.9473,
      "mean_token_accuracy": 0.20743325650691985,
      "num_tokens": 65980829.0,
      "step": 38040
    },
    {
      "entropy": 5.367606163024902,
      "epoch": 2.960046683524606,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000412627108950239,
      "loss": 4.8762,
      "mean_token_accuracy": 0.20567074716091155,
      "num_tokens": 65989412.0,
      "step": 38045
    },
    {
      "entropy": 5.3479413986206055,
      "epoch": 2.9604357128963237,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004126051818733385,
      "loss": 4.966,
      "mean_token_accuracy": 0.2013981282711029,
      "num_tokens": 65998887.0,
      "step": 38050
    },
    {
      "entropy": 5.431630325317383,
      "epoch": 2.9608247422680414,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004125832527084984,
      "loss": 5.0082,
      "mean_token_accuracy": 0.20024715960025788,
      "num_tokens": 66007266.0,
      "step": 38055
    },
    {
      "entropy": 5.393360614776611,
      "epoch": 2.961213771639759,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041256132145605143,
      "loss": 4.9316,
      "mean_token_accuracy": 0.20378567427396774,
      "num_tokens": 66015968.0,
      "step": 38060
    },
    {
      "entropy": 5.427715158462524,
      "epoch": 2.9616028010114763,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004125393881163304,
      "loss": 5.0222,
      "mean_token_accuracy": 0.1934209167957306,
      "num_tokens": 66025128.0,
      "step": 38065
    },
    {
      "entropy": 5.3826963901519775,
      "epoch": 2.961991830383194,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041251745268966797,
      "loss": 4.9153,
      "mean_token_accuracy": 0.20436899214982987,
      "num_tokens": 66033730.0,
      "step": 38070
    },
    {
      "entropy": 5.27549524307251,
      "epoch": 2.9623808597549113,
      "grad_norm": 1.125,
      "learning_rate": 0.0004124955151763971,
      "loss": 4.8628,
      "mean_token_accuracy": 0.21150993108749389,
      "num_tokens": 66042711.0,
      "step": 38075
    },
    {
      "entropy": 5.443307018280029,
      "epoch": 2.962769889126629,
      "grad_norm": 1.125,
      "learning_rate": 0.0004124735755768506,
      "loss": 4.9639,
      "mean_token_accuracy": 0.2065073072910309,
      "num_tokens": 66051107.0,
      "step": 38080
    },
    {
      "entropy": 5.391071796417236,
      "epoch": 2.9631589184983467,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004124516338913615,
      "loss": 4.9675,
      "mean_token_accuracy": 0.20651648938655853,
      "num_tokens": 66059532.0,
      "step": 38085
    },
    {
      "entropy": 5.3682457447052006,
      "epoch": 2.9635479478700644,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004124296901202626,
      "loss": 4.9938,
      "mean_token_accuracy": 0.1968667760491371,
      "num_tokens": 66068034.0,
      "step": 38090
    },
    {
      "entropy": 5.427253723144531,
      "epoch": 2.9639369772417816,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041240774426388685,
      "loss": 5.0256,
      "mean_token_accuracy": 0.19464122951030732,
      "num_tokens": 66077403.0,
      "step": 38095
    },
    {
      "entropy": 5.428533601760864,
      "epoch": 2.9643260066134993,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004123857963225673,
      "loss": 5.0008,
      "mean_token_accuracy": 0.19266920089721679,
      "num_tokens": 66085930.0,
      "step": 38100
    },
    {
      "entropy": 5.411783075332641,
      "epoch": 2.964715035985217,
      "grad_norm": 1.203125,
      "learning_rate": 0.000412363846296637,
      "loss": 4.9406,
      "mean_token_accuracy": 0.20242844820022582,
      "num_tokens": 66094671.0,
      "step": 38105
    },
    {
      "entropy": 5.395580148696899,
      "epoch": 2.9651040653569343,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000412341894186429,
      "loss": 5.057,
      "mean_token_accuracy": 0.1989541009068489,
      "num_tokens": 66103250.0,
      "step": 38110
    },
    {
      "entropy": 5.3936292171478275,
      "epoch": 2.965493094728652,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004123199399922763,
      "loss": 4.9664,
      "mean_token_accuracy": 0.20479229539632798,
      "num_tokens": 66111073.0,
      "step": 38115
    },
    {
      "entropy": 5.339973211288452,
      "epoch": 2.9658821241003697,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041229798371451203,
      "loss": 4.9001,
      "mean_token_accuracy": 0.20256479680538178,
      "num_tokens": 66119514.0,
      "step": 38120
    },
    {
      "entropy": 5.361181163787842,
      "epoch": 2.9662711534720874,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004122760253534695,
      "loss": 4.9288,
      "mean_token_accuracy": 0.20087486356496811,
      "num_tokens": 66128804.0,
      "step": 38125
    },
    {
      "entropy": 5.392936277389526,
      "epoch": 2.9666601828438046,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004122540649094817,
      "loss": 4.9648,
      "mean_token_accuracy": 0.20542702972888946,
      "num_tokens": 66137683.0,
      "step": 38130
    },
    {
      "entropy": 5.4481250762939455,
      "epoch": 2.9670492122155223,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041223210238288196,
      "loss": 5.0267,
      "mean_token_accuracy": 0.19689006209373475,
      "num_tokens": 66147135.0,
      "step": 38135
    },
    {
      "entropy": 5.411341667175293,
      "epoch": 2.9674382415872396,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004122101377740035,
      "loss": 4.9148,
      "mean_token_accuracy": 0.206108221411705,
      "num_tokens": 66155534.0,
      "step": 38140
    },
    {
      "entropy": 5.391073083877563,
      "epoch": 2.9678272709589573,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004121881710831796,
      "loss": 4.908,
      "mean_token_accuracy": 0.20672812014818193,
      "num_tokens": 66164413.0,
      "step": 38145
    },
    {
      "entropy": 5.382673263549805,
      "epoch": 2.968216300330675,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004121662023107435,
      "loss": 4.9808,
      "mean_token_accuracy": 0.20280151963233947,
      "num_tokens": 66173469.0,
      "step": 38150
    },
    {
      "entropy": 5.351093006134033,
      "epoch": 2.9686053297023927,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004121442314570286,
      "loss": 4.9007,
      "mean_token_accuracy": 0.2045367792248726,
      "num_tokens": 66182183.0,
      "step": 38155
    },
    {
      "entropy": 5.394851446151733,
      "epoch": 2.9689943590741104,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041212225852236834,
      "loss": 4.9408,
      "mean_token_accuracy": 0.20108719170093536,
      "num_tokens": 66190647.0,
      "step": 38160
    },
    {
      "entropy": 5.274054670333863,
      "epoch": 2.9693833884458276,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004121002835070961,
      "loss": 4.816,
      "mean_token_accuracy": 0.20558966994285582,
      "num_tokens": 66198536.0,
      "step": 38165
    },
    {
      "entropy": 5.439622306823731,
      "epoch": 2.9697724178175453,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004120783064115452,
      "loss": 5.1272,
      "mean_token_accuracy": 0.18706280291080474,
      "num_tokens": 66207772.0,
      "step": 38170
    },
    {
      "entropy": 5.4685387134552,
      "epoch": 2.9701614471892626,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004120563272360492,
      "loss": 5.0428,
      "mean_token_accuracy": 0.19492299854755402,
      "num_tokens": 66216862.0,
      "step": 38175
    },
    {
      "entropy": 5.456356954574585,
      "epoch": 2.9705504765609803,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004120343459809416,
      "loss": 5.0112,
      "mean_token_accuracy": 0.19915633648633957,
      "num_tokens": 66225817.0,
      "step": 38180
    },
    {
      "entropy": 5.39089183807373,
      "epoch": 2.970939505932698,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004120123626465559,
      "loss": 5.0124,
      "mean_token_accuracy": 0.19599299877882004,
      "num_tokens": 66234894.0,
      "step": 38185
    },
    {
      "entropy": 5.356740522384643,
      "epoch": 2.9713285353044157,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041199037723322566,
      "loss": 4.8964,
      "mean_token_accuracy": 0.20243219286203384,
      "num_tokens": 66243800.0,
      "step": 38190
    },
    {
      "entropy": 5.359181118011475,
      "epoch": 2.971717564676133,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041196838974128446,
      "loss": 4.9094,
      "mean_token_accuracy": 0.20440909117460251,
      "num_tokens": 66253192.0,
      "step": 38195
    },
    {
      "entropy": 5.345286750793457,
      "epoch": 2.9721065940478506,
      "grad_norm": 1.109375,
      "learning_rate": 0.000411946400171066,
      "loss": 4.9202,
      "mean_token_accuracy": 0.20844102054834365,
      "num_tokens": 66261775.0,
      "step": 38200
    },
    {
      "entropy": 5.356602430343628,
      "epoch": 2.9724956234195683,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041192440852290385,
      "loss": 4.9532,
      "mean_token_accuracy": 0.20127539187669755,
      "num_tokens": 66271069.0,
      "step": 38205
    },
    {
      "entropy": 5.4719208717346195,
      "epoch": 2.9728846527912856,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004119024147971318,
      "loss": 5.1022,
      "mean_token_accuracy": 0.19554950296878815,
      "num_tokens": 66280021.0,
      "step": 38210
    },
    {
      "entropy": 5.39001317024231,
      "epoch": 2.9732736821630033,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041188041899408345,
      "loss": 4.9418,
      "mean_token_accuracy": 0.20625476986169816,
      "num_tokens": 66288782.0,
      "step": 38215
    },
    {
      "entropy": 5.4133189678192135,
      "epoch": 2.973662711534721,
      "grad_norm": 1.25,
      "learning_rate": 0.0004118584211140926,
      "loss": 4.9971,
      "mean_token_accuracy": 0.1953978568315506,
      "num_tokens": 66297716.0,
      "step": 38220
    },
    {
      "entropy": 5.468969392776489,
      "epoch": 2.9740517409064386,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041183642115749316,
      "loss": 5.002,
      "mean_token_accuracy": 0.1991521328687668,
      "num_tokens": 66305977.0,
      "step": 38225
    },
    {
      "entropy": 5.416920280456543,
      "epoch": 2.974440770278156,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041181441912461873,
      "loss": 5.042,
      "mean_token_accuracy": 0.1990526795387268,
      "num_tokens": 66314344.0,
      "step": 38230
    },
    {
      "entropy": 5.326670074462891,
      "epoch": 2.9748297996498736,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004117924150158032,
      "loss": 4.9331,
      "mean_token_accuracy": 0.20160309672355653,
      "num_tokens": 66322776.0,
      "step": 38235
    },
    {
      "entropy": 5.42648024559021,
      "epoch": 2.9752188290215913,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041177040883138064,
      "loss": 5.0075,
      "mean_token_accuracy": 0.20792577862739564,
      "num_tokens": 66331161.0,
      "step": 38240
    },
    {
      "entropy": 5.418236875534058,
      "epoch": 2.9756078583933085,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041174840057168474,
      "loss": 4.9853,
      "mean_token_accuracy": 0.20161716789007186,
      "num_tokens": 66339579.0,
      "step": 38245
    },
    {
      "entropy": 5.229699230194091,
      "epoch": 2.9759968877650262,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004117263902370495,
      "loss": 4.8677,
      "mean_token_accuracy": 0.20782626569271087,
      "num_tokens": 66348716.0,
      "step": 38250
    },
    {
      "entropy": 5.379639101028443,
      "epoch": 2.976385917136744,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041170437782780896,
      "loss": 4.9361,
      "mean_token_accuracy": 0.19685055017471315,
      "num_tokens": 66357546.0,
      "step": 38255
    },
    {
      "entropy": 5.351508092880249,
      "epoch": 2.9767749465084616,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000411682363344297,
      "loss": 4.8539,
      "mean_token_accuracy": 0.2113683819770813,
      "num_tokens": 66366024.0,
      "step": 38260
    },
    {
      "entropy": 5.393351888656616,
      "epoch": 2.977163975880179,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041166034678684777,
      "loss": 4.9284,
      "mean_token_accuracy": 0.20496572256088258,
      "num_tokens": 66374413.0,
      "step": 38265
    },
    {
      "entropy": 5.38733024597168,
      "epoch": 2.9775530052518966,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004116383281557953,
      "loss": 5.0058,
      "mean_token_accuracy": 0.19440943449735643,
      "num_tokens": 66383131.0,
      "step": 38270
    },
    {
      "entropy": 5.385790205001831,
      "epoch": 2.977942034623614,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041161630745147374,
      "loss": 5.0578,
      "mean_token_accuracy": 0.1938394159078598,
      "num_tokens": 66392227.0,
      "step": 38275
    },
    {
      "entropy": 5.363867902755738,
      "epoch": 2.9783310639953315,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004115942846742171,
      "loss": 4.9721,
      "mean_token_accuracy": 0.20194346010684966,
      "num_tokens": 66401384.0,
      "step": 38280
    },
    {
      "entropy": 5.393619155883789,
      "epoch": 2.9787200933670492,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004115722598243596,
      "loss": 4.9466,
      "mean_token_accuracy": 0.20679647624492645,
      "num_tokens": 66409936.0,
      "step": 38285
    },
    {
      "entropy": 5.356252002716064,
      "epoch": 2.979109122738767,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041155023290223537,
      "loss": 4.8869,
      "mean_token_accuracy": 0.2110457330942154,
      "num_tokens": 66418444.0,
      "step": 38290
    },
    {
      "entropy": 5.346478080749511,
      "epoch": 2.979498152110484,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004115282039081788,
      "loss": 4.9364,
      "mean_token_accuracy": 0.20627668499946594,
      "num_tokens": 66427211.0,
      "step": 38295
    },
    {
      "entropy": 5.350663375854492,
      "epoch": 2.979887181482202,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041150617284252405,
      "loss": 4.9144,
      "mean_token_accuracy": 0.20945375859737397,
      "num_tokens": 66435767.0,
      "step": 38300
    },
    {
      "entropy": 5.420959854125977,
      "epoch": 2.9802762108539196,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004114841397056053,
      "loss": 4.9681,
      "mean_token_accuracy": 0.1968962788581848,
      "num_tokens": 66443647.0,
      "step": 38305
    },
    {
      "entropy": 5.357402706146241,
      "epoch": 2.980665240225637,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041146210449775704,
      "loss": 4.9147,
      "mean_token_accuracy": 0.20758935064077377,
      "num_tokens": 66452011.0,
      "step": 38310
    },
    {
      "entropy": 5.431408214569092,
      "epoch": 2.9810542695973545,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004114400672193136,
      "loss": 5.0179,
      "mean_token_accuracy": 0.1949569046497345,
      "num_tokens": 66460151.0,
      "step": 38315
    },
    {
      "entropy": 5.443654155731201,
      "epoch": 2.9814432989690722,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041141802787060923,
      "loss": 4.9999,
      "mean_token_accuracy": 0.20273821651935578,
      "num_tokens": 66468409.0,
      "step": 38320
    },
    {
      "entropy": 5.4179483413696286,
      "epoch": 2.98183232834079,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004113959864519784,
      "loss": 5.0791,
      "mean_token_accuracy": 0.1896918535232544,
      "num_tokens": 66477765.0,
      "step": 38325
    },
    {
      "entropy": 5.45331039428711,
      "epoch": 2.982221357712507,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004113739429637556,
      "loss": 5.0787,
      "mean_token_accuracy": 0.19466610997915268,
      "num_tokens": 66486726.0,
      "step": 38330
    },
    {
      "entropy": 5.4840254306793215,
      "epoch": 2.982610387084225,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004113518974062753,
      "loss": 5.032,
      "mean_token_accuracy": 0.1924159497022629,
      "num_tokens": 66495994.0,
      "step": 38335
    },
    {
      "entropy": 5.458738470077515,
      "epoch": 2.9829994164559426,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000411329849779872,
      "loss": 5.0143,
      "mean_token_accuracy": 0.19422049522399903,
      "num_tokens": 66504515.0,
      "step": 38340
    },
    {
      "entropy": 5.350718259811401,
      "epoch": 2.98338844582766,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041130780008488015,
      "loss": 4.8964,
      "mean_token_accuracy": 0.20535831898450851,
      "num_tokens": 66513637.0,
      "step": 38345
    },
    {
      "entropy": 5.332370948791504,
      "epoch": 2.9837774751993775,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004112857483216344,
      "loss": 4.8669,
      "mean_token_accuracy": 0.2118707627058029,
      "num_tokens": 66521453.0,
      "step": 38350
    },
    {
      "entropy": 5.2376385688781735,
      "epoch": 2.9841665045710952,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041126369449046935,
      "loss": 4.7757,
      "mean_token_accuracy": 0.2203754186630249,
      "num_tokens": 66530888.0,
      "step": 38355
    },
    {
      "entropy": 5.380611181259155,
      "epoch": 2.984555533942813,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041124163859171967,
      "loss": 5.037,
      "mean_token_accuracy": 0.19566044509410857,
      "num_tokens": 66538758.0,
      "step": 38360
    },
    {
      "entropy": 5.435381174087524,
      "epoch": 2.98494456331453,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004112195806257199,
      "loss": 5.0091,
      "mean_token_accuracy": 0.19955134242773057,
      "num_tokens": 66546981.0,
      "step": 38365
    },
    {
      "entropy": 5.405702543258667,
      "epoch": 2.985333592686248,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004111975205928049,
      "loss": 4.9648,
      "mean_token_accuracy": 0.2003589853644371,
      "num_tokens": 66555375.0,
      "step": 38370
    },
    {
      "entropy": 5.385578584671021,
      "epoch": 2.985722622057965,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041117545849330924,
      "loss": 4.9416,
      "mean_token_accuracy": 0.20266453325748443,
      "num_tokens": 66564581.0,
      "step": 38375
    },
    {
      "entropy": 5.346369504928589,
      "epoch": 2.986111651429683,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041115339432756776,
      "loss": 4.9908,
      "mean_token_accuracy": 0.19918077439069748,
      "num_tokens": 66573876.0,
      "step": 38380
    },
    {
      "entropy": 5.380865907669067,
      "epoch": 2.9865006808014005,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004111313280959152,
      "loss": 4.9876,
      "mean_token_accuracy": 0.21135186403989792,
      "num_tokens": 66582039.0,
      "step": 38385
    },
    {
      "entropy": 5.333020305633545,
      "epoch": 2.986889710173118,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041110925979868647,
      "loss": 4.8673,
      "mean_token_accuracy": 0.21232945322990418,
      "num_tokens": 66590188.0,
      "step": 38390
    },
    {
      "entropy": 5.334687757492065,
      "epoch": 2.9872787395448355,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004110871894362163,
      "loss": 4.9455,
      "mean_token_accuracy": 0.2013239547610283,
      "num_tokens": 66598889.0,
      "step": 38395
    },
    {
      "entropy": 5.365055894851684,
      "epoch": 2.987667768916553,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041106511700883975,
      "loss": 4.9519,
      "mean_token_accuracy": 0.20141951143741607,
      "num_tokens": 66606909.0,
      "step": 38400
    },
    {
      "entropy": 5.396669578552246,
      "epoch": 2.988056798288271,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041104304251689147,
      "loss": 4.972,
      "mean_token_accuracy": 0.2027745798230171,
      "num_tokens": 66615644.0,
      "step": 38405
    },
    {
      "entropy": 5.45669116973877,
      "epoch": 2.988445827659988,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041102096596070666,
      "loss": 5.1039,
      "mean_token_accuracy": 0.19714124202728273,
      "num_tokens": 66624466.0,
      "step": 38410
    },
    {
      "entropy": 5.490889167785644,
      "epoch": 2.988834857031706,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041099888734062013,
      "loss": 5.124,
      "mean_token_accuracy": 0.19706136882305145,
      "num_tokens": 66633907.0,
      "step": 38415
    },
    {
      "entropy": 5.438182592391968,
      "epoch": 2.9892238864034235,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000410976806656967,
      "loss": 5.0653,
      "mean_token_accuracy": 0.2002486914396286,
      "num_tokens": 66643652.0,
      "step": 38420
    },
    {
      "entropy": 5.420694732666016,
      "epoch": 2.989612915775141,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004109547239100822,
      "loss": 4.9951,
      "mean_token_accuracy": 0.20185774564743042,
      "num_tokens": 66652440.0,
      "step": 38425
    },
    {
      "entropy": 5.388789510726928,
      "epoch": 2.9900019451468585,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004109326391003009,
      "loss": 4.9322,
      "mean_token_accuracy": 0.20394351333379745,
      "num_tokens": 66661660.0,
      "step": 38430
    },
    {
      "entropy": 5.419493579864502,
      "epoch": 2.990390974518576,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041091055222795814,
      "loss": 4.9586,
      "mean_token_accuracy": 0.20560043603181838,
      "num_tokens": 66670289.0,
      "step": 38435
    },
    {
      "entropy": 5.407018852233887,
      "epoch": 2.990780003890294,
      "grad_norm": 1.125,
      "learning_rate": 0.0004108884632933891,
      "loss": 4.9847,
      "mean_token_accuracy": 0.20684262365102768,
      "num_tokens": 66679777.0,
      "step": 38440
    },
    {
      "entropy": 5.476563882827759,
      "epoch": 2.991169033262011,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041086637229692893,
      "loss": 4.9956,
      "mean_token_accuracy": 0.19661099463701248,
      "num_tokens": 66689264.0,
      "step": 38445
    },
    {
      "entropy": 5.420726203918457,
      "epoch": 2.991558062633729,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041084427923891275,
      "loss": 5.0124,
      "mean_token_accuracy": 0.2009250059723854,
      "num_tokens": 66697632.0,
      "step": 38450
    },
    {
      "entropy": 5.37018051147461,
      "epoch": 2.9919470920054465,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041082218411967586,
      "loss": 4.9628,
      "mean_token_accuracy": 0.19962136894464494,
      "num_tokens": 66706312.0,
      "step": 38455
    },
    {
      "entropy": 5.424208450317383,
      "epoch": 2.992336121377164,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041080008693955355,
      "loss": 4.9168,
      "mean_token_accuracy": 0.20134995132684708,
      "num_tokens": 66714322.0,
      "step": 38460
    },
    {
      "entropy": 5.32239875793457,
      "epoch": 2.9927251507488815,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004107779876988811,
      "loss": 4.9388,
      "mean_token_accuracy": 0.20026778131723405,
      "num_tokens": 66723039.0,
      "step": 38465
    },
    {
      "entropy": 5.3899181365966795,
      "epoch": 2.993114180120599,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041075588639799375,
      "loss": 4.9459,
      "mean_token_accuracy": 0.2114546686410904,
      "num_tokens": 66731414.0,
      "step": 38470
    },
    {
      "entropy": 5.4143274307250975,
      "epoch": 2.9935032094923164,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004107337830372269,
      "loss": 4.9518,
      "mean_token_accuracy": 0.20306631028652192,
      "num_tokens": 66740664.0,
      "step": 38475
    },
    {
      "entropy": 5.347423315048218,
      "epoch": 2.993892238864034,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000410711677616916,
      "loss": 4.8859,
      "mean_token_accuracy": 0.19792909324169158,
      "num_tokens": 66749073.0,
      "step": 38480
    },
    {
      "entropy": 5.311897802352905,
      "epoch": 2.994281268235752,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004106895701373963,
      "loss": 4.9408,
      "mean_token_accuracy": 0.20299936980009078,
      "num_tokens": 66757734.0,
      "step": 38485
    },
    {
      "entropy": 5.344456911087036,
      "epoch": 2.9946702976074695,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041066746059900336,
      "loss": 4.9175,
      "mean_token_accuracy": 0.1996750771999359,
      "num_tokens": 66766699.0,
      "step": 38490
    },
    {
      "entropy": 5.369357872009277,
      "epoch": 2.995059326979187,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004106453490020727,
      "loss": 5.0767,
      "mean_token_accuracy": 0.19289746582508088,
      "num_tokens": 66775398.0,
      "step": 38495
    },
    {
      "entropy": 5.337516641616821,
      "epoch": 2.9954483563509045,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041062323534693973,
      "loss": 4.8621,
      "mean_token_accuracy": 0.20904262065887452,
      "num_tokens": 66783157.0,
      "step": 38500
    },
    {
      "entropy": 5.311770391464234,
      "epoch": 2.995837385722622,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041060111963394004,
      "loss": 4.9647,
      "mean_token_accuracy": 0.19998246133327485,
      "num_tokens": 66791293.0,
      "step": 38505
    },
    {
      "entropy": 5.2920369625091555,
      "epoch": 2.9962264150943394,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041057900186340917,
      "loss": 4.8721,
      "mean_token_accuracy": 0.20782489478588104,
      "num_tokens": 66800212.0,
      "step": 38510
    },
    {
      "entropy": 5.497542667388916,
      "epoch": 2.996615444466057,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041055688203568274,
      "loss": 5.0456,
      "mean_token_accuracy": 0.1956419438123703,
      "num_tokens": 66808050.0,
      "step": 38515
    },
    {
      "entropy": 5.424835872650147,
      "epoch": 2.997004473837775,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004105347601510964,
      "loss": 4.9504,
      "mean_token_accuracy": 0.21011705994606017,
      "num_tokens": 66816481.0,
      "step": 38520
    },
    {
      "entropy": 5.328875923156739,
      "epoch": 2.9973935032094925,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004105126362099859,
      "loss": 4.9474,
      "mean_token_accuracy": 0.20159079283475875,
      "num_tokens": 66825294.0,
      "step": 38525
    },
    {
      "entropy": 5.357822322845459,
      "epoch": 2.9977825325812097,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004104905102126867,
      "loss": 4.9069,
      "mean_token_accuracy": 0.2029980644583702,
      "num_tokens": 66833818.0,
      "step": 38530
    },
    {
      "entropy": 5.435402917861938,
      "epoch": 2.9981715619529274,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004104683821595347,
      "loss": 5.0168,
      "mean_token_accuracy": 0.19405477643013,
      "num_tokens": 66842604.0,
      "step": 38535
    },
    {
      "entropy": 5.453289270401001,
      "epoch": 2.998560591324645,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041044625205086566,
      "loss": 5.0817,
      "mean_token_accuracy": 0.19854393154382705,
      "num_tokens": 66850974.0,
      "step": 38540
    },
    {
      "entropy": 5.3967334747314455,
      "epoch": 2.9989496206963624,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041042411988701535,
      "loss": 4.9243,
      "mean_token_accuracy": 0.21168268471956253,
      "num_tokens": 66859671.0,
      "step": 38545
    },
    {
      "entropy": 5.410159921646118,
      "epoch": 2.99933865006808,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041040198566831944,
      "loss": 5.007,
      "mean_token_accuracy": 0.20219929665327072,
      "num_tokens": 66868864.0,
      "step": 38550
    },
    {
      "entropy": 5.320473957061767,
      "epoch": 2.999727679439798,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004103798493951141,
      "loss": 4.9439,
      "mean_token_accuracy": 0.20341742634773255,
      "num_tokens": 66877547.0,
      "step": 38555
    },
    {
      "entropy": 5.36516793568929,
      "epoch": 3.0000778058743434,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041035771106773483,
      "loss": 4.8523,
      "mean_token_accuracy": 0.2093366781870524,
      "num_tokens": 66884854.0,
      "step": 38560
    },
    {
      "entropy": 5.421144628524781,
      "epoch": 3.000466835246061,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004103355706865179,
      "loss": 4.8901,
      "mean_token_accuracy": 0.19692652374505998,
      "num_tokens": 66893963.0,
      "step": 38565
    },
    {
      "entropy": 5.311478090286255,
      "epoch": 3.0008558646177788,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000410313428251799,
      "loss": 4.8479,
      "mean_token_accuracy": 0.2020140618085861,
      "num_tokens": 66902788.0,
      "step": 38570
    },
    {
      "entropy": 5.378206491470337,
      "epoch": 3.001244893989496,
      "grad_norm": 1.25,
      "learning_rate": 0.00041029128376391424,
      "loss": 4.9348,
      "mean_token_accuracy": 0.20524081736803054,
      "num_tokens": 66912294.0,
      "step": 38575
    },
    {
      "entropy": 5.335737466812134,
      "epoch": 3.0016339233612137,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041026913722319954,
      "loss": 4.7983,
      "mean_token_accuracy": 0.21404553055763245,
      "num_tokens": 66920080.0,
      "step": 38580
    },
    {
      "entropy": 5.3069007396698,
      "epoch": 3.0020229527329314,
      "grad_norm": 1.15625,
      "learning_rate": 0.000410246988629991,
      "loss": 4.8393,
      "mean_token_accuracy": 0.20780400335788726,
      "num_tokens": 66928404.0,
      "step": 38585
    },
    {
      "entropy": 5.387386894226074,
      "epoch": 3.002411982104649,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041022483798462466,
      "loss": 4.9528,
      "mean_token_accuracy": 0.20186459869146348,
      "num_tokens": 66938040.0,
      "step": 38590
    },
    {
      "entropy": 5.407912349700927,
      "epoch": 3.0028010114763664,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004102026852874367,
      "loss": 4.918,
      "mean_token_accuracy": 0.21327041685581208,
      "num_tokens": 66946605.0,
      "step": 38595
    },
    {
      "entropy": 5.417638301849365,
      "epoch": 3.003190040848084,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041018053053876316,
      "loss": 4.9329,
      "mean_token_accuracy": 0.20908523350954056,
      "num_tokens": 66955276.0,
      "step": 38600
    },
    {
      "entropy": 5.441386079788208,
      "epoch": 3.0035790702198017,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004101583737389402,
      "loss": 4.9035,
      "mean_token_accuracy": 0.2020650878548622,
      "num_tokens": 66963950.0,
      "step": 38605
    },
    {
      "entropy": 5.3821440696716305,
      "epoch": 3.003968099591519,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041013621488830406,
      "loss": 4.8666,
      "mean_token_accuracy": 0.20635460764169694,
      "num_tokens": 66972101.0,
      "step": 38610
    },
    {
      "entropy": 5.294893550872803,
      "epoch": 3.0043571289632367,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041011405398719096,
      "loss": 4.8051,
      "mean_token_accuracy": 0.20820271223783493,
      "num_tokens": 66981251.0,
      "step": 38615
    },
    {
      "entropy": 5.442904710769653,
      "epoch": 3.0047461583349544,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041009189103593716,
      "loss": 4.9227,
      "mean_token_accuracy": 0.2072306215763092,
      "num_tokens": 66990292.0,
      "step": 38620
    },
    {
      "entropy": 5.432159280776977,
      "epoch": 3.0051351877066717,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004100697260348789,
      "loss": 4.9623,
      "mean_token_accuracy": 0.20261773318052292,
      "num_tokens": 66999699.0,
      "step": 38625
    },
    {
      "entropy": 5.384227132797241,
      "epoch": 3.0055242170783893,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041004755898435254,
      "loss": 4.8466,
      "mean_token_accuracy": 0.20569859594106674,
      "num_tokens": 67008272.0,
      "step": 38630
    },
    {
      "entropy": 5.314862632751465,
      "epoch": 3.005913246450107,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004100253898846945,
      "loss": 4.8961,
      "mean_token_accuracy": 0.20325679332017899,
      "num_tokens": 67016865.0,
      "step": 38635
    },
    {
      "entropy": 5.375717401504517,
      "epoch": 3.0063022758218247,
      "grad_norm": 1.21875,
      "learning_rate": 0.000410003218736241,
      "loss": 4.9081,
      "mean_token_accuracy": 0.20962373465299605,
      "num_tokens": 67025791.0,
      "step": 38640
    },
    {
      "entropy": 5.31078839302063,
      "epoch": 3.006691305193542,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00040998104553932856,
      "loss": 4.7957,
      "mean_token_accuracy": 0.20929372906684876,
      "num_tokens": 67034872.0,
      "step": 38645
    },
    {
      "entropy": 5.319442844390869,
      "epoch": 3.0070803345652597,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004099588702942936,
      "loss": 4.8828,
      "mean_token_accuracy": 0.20637225210666657,
      "num_tokens": 67043350.0,
      "step": 38650
    },
    {
      "entropy": 5.353558588027954,
      "epoch": 3.0074693639369774,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040993669300147257,
      "loss": 4.9593,
      "mean_token_accuracy": 0.20271400064229966,
      "num_tokens": 67051593.0,
      "step": 38655
    },
    {
      "entropy": 5.433165407180786,
      "epoch": 3.0078583933086946,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040991451366120205,
      "loss": 4.9545,
      "mean_token_accuracy": 0.20145209282636642,
      "num_tokens": 67060024.0,
      "step": 38660
    },
    {
      "entropy": 5.5142662048339846,
      "epoch": 3.0082474226804123,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004098923322738185,
      "loss": 5.0027,
      "mean_token_accuracy": 0.2021059662103653,
      "num_tokens": 67069136.0,
      "step": 38665
    },
    {
      "entropy": 5.2748268127441404,
      "epoch": 3.00863645205213,
      "grad_norm": 1.390625,
      "learning_rate": 0.00040987014883965843,
      "loss": 4.7412,
      "mean_token_accuracy": 0.21629730015993118,
      "num_tokens": 67078209.0,
      "step": 38670
    },
    {
      "entropy": 5.274156332015991,
      "epoch": 3.0090254814238473,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040984796335905867,
      "loss": 4.8621,
      "mean_token_accuracy": 0.20710614174604416,
      "num_tokens": 67087676.0,
      "step": 38675
    },
    {
      "entropy": 5.376067113876343,
      "epoch": 3.009414510795565,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004098257758323556,
      "loss": 4.9415,
      "mean_token_accuracy": 0.20264537632465363,
      "num_tokens": 67096730.0,
      "step": 38680
    },
    {
      "entropy": 5.518951940536499,
      "epoch": 3.0098035401672827,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004098035862598859,
      "loss": 5.0267,
      "mean_token_accuracy": 0.19334909468889236,
      "num_tokens": 67106017.0,
      "step": 38685
    },
    {
      "entropy": 5.374785089492798,
      "epoch": 3.0101925695390004,
      "grad_norm": 1.125,
      "learning_rate": 0.0004097813946419865,
      "loss": 4.7811,
      "mean_token_accuracy": 0.21477905362844468,
      "num_tokens": 67114708.0,
      "step": 38690
    },
    {
      "entropy": 5.367517375946045,
      "epoch": 3.0105815989107176,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004097592009789939,
      "loss": 4.9268,
      "mean_token_accuracy": 0.1958172455430031,
      "num_tokens": 67124257.0,
      "step": 38695
    },
    {
      "entropy": 5.402484989166259,
      "epoch": 3.0109706282824353,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040973700527124494,
      "loss": 4.9219,
      "mean_token_accuracy": 0.19898533970117568,
      "num_tokens": 67133066.0,
      "step": 38700
    },
    {
      "entropy": 5.454821968078614,
      "epoch": 3.011359657654153,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040971480751907633,
      "loss": 4.9179,
      "mean_token_accuracy": 0.2036113455891609,
      "num_tokens": 67141906.0,
      "step": 38705
    },
    {
      "entropy": 5.393882608413696,
      "epoch": 3.0117486870258703,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040969260772282494,
      "loss": 4.8748,
      "mean_token_accuracy": 0.203598515689373,
      "num_tokens": 67150533.0,
      "step": 38710
    },
    {
      "entropy": 5.386877918243409,
      "epoch": 3.012137716397588,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004096704058828277,
      "loss": 4.9031,
      "mean_token_accuracy": 0.20711033046245575,
      "num_tokens": 67159130.0,
      "step": 38715
    },
    {
      "entropy": 5.393054056167602,
      "epoch": 3.0125267457693057,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040964820199942135,
      "loss": 4.8315,
      "mean_token_accuracy": 0.21248968839645385,
      "num_tokens": 67168075.0,
      "step": 38720
    },
    {
      "entropy": 5.41006236076355,
      "epoch": 3.012915775141023,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004096259960729428,
      "loss": 4.8822,
      "mean_token_accuracy": 0.21490530669689178,
      "num_tokens": 67176972.0,
      "step": 38725
    },
    {
      "entropy": 5.410172939300537,
      "epoch": 3.0133048045127406,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040960378810372906,
      "loss": 4.9664,
      "mean_token_accuracy": 0.20212098211050034,
      "num_tokens": 67185091.0,
      "step": 38730
    },
    {
      "entropy": 5.3608965396881105,
      "epoch": 3.0136938338844583,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004095815780921171,
      "loss": 4.8964,
      "mean_token_accuracy": 0.21032601445913315,
      "num_tokens": 67193345.0,
      "step": 38735
    },
    {
      "entropy": 5.321224212646484,
      "epoch": 3.014082863256176,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004095593660384439,
      "loss": 4.8204,
      "mean_token_accuracy": 0.21454232633113862,
      "num_tokens": 67201893.0,
      "step": 38740
    },
    {
      "entropy": 5.424612522125244,
      "epoch": 3.0144718926278933,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004095371519430465,
      "loss": 4.9915,
      "mean_token_accuracy": 0.2006002515554428,
      "num_tokens": 67211112.0,
      "step": 38745
    },
    {
      "entropy": 5.3595709800720215,
      "epoch": 3.014860921999611,
      "grad_norm": 1.125,
      "learning_rate": 0.00040951493580626195,
      "loss": 4.7966,
      "mean_token_accuracy": 0.2097976341843605,
      "num_tokens": 67220078.0,
      "step": 38750
    },
    {
      "entropy": 5.352951097488403,
      "epoch": 3.0152499513713287,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004094927176284273,
      "loss": 4.9317,
      "mean_token_accuracy": 0.20370444655418396,
      "num_tokens": 67228883.0,
      "step": 38755
    },
    {
      "entropy": 5.447063016891479,
      "epoch": 3.015638980743046,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040947049740987977,
      "loss": 4.9989,
      "mean_token_accuracy": 0.19562507718801497,
      "num_tokens": 67237377.0,
      "step": 38760
    },
    {
      "entropy": 5.364328050613404,
      "epoch": 3.0160280101147636,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004094482751509565,
      "loss": 4.8428,
      "mean_token_accuracy": 0.2077786237001419,
      "num_tokens": 67245488.0,
      "step": 38765
    },
    {
      "entropy": 5.342042255401611,
      "epoch": 3.0164170394864813,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004094260508519947,
      "loss": 4.8691,
      "mean_token_accuracy": 0.21616282761096955,
      "num_tokens": 67253791.0,
      "step": 38770
    },
    {
      "entropy": 5.283930492401123,
      "epoch": 3.0168060688581986,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040940382451333144,
      "loss": 4.815,
      "mean_token_accuracy": 0.2101530283689499,
      "num_tokens": 67262373.0,
      "step": 38775
    },
    {
      "entropy": 5.300125980377198,
      "epoch": 3.0171950982299163,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040938159613530405,
      "loss": 4.831,
      "mean_token_accuracy": 0.21318756490945817,
      "num_tokens": 67270946.0,
      "step": 38780
    },
    {
      "entropy": 5.39377875328064,
      "epoch": 3.017584127601634,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004093593657182499,
      "loss": 4.9544,
      "mean_token_accuracy": 0.20606863647699356,
      "num_tokens": 67280150.0,
      "step": 38785
    },
    {
      "entropy": 5.467695569992065,
      "epoch": 3.0179731569733517,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004093371332625062,
      "loss": 4.8735,
      "mean_token_accuracy": 0.20999743789434433,
      "num_tokens": 67288109.0,
      "step": 38790
    },
    {
      "entropy": 5.314780282974243,
      "epoch": 3.018362186345069,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004093148987684103,
      "loss": 4.8275,
      "mean_token_accuracy": 0.20847830474376677,
      "num_tokens": 67296538.0,
      "step": 38795
    },
    {
      "entropy": 5.340438461303711,
      "epoch": 3.0187512157167866,
      "grad_norm": 1.125,
      "learning_rate": 0.00040929266223629966,
      "loss": 4.8447,
      "mean_token_accuracy": 0.20686030387878418,
      "num_tokens": 67306088.0,
      "step": 38800
    },
    {
      "entropy": 5.357239437103272,
      "epoch": 3.0191402450885043,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040927042366651164,
      "loss": 4.8695,
      "mean_token_accuracy": 0.20415499061346054,
      "num_tokens": 67315101.0,
      "step": 38805
    },
    {
      "entropy": 5.297847700119019,
      "epoch": 3.0195292744602216,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040924818305938354,
      "loss": 4.7908,
      "mean_token_accuracy": 0.2102730467915535,
      "num_tokens": 67323309.0,
      "step": 38810
    },
    {
      "entropy": 5.361377286911011,
      "epoch": 3.0199183038319393,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040922594041525303,
      "loss": 4.9319,
      "mean_token_accuracy": 0.20306951254606248,
      "num_tokens": 67331893.0,
      "step": 38815
    },
    {
      "entropy": 5.2935832977294925,
      "epoch": 3.020307333203657,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004092036957344574,
      "loss": 4.8043,
      "mean_token_accuracy": 0.21503401398658753,
      "num_tokens": 67339804.0,
      "step": 38820
    },
    {
      "entropy": 5.339497804641724,
      "epoch": 3.0206963625753747,
      "grad_norm": 1.125,
      "learning_rate": 0.00040918144901733443,
      "loss": 4.8466,
      "mean_token_accuracy": 0.2111998289823532,
      "num_tokens": 67348306.0,
      "step": 38825
    },
    {
      "entropy": 5.391669225692749,
      "epoch": 3.021085391947092,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004091592002642215,
      "loss": 4.9183,
      "mean_token_accuracy": 0.20186142176389693,
      "num_tokens": 67356601.0,
      "step": 38830
    },
    {
      "entropy": 5.285781335830689,
      "epoch": 3.0214744213188096,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004091369494754562,
      "loss": 4.8705,
      "mean_token_accuracy": 0.2093329533934593,
      "num_tokens": 67364990.0,
      "step": 38835
    },
    {
      "entropy": 5.44060001373291,
      "epoch": 3.0218634506905273,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004091146966513763,
      "loss": 4.9452,
      "mean_token_accuracy": 0.20239161998033522,
      "num_tokens": 67373165.0,
      "step": 38840
    },
    {
      "entropy": 5.403328990936279,
      "epoch": 3.0222524800622446,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040909244179231926,
      "loss": 4.9279,
      "mean_token_accuracy": 0.20034720748662949,
      "num_tokens": 67381627.0,
      "step": 38845
    },
    {
      "entropy": 5.303368186950683,
      "epoch": 3.0226415094339623,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004090701848986229,
      "loss": 4.7896,
      "mean_token_accuracy": 0.21321910619735718,
      "num_tokens": 67389928.0,
      "step": 38850
    },
    {
      "entropy": 5.344474411010742,
      "epoch": 3.02303053880568,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004090479259706247,
      "loss": 4.8431,
      "mean_token_accuracy": 0.209566567838192,
      "num_tokens": 67398377.0,
      "step": 38855
    },
    {
      "entropy": 5.31518406867981,
      "epoch": 3.023419568177397,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004090256650086628,
      "loss": 4.761,
      "mean_token_accuracy": 0.2134068340063095,
      "num_tokens": 67406490.0,
      "step": 38860
    },
    {
      "entropy": 5.314701795578003,
      "epoch": 3.023808597549115,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040900340201307475,
      "loss": 4.8686,
      "mean_token_accuracy": 0.2088356539607048,
      "num_tokens": 67415178.0,
      "step": 38865
    },
    {
      "entropy": 5.3110496520996096,
      "epoch": 3.0241976269208326,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040898113698419823,
      "loss": 4.8866,
      "mean_token_accuracy": 0.20884716212749482,
      "num_tokens": 67423770.0,
      "step": 38870
    },
    {
      "entropy": 5.37125597000122,
      "epoch": 3.0245866562925503,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004089588699223713,
      "loss": 4.9701,
      "mean_token_accuracy": 0.20181357860565186,
      "num_tokens": 67432771.0,
      "step": 38875
    },
    {
      "entropy": 5.363561725616455,
      "epoch": 3.0249756856642676,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040893660082793167,
      "loss": 4.9035,
      "mean_token_accuracy": 0.20304954499006272,
      "num_tokens": 67441697.0,
      "step": 38880
    },
    {
      "entropy": 5.390934038162231,
      "epoch": 3.0253647150359853,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040891432970121736,
      "loss": 4.9809,
      "mean_token_accuracy": 0.194979690015316,
      "num_tokens": 67450950.0,
      "step": 38885
    },
    {
      "entropy": 5.333687210083008,
      "epoch": 3.025753744407703,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040889205654256615,
      "loss": 4.812,
      "mean_token_accuracy": 0.21041290163993837,
      "num_tokens": 67459863.0,
      "step": 38890
    },
    {
      "entropy": 5.397540283203125,
      "epoch": 3.02614277377942,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040886978135231625,
      "loss": 4.8983,
      "mean_token_accuracy": 0.20955174863338472,
      "num_tokens": 67468829.0,
      "step": 38895
    },
    {
      "entropy": 5.323147869110107,
      "epoch": 3.026531803151138,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040884750413080535,
      "loss": 4.8122,
      "mean_token_accuracy": 0.2131826788187027,
      "num_tokens": 67477979.0,
      "step": 38900
    },
    {
      "entropy": 5.378287696838379,
      "epoch": 3.0269208325228556,
      "grad_norm": 1.125,
      "learning_rate": 0.00040882522487837176,
      "loss": 4.9376,
      "mean_token_accuracy": 0.20290352553129196,
      "num_tokens": 67486929.0,
      "step": 38905
    },
    {
      "entropy": 5.330934715270996,
      "epoch": 3.027309861894573,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040880294359535327,
      "loss": 4.8127,
      "mean_token_accuracy": 0.20899835079908372,
      "num_tokens": 67495434.0,
      "step": 38910
    },
    {
      "entropy": 5.374479818344116,
      "epoch": 3.0276988912662905,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040878066028208804,
      "loss": 4.8956,
      "mean_token_accuracy": 0.20191641598939897,
      "num_tokens": 67504437.0,
      "step": 38915
    },
    {
      "entropy": 5.328443574905395,
      "epoch": 3.0280879206380082,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040875837493891434,
      "loss": 4.9074,
      "mean_token_accuracy": 0.20944027304649354,
      "num_tokens": 67512899.0,
      "step": 38920
    },
    {
      "entropy": 5.396648979187011,
      "epoch": 3.028476950009726,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004087360875661701,
      "loss": 4.9399,
      "mean_token_accuracy": 0.19821995943784715,
      "num_tokens": 67521602.0,
      "step": 38925
    },
    {
      "entropy": 5.403042984008789,
      "epoch": 3.028865979381443,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040871379816419376,
      "loss": 4.909,
      "mean_token_accuracy": 0.20549943149089814,
      "num_tokens": 67530808.0,
      "step": 38930
    },
    {
      "entropy": 5.436919832229615,
      "epoch": 3.029255008753161,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004086915067333232,
      "loss": 4.8586,
      "mean_token_accuracy": 0.20334610491991043,
      "num_tokens": 67539216.0,
      "step": 38935
    },
    {
      "entropy": 5.329461574554443,
      "epoch": 3.0296440381248786,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004086692132738968,
      "loss": 4.852,
      "mean_token_accuracy": 0.20223007649183272,
      "num_tokens": 67547202.0,
      "step": 38940
    },
    {
      "entropy": 5.406212186813354,
      "epoch": 3.030033067496596,
      "grad_norm": 1.046875,
      "learning_rate": 0.000408646917786253,
      "loss": 5.0343,
      "mean_token_accuracy": 0.19755114018917083,
      "num_tokens": 67556427.0,
      "step": 38945
    },
    {
      "entropy": 5.453958511352539,
      "epoch": 3.0304220968683135,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004086246202707298,
      "loss": 4.9598,
      "mean_token_accuracy": 0.20237015336751937,
      "num_tokens": 67564945.0,
      "step": 38950
    },
    {
      "entropy": 5.438914775848389,
      "epoch": 3.0308111262400312,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040860232072766577,
      "loss": 5.005,
      "mean_token_accuracy": 0.20529320538043977,
      "num_tokens": 67574336.0,
      "step": 38955
    },
    {
      "entropy": 5.428966379165649,
      "epoch": 3.0312001556117485,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004085800191573991,
      "loss": 4.9271,
      "mean_token_accuracy": 0.20519212037324905,
      "num_tokens": 67583628.0,
      "step": 38960
    },
    {
      "entropy": 5.378624153137207,
      "epoch": 3.031589184983466,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004085577155602683,
      "loss": 4.8963,
      "mean_token_accuracy": 0.20692189186811447,
      "num_tokens": 67592462.0,
      "step": 38965
    },
    {
      "entropy": 5.404040718078614,
      "epoch": 3.031978214355184,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040853540993661175,
      "loss": 4.8707,
      "mean_token_accuracy": 0.2100987493991852,
      "num_tokens": 67601377.0,
      "step": 38970
    },
    {
      "entropy": 5.275073385238647,
      "epoch": 3.0323672437269016,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040851310228676787,
      "loss": 4.772,
      "mean_token_accuracy": 0.215145543217659,
      "num_tokens": 67610253.0,
      "step": 38975
    },
    {
      "entropy": 5.322058868408203,
      "epoch": 3.032756273098619,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004084907926110752,
      "loss": 4.9202,
      "mean_token_accuracy": 0.20278232991695405,
      "num_tokens": 67618092.0,
      "step": 38980
    },
    {
      "entropy": 5.343905687332153,
      "epoch": 3.0331453024703365,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004084684809098722,
      "loss": 4.8529,
      "mean_token_accuracy": 0.21086602061986923,
      "num_tokens": 67626547.0,
      "step": 38985
    },
    {
      "entropy": 5.447866153717041,
      "epoch": 3.0335343318420542,
      "grad_norm": 1.125,
      "learning_rate": 0.00040844616718349743,
      "loss": 4.9869,
      "mean_token_accuracy": 0.1968732863664627,
      "num_tokens": 67635891.0,
      "step": 38990
    },
    {
      "entropy": 5.385292434692383,
      "epoch": 3.0339233612137715,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040842385143228946,
      "loss": 4.8402,
      "mean_token_accuracy": 0.20671410262584686,
      "num_tokens": 67645030.0,
      "step": 38995
    },
    {
      "entropy": 5.33613862991333,
      "epoch": 3.034312390585489,
      "grad_norm": 1.328125,
      "learning_rate": 0.00040840153365658693,
      "loss": 4.9105,
      "mean_token_accuracy": 0.20903646796941758,
      "num_tokens": 67652941.0,
      "step": 39000
    },
    {
      "epoch": 3.034312390585489,
      "eval_entropy": 5.113205782006404,
      "eval_loss": 5.0559210777282715,
      "eval_mean_token_accuracy": 0.20652723389536976,
      "eval_num_tokens": 67652941.0,
      "eval_runtime": 28.7579,
      "eval_samples_per_second": 1445.097,
      "eval_steps_per_second": 180.646,
      "step": 39000
    },
    {
      "entropy": 5.350287580490113,
      "epoch": 3.034701419957207,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040837921385672855,
      "loss": 4.8354,
      "mean_token_accuracy": 0.20858560204505922,
      "num_tokens": 67661600.0,
      "step": 39005
    },
    {
      "entropy": 5.396736669540405,
      "epoch": 3.035090449328924,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040835689203305275,
      "loss": 4.9488,
      "mean_token_accuracy": 0.19933412373065948,
      "num_tokens": 67670317.0,
      "step": 39010
    },
    {
      "entropy": 5.337585306167602,
      "epoch": 3.035479478700642,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040833456818589845,
      "loss": 4.8631,
      "mean_token_accuracy": 0.20599283576011657,
      "num_tokens": 67679065.0,
      "step": 39015
    },
    {
      "entropy": 5.377735233306884,
      "epoch": 3.0358685080723595,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040831224231560424,
      "loss": 4.8546,
      "mean_token_accuracy": 0.2172616735100746,
      "num_tokens": 67687724.0,
      "step": 39020
    },
    {
      "entropy": 5.3365734100341795,
      "epoch": 3.0362575374440772,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000408289914422509,
      "loss": 4.8566,
      "mean_token_accuracy": 0.21153998672962188,
      "num_tokens": 67695846.0,
      "step": 39025
    },
    {
      "entropy": 5.381910514831543,
      "epoch": 3.0366465668157945,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004082675845069515,
      "loss": 4.8807,
      "mean_token_accuracy": 0.21139136105775833,
      "num_tokens": 67704675.0,
      "step": 39030
    },
    {
      "entropy": 5.365243911743164,
      "epoch": 3.037035596187512,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004082452525692704,
      "loss": 4.8431,
      "mean_token_accuracy": 0.21302517652511596,
      "num_tokens": 67713836.0,
      "step": 39035
    },
    {
      "entropy": 5.324962997436524,
      "epoch": 3.03742462555923,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040822291860980484,
      "loss": 4.8795,
      "mean_token_accuracy": 0.20974415838718413,
      "num_tokens": 67722920.0,
      "step": 39040
    },
    {
      "entropy": 5.3505021095275875,
      "epoch": 3.037813654930947,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040820058262889343,
      "loss": 4.945,
      "mean_token_accuracy": 0.2067013993859291,
      "num_tokens": 67732285.0,
      "step": 39045
    },
    {
      "entropy": 5.435254144668579,
      "epoch": 3.038202684302665,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040817824462687516,
      "loss": 4.868,
      "mean_token_accuracy": 0.21356982439756395,
      "num_tokens": 67739836.0,
      "step": 39050
    },
    {
      "entropy": 5.314601135253906,
      "epoch": 3.0385917136743825,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040815590460408906,
      "loss": 4.8494,
      "mean_token_accuracy": 0.20694127678871155,
      "num_tokens": 67748833.0,
      "step": 39055
    },
    {
      "entropy": 5.248241901397705,
      "epoch": 3.0389807430460998,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040813356256087395,
      "loss": 4.7656,
      "mean_token_accuracy": 0.21697914153337478,
      "num_tokens": 67756778.0,
      "step": 39060
    },
    {
      "entropy": 5.390316534042358,
      "epoch": 3.0393697724178175,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004081112184975691,
      "loss": 4.8888,
      "mean_token_accuracy": 0.20254572629928588,
      "num_tokens": 67765691.0,
      "step": 39065
    },
    {
      "entropy": 5.3137260437011715,
      "epoch": 3.039758801789535,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004080888724145132,
      "loss": 4.8023,
      "mean_token_accuracy": 0.21399228423833846,
      "num_tokens": 67774685.0,
      "step": 39070
    },
    {
      "entropy": 5.346880578994751,
      "epoch": 3.040147831161253,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004080665243120456,
      "loss": 4.966,
      "mean_token_accuracy": 0.19960699528455733,
      "num_tokens": 67783928.0,
      "step": 39075
    },
    {
      "entropy": 5.382338285446167,
      "epoch": 3.04053686053297,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004080441741905053,
      "loss": 4.8872,
      "mean_token_accuracy": 0.21014311760663987,
      "num_tokens": 67792793.0,
      "step": 39080
    },
    {
      "entropy": 5.368513488769532,
      "epoch": 3.040925889904688,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004080218220502314,
      "loss": 4.8743,
      "mean_token_accuracy": 0.2098890721797943,
      "num_tokens": 67801648.0,
      "step": 39085
    },
    {
      "entropy": 5.438955354690552,
      "epoch": 3.0413149192764055,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040799946789156304,
      "loss": 5.0108,
      "mean_token_accuracy": 0.19016854614019393,
      "num_tokens": 67810390.0,
      "step": 39090
    },
    {
      "entropy": 5.3781956195831295,
      "epoch": 3.0417039486481228,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040797711171483954,
      "loss": 4.8545,
      "mean_token_accuracy": 0.20942164212465286,
      "num_tokens": 67819091.0,
      "step": 39095
    },
    {
      "entropy": 5.388621234893799,
      "epoch": 3.0420929780198405,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004079547535203999,
      "loss": 4.8963,
      "mean_token_accuracy": 0.20105807334184647,
      "num_tokens": 67827901.0,
      "step": 39100
    },
    {
      "entropy": 5.368036079406738,
      "epoch": 3.042482007391558,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004079323933085836,
      "loss": 4.8627,
      "mean_token_accuracy": 0.20998878479003907,
      "num_tokens": 67836221.0,
      "step": 39105
    },
    {
      "entropy": 5.384793758392334,
      "epoch": 3.0428710367632754,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040791003107972976,
      "loss": 4.9343,
      "mean_token_accuracy": 0.2035637229681015,
      "num_tokens": 67845565.0,
      "step": 39110
    },
    {
      "entropy": 5.406043863296508,
      "epoch": 3.043260066134993,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004078876668341778,
      "loss": 4.9866,
      "mean_token_accuracy": 0.20261608213186263,
      "num_tokens": 67854433.0,
      "step": 39115
    },
    {
      "entropy": 5.40343542098999,
      "epoch": 3.043649095506711,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004078653005722669,
      "loss": 4.8605,
      "mean_token_accuracy": 0.2104961797595024,
      "num_tokens": 67862511.0,
      "step": 39120
    },
    {
      "entropy": 5.387297296524048,
      "epoch": 3.0440381248784285,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004078429322943366,
      "loss": 4.8825,
      "mean_token_accuracy": 0.20340321958065033,
      "num_tokens": 67871145.0,
      "step": 39125
    },
    {
      "entropy": 5.27274751663208,
      "epoch": 3.0444271542501458,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040782056200072627,
      "loss": 4.8138,
      "mean_token_accuracy": 0.21231719255447387,
      "num_tokens": 67879377.0,
      "step": 39130
    },
    {
      "entropy": 5.303968811035157,
      "epoch": 3.0448161836218635,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040779818969177527,
      "loss": 4.8121,
      "mean_token_accuracy": 0.21548170894384383,
      "num_tokens": 67888073.0,
      "step": 39135
    },
    {
      "entropy": 5.414448547363281,
      "epoch": 3.045205212993581,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040777581536782315,
      "loss": 4.9558,
      "mean_token_accuracy": 0.1991024613380432,
      "num_tokens": 67896426.0,
      "step": 39140
    },
    {
      "entropy": 5.356995582580566,
      "epoch": 3.0455942423652984,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004077534390292093,
      "loss": 4.831,
      "mean_token_accuracy": 0.21296161264181138,
      "num_tokens": 67904744.0,
      "step": 39145
    },
    {
      "entropy": 5.322750806808472,
      "epoch": 3.045983271737016,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040773106067627335,
      "loss": 4.8225,
      "mean_token_accuracy": 0.20914295315742493,
      "num_tokens": 67912893.0,
      "step": 39150
    },
    {
      "entropy": 5.33179497718811,
      "epoch": 3.046372301108734,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040770868030935486,
      "loss": 4.8789,
      "mean_token_accuracy": 0.203414948284626,
      "num_tokens": 67921150.0,
      "step": 39155
    },
    {
      "entropy": 5.375274276733398,
      "epoch": 3.046761330480451,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040768629792879344,
      "loss": 4.8831,
      "mean_token_accuracy": 0.20690160542726516,
      "num_tokens": 67929686.0,
      "step": 39160
    },
    {
      "entropy": 5.440552663803101,
      "epoch": 3.0471503598521688,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004076639135349284,
      "loss": 4.9485,
      "mean_token_accuracy": 0.2051393285393715,
      "num_tokens": 67937523.0,
      "step": 39165
    },
    {
      "entropy": 5.367649126052856,
      "epoch": 3.0475393892238865,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040764152712809983,
      "loss": 4.8495,
      "mean_token_accuracy": 0.20849025696516038,
      "num_tokens": 67945368.0,
      "step": 39170
    },
    {
      "entropy": 5.4230005741119385,
      "epoch": 3.047928418595604,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004076191387086472,
      "loss": 4.9603,
      "mean_token_accuracy": 0.20741447508335115,
      "num_tokens": 67953769.0,
      "step": 39175
    },
    {
      "entropy": 5.417774248123169,
      "epoch": 3.0483174479673214,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040759674827691014,
      "loss": 4.8883,
      "mean_token_accuracy": 0.20528468042612075,
      "num_tokens": 67962131.0,
      "step": 39180
    },
    {
      "entropy": 5.321724796295166,
      "epoch": 3.048706477339039,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004075743558332285,
      "loss": 4.8746,
      "mean_token_accuracy": 0.2095373287796974,
      "num_tokens": 67971344.0,
      "step": 39185
    },
    {
      "entropy": 5.326876640319824,
      "epoch": 3.049095506710757,
      "grad_norm": 1.109375,
      "learning_rate": 0.000407551961377942,
      "loss": 4.8636,
      "mean_token_accuracy": 0.20781180262565613,
      "num_tokens": 67979553.0,
      "step": 39190
    },
    {
      "entropy": 5.35449185371399,
      "epoch": 3.049484536082474,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004075295649113905,
      "loss": 4.9631,
      "mean_token_accuracy": 0.20586841851472854,
      "num_tokens": 67989821.0,
      "step": 39195
    },
    {
      "entropy": 5.476725482940674,
      "epoch": 3.0498735654541917,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00040750716643391384,
      "loss": 4.9781,
      "mean_token_accuracy": 0.20524748265743256,
      "num_tokens": 67998727.0,
      "step": 39200
    },
    {
      "entropy": 5.399300289154053,
      "epoch": 3.0502625948259094,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004074847659458518,
      "loss": 4.8075,
      "mean_token_accuracy": 0.20670504868030548,
      "num_tokens": 68007158.0,
      "step": 39205
    },
    {
      "entropy": 5.292526721954346,
      "epoch": 3.050651624197627,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040746236344754423,
      "loss": 4.7599,
      "mean_token_accuracy": 0.21635707020759581,
      "num_tokens": 68015196.0,
      "step": 39210
    },
    {
      "entropy": 5.296192359924317,
      "epoch": 3.0510406535693444,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004074399589393312,
      "loss": 4.8533,
      "mean_token_accuracy": 0.20391743779182434,
      "num_tokens": 68023983.0,
      "step": 39215
    },
    {
      "entropy": 5.366579246520996,
      "epoch": 3.051429682941062,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004074175524215526,
      "loss": 4.868,
      "mean_token_accuracy": 0.20453790724277496,
      "num_tokens": 68032347.0,
      "step": 39220
    },
    {
      "entropy": 5.333183717727661,
      "epoch": 3.05181871231278,
      "grad_norm": 1.125,
      "learning_rate": 0.0004073951438945484,
      "loss": 4.8559,
      "mean_token_accuracy": 0.2089476466178894,
      "num_tokens": 68041463.0,
      "step": 39225
    },
    {
      "entropy": 5.430872440338135,
      "epoch": 3.052207741684497,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040737273335865866,
      "loss": 4.9826,
      "mean_token_accuracy": 0.20468396544456482,
      "num_tokens": 68050647.0,
      "step": 39230
    },
    {
      "entropy": 5.339348030090332,
      "epoch": 3.0525967710562147,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040735032081422334,
      "loss": 4.7646,
      "mean_token_accuracy": 0.21750395596027375,
      "num_tokens": 68059488.0,
      "step": 39235
    },
    {
      "entropy": 5.302882957458496,
      "epoch": 3.0529858004279324,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040732790626158257,
      "loss": 4.853,
      "mean_token_accuracy": 0.20718845427036287,
      "num_tokens": 68068451.0,
      "step": 39240
    },
    {
      "entropy": 5.3747971534729,
      "epoch": 3.0533748297996497,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040730548970107646,
      "loss": 4.8728,
      "mean_token_accuracy": 0.20101031810045242,
      "num_tokens": 68077556.0,
      "step": 39245
    },
    {
      "entropy": 5.355021142959595,
      "epoch": 3.0537638591713674,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004072830711330451,
      "loss": 4.8149,
      "mean_token_accuracy": 0.20541394203901292,
      "num_tokens": 68086677.0,
      "step": 39250
    },
    {
      "entropy": 5.311465740203857,
      "epoch": 3.054152888543085,
      "grad_norm": 1.125,
      "learning_rate": 0.0004072606505578287,
      "loss": 4.829,
      "mean_token_accuracy": 0.20261309742927552,
      "num_tokens": 68094968.0,
      "step": 39255
    },
    {
      "entropy": 5.310690546035767,
      "epoch": 3.054541917914803,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040723822797576737,
      "loss": 4.85,
      "mean_token_accuracy": 0.20886177569627762,
      "num_tokens": 68103563.0,
      "step": 39260
    },
    {
      "entropy": 5.370297431945801,
      "epoch": 3.05493094728652,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004072158033872015,
      "loss": 4.9116,
      "mean_token_accuracy": 0.2063387632369995,
      "num_tokens": 68111571.0,
      "step": 39265
    },
    {
      "entropy": 5.494987487792969,
      "epoch": 3.0553199766582377,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004071933767924713,
      "loss": 5.0806,
      "mean_token_accuracy": 0.1934620812535286,
      "num_tokens": 68119955.0,
      "step": 39270
    },
    {
      "entropy": 5.40981125831604,
      "epoch": 3.0557090060299554,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040717094819191685,
      "loss": 4.845,
      "mean_token_accuracy": 0.21306700259447098,
      "num_tokens": 68127989.0,
      "step": 39275
    },
    {
      "entropy": 5.31781268119812,
      "epoch": 3.0560980354016727,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040714851758587875,
      "loss": 4.7485,
      "mean_token_accuracy": 0.20780339390039443,
      "num_tokens": 68136291.0,
      "step": 39280
    },
    {
      "entropy": 5.384258174896241,
      "epoch": 3.0564870647733904,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040712608497469715,
      "loss": 5.051,
      "mean_token_accuracy": 0.1990741476416588,
      "num_tokens": 68145671.0,
      "step": 39285
    },
    {
      "entropy": 5.346583652496338,
      "epoch": 3.056876094145108,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004071036503587126,
      "loss": 4.7801,
      "mean_token_accuracy": 0.2134235084056854,
      "num_tokens": 68154183.0,
      "step": 39290
    },
    {
      "entropy": 5.433489608764648,
      "epoch": 3.0572651235168253,
      "grad_norm": 1.25,
      "learning_rate": 0.0004070812137382654,
      "loss": 4.9845,
      "mean_token_accuracy": 0.19338323622941972,
      "num_tokens": 68163537.0,
      "step": 39295
    },
    {
      "entropy": 5.360364389419556,
      "epoch": 3.057654152888543,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000407058775113696,
      "loss": 4.8672,
      "mean_token_accuracy": 0.20649701952934266,
      "num_tokens": 68172052.0,
      "step": 39300
    },
    {
      "entropy": 5.370936059951783,
      "epoch": 3.0580431822602607,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004070363344853449,
      "loss": 4.8867,
      "mean_token_accuracy": 0.20461745858192443,
      "num_tokens": 68180807.0,
      "step": 39305
    },
    {
      "entropy": 5.392648410797119,
      "epoch": 3.0584322116319784,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040701389185355256,
      "loss": 4.9076,
      "mean_token_accuracy": 0.20824662148952483,
      "num_tokens": 68189772.0,
      "step": 39310
    },
    {
      "entropy": 5.5143177032470705,
      "epoch": 3.0588212410036957,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004069914472186595,
      "loss": 5.004,
      "mean_token_accuracy": 0.20063359886407853,
      "num_tokens": 68197816.0,
      "step": 39315
    },
    {
      "entropy": 5.3398984432220455,
      "epoch": 3.0592102703754134,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004069690005810063,
      "loss": 4.8075,
      "mean_token_accuracy": 0.21724888682365417,
      "num_tokens": 68206799.0,
      "step": 39320
    },
    {
      "entropy": 5.459707021713257,
      "epoch": 3.059599299747131,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040694655194093365,
      "loss": 5.107,
      "mean_token_accuracy": 0.19280206710100173,
      "num_tokens": 68217107.0,
      "step": 39325
    },
    {
      "entropy": 5.385234689712524,
      "epoch": 3.0599883291188483,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040692410129878205,
      "loss": 4.9101,
      "mean_token_accuracy": 0.20619029700756072,
      "num_tokens": 68225587.0,
      "step": 39330
    },
    {
      "entropy": 5.456292915344238,
      "epoch": 3.060377358490566,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004069016486548922,
      "loss": 4.9531,
      "mean_token_accuracy": 0.20092508643865586,
      "num_tokens": 68234090.0,
      "step": 39335
    },
    {
      "entropy": 5.296092462539673,
      "epoch": 3.0607663878622837,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004068791940096048,
      "loss": 4.8427,
      "mean_token_accuracy": 0.21250622421503068,
      "num_tokens": 68242904.0,
      "step": 39340
    },
    {
      "entropy": 5.3931621551513675,
      "epoch": 3.061155417234001,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040685673736326044,
      "loss": 4.9701,
      "mean_token_accuracy": 0.20412750095129012,
      "num_tokens": 68251232.0,
      "step": 39345
    },
    {
      "entropy": 5.4876807689666744,
      "epoch": 3.0615444466057187,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004068342787162001,
      "loss": 4.9557,
      "mean_token_accuracy": 0.2000432327389717,
      "num_tokens": 68259658.0,
      "step": 39350
    },
    {
      "entropy": 5.345956230163575,
      "epoch": 3.0619334759774364,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040681181806876435,
      "loss": 4.8233,
      "mean_token_accuracy": 0.2140554592013359,
      "num_tokens": 68268421.0,
      "step": 39355
    },
    {
      "entropy": 5.3152477741241455,
      "epoch": 3.062322505349154,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000406789355421294,
      "loss": 4.9453,
      "mean_token_accuracy": 0.20442201495170592,
      "num_tokens": 68277564.0,
      "step": 39360
    },
    {
      "entropy": 5.4056840419769285,
      "epoch": 3.0627115347208713,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004067668907741301,
      "loss": 4.9581,
      "mean_token_accuracy": 0.2014770969748497,
      "num_tokens": 68286305.0,
      "step": 39365
    },
    {
      "entropy": 5.419165372848511,
      "epoch": 3.063100564092589,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004067444241276132,
      "loss": 4.8946,
      "mean_token_accuracy": 0.20499465614557266,
      "num_tokens": 68294536.0,
      "step": 39370
    },
    {
      "entropy": 5.295297002792358,
      "epoch": 3.0634895934643067,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040672195548208443,
      "loss": 4.8431,
      "mean_token_accuracy": 0.2146304503083229,
      "num_tokens": 68302995.0,
      "step": 39375
    },
    {
      "entropy": 5.283709764480591,
      "epoch": 3.063878622836024,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004066994848378846,
      "loss": 4.8304,
      "mean_token_accuracy": 0.21160885095596313,
      "num_tokens": 68311643.0,
      "step": 39380
    },
    {
      "entropy": 5.320725059509277,
      "epoch": 3.0642676522077417,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040667701219535473,
      "loss": 4.8142,
      "mean_token_accuracy": 0.21575698107481003,
      "num_tokens": 68319317.0,
      "step": 39385
    },
    {
      "entropy": 5.32264404296875,
      "epoch": 3.0646566815794594,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004066545375548358,
      "loss": 4.8198,
      "mean_token_accuracy": 0.2110499322414398,
      "num_tokens": 68328116.0,
      "step": 39390
    },
    {
      "entropy": 5.3408191204071045,
      "epoch": 3.0650457109511766,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004066320609166689,
      "loss": 4.967,
      "mean_token_accuracy": 0.2038261592388153,
      "num_tokens": 68336953.0,
      "step": 39395
    },
    {
      "entropy": 5.363026809692383,
      "epoch": 3.0654347403228943,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004066095822811949,
      "loss": 4.9305,
      "mean_token_accuracy": 0.2025128111243248,
      "num_tokens": 68346152.0,
      "step": 39400
    },
    {
      "entropy": 5.446612215042114,
      "epoch": 3.065823769694612,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000406587101648755,
      "loss": 4.9631,
      "mean_token_accuracy": 0.20267566293478012,
      "num_tokens": 68354785.0,
      "step": 39405
    },
    {
      "entropy": 5.3995355606079105,
      "epoch": 3.0662127990663297,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004065646190196902,
      "loss": 4.8641,
      "mean_token_accuracy": 0.2094322293996811,
      "num_tokens": 68363323.0,
      "step": 39410
    },
    {
      "entropy": 5.3029882431030275,
      "epoch": 3.066601828438047,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004065421343943417,
      "loss": 4.8844,
      "mean_token_accuracy": 0.2083094373345375,
      "num_tokens": 68371461.0,
      "step": 39415
    },
    {
      "entropy": 5.400768852233886,
      "epoch": 3.0669908578097647,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040651964777305074,
      "loss": 4.9174,
      "mean_token_accuracy": 0.20454638749361037,
      "num_tokens": 68380356.0,
      "step": 39420
    },
    {
      "entropy": 5.40072808265686,
      "epoch": 3.0673798871814824,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004064971591561585,
      "loss": 4.9335,
      "mean_token_accuracy": 0.20237781405448912,
      "num_tokens": 68389029.0,
      "step": 39425
    },
    {
      "entropy": 5.291285991668701,
      "epoch": 3.0677689165531996,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040647466854400606,
      "loss": 4.7826,
      "mean_token_accuracy": 0.21494049280881883,
      "num_tokens": 68397481.0,
      "step": 39430
    },
    {
      "entropy": 5.2940027713775635,
      "epoch": 3.0681579459249173,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004064521759369349,
      "loss": 4.8442,
      "mean_token_accuracy": 0.21035151034593583,
      "num_tokens": 68405459.0,
      "step": 39435
    },
    {
      "entropy": 5.243048095703125,
      "epoch": 3.068546975296635,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040642968133528613,
      "loss": 4.8084,
      "mean_token_accuracy": 0.21240645945072173,
      "num_tokens": 68413840.0,
      "step": 39440
    },
    {
      "entropy": 5.3717375755310055,
      "epoch": 3.0689360046683523,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004064071847394012,
      "loss": 4.9545,
      "mean_token_accuracy": 0.2029716983437538,
      "num_tokens": 68422608.0,
      "step": 39445
    },
    {
      "entropy": 5.36797513961792,
      "epoch": 3.06932503404007,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004063846861496213,
      "loss": 4.8751,
      "mean_token_accuracy": 0.20841484665870666,
      "num_tokens": 68431484.0,
      "step": 39450
    },
    {
      "entropy": 5.405462408065796,
      "epoch": 3.0697140634117877,
      "grad_norm": 1.15625,
      "learning_rate": 0.000406362185566288,
      "loss": 4.8913,
      "mean_token_accuracy": 0.20354299247264862,
      "num_tokens": 68440031.0,
      "step": 39455
    },
    {
      "entropy": 5.371223449707031,
      "epoch": 3.0701030927835053,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004063396829897426,
      "loss": 4.9783,
      "mean_token_accuracy": 0.20123460292816162,
      "num_tokens": 68449207.0,
      "step": 39460
    },
    {
      "entropy": 5.417889165878296,
      "epoch": 3.0704921221552226,
      "grad_norm": 1.125,
      "learning_rate": 0.0004063171784203265,
      "loss": 4.8596,
      "mean_token_accuracy": 0.2032051295042038,
      "num_tokens": 68457985.0,
      "step": 39465
    },
    {
      "entropy": 5.371247673034668,
      "epoch": 3.0708811515269403,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004062946718583813,
      "loss": 4.9342,
      "mean_token_accuracy": 0.20044719874858857,
      "num_tokens": 68466364.0,
      "step": 39470
    },
    {
      "entropy": 5.460356616973877,
      "epoch": 3.071270180898658,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040627216330424844,
      "loss": 4.9465,
      "mean_token_accuracy": 0.20424651950597764,
      "num_tokens": 68475250.0,
      "step": 39475
    },
    {
      "entropy": 5.410144901275634,
      "epoch": 3.0716592102703753,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004062496527582694,
      "loss": 4.9488,
      "mean_token_accuracy": 0.20227152109146118,
      "num_tokens": 68483696.0,
      "step": 39480
    },
    {
      "entropy": 5.371443128585815,
      "epoch": 3.072048239642093,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004062271402207858,
      "loss": 4.9115,
      "mean_token_accuracy": 0.20254356265068055,
      "num_tokens": 68491988.0,
      "step": 39485
    },
    {
      "entropy": 5.424184703826905,
      "epoch": 3.0724372690138106,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004062046256921391,
      "loss": 4.9799,
      "mean_token_accuracy": 0.19297416210174562,
      "num_tokens": 68500477.0,
      "step": 39490
    },
    {
      "entropy": 5.369602823257447,
      "epoch": 3.072826298385528,
      "grad_norm": 1.125,
      "learning_rate": 0.00040618210917267117,
      "loss": 4.8572,
      "mean_token_accuracy": 0.21553044021129608,
      "num_tokens": 68509363.0,
      "step": 39495
    },
    {
      "entropy": 5.344871187210083,
      "epoch": 3.0732153277572456,
      "grad_norm": 1.125,
      "learning_rate": 0.00040615959066272355,
      "loss": 4.8625,
      "mean_token_accuracy": 0.20982418209314346,
      "num_tokens": 68518485.0,
      "step": 39500
    },
    {
      "entropy": 5.305203104019165,
      "epoch": 3.0736043571289633,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004061370701626379,
      "loss": 4.7165,
      "mean_token_accuracy": 0.21390097141265868,
      "num_tokens": 68526913.0,
      "step": 39505
    },
    {
      "entropy": 5.365344524383545,
      "epoch": 3.073993386500681,
      "grad_norm": 1.125,
      "learning_rate": 0.0004061145476727559,
      "loss": 4.8962,
      "mean_token_accuracy": 0.21064141988754273,
      "num_tokens": 68535594.0,
      "step": 39510
    },
    {
      "entropy": 5.358437490463257,
      "epoch": 3.0743824158723982,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004060920231934193,
      "loss": 4.8861,
      "mean_token_accuracy": 0.20125058442354202,
      "num_tokens": 68544380.0,
      "step": 39515
    },
    {
      "entropy": 5.382339668273926,
      "epoch": 3.074771445244116,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004060694967249699,
      "loss": 4.8427,
      "mean_token_accuracy": 0.2104205533862114,
      "num_tokens": 68552535.0,
      "step": 39520
    },
    {
      "entropy": 5.42639389038086,
      "epoch": 3.0751604746158336,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004060469682677496,
      "loss": 5.0394,
      "mean_token_accuracy": 0.20230480879545212,
      "num_tokens": 68561920.0,
      "step": 39525
    },
    {
      "entropy": 5.340296411514283,
      "epoch": 3.075549503987551,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040602443782210003,
      "loss": 4.856,
      "mean_token_accuracy": 0.2111328735947609,
      "num_tokens": 68571053.0,
      "step": 39530
    },
    {
      "entropy": 5.394915676116943,
      "epoch": 3.0759385333592686,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004060019053883632,
      "loss": 4.9075,
      "mean_token_accuracy": 0.20821771025657654,
      "num_tokens": 68579716.0,
      "step": 39535
    },
    {
      "entropy": 5.352871036529541,
      "epoch": 3.0763275627309863,
      "grad_norm": 1.171875,
      "learning_rate": 0.000405979370966881,
      "loss": 4.8275,
      "mean_token_accuracy": 0.21041016429662704,
      "num_tokens": 68587982.0,
      "step": 39540
    },
    {
      "entropy": 5.365692472457885,
      "epoch": 3.0767165921027035,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004059568345579953,
      "loss": 4.9085,
      "mean_token_accuracy": 0.19981757253408433,
      "num_tokens": 68596483.0,
      "step": 39545
    },
    {
      "entropy": 5.402635908126831,
      "epoch": 3.0771056214744212,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040593429616204806,
      "loss": 4.9698,
      "mean_token_accuracy": 0.2034299701452255,
      "num_tokens": 68605841.0,
      "step": 39550
    },
    {
      "entropy": 5.323451375961303,
      "epoch": 3.077494650846139,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040591175577938125,
      "loss": 4.9287,
      "mean_token_accuracy": 0.20032303929328918,
      "num_tokens": 68613912.0,
      "step": 39555
    },
    {
      "entropy": 5.362386274337768,
      "epoch": 3.0778836802178566,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040588921341033697,
      "loss": 4.897,
      "mean_token_accuracy": 0.20890214145183564,
      "num_tokens": 68623383.0,
      "step": 39560
    },
    {
      "entropy": 5.4454751968383786,
      "epoch": 3.078272709589574,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004058666690552572,
      "loss": 4.9303,
      "mean_token_accuracy": 0.20290555953979492,
      "num_tokens": 68632099.0,
      "step": 39565
    },
    {
      "entropy": 5.393856191635132,
      "epoch": 3.0786617389612916,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040584412271448407,
      "loss": 4.8975,
      "mean_token_accuracy": 0.20557726472616195,
      "num_tokens": 68640966.0,
      "step": 39570
    },
    {
      "entropy": 5.356562280654908,
      "epoch": 3.0790507683330093,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004058215743883595,
      "loss": 4.9017,
      "mean_token_accuracy": 0.19891307502985,
      "num_tokens": 68649608.0,
      "step": 39575
    },
    {
      "entropy": 5.360400772094726,
      "epoch": 3.0794397977047265,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004057990240772259,
      "loss": 4.8625,
      "mean_token_accuracy": 0.22148149609565734,
      "num_tokens": 68659478.0,
      "step": 39580
    },
    {
      "entropy": 5.422684907913208,
      "epoch": 3.0798288270764442,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040577647178142523,
      "loss": 4.9439,
      "mean_token_accuracy": 0.20765017867088317,
      "num_tokens": 68668401.0,
      "step": 39585
    },
    {
      "entropy": 5.357463645935058,
      "epoch": 3.080217856448162,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004057539175012998,
      "loss": 4.8249,
      "mean_token_accuracy": 0.20827181190252303,
      "num_tokens": 68676928.0,
      "step": 39590
    },
    {
      "entropy": 5.310135269165039,
      "epoch": 3.080606885819879,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040573136123719174,
      "loss": 4.9421,
      "mean_token_accuracy": 0.2026720717549324,
      "num_tokens": 68685834.0,
      "step": 39595
    },
    {
      "entropy": 5.341330480575562,
      "epoch": 3.080995915191597,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040570880298944334,
      "loss": 4.8531,
      "mean_token_accuracy": 0.21399440616369247,
      "num_tokens": 68694576.0,
      "step": 39600
    },
    {
      "entropy": 5.363631820678711,
      "epoch": 3.0813849445633146,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004056862427583969,
      "loss": 4.8631,
      "mean_token_accuracy": 0.21411360651254654,
      "num_tokens": 68703660.0,
      "step": 39605
    },
    {
      "entropy": 5.285452747344971,
      "epoch": 3.0817739739350323,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004056636805443948,
      "loss": 4.9333,
      "mean_token_accuracy": 0.20576193928718567,
      "num_tokens": 68713263.0,
      "step": 39610
    },
    {
      "entropy": 5.36211142539978,
      "epoch": 3.0821630033067495,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040564111634777927,
      "loss": 4.8564,
      "mean_token_accuracy": 0.2024134397506714,
      "num_tokens": 68721856.0,
      "step": 39615
    },
    {
      "entropy": 5.392276954650879,
      "epoch": 3.0825520326784672,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040561855016889276,
      "loss": 4.8947,
      "mean_token_accuracy": 0.20623093992471694,
      "num_tokens": 68729887.0,
      "step": 39620
    },
    {
      "entropy": 5.357159852981567,
      "epoch": 3.082941062050185,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040559598200807763,
      "loss": 4.9214,
      "mean_token_accuracy": 0.21215338557958602,
      "num_tokens": 68739154.0,
      "step": 39625
    },
    {
      "entropy": 5.382906055450439,
      "epoch": 3.083330091421902,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004055734118656764,
      "loss": 4.93,
      "mean_token_accuracy": 0.20204052627086638,
      "num_tokens": 68748493.0,
      "step": 39630
    },
    {
      "entropy": 5.456566667556762,
      "epoch": 3.08371912079362,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004055508397420314,
      "loss": 4.9475,
      "mean_token_accuracy": 0.20301341116428376,
      "num_tokens": 68757479.0,
      "step": 39635
    },
    {
      "entropy": 5.354962539672852,
      "epoch": 3.0841081501653376,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040552826563748526,
      "loss": 4.8898,
      "mean_token_accuracy": 0.20634498596191406,
      "num_tokens": 68766438.0,
      "step": 39640
    },
    {
      "entropy": 5.415283632278443,
      "epoch": 3.0844971795370553,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004055056895523804,
      "loss": 5.0078,
      "mean_token_accuracy": 0.20093509256839753,
      "num_tokens": 68774436.0,
      "step": 39645
    },
    {
      "entropy": 5.390762042999268,
      "epoch": 3.0848862089087725,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040548311148705943,
      "loss": 4.8544,
      "mean_token_accuracy": 0.2153600811958313,
      "num_tokens": 68783576.0,
      "step": 39650
    },
    {
      "entropy": 5.365612125396728,
      "epoch": 3.08527523828049,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004054605314418649,
      "loss": 4.8555,
      "mean_token_accuracy": 0.20721151828765869,
      "num_tokens": 68792793.0,
      "step": 39655
    },
    {
      "entropy": 5.328716611862182,
      "epoch": 3.085664267652208,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040543794941713944,
      "loss": 4.8534,
      "mean_token_accuracy": 0.20648181587457656,
      "num_tokens": 68801506.0,
      "step": 39660
    },
    {
      "entropy": 5.376133251190185,
      "epoch": 3.086053297023925,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004054153654132258,
      "loss": 4.9391,
      "mean_token_accuracy": 0.20122779458761214,
      "num_tokens": 68810999.0,
      "step": 39665
    },
    {
      "entropy": 5.398913860321045,
      "epoch": 3.086442326395643,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004053927794304665,
      "loss": 4.9,
      "mean_token_accuracy": 0.2064869448542595,
      "num_tokens": 68819688.0,
      "step": 39670
    },
    {
      "entropy": 5.388112783432007,
      "epoch": 3.0868313557673606,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004053701914692043,
      "loss": 4.9028,
      "mean_token_accuracy": 0.19929781556129456,
      "num_tokens": 68828079.0,
      "step": 39675
    },
    {
      "entropy": 5.324591493606567,
      "epoch": 3.087220385139078,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000405347601529782,
      "loss": 4.7996,
      "mean_token_accuracy": 0.21183450371026993,
      "num_tokens": 68836948.0,
      "step": 39680
    },
    {
      "entropy": 5.318232202529908,
      "epoch": 3.0876094145107955,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004053250096125423,
      "loss": 4.8791,
      "mean_token_accuracy": 0.20922672301530837,
      "num_tokens": 68845023.0,
      "step": 39685
    },
    {
      "entropy": 5.3375632762908936,
      "epoch": 3.087998443882513,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004053024157178281,
      "loss": 4.9211,
      "mean_token_accuracy": 0.20704969763755798,
      "num_tokens": 68853496.0,
      "step": 39690
    },
    {
      "entropy": 5.39705114364624,
      "epoch": 3.088387473254231,
      "grad_norm": 1.328125,
      "learning_rate": 0.000405279819845982,
      "loss": 4.9523,
      "mean_token_accuracy": 0.20426083654165267,
      "num_tokens": 68861593.0,
      "step": 39695
    },
    {
      "entropy": 5.291562747955322,
      "epoch": 3.088776502625948,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040525722199734706,
      "loss": 4.8157,
      "mean_token_accuracy": 0.2140125498175621,
      "num_tokens": 68870477.0,
      "step": 39700
    },
    {
      "entropy": 5.389525270462036,
      "epoch": 3.089165531997666,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004052346221722661,
      "loss": 4.9433,
      "mean_token_accuracy": 0.19924015998840333,
      "num_tokens": 68878353.0,
      "step": 39705
    },
    {
      "entropy": 5.402024364471435,
      "epoch": 3.0895545613693836,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040521202037108205,
      "loss": 4.8909,
      "mean_token_accuracy": 0.20416449755430222,
      "num_tokens": 68886479.0,
      "step": 39710
    },
    {
      "entropy": 5.33320164680481,
      "epoch": 3.089943590741101,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004051894165941378,
      "loss": 4.8966,
      "mean_token_accuracy": 0.20517400950193404,
      "num_tokens": 68895538.0,
      "step": 39715
    },
    {
      "entropy": 5.323169279098511,
      "epoch": 3.0903326201128185,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040516681084177636,
      "loss": 4.8245,
      "mean_token_accuracy": 0.21246874928474427,
      "num_tokens": 68903893.0,
      "step": 39720
    },
    {
      "entropy": 5.366964435577392,
      "epoch": 3.090721649484536,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040514420311434085,
      "loss": 4.9123,
      "mean_token_accuracy": 0.20191441029310225,
      "num_tokens": 68913008.0,
      "step": 39725
    },
    {
      "entropy": 5.241947507858276,
      "epoch": 3.0911106788562535,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040512159341217415,
      "loss": 4.8034,
      "mean_token_accuracy": 0.2051450490951538,
      "num_tokens": 68921405.0,
      "step": 39730
    },
    {
      "entropy": 5.393673229217529,
      "epoch": 3.091499708227971,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040509898173561934,
      "loss": 4.9666,
      "mean_token_accuracy": 0.20363552272319793,
      "num_tokens": 68930205.0,
      "step": 39735
    },
    {
      "entropy": 5.348920106887817,
      "epoch": 3.091888737599689,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004050763680850196,
      "loss": 4.8005,
      "mean_token_accuracy": 0.21652241349220275,
      "num_tokens": 68938069.0,
      "step": 39740
    },
    {
      "entropy": 5.325261878967285,
      "epoch": 3.0922777669714065,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004050537524607179,
      "loss": 4.8872,
      "mean_token_accuracy": 0.20332405865192413,
      "num_tokens": 68947229.0,
      "step": 39745
    },
    {
      "entropy": 5.3468352317810055,
      "epoch": 3.092666796343124,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040503113486305766,
      "loss": 4.8149,
      "mean_token_accuracy": 0.21073764115571975,
      "num_tokens": 68955773.0,
      "step": 39750
    },
    {
      "entropy": 5.354019117355347,
      "epoch": 3.0930558257148415,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040500851529238184,
      "loss": 4.861,
      "mean_token_accuracy": 0.21193266659975052,
      "num_tokens": 68964894.0,
      "step": 39755
    },
    {
      "entropy": 5.472339487075805,
      "epoch": 3.093444855086559,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004049858937490337,
      "loss": 5.0135,
      "mean_token_accuracy": 0.2013378232717514,
      "num_tokens": 68973965.0,
      "step": 39760
    },
    {
      "entropy": 5.346300220489502,
      "epoch": 3.0938338844582765,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004049632702333566,
      "loss": 4.8149,
      "mean_token_accuracy": 0.21466439068317414,
      "num_tokens": 68982166.0,
      "step": 39765
    },
    {
      "entropy": 5.368621730804444,
      "epoch": 3.094222913829994,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040494064474569375,
      "loss": 4.8684,
      "mean_token_accuracy": 0.2058524578809738,
      "num_tokens": 68990941.0,
      "step": 39770
    },
    {
      "entropy": 5.358957195281983,
      "epoch": 3.094611943201712,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040491801728638835,
      "loss": 4.8844,
      "mean_token_accuracy": 0.21037195920944213,
      "num_tokens": 68999243.0,
      "step": 39775
    },
    {
      "entropy": 5.463205289840698,
      "epoch": 3.095000972573429,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040489538785578385,
      "loss": 4.8445,
      "mean_token_accuracy": 0.212743116915226,
      "num_tokens": 69007547.0,
      "step": 39780
    },
    {
      "entropy": 5.322212076187133,
      "epoch": 3.095390001945147,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040487275645422356,
      "loss": 4.8184,
      "mean_token_accuracy": 0.20714086592197417,
      "num_tokens": 69017216.0,
      "step": 39785
    },
    {
      "entropy": 5.341900062561035,
      "epoch": 3.0957790313168645,
      "grad_norm": 1.15625,
      "learning_rate": 0.000404850123082051,
      "loss": 4.8323,
      "mean_token_accuracy": 0.21671584099531174,
      "num_tokens": 69025482.0,
      "step": 39790
    },
    {
      "entropy": 5.3719336032867435,
      "epoch": 3.096168060688582,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004048274877396094,
      "loss": 4.9551,
      "mean_token_accuracy": 0.20287248343229294,
      "num_tokens": 69034027.0,
      "step": 39795
    },
    {
      "entropy": 5.363114500045777,
      "epoch": 3.0965570900602994,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004048048504272424,
      "loss": 4.9178,
      "mean_token_accuracy": 0.20182660669088365,
      "num_tokens": 69042609.0,
      "step": 39800
    },
    {
      "entropy": 5.400418853759765,
      "epoch": 3.096946119432017,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004047822111452933,
      "loss": 4.9121,
      "mean_token_accuracy": 0.2024646908044815,
      "num_tokens": 69051277.0,
      "step": 39805
    },
    {
      "entropy": 5.405140542984009,
      "epoch": 3.097335148803735,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040475956989410584,
      "loss": 4.8541,
      "mean_token_accuracy": 0.21522360891103745,
      "num_tokens": 69059424.0,
      "step": 39810
    },
    {
      "entropy": 5.4380302906036375,
      "epoch": 3.097724178175452,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004047369266740233,
      "loss": 5.0139,
      "mean_token_accuracy": 0.20082610994577407,
      "num_tokens": 69068198.0,
      "step": 39815
    },
    {
      "entropy": 5.436377286911011,
      "epoch": 3.09811320754717,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040471428148538944,
      "loss": 4.9603,
      "mean_token_accuracy": 0.2011513441801071,
      "num_tokens": 69077477.0,
      "step": 39820
    },
    {
      "entropy": 5.359043455123901,
      "epoch": 3.0985022369188875,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004046916343285479,
      "loss": 4.8567,
      "mean_token_accuracy": 0.21438123285770416,
      "num_tokens": 69086452.0,
      "step": 39825
    },
    {
      "entropy": 5.383451509475708,
      "epoch": 3.0988912662906047,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040466898520384213,
      "loss": 4.9639,
      "mean_token_accuracy": 0.20630960762500763,
      "num_tokens": 69095191.0,
      "step": 39830
    },
    {
      "entropy": 5.371219682693481,
      "epoch": 3.0992802956623224,
      "grad_norm": 1.109375,
      "learning_rate": 0.000404646334111616,
      "loss": 4.9539,
      "mean_token_accuracy": 0.20594960153102876,
      "num_tokens": 69104079.0,
      "step": 39835
    },
    {
      "entropy": 5.304137897491455,
      "epoch": 3.09966932503404,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004046236810522131,
      "loss": 4.8889,
      "mean_token_accuracy": 0.20246531665325165,
      "num_tokens": 69113593.0,
      "step": 39840
    },
    {
      "entropy": 5.336619281768799,
      "epoch": 3.100058354405758,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004046010260259771,
      "loss": 4.9223,
      "mean_token_accuracy": 0.200758096575737,
      "num_tokens": 69122610.0,
      "step": 39845
    },
    {
      "entropy": 5.428775930404663,
      "epoch": 3.100447383777475,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040457836903325177,
      "loss": 4.929,
      "mean_token_accuracy": 0.20713288187980652,
      "num_tokens": 69130755.0,
      "step": 39850
    },
    {
      "entropy": 5.448776769638061,
      "epoch": 3.100836413149193,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040455571007438104,
      "loss": 4.8921,
      "mean_token_accuracy": 0.20961010605096816,
      "num_tokens": 69139008.0,
      "step": 39855
    },
    {
      "entropy": 5.3793620586395265,
      "epoch": 3.1012254425209105,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004045330491497085,
      "loss": 4.9162,
      "mean_token_accuracy": 0.20687601417303086,
      "num_tokens": 69148117.0,
      "step": 39860
    },
    {
      "entropy": 5.325445556640625,
      "epoch": 3.1016144718926277,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040451038625957827,
      "loss": 4.9138,
      "mean_token_accuracy": 0.20354344546794892,
      "num_tokens": 69156072.0,
      "step": 39865
    },
    {
      "entropy": 5.306148481369019,
      "epoch": 3.1020035012643454,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004044877214043339,
      "loss": 4.7778,
      "mean_token_accuracy": 0.21406500041484833,
      "num_tokens": 69165026.0,
      "step": 39870
    },
    {
      "entropy": 5.425632286071777,
      "epoch": 3.102392530636063,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004044650545843196,
      "loss": 4.9617,
      "mean_token_accuracy": 0.19702733904123307,
      "num_tokens": 69174283.0,
      "step": 39875
    },
    {
      "entropy": 5.370766973495483,
      "epoch": 3.1027815600077804,
      "grad_norm": 1.109375,
      "learning_rate": 0.000404442385799879,
      "loss": 4.9908,
      "mean_token_accuracy": 0.19917769283056258,
      "num_tokens": 69183335.0,
      "step": 39880
    },
    {
      "entropy": 5.469532060623169,
      "epoch": 3.103170589379498,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004044197150513564,
      "loss": 4.98,
      "mean_token_accuracy": 0.1970291718840599,
      "num_tokens": 69192272.0,
      "step": 39885
    },
    {
      "entropy": 5.412269020080567,
      "epoch": 3.1035596187512158,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004043970423390955,
      "loss": 4.9192,
      "mean_token_accuracy": 0.20483305305242538,
      "num_tokens": 69201274.0,
      "step": 39890
    },
    {
      "entropy": 5.433551883697509,
      "epoch": 3.1039486481229335,
      "grad_norm": 1.125,
      "learning_rate": 0.0004043743676634405,
      "loss": 4.9587,
      "mean_token_accuracy": 0.2022848382592201,
      "num_tokens": 69210112.0,
      "step": 39895
    },
    {
      "entropy": 5.46302285194397,
      "epoch": 3.1043376774946507,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004043516910247353,
      "loss": 4.9085,
      "mean_token_accuracy": 0.2100842833518982,
      "num_tokens": 69218822.0,
      "step": 39900
    },
    {
      "entropy": 5.377660036087036,
      "epoch": 3.1047267068663684,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004043290124233242,
      "loss": 4.9178,
      "mean_token_accuracy": 0.2036944955587387,
      "num_tokens": 69227683.0,
      "step": 39905
    },
    {
      "entropy": 5.30357985496521,
      "epoch": 3.105115736238086,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004043063318595511,
      "loss": 4.8623,
      "mean_token_accuracy": 0.20263220965862275,
      "num_tokens": 69236338.0,
      "step": 39910
    },
    {
      "entropy": 5.364724159240723,
      "epoch": 3.1055047656098034,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004042836493337603,
      "loss": 4.9256,
      "mean_token_accuracy": 0.20477401912212373,
      "num_tokens": 69245094.0,
      "step": 39915
    },
    {
      "entropy": 5.354433393478393,
      "epoch": 3.105893794981521,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004042609648462959,
      "loss": 4.8433,
      "mean_token_accuracy": 0.2097805604338646,
      "num_tokens": 69253791.0,
      "step": 39920
    },
    {
      "entropy": 5.396326541900635,
      "epoch": 3.1062828243532388,
      "grad_norm": 1.09375,
      "learning_rate": 0.000404238278397502,
      "loss": 4.9237,
      "mean_token_accuracy": 0.20093152523040772,
      "num_tokens": 69262410.0,
      "step": 39925
    },
    {
      "entropy": 5.352022361755371,
      "epoch": 3.106671853724956,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040421558998772293,
      "loss": 4.923,
      "mean_token_accuracy": 0.20439564287662507,
      "num_tokens": 69270840.0,
      "step": 39930
    },
    {
      "entropy": 5.349221086502075,
      "epoch": 3.1070608830966737,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000404192899617303,
      "loss": 4.7998,
      "mean_token_accuracy": 0.2103898838162422,
      "num_tokens": 69279385.0,
      "step": 39935
    },
    {
      "entropy": 5.319099760055542,
      "epoch": 3.1074499124683914,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004041702072865864,
      "loss": 4.7846,
      "mean_token_accuracy": 0.21359027922153473,
      "num_tokens": 69287927.0,
      "step": 39940
    },
    {
      "entropy": 5.398356914520264,
      "epoch": 3.107838941840109,
      "grad_norm": 1.25,
      "learning_rate": 0.0004041475129959176,
      "loss": 4.8914,
      "mean_token_accuracy": 0.21204473674297333,
      "num_tokens": 69297282.0,
      "step": 39945
    },
    {
      "entropy": 5.4548969745635985,
      "epoch": 3.1082279712118264,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004041248167456408,
      "loss": 4.9593,
      "mean_token_accuracy": 0.21003576815128328,
      "num_tokens": 69305290.0,
      "step": 39950
    },
    {
      "entropy": 5.349700307846069,
      "epoch": 3.108617000583544,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004041021185361004,
      "loss": 4.8204,
      "mean_token_accuracy": 0.21461442708969117,
      "num_tokens": 69313971.0,
      "step": 39955
    },
    {
      "entropy": 5.324366855621338,
      "epoch": 3.1090060299552618,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004040794183676409,
      "loss": 4.8916,
      "mean_token_accuracy": 0.2093067556619644,
      "num_tokens": 69322117.0,
      "step": 39960
    },
    {
      "entropy": 5.3449869632720945,
      "epoch": 3.109395059326979,
      "grad_norm": 1.125,
      "learning_rate": 0.0004040567162406066,
      "loss": 4.8791,
      "mean_token_accuracy": 0.2049208551645279,
      "num_tokens": 69330910.0,
      "step": 39965
    },
    {
      "entropy": 5.418989276885986,
      "epoch": 3.1097840886986967,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004040340121553421,
      "loss": 4.9383,
      "mean_token_accuracy": 0.1999283879995346,
      "num_tokens": 69339917.0,
      "step": 39970
    },
    {
      "entropy": 5.413790464401245,
      "epoch": 3.1101731180704144,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004040113061121918,
      "loss": 4.8935,
      "mean_token_accuracy": 0.2031472384929657,
      "num_tokens": 69348881.0,
      "step": 39975
    },
    {
      "entropy": 5.423224830627442,
      "epoch": 3.110562147442132,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040398859811150036,
      "loss": 4.9436,
      "mean_token_accuracy": 0.19647252708673477,
      "num_tokens": 69358451.0,
      "step": 39980
    },
    {
      "entropy": 5.3403627395629885,
      "epoch": 3.1109511768138494,
      "grad_norm": 1.125,
      "learning_rate": 0.0004039658881536122,
      "loss": 4.8804,
      "mean_token_accuracy": 0.20627961456775665,
      "num_tokens": 69367899.0,
      "step": 39985
    },
    {
      "entropy": 5.3092090606689455,
      "epoch": 3.111340206185567,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004039431762388719,
      "loss": 4.8926,
      "mean_token_accuracy": 0.205707448720932,
      "num_tokens": 69377862.0,
      "step": 39990
    },
    {
      "entropy": 5.392904567718506,
      "epoch": 3.1117292355572848,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004039204623676242,
      "loss": 4.8828,
      "mean_token_accuracy": 0.21372638046741485,
      "num_tokens": 69386246.0,
      "step": 39995
    },
    {
      "entropy": 5.319687843322754,
      "epoch": 3.112118264929002,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040389774654021367,
      "loss": 4.7913,
      "mean_token_accuracy": 0.22193794548511506,
      "num_tokens": 69394616.0,
      "step": 40000
    },
    {
      "entropy": 5.306745910644532,
      "epoch": 3.1125072943007197,
      "grad_norm": 1.171875,
      "learning_rate": 0.000403875028756985,
      "loss": 4.8055,
      "mean_token_accuracy": 0.2160838097333908,
      "num_tokens": 69403081.0,
      "step": 40005
    },
    {
      "entropy": 5.3664649486541744,
      "epoch": 3.1128963236724374,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004038523090182829,
      "loss": 4.9361,
      "mean_token_accuracy": 0.2002573847770691,
      "num_tokens": 69411819.0,
      "step": 40010
    },
    {
      "entropy": 5.367674541473389,
      "epoch": 3.1132853530441547,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000403829587324452,
      "loss": 4.8427,
      "mean_token_accuracy": 0.21124423742294313,
      "num_tokens": 69419836.0,
      "step": 40015
    },
    {
      "entropy": 5.367705059051514,
      "epoch": 3.1136743824158724,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004038068636758373,
      "loss": 4.8946,
      "mean_token_accuracy": 0.20429813414812087,
      "num_tokens": 69428524.0,
      "step": 40020
    },
    {
      "entropy": 5.335054397583008,
      "epoch": 3.11406341178759,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004037841380727834,
      "loss": 4.8633,
      "mean_token_accuracy": 0.20352984070777894,
      "num_tokens": 69436855.0,
      "step": 40025
    },
    {
      "entropy": 5.362209844589233,
      "epoch": 3.1144524411593073,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004037614105156352,
      "loss": 4.9405,
      "mean_token_accuracy": 0.20062340945005416,
      "num_tokens": 69446039.0,
      "step": 40030
    },
    {
      "entropy": 5.361679029464722,
      "epoch": 3.114841470531025,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004037386810047376,
      "loss": 4.923,
      "mean_token_accuracy": 0.20735732614994049,
      "num_tokens": 69454709.0,
      "step": 40035
    },
    {
      "entropy": 5.440235137939453,
      "epoch": 3.1152304999027427,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004037159495404354,
      "loss": 4.8878,
      "mean_token_accuracy": 0.21152849793434142,
      "num_tokens": 69463983.0,
      "step": 40040
    },
    {
      "entropy": 5.335074520111084,
      "epoch": 3.1156195292744604,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040369321612307343,
      "loss": 4.8501,
      "mean_token_accuracy": 0.20511524528265,
      "num_tokens": 69472073.0,
      "step": 40045
    },
    {
      "entropy": 5.270042514801025,
      "epoch": 3.1160085586461777,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040367048075299684,
      "loss": 4.8553,
      "mean_token_accuracy": 0.2036452114582062,
      "num_tokens": 69480119.0,
      "step": 40050
    },
    {
      "entropy": 5.377756929397583,
      "epoch": 3.1163975880178953,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004036477434305505,
      "loss": 4.9372,
      "mean_token_accuracy": 0.19983848184347153,
      "num_tokens": 69488517.0,
      "step": 40055
    },
    {
      "entropy": 5.488611173629761,
      "epoch": 3.116786617389613,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004036250041560794,
      "loss": 4.9578,
      "mean_token_accuracy": 0.20647504925727844,
      "num_tokens": 69496774.0,
      "step": 40060
    },
    {
      "entropy": 5.418052768707275,
      "epoch": 3.1171756467613303,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004036022629299286,
      "loss": 4.9777,
      "mean_token_accuracy": 0.2017388015985489,
      "num_tokens": 69506015.0,
      "step": 40065
    },
    {
      "entropy": 5.343689441680908,
      "epoch": 3.117564676133048,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004035795197524431,
      "loss": 4.9231,
      "mean_token_accuracy": 0.20114083737134933,
      "num_tokens": 69514392.0,
      "step": 40070
    },
    {
      "entropy": 5.353210020065307,
      "epoch": 3.1179537055047657,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004035567746239681,
      "loss": 4.8059,
      "mean_token_accuracy": 0.2126919984817505,
      "num_tokens": 69523186.0,
      "step": 40075
    },
    {
      "entropy": 5.336862230300904,
      "epoch": 3.1183427348764834,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040353402754484866,
      "loss": 4.9359,
      "mean_token_accuracy": 0.20200885236263275,
      "num_tokens": 69532128.0,
      "step": 40080
    },
    {
      "entropy": 5.332799196243286,
      "epoch": 3.1187317642482006,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004035112785154299,
      "loss": 4.8677,
      "mean_token_accuracy": 0.20682823807001113,
      "num_tokens": 69539846.0,
      "step": 40085
    },
    {
      "entropy": 5.308258104324341,
      "epoch": 3.1191207936199183,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040348852753605707,
      "loss": 4.8073,
      "mean_token_accuracy": 0.2081013098359108,
      "num_tokens": 69548238.0,
      "step": 40090
    },
    {
      "entropy": 5.431467294692993,
      "epoch": 3.119509822991636,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040346577460707535,
      "loss": 4.9708,
      "mean_token_accuracy": 0.2026794895529747,
      "num_tokens": 69557574.0,
      "step": 40095
    },
    {
      "entropy": 5.386212968826294,
      "epoch": 3.1198988523633533,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040344301972882997,
      "loss": 4.8441,
      "mean_token_accuracy": 0.2082304388284683,
      "num_tokens": 69565711.0,
      "step": 40100
    },
    {
      "entropy": 5.259875345230102,
      "epoch": 3.120287881735071,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004034202629016662,
      "loss": 4.8248,
      "mean_token_accuracy": 0.20505126118659972,
      "num_tokens": 69574589.0,
      "step": 40105
    },
    {
      "entropy": 5.349554014205933,
      "epoch": 3.1206769111067887,
      "grad_norm": 1.25,
      "learning_rate": 0.00040339750412592937,
      "loss": 4.8654,
      "mean_token_accuracy": 0.2117622449994087,
      "num_tokens": 69583229.0,
      "step": 40110
    },
    {
      "entropy": 5.3815450191497805,
      "epoch": 3.121065940478506,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004033747434019647,
      "loss": 4.9407,
      "mean_token_accuracy": 0.20640986561775207,
      "num_tokens": 69591727.0,
      "step": 40115
    },
    {
      "entropy": 5.337004804611206,
      "epoch": 3.1214549698502236,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040335198073011765,
      "loss": 4.8375,
      "mean_token_accuracy": 0.20756037682294845,
      "num_tokens": 69600192.0,
      "step": 40120
    },
    {
      "entropy": 5.29621696472168,
      "epoch": 3.1218439992219413,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004033292161107336,
      "loss": 4.8558,
      "mean_token_accuracy": 0.21316477209329604,
      "num_tokens": 69608373.0,
      "step": 40125
    },
    {
      "entropy": 5.334916019439698,
      "epoch": 3.122233028593659,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040330644954415795,
      "loss": 4.8743,
      "mean_token_accuracy": 0.20358263999223708,
      "num_tokens": 69616140.0,
      "step": 40130
    },
    {
      "entropy": 5.331064033508301,
      "epoch": 3.1226220579653763,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004032836810307362,
      "loss": 4.8826,
      "mean_token_accuracy": 0.21480502039194108,
      "num_tokens": 69624453.0,
      "step": 40135
    },
    {
      "entropy": 5.403168058395385,
      "epoch": 3.123011087337094,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004032609105708137,
      "loss": 4.9296,
      "mean_token_accuracy": 0.204910109937191,
      "num_tokens": 69633483.0,
      "step": 40140
    },
    {
      "entropy": 5.41729416847229,
      "epoch": 3.1234001167088117,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040323813816473604,
      "loss": 4.8604,
      "mean_token_accuracy": 0.20464650690555572,
      "num_tokens": 69641934.0,
      "step": 40145
    },
    {
      "entropy": 5.285771274566651,
      "epoch": 3.123789146080529,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004032153638128488,
      "loss": 4.8498,
      "mean_token_accuracy": 0.20453405976295472,
      "num_tokens": 69650694.0,
      "step": 40150
    },
    {
      "entropy": 5.298635196685791,
      "epoch": 3.1241781754522466,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040319258751549743,
      "loss": 4.8507,
      "mean_token_accuracy": 0.20757994055747986,
      "num_tokens": 69660138.0,
      "step": 40155
    },
    {
      "entropy": 5.431415367126465,
      "epoch": 3.1245672048239643,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040316980927302765,
      "loss": 4.908,
      "mean_token_accuracy": 0.2061296135187149,
      "num_tokens": 69668238.0,
      "step": 40160
    },
    {
      "entropy": 5.390276956558227,
      "epoch": 3.1249562341956816,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040314702908578496,
      "loss": 4.9349,
      "mean_token_accuracy": 0.19668539762496948,
      "num_tokens": 69677228.0,
      "step": 40165
    },
    {
      "entropy": 5.31136679649353,
      "epoch": 3.1253452635673993,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004031242469541151,
      "loss": 4.8576,
      "mean_token_accuracy": 0.20910992175340654,
      "num_tokens": 69686337.0,
      "step": 40170
    },
    {
      "entropy": 5.409294176101684,
      "epoch": 3.125734292939117,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040310146287836373,
      "loss": 4.8825,
      "mean_token_accuracy": 0.20549166798591614,
      "num_tokens": 69695368.0,
      "step": 40175
    },
    {
      "entropy": 5.376325178146362,
      "epoch": 3.1261233223108347,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040307867685887653,
      "loss": 4.872,
      "mean_token_accuracy": 0.20605642199516297,
      "num_tokens": 69703811.0,
      "step": 40180
    },
    {
      "entropy": 5.252239561080932,
      "epoch": 3.126512351682552,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040305588889599927,
      "loss": 4.7211,
      "mean_token_accuracy": 0.22173484712839125,
      "num_tokens": 69712619.0,
      "step": 40185
    },
    {
      "entropy": 5.325303363800049,
      "epoch": 3.1269013810542696,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004030330989900777,
      "loss": 4.9478,
      "mean_token_accuracy": 0.20044964998960496,
      "num_tokens": 69720945.0,
      "step": 40190
    },
    {
      "entropy": 5.435241794586181,
      "epoch": 3.1272904104259873,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040301030714145764,
      "loss": 4.9085,
      "mean_token_accuracy": 0.20907322466373443,
      "num_tokens": 69729827.0,
      "step": 40195
    },
    {
      "entropy": 5.389410734176636,
      "epoch": 3.1276794397977046,
      "grad_norm": 1.125,
      "learning_rate": 0.000402987513350485,
      "loss": 4.8806,
      "mean_token_accuracy": 0.20707599073648453,
      "num_tokens": 69738719.0,
      "step": 40200
    },
    {
      "entropy": 5.353525924682617,
      "epoch": 3.1280684691694223,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004029647176175055,
      "loss": 4.8791,
      "mean_token_accuracy": 0.20640521496534348,
      "num_tokens": 69747379.0,
      "step": 40205
    },
    {
      "entropy": 5.289103889465332,
      "epoch": 3.12845749854114,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040294191994286513,
      "loss": 4.7964,
      "mean_token_accuracy": 0.217774897813797,
      "num_tokens": 69755926.0,
      "step": 40210
    },
    {
      "entropy": 5.423659753799439,
      "epoch": 3.1288465279128572,
      "grad_norm": 1.125,
      "learning_rate": 0.00040291912032690963,
      "loss": 4.9151,
      "mean_token_accuracy": 0.19890027046203612,
      "num_tokens": 69763851.0,
      "step": 40215
    },
    {
      "entropy": 5.380201387405395,
      "epoch": 3.129235557284575,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004028963187699852,
      "loss": 4.8452,
      "mean_token_accuracy": 0.21401526629924775,
      "num_tokens": 69771332.0,
      "step": 40220
    },
    {
      "entropy": 5.324225664138794,
      "epoch": 3.1296245866562926,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004028735152724377,
      "loss": 4.8299,
      "mean_token_accuracy": 0.21453750729560853,
      "num_tokens": 69779878.0,
      "step": 40225
    },
    {
      "entropy": 5.353233528137207,
      "epoch": 3.1300136160280103,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040285070983461313,
      "loss": 4.9609,
      "mean_token_accuracy": 0.2039760172367096,
      "num_tokens": 69788564.0,
      "step": 40230
    },
    {
      "entropy": 5.365759706497192,
      "epoch": 3.1304026453997276,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040282790245685753,
      "loss": 4.9381,
      "mean_token_accuracy": 0.20766042321920394,
      "num_tokens": 69797512.0,
      "step": 40235
    },
    {
      "entropy": 5.394866323471069,
      "epoch": 3.1307916747714453,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000402805093139517,
      "loss": 4.8655,
      "mean_token_accuracy": 0.21376823931932448,
      "num_tokens": 69806103.0,
      "step": 40240
    },
    {
      "entropy": 5.297544622421265,
      "epoch": 3.131180704143163,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004027822818829375,
      "loss": 4.8966,
      "mean_token_accuracy": 0.2064782425761223,
      "num_tokens": 69814881.0,
      "step": 40245
    },
    {
      "entropy": 5.395734977722168,
      "epoch": 3.13156973351488,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004027594686874653,
      "loss": 4.9208,
      "mean_token_accuracy": 0.20460042655467986,
      "num_tokens": 69823507.0,
      "step": 40250
    },
    {
      "entropy": 5.342482852935791,
      "epoch": 3.131958762886598,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040273665355344655,
      "loss": 4.8448,
      "mean_token_accuracy": 0.2050609052181244,
      "num_tokens": 69832960.0,
      "step": 40255
    },
    {
      "entropy": 5.3982704162597654,
      "epoch": 3.1323477922583156,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004027138364812274,
      "loss": 4.9016,
      "mean_token_accuracy": 0.2055274710059166,
      "num_tokens": 69841216.0,
      "step": 40260
    },
    {
      "entropy": 5.3960167407989506,
      "epoch": 3.132736821630033,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004026910174711541,
      "loss": 4.956,
      "mean_token_accuracy": 0.20234512388706208,
      "num_tokens": 69849847.0,
      "step": 40265
    },
    {
      "entropy": 5.442897939682007,
      "epoch": 3.1331258510017506,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040266819652357266,
      "loss": 5.0305,
      "mean_token_accuracy": 0.1991891771554947,
      "num_tokens": 69859616.0,
      "step": 40270
    },
    {
      "entropy": 5.414956521987915,
      "epoch": 3.1335148803734683,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040264537363882976,
      "loss": 4.9392,
      "mean_token_accuracy": 0.2037709727883339,
      "num_tokens": 69869203.0,
      "step": 40275
    },
    {
      "entropy": 5.4073206901550295,
      "epoch": 3.133903909745186,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040262254881727135,
      "loss": 4.8025,
      "mean_token_accuracy": 0.21401309221982956,
      "num_tokens": 69877127.0,
      "step": 40280
    },
    {
      "entropy": 5.444560813903808,
      "epoch": 3.134292939116903,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040259972205924395,
      "loss": 5.0473,
      "mean_token_accuracy": 0.19441619962453843,
      "num_tokens": 69885676.0,
      "step": 40285
    },
    {
      "entropy": 5.399515628814697,
      "epoch": 3.134681968488621,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040257689336509385,
      "loss": 4.9289,
      "mean_token_accuracy": 0.20139123052358626,
      "num_tokens": 69894818.0,
      "step": 40290
    },
    {
      "entropy": 5.3704366207122805,
      "epoch": 3.1350709978603386,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004025540627351675,
      "loss": 4.8679,
      "mean_token_accuracy": 0.21071064621210098,
      "num_tokens": 69902786.0,
      "step": 40295
    },
    {
      "entropy": 5.388719463348389,
      "epoch": 3.135460027232056,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004025312301698112,
      "loss": 4.8788,
      "mean_token_accuracy": 0.21232085227966307,
      "num_tokens": 69911343.0,
      "step": 40300
    },
    {
      "entropy": 5.392844104766846,
      "epoch": 3.1358490566037736,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004025083956693715,
      "loss": 4.9044,
      "mean_token_accuracy": 0.2013694539666176,
      "num_tokens": 69920234.0,
      "step": 40305
    },
    {
      "entropy": 5.415490198135376,
      "epoch": 3.1362380859754913,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004024855592341948,
      "loss": 4.8916,
      "mean_token_accuracy": 0.2073521226644516,
      "num_tokens": 69928680.0,
      "step": 40310
    },
    {
      "entropy": 5.342182397842407,
      "epoch": 3.136627115347209,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004024627208646277,
      "loss": 4.8729,
      "mean_token_accuracy": 0.21834908276796341,
      "num_tokens": 69937020.0,
      "step": 40315
    },
    {
      "entropy": 5.393909740447998,
      "epoch": 3.137016144718926,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040243988056101667,
      "loss": 4.8958,
      "mean_token_accuracy": 0.20862918198108674,
      "num_tokens": 69946402.0,
      "step": 40320
    },
    {
      "entropy": 5.4395952224731445,
      "epoch": 3.137405174090644,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004024170383237083,
      "loss": 4.9043,
      "mean_token_accuracy": 0.21211323738098145,
      "num_tokens": 69954386.0,
      "step": 40325
    },
    {
      "entropy": 5.281073522567749,
      "epoch": 3.1377942034623616,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004023941941530492,
      "loss": 4.7498,
      "mean_token_accuracy": 0.21993305087089537,
      "num_tokens": 69962825.0,
      "step": 40330
    },
    {
      "entropy": 5.391599225997925,
      "epoch": 3.138183232834079,
      "grad_norm": 1.15625,
      "learning_rate": 0.000402371348049386,
      "loss": 4.9592,
      "mean_token_accuracy": 0.20974842309951783,
      "num_tokens": 69971726.0,
      "step": 40335
    },
    {
      "entropy": 5.398697710037231,
      "epoch": 3.1385722622057965,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004023485000130653,
      "loss": 4.8945,
      "mean_token_accuracy": 0.2095278486609459,
      "num_tokens": 69979891.0,
      "step": 40340
    },
    {
      "entropy": 5.386595582962036,
      "epoch": 3.1389612915775142,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040232565004443373,
      "loss": 4.8299,
      "mean_token_accuracy": 0.2102814570069313,
      "num_tokens": 69988052.0,
      "step": 40345
    },
    {
      "entropy": 5.343291521072388,
      "epoch": 3.1393503209492315,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004023027981438382,
      "loss": 4.8541,
      "mean_token_accuracy": 0.20610954910516738,
      "num_tokens": 69996221.0,
      "step": 40350
    },
    {
      "entropy": 5.380416631698608,
      "epoch": 3.139739350320949,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040227994431162517,
      "loss": 4.9506,
      "mean_token_accuracy": 0.20314665287733077,
      "num_tokens": 70004928.0,
      "step": 40355
    },
    {
      "entropy": 5.327496004104614,
      "epoch": 3.140128379692667,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040225708854814174,
      "loss": 4.8891,
      "mean_token_accuracy": 0.20487699806690216,
      "num_tokens": 70013525.0,
      "step": 40360
    },
    {
      "entropy": 5.409126234054566,
      "epoch": 3.140517409064384,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004022342308537345,
      "loss": 4.9377,
      "mean_token_accuracy": 0.20181942582130433,
      "num_tokens": 70021845.0,
      "step": 40365
    },
    {
      "entropy": 5.3865447521209715,
      "epoch": 3.140906438436102,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004022113712287503,
      "loss": 4.8947,
      "mean_token_accuracy": 0.20183716267347335,
      "num_tokens": 70030370.0,
      "step": 40370
    },
    {
      "entropy": 5.364449691772461,
      "epoch": 3.1412954678078195,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000402188509673536,
      "loss": 4.9027,
      "mean_token_accuracy": 0.20568051934242249,
      "num_tokens": 70038555.0,
      "step": 40375
    },
    {
      "entropy": 5.368240022659302,
      "epoch": 3.1416844971795372,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040216564618843854,
      "loss": 4.8879,
      "mean_token_accuracy": 0.2091048151254654,
      "num_tokens": 70047416.0,
      "step": 40380
    },
    {
      "entropy": 5.320351839065552,
      "epoch": 3.1420735265512545,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004021427807738048,
      "loss": 4.9194,
      "mean_token_accuracy": 0.19994339048862458,
      "num_tokens": 70056058.0,
      "step": 40385
    },
    {
      "entropy": 5.392819786071778,
      "epoch": 3.142462555922972,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004021199134299817,
      "loss": 4.9016,
      "mean_token_accuracy": 0.20911280512809755,
      "num_tokens": 70064020.0,
      "step": 40390
    },
    {
      "entropy": 5.298990774154663,
      "epoch": 3.14285158529469,
      "grad_norm": 1.125,
      "learning_rate": 0.00040209704415731623,
      "loss": 4.867,
      "mean_token_accuracy": 0.211707603931427,
      "num_tokens": 70072675.0,
      "step": 40395
    },
    {
      "entropy": 5.384028482437134,
      "epoch": 3.143240614666407,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040207417295615546,
      "loss": 4.8435,
      "mean_token_accuracy": 0.21490723192691802,
      "num_tokens": 70081061.0,
      "step": 40400
    },
    {
      "entropy": 5.3710103034973145,
      "epoch": 3.143629644038125,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040205129982684635,
      "loss": 4.8469,
      "mean_token_accuracy": 0.20873931497335435,
      "num_tokens": 70089837.0,
      "step": 40405
    },
    {
      "entropy": 5.2984226703643795,
      "epoch": 3.1440186734098425,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040202842476973606,
      "loss": 4.882,
      "mean_token_accuracy": 0.2059580534696579,
      "num_tokens": 70099638.0,
      "step": 40410
    },
    {
      "entropy": 5.394252586364746,
      "epoch": 3.1444077027815602,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040200554778517154,
      "loss": 4.9322,
      "mean_token_accuracy": 0.20417493283748628,
      "num_tokens": 70108156.0,
      "step": 40415
    },
    {
      "entropy": 5.449779844284057,
      "epoch": 3.1447967321532775,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040198266887349994,
      "loss": 4.9387,
      "mean_token_accuracy": 0.21387397795915603,
      "num_tokens": 70116627.0,
      "step": 40420
    },
    {
      "entropy": 5.392884111404419,
      "epoch": 3.145185761524995,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040195978803506853,
      "loss": 4.9468,
      "mean_token_accuracy": 0.20222701132297516,
      "num_tokens": 70124796.0,
      "step": 40425
    },
    {
      "entropy": 5.401149415969849,
      "epoch": 3.145574790896713,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040193690527022443,
      "loss": 4.94,
      "mean_token_accuracy": 0.2050477609038353,
      "num_tokens": 70134047.0,
      "step": 40430
    },
    {
      "entropy": 5.333043336868286,
      "epoch": 3.14596382026843,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040191402057931474,
      "loss": 4.8143,
      "mean_token_accuracy": 0.20831589698791503,
      "num_tokens": 70142706.0,
      "step": 40435
    },
    {
      "entropy": 5.4017455101013185,
      "epoch": 3.146352849640148,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004018911339626869,
      "loss": 4.9241,
      "mean_token_accuracy": 0.2054540127515793,
      "num_tokens": 70151775.0,
      "step": 40440
    },
    {
      "entropy": 5.3195658206939695,
      "epoch": 3.1467418790118655,
      "grad_norm": 1.09375,
      "learning_rate": 0.000401868245420688,
      "loss": 4.8675,
      "mean_token_accuracy": 0.20930508971214296,
      "num_tokens": 70160208.0,
      "step": 40445
    },
    {
      "entropy": 5.380018663406372,
      "epoch": 3.147130908383583,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040184535495366543,
      "loss": 4.9361,
      "mean_token_accuracy": 0.20575164556503295,
      "num_tokens": 70168881.0,
      "step": 40450
    },
    {
      "entropy": 5.40568733215332,
      "epoch": 3.1475199377553005,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004018224625619664,
      "loss": 4.8702,
      "mean_token_accuracy": 0.21192592680454253,
      "num_tokens": 70176862.0,
      "step": 40455
    },
    {
      "entropy": 5.359659814834595,
      "epoch": 3.147908967127018,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040179956824593847,
      "loss": 4.8057,
      "mean_token_accuracy": 0.21260741204023362,
      "num_tokens": 70185199.0,
      "step": 40460
    },
    {
      "entropy": 5.2732336044311525,
      "epoch": 3.1482979964987354,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040177667200592894,
      "loss": 4.8208,
      "mean_token_accuracy": 0.2140389859676361,
      "num_tokens": 70193323.0,
      "step": 40465
    },
    {
      "entropy": 5.375649929046631,
      "epoch": 3.148687025870453,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004017537738422852,
      "loss": 4.8747,
      "mean_token_accuracy": 0.20643111169338227,
      "num_tokens": 70202082.0,
      "step": 40470
    },
    {
      "entropy": 5.384236288070679,
      "epoch": 3.149076055242171,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040173087375535457,
      "loss": 4.9314,
      "mean_token_accuracy": 0.20547318309545518,
      "num_tokens": 70211504.0,
      "step": 40475
    },
    {
      "entropy": 5.484420967102051,
      "epoch": 3.1494650846138885,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040170797174548476,
      "loss": 5.0041,
      "mean_token_accuracy": 0.19937659054994583,
      "num_tokens": 70219717.0,
      "step": 40480
    },
    {
      "entropy": 5.3837755680084225,
      "epoch": 3.1498541139856058,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040168506781302303,
      "loss": 4.9273,
      "mean_token_accuracy": 0.2100866988301277,
      "num_tokens": 70228877.0,
      "step": 40485
    },
    {
      "entropy": 5.319440031051636,
      "epoch": 3.1502431433573235,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004016621619583171,
      "loss": 4.8403,
      "mean_token_accuracy": 0.20495802462100982,
      "num_tokens": 70237538.0,
      "step": 40490
    },
    {
      "entropy": 5.340173530578613,
      "epoch": 3.150632172729041,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040163925418171454,
      "loss": 4.8656,
      "mean_token_accuracy": 0.20959947556257247,
      "num_tokens": 70245977.0,
      "step": 40495
    },
    {
      "entropy": 5.342642498016358,
      "epoch": 3.1510212021007584,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004016163444835628,
      "loss": 4.8916,
      "mean_token_accuracy": 0.20941452085971832,
      "num_tokens": 70255262.0,
      "step": 40500
    },
    {
      "entropy": 5.341546440124512,
      "epoch": 3.151410231472476,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004015934328642096,
      "loss": 4.7383,
      "mean_token_accuracy": 0.2134230449795723,
      "num_tokens": 70263999.0,
      "step": 40505
    },
    {
      "entropy": 5.340118074417115,
      "epoch": 3.151799260844194,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040157051932400244,
      "loss": 4.8608,
      "mean_token_accuracy": 0.20276293456554412,
      "num_tokens": 70272228.0,
      "step": 40510
    },
    {
      "entropy": 5.309202480316162,
      "epoch": 3.1521882902159115,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004015476038632892,
      "loss": 4.9147,
      "mean_token_accuracy": 0.20733501017093658,
      "num_tokens": 70281818.0,
      "step": 40515
    },
    {
      "entropy": 5.36655330657959,
      "epoch": 3.1525773195876288,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004015246864824175,
      "loss": 4.9278,
      "mean_token_accuracy": 0.20221686065196992,
      "num_tokens": 70290298.0,
      "step": 40520
    },
    {
      "entropy": 5.30918779373169,
      "epoch": 3.1529663489593465,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040150176718173504,
      "loss": 4.8038,
      "mean_token_accuracy": 0.2186904400587082,
      "num_tokens": 70299166.0,
      "step": 40525
    },
    {
      "entropy": 5.33083758354187,
      "epoch": 3.153355378331064,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004014788459615896,
      "loss": 4.8398,
      "mean_token_accuracy": 0.20933312326669692,
      "num_tokens": 70308110.0,
      "step": 40530
    },
    {
      "entropy": 5.324524021148681,
      "epoch": 3.1537444077027814,
      "grad_norm": 1.125,
      "learning_rate": 0.0004014559228223291,
      "loss": 4.7801,
      "mean_token_accuracy": 0.2150770455598831,
      "num_tokens": 70317270.0,
      "step": 40535
    },
    {
      "entropy": 5.420511913299561,
      "epoch": 3.154133437074499,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040143299776430115,
      "loss": 4.9595,
      "mean_token_accuracy": 0.20503859370946884,
      "num_tokens": 70326078.0,
      "step": 40540
    },
    {
      "entropy": 5.37081503868103,
      "epoch": 3.154522466446217,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004014100707878536,
      "loss": 4.8871,
      "mean_token_accuracy": 0.21537187397480012,
      "num_tokens": 70334388.0,
      "step": 40545
    },
    {
      "entropy": 5.389930963516235,
      "epoch": 3.154911495817934,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040138714189333455,
      "loss": 4.8869,
      "mean_token_accuracy": 0.21046380698680878,
      "num_tokens": 70342844.0,
      "step": 40550
    },
    {
      "entropy": 5.4401164054870605,
      "epoch": 3.1553005251896518,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040136421108109176,
      "loss": 5.0011,
      "mean_token_accuracy": 0.20090001076459885,
      "num_tokens": 70351331.0,
      "step": 40555
    },
    {
      "entropy": 5.395886182785034,
      "epoch": 3.1556895545613695,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004013412783514732,
      "loss": 4.9119,
      "mean_token_accuracy": 0.20311810374259948,
      "num_tokens": 70359573.0,
      "step": 40560
    },
    {
      "entropy": 5.4086761474609375,
      "epoch": 3.156078583933087,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004013183437048268,
      "loss": 4.8716,
      "mean_token_accuracy": 0.2180813029408455,
      "num_tokens": 70367586.0,
      "step": 40565
    },
    {
      "entropy": 5.337568521499634,
      "epoch": 3.1564676133048044,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040129540714150066,
      "loss": 4.8337,
      "mean_token_accuracy": 0.21647854298353195,
      "num_tokens": 70376399.0,
      "step": 40570
    },
    {
      "entropy": 5.370188903808594,
      "epoch": 3.156856642676522,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004012724686618427,
      "loss": 4.9439,
      "mean_token_accuracy": 0.19555094689130784,
      "num_tokens": 70385221.0,
      "step": 40575
    },
    {
      "entropy": 5.395825529098511,
      "epoch": 3.15724567204824,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000401249528266201,
      "loss": 4.931,
      "mean_token_accuracy": 0.20272525697946547,
      "num_tokens": 70393844.0,
      "step": 40580
    },
    {
      "entropy": 5.366936206817627,
      "epoch": 3.157634701419957,
      "grad_norm": 1.25,
      "learning_rate": 0.00040122658595492366,
      "loss": 4.8098,
      "mean_token_accuracy": 0.21259437948465348,
      "num_tokens": 70402417.0,
      "step": 40585
    },
    {
      "entropy": 5.390923738479614,
      "epoch": 3.1580237307916748,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004012036417283588,
      "loss": 4.829,
      "mean_token_accuracy": 0.21449365615844726,
      "num_tokens": 70410932.0,
      "step": 40590
    },
    {
      "entropy": 5.303453588485718,
      "epoch": 3.1584127601633925,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040118069558685456,
      "loss": 4.8188,
      "mean_token_accuracy": 0.2064634919166565,
      "num_tokens": 70419528.0,
      "step": 40595
    },
    {
      "entropy": 5.31270809173584,
      "epoch": 3.1588017895351097,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040115774753075905,
      "loss": 4.8965,
      "mean_token_accuracy": 0.2104461297392845,
      "num_tokens": 70428220.0,
      "step": 40600
    },
    {
      "entropy": 5.290410327911377,
      "epoch": 3.1591908189068274,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004011347975604206,
      "loss": 4.7685,
      "mean_token_accuracy": 0.22126491218805314,
      "num_tokens": 70436155.0,
      "step": 40605
    },
    {
      "entropy": 5.371480274200439,
      "epoch": 3.159579848278545,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004011118456761873,
      "loss": 4.9479,
      "mean_token_accuracy": 0.20990384221076966,
      "num_tokens": 70445618.0,
      "step": 40610
    },
    {
      "entropy": 5.355297660827636,
      "epoch": 3.159968877650263,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004010888918784075,
      "loss": 4.8408,
      "mean_token_accuracy": 0.21679199635982513,
      "num_tokens": 70454681.0,
      "step": 40615
    },
    {
      "entropy": 5.301231813430786,
      "epoch": 3.16035790702198,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004010659361674294,
      "loss": 4.8111,
      "mean_token_accuracy": 0.21232539564371108,
      "num_tokens": 70463408.0,
      "step": 40620
    },
    {
      "entropy": 5.405141735076905,
      "epoch": 3.1607469363936977,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040104297854360145,
      "loss": 4.9466,
      "mean_token_accuracy": 0.20330903679132462,
      "num_tokens": 70472010.0,
      "step": 40625
    },
    {
      "entropy": 5.415031385421753,
      "epoch": 3.1611359657654154,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004010200190072719,
      "loss": 4.8986,
      "mean_token_accuracy": 0.20761923342943192,
      "num_tokens": 70480869.0,
      "step": 40630
    },
    {
      "entropy": 5.373095226287842,
      "epoch": 3.1615249951371327,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004009970575587891,
      "loss": 4.8893,
      "mean_token_accuracy": 0.20356594771146774,
      "num_tokens": 70489651.0,
      "step": 40635
    },
    {
      "entropy": 5.298930358886719,
      "epoch": 3.1619140245088504,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004009740941985016,
      "loss": 4.8523,
      "mean_token_accuracy": 0.2056410938501358,
      "num_tokens": 70498332.0,
      "step": 40640
    },
    {
      "entropy": 5.42194356918335,
      "epoch": 3.162303053880568,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004009511289267576,
      "loss": 4.998,
      "mean_token_accuracy": 0.19585587829351425,
      "num_tokens": 70507504.0,
      "step": 40645
    },
    {
      "entropy": 5.422905826568604,
      "epoch": 3.1626920832522853,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004009281617439058,
      "loss": 4.9256,
      "mean_token_accuracy": 0.20331190079450606,
      "num_tokens": 70516560.0,
      "step": 40650
    },
    {
      "entropy": 5.370505046844483,
      "epoch": 3.163081112624003,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004009051926502945,
      "loss": 4.8869,
      "mean_token_accuracy": 0.21485923379659652,
      "num_tokens": 70524387.0,
      "step": 40655
    },
    {
      "entropy": 5.34507999420166,
      "epoch": 3.1634701419957207,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040088222164627233,
      "loss": 4.9253,
      "mean_token_accuracy": 0.2156902953982353,
      "num_tokens": 70533393.0,
      "step": 40660
    },
    {
      "entropy": 5.340969133377075,
      "epoch": 3.1638591713674384,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040085924873218783,
      "loss": 4.8214,
      "mean_token_accuracy": 0.20664364099502563,
      "num_tokens": 70542353.0,
      "step": 40665
    },
    {
      "entropy": 5.332652044296265,
      "epoch": 3.1642482007391557,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004008362739083895,
      "loss": 4.8406,
      "mean_token_accuracy": 0.20525905191898347,
      "num_tokens": 70550543.0,
      "step": 40670
    },
    {
      "entropy": 5.366426849365235,
      "epoch": 3.1646372301108734,
      "grad_norm": 1.140625,
      "learning_rate": 0.000400813297175226,
      "loss": 4.8544,
      "mean_token_accuracy": 0.20537822395563127,
      "num_tokens": 70558671.0,
      "step": 40675
    },
    {
      "entropy": 5.275816869735718,
      "epoch": 3.165026259482591,
      "grad_norm": 1.203125,
      "learning_rate": 0.000400790318533046,
      "loss": 4.7717,
      "mean_token_accuracy": 0.21961667239665986,
      "num_tokens": 70566938.0,
      "step": 40680
    },
    {
      "entropy": 5.4340015888214115,
      "epoch": 3.1654152888543083,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040076733798219817,
      "loss": 4.9479,
      "mean_token_accuracy": 0.2096466451883316,
      "num_tokens": 70575281.0,
      "step": 40685
    },
    {
      "entropy": 5.315427827835083,
      "epoch": 3.165804318226026,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004007443555230312,
      "loss": 4.8618,
      "mean_token_accuracy": 0.20321063399314881,
      "num_tokens": 70584543.0,
      "step": 40690
    },
    {
      "entropy": 5.388792228698731,
      "epoch": 3.1661933475977437,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040072137115589366,
      "loss": 4.9507,
      "mean_token_accuracy": 0.19957754164934158,
      "num_tokens": 70594385.0,
      "step": 40695
    },
    {
      "entropy": 5.388979148864746,
      "epoch": 3.166582376969461,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004006983848811345,
      "loss": 4.8667,
      "mean_token_accuracy": 0.21051215082407,
      "num_tokens": 70602842.0,
      "step": 40700
    },
    {
      "entropy": 5.379982662200928,
      "epoch": 3.1669714063411787,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040067539669910235,
      "loss": 4.9211,
      "mean_token_accuracy": 0.2004130706191063,
      "num_tokens": 70611845.0,
      "step": 40705
    },
    {
      "entropy": 5.403703451156616,
      "epoch": 3.1673604357128964,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004006524066101461,
      "loss": 4.9615,
      "mean_token_accuracy": 0.19907924979925157,
      "num_tokens": 70621351.0,
      "step": 40710
    },
    {
      "entropy": 5.422404766082764,
      "epoch": 3.167749465084614,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004006294146146146,
      "loss": 4.9207,
      "mean_token_accuracy": 0.20416584610939026,
      "num_tokens": 70629765.0,
      "step": 40715
    },
    {
      "entropy": 5.502452802658081,
      "epoch": 3.1681384944563313,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004006064207128567,
      "loss": 4.986,
      "mean_token_accuracy": 0.20159444510936736,
      "num_tokens": 70637812.0,
      "step": 40720
    },
    {
      "entropy": 5.407199192047119,
      "epoch": 3.168527523828049,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040058342490522134,
      "loss": 4.9075,
      "mean_token_accuracy": 0.20362248569726943,
      "num_tokens": 70647268.0,
      "step": 40725
    },
    {
      "entropy": 5.257736539840698,
      "epoch": 3.1689165531997667,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004005604271920573,
      "loss": 4.7246,
      "mean_token_accuracy": 0.21818455755710603,
      "num_tokens": 70655911.0,
      "step": 40730
    },
    {
      "entropy": 5.282678031921387,
      "epoch": 3.169305582571484,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040053742757371365,
      "loss": 4.8457,
      "mean_token_accuracy": 0.21023205816745758,
      "num_tokens": 70664323.0,
      "step": 40735
    },
    {
      "entropy": 5.304571104049683,
      "epoch": 3.1696946119432017,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004005144260505394,
      "loss": 4.8334,
      "mean_token_accuracy": 0.2170025959610939,
      "num_tokens": 70673081.0,
      "step": 40740
    },
    {
      "entropy": 5.367451810836792,
      "epoch": 3.1700836413149194,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004004914226228834,
      "loss": 4.8826,
      "mean_token_accuracy": 0.2148715525865555,
      "num_tokens": 70681668.0,
      "step": 40745
    },
    {
      "entropy": 5.3943780899047855,
      "epoch": 3.170472670686637,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004004684172910949,
      "loss": 4.8526,
      "mean_token_accuracy": 0.2119895726442337,
      "num_tokens": 70689875.0,
      "step": 40750
    },
    {
      "entropy": 5.383929920196533,
      "epoch": 3.1708617000583543,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004004454100555229,
      "loss": 4.9825,
      "mean_token_accuracy": 0.20376113206148147,
      "num_tokens": 70699634.0,
      "step": 40755
    },
    {
      "entropy": 5.457249116897583,
      "epoch": 3.171250729430072,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040042240091651645,
      "loss": 4.966,
      "mean_token_accuracy": 0.20787006318569184,
      "num_tokens": 70708951.0,
      "step": 40760
    },
    {
      "entropy": 5.404014015197754,
      "epoch": 3.1716397588017897,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040039938987442464,
      "loss": 4.8721,
      "mean_token_accuracy": 0.21195762753486633,
      "num_tokens": 70717317.0,
      "step": 40765
    },
    {
      "entropy": 5.420404148101807,
      "epoch": 3.172028788173507,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040037637692959667,
      "loss": 4.956,
      "mean_token_accuracy": 0.20098093748092652,
      "num_tokens": 70725867.0,
      "step": 40770
    },
    {
      "entropy": 5.29369535446167,
      "epoch": 3.1724178175452247,
      "grad_norm": 1.125,
      "learning_rate": 0.00040035336208238176,
      "loss": 4.8479,
      "mean_token_accuracy": 0.21436292082071304,
      "num_tokens": 70734774.0,
      "step": 40775
    },
    {
      "entropy": 5.447038221359253,
      "epoch": 3.1728068469169424,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040033034533312913,
      "loss": 4.9017,
      "mean_token_accuracy": 0.1999308481812477,
      "num_tokens": 70743420.0,
      "step": 40780
    },
    {
      "entropy": 5.4357240200042725,
      "epoch": 3.1731958762886596,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000400307326682188,
      "loss": 4.9133,
      "mean_token_accuracy": 0.20987289249897004,
      "num_tokens": 70751979.0,
      "step": 40785
    },
    {
      "entropy": 5.355973958969116,
      "epoch": 3.1735849056603773,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040028430612990755,
      "loss": 4.8846,
      "mean_token_accuracy": 0.20614465028047563,
      "num_tokens": 70760652.0,
      "step": 40790
    },
    {
      "entropy": 5.370217418670654,
      "epoch": 3.173973935032095,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040026128367663726,
      "loss": 4.9261,
      "mean_token_accuracy": 0.20652822107076646,
      "num_tokens": 70768997.0,
      "step": 40795
    },
    {
      "entropy": 5.3820442199707035,
      "epoch": 3.1743629644038123,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040023825932272635,
      "loss": 4.9228,
      "mean_token_accuracy": 0.2064330905675888,
      "num_tokens": 70777875.0,
      "step": 40800
    },
    {
      "entropy": 5.336959409713745,
      "epoch": 3.17475199377553,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040021523306852416,
      "loss": 4.8333,
      "mean_token_accuracy": 0.21253376454114914,
      "num_tokens": 70786592.0,
      "step": 40805
    },
    {
      "entropy": 5.34142370223999,
      "epoch": 3.1751410231472477,
      "grad_norm": 1.125,
      "learning_rate": 0.00040019220491438007,
      "loss": 4.9703,
      "mean_token_accuracy": 0.20435669720172883,
      "num_tokens": 70795689.0,
      "step": 40810
    },
    {
      "entropy": 5.425786733627319,
      "epoch": 3.1755300525189654,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004001691748606436,
      "loss": 4.9949,
      "mean_token_accuracy": 0.19736155718564988,
      "num_tokens": 70805078.0,
      "step": 40815
    },
    {
      "entropy": 5.418128108978271,
      "epoch": 3.1759190818906826,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004001461429076641,
      "loss": 4.828,
      "mean_token_accuracy": 0.2113950654864311,
      "num_tokens": 70813820.0,
      "step": 40820
    },
    {
      "entropy": 5.383731889724731,
      "epoch": 3.1763081112624003,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004001231090557912,
      "loss": 4.9021,
      "mean_token_accuracy": 0.20247779041528702,
      "num_tokens": 70821930.0,
      "step": 40825
    },
    {
      "entropy": 5.312462520599365,
      "epoch": 3.176697140634118,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040010007330537405,
      "loss": 4.8676,
      "mean_token_accuracy": 0.20845767110586166,
      "num_tokens": 70830158.0,
      "step": 40830
    },
    {
      "entropy": 5.480395364761352,
      "epoch": 3.1770861700058353,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040007703565676263,
      "loss": 4.9639,
      "mean_token_accuracy": 0.1950363412499428,
      "num_tokens": 70838774.0,
      "step": 40835
    },
    {
      "entropy": 5.393593502044678,
      "epoch": 3.177475199377553,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040005399611030615,
      "loss": 4.9406,
      "mean_token_accuracy": 0.20137183666229247,
      "num_tokens": 70848027.0,
      "step": 40840
    },
    {
      "entropy": 5.395750665664673,
      "epoch": 3.1778642287492707,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004000309546663543,
      "loss": 4.9256,
      "mean_token_accuracy": 0.19920625984668733,
      "num_tokens": 70856629.0,
      "step": 40845
    },
    {
      "entropy": 5.319666051864624,
      "epoch": 3.1782532581209884,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040000791132525676,
      "loss": 4.842,
      "mean_token_accuracy": 0.21338384300470353,
      "num_tokens": 70865248.0,
      "step": 40850
    },
    {
      "entropy": 5.354864883422851,
      "epoch": 3.1786422874927056,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039998486608736305,
      "loss": 4.8133,
      "mean_token_accuracy": 0.21008565574884414,
      "num_tokens": 70874351.0,
      "step": 40855
    },
    {
      "entropy": 5.374858140945435,
      "epoch": 3.1790313168644233,
      "grad_norm": 1.25,
      "learning_rate": 0.0003999618189530231,
      "loss": 4.7486,
      "mean_token_accuracy": 0.21425808668136598,
      "num_tokens": 70882478.0,
      "step": 40860
    },
    {
      "entropy": 5.382241487503052,
      "epoch": 3.179420346236141,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003999387699225863,
      "loss": 4.914,
      "mean_token_accuracy": 0.21170974820852279,
      "num_tokens": 70890461.0,
      "step": 40865
    },
    {
      "entropy": 5.353551959991455,
      "epoch": 3.1798093756078583,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003999157189964026,
      "loss": 4.896,
      "mean_token_accuracy": 0.21025590896606444,
      "num_tokens": 70898909.0,
      "step": 40870
    },
    {
      "entropy": 5.305985641479492,
      "epoch": 3.180198404979576,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003998926661748217,
      "loss": 4.8336,
      "mean_token_accuracy": 0.2109777733683586,
      "num_tokens": 70907497.0,
      "step": 40875
    },
    {
      "entropy": 5.421080255508423,
      "epoch": 3.1805874343512937,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003998696114581933,
      "loss": 4.9487,
      "mean_token_accuracy": 0.20963205248117447,
      "num_tokens": 70916018.0,
      "step": 40880
    },
    {
      "entropy": 5.479217529296875,
      "epoch": 3.180976463723011,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003998465548468674,
      "loss": 4.918,
      "mean_token_accuracy": 0.20002132654190063,
      "num_tokens": 70924876.0,
      "step": 40885
    },
    {
      "entropy": 5.36046576499939,
      "epoch": 3.1813654930947286,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003998234963411936,
      "loss": 4.8468,
      "mean_token_accuracy": 0.21071468591690062,
      "num_tokens": 70933113.0,
      "step": 40890
    },
    {
      "entropy": 5.359946918487549,
      "epoch": 3.1817545224664463,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003998004359415221,
      "loss": 4.8965,
      "mean_token_accuracy": 0.2105015844106674,
      "num_tokens": 70941916.0,
      "step": 40895
    },
    {
      "entropy": 5.416924619674683,
      "epoch": 3.1821435518381636,
      "grad_norm": 1.125,
      "learning_rate": 0.00039977737364820247,
      "loss": 4.9325,
      "mean_token_accuracy": 0.2027293249964714,
      "num_tokens": 70950110.0,
      "step": 40900
    },
    {
      "entropy": 5.416379070281982,
      "epoch": 3.1825325812098813,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039975430946158493,
      "loss": 4.9823,
      "mean_token_accuracy": 0.20106424391269684,
      "num_tokens": 70958801.0,
      "step": 40905
    },
    {
      "entropy": 5.425722122192383,
      "epoch": 3.182921610581599,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003997312433820192,
      "loss": 4.9049,
      "mean_token_accuracy": 0.20908755362033843,
      "num_tokens": 70967552.0,
      "step": 40910
    },
    {
      "entropy": 5.413363790512085,
      "epoch": 3.1833106399533166,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039970817540985545,
      "loss": 5.0266,
      "mean_token_accuracy": 0.2080891340970993,
      "num_tokens": 70975924.0,
      "step": 40915
    },
    {
      "entropy": 5.368432331085205,
      "epoch": 3.183699669325034,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003996851055454436,
      "loss": 4.8782,
      "mean_token_accuracy": 0.20943473577499389,
      "num_tokens": 70984954.0,
      "step": 40920
    },
    {
      "entropy": 5.497927951812744,
      "epoch": 3.1840886986967516,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003996620337891337,
      "loss": 5.0031,
      "mean_token_accuracy": 0.1985649585723877,
      "num_tokens": 70994773.0,
      "step": 40925
    },
    {
      "entropy": 5.488444805145264,
      "epoch": 3.1844777280684693,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000399638960141276,
      "loss": 4.9394,
      "mean_token_accuracy": 0.2036571368575096,
      "num_tokens": 71003173.0,
      "step": 40930
    },
    {
      "entropy": 5.357963132858276,
      "epoch": 3.1848667574401865,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039961588460222033,
      "loss": 4.8338,
      "mean_token_accuracy": 0.21225104182958604,
      "num_tokens": 71011729.0,
      "step": 40935
    },
    {
      "entropy": 5.402867460250855,
      "epoch": 3.1852557868119042,
      "grad_norm": 1.1875,
      "learning_rate": 0.000399592807172317,
      "loss": 4.9577,
      "mean_token_accuracy": 0.20298743844032288,
      "num_tokens": 71019876.0,
      "step": 40940
    },
    {
      "entropy": 5.342504692077637,
      "epoch": 3.185644816183622,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039956972785191606,
      "loss": 4.8715,
      "mean_token_accuracy": 0.20340289771556855,
      "num_tokens": 71028787.0,
      "step": 40945
    },
    {
      "entropy": 5.336028003692627,
      "epoch": 3.1860338455553396,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039954664664136787,
      "loss": 4.8239,
      "mean_token_accuracy": 0.21322612911462785,
      "num_tokens": 71037173.0,
      "step": 40950
    },
    {
      "entropy": 5.421294355392456,
      "epoch": 3.186422874927057,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003995235635410225,
      "loss": 4.8787,
      "mean_token_accuracy": 0.20794139206409454,
      "num_tokens": 71045486.0,
      "step": 40955
    },
    {
      "entropy": 5.443972778320313,
      "epoch": 3.1868119042987746,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003995004785512302,
      "loss": 4.9078,
      "mean_token_accuracy": 0.207851405441761,
      "num_tokens": 71053970.0,
      "step": 40960
    },
    {
      "entropy": 5.3299414157867435,
      "epoch": 3.1872009336704923,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003994773916723414,
      "loss": 4.8232,
      "mean_token_accuracy": 0.2099948987364769,
      "num_tokens": 71062381.0,
      "step": 40965
    },
    {
      "entropy": 5.369425344467163,
      "epoch": 3.1875899630422095,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003994543029047063,
      "loss": 4.763,
      "mean_token_accuracy": 0.2149786874651909,
      "num_tokens": 71069974.0,
      "step": 40970
    },
    {
      "entropy": 5.422058200836181,
      "epoch": 3.1879789924139272,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003994312122486752,
      "loss": 4.9448,
      "mean_token_accuracy": 0.20070834159851075,
      "num_tokens": 71078985.0,
      "step": 40975
    },
    {
      "entropy": 5.317565011978149,
      "epoch": 3.188368021785645,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003994081197045985,
      "loss": 4.9107,
      "mean_token_accuracy": 0.2082419604063034,
      "num_tokens": 71087458.0,
      "step": 40980
    },
    {
      "entropy": 5.461523628234863,
      "epoch": 3.188757051157362,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003993850252728267,
      "loss": 4.9867,
      "mean_token_accuracy": 0.2013772562146187,
      "num_tokens": 71096558.0,
      "step": 40985
    },
    {
      "entropy": 5.433375549316406,
      "epoch": 3.18914608052908,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039936192895371006,
      "loss": 4.9297,
      "mean_token_accuracy": 0.20648936927318573,
      "num_tokens": 71105507.0,
      "step": 40990
    },
    {
      "entropy": 5.312155532836914,
      "epoch": 3.1895351099007976,
      "grad_norm": 1.09375,
      "learning_rate": 0.000399338830747599,
      "loss": 4.8707,
      "mean_token_accuracy": 0.21176324784755707,
      "num_tokens": 71114237.0,
      "step": 40995
    },
    {
      "entropy": 5.373797988891601,
      "epoch": 3.1899241392725153,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003993157306548442,
      "loss": 4.9344,
      "mean_token_accuracy": 0.20594169199466705,
      "num_tokens": 71123315.0,
      "step": 41000
    },
    {
      "entropy": 5.363411235809326,
      "epoch": 3.1903131686442325,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039929262867579603,
      "loss": 4.8984,
      "mean_token_accuracy": 0.2107297047972679,
      "num_tokens": 71132081.0,
      "step": 41005
    },
    {
      "entropy": 5.391424655914307,
      "epoch": 3.1907021980159502,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039926952481080495,
      "loss": 4.8768,
      "mean_token_accuracy": 0.2033358931541443,
      "num_tokens": 71140558.0,
      "step": 41010
    },
    {
      "entropy": 5.414814043045044,
      "epoch": 3.191091227387668,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039924641906022176,
      "loss": 4.8969,
      "mean_token_accuracy": 0.20270241647958756,
      "num_tokens": 71149474.0,
      "step": 41015
    },
    {
      "entropy": 5.375221204757691,
      "epoch": 3.191480256759385,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003992233114243969,
      "loss": 4.931,
      "mean_token_accuracy": 0.1995809033513069,
      "num_tokens": 71158096.0,
      "step": 41020
    },
    {
      "entropy": 5.350184488296509,
      "epoch": 3.191869286131103,
      "grad_norm": 1.265625,
      "learning_rate": 0.000399200201903681,
      "loss": 4.8799,
      "mean_token_accuracy": 0.20536412298679352,
      "num_tokens": 71166626.0,
      "step": 41025
    },
    {
      "entropy": 5.421692895889282,
      "epoch": 3.1922583155028206,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003991770904984247,
      "loss": 4.9652,
      "mean_token_accuracy": 0.20443134903907775,
      "num_tokens": 71175390.0,
      "step": 41030
    },
    {
      "entropy": 5.39427604675293,
      "epoch": 3.192647344874538,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003991539772089787,
      "loss": 4.9138,
      "mean_token_accuracy": 0.2091783106327057,
      "num_tokens": 71183851.0,
      "step": 41035
    },
    {
      "entropy": 5.367454385757446,
      "epoch": 3.1930363742462555,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039913086203569373,
      "loss": 4.9358,
      "mean_token_accuracy": 0.2092101827263832,
      "num_tokens": 71192592.0,
      "step": 41040
    },
    {
      "entropy": 5.375721025466919,
      "epoch": 3.1934254036179732,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039910774497892046,
      "loss": 4.9329,
      "mean_token_accuracy": 0.21254641711711883,
      "num_tokens": 71201139.0,
      "step": 41045
    },
    {
      "entropy": 5.286424350738526,
      "epoch": 3.193814432989691,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039908462603900977,
      "loss": 4.7768,
      "mean_token_accuracy": 0.2172835499048233,
      "num_tokens": 71209400.0,
      "step": 41050
    },
    {
      "entropy": 5.304551267623902,
      "epoch": 3.194203462361408,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003990615052163123,
      "loss": 4.8599,
      "mean_token_accuracy": 0.21378054469823837,
      "num_tokens": 71218041.0,
      "step": 41055
    },
    {
      "entropy": 5.408150005340576,
      "epoch": 3.194592491733126,
      "grad_norm": 1.234375,
      "learning_rate": 0.000399038382511179,
      "loss": 4.9054,
      "mean_token_accuracy": 0.20234844982624053,
      "num_tokens": 71227323.0,
      "step": 41060
    },
    {
      "entropy": 5.454283857345581,
      "epoch": 3.1949815211048436,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003990152579239607,
      "loss": 4.9155,
      "mean_token_accuracy": 0.2093242809176445,
      "num_tokens": 71235410.0,
      "step": 41065
    },
    {
      "entropy": 5.3518414974212645,
      "epoch": 3.195370550476561,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039899213145500815,
      "loss": 4.8632,
      "mean_token_accuracy": 0.20371057242155075,
      "num_tokens": 71243577.0,
      "step": 41070
    },
    {
      "entropy": 5.392955350875854,
      "epoch": 3.1957595798482785,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039896900310467244,
      "loss": 5.1062,
      "mean_token_accuracy": 0.19635103940963744,
      "num_tokens": 71252857.0,
      "step": 41075
    },
    {
      "entropy": 5.387083387374878,
      "epoch": 3.196148609219996,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039894587287330434,
      "loss": 4.9608,
      "mean_token_accuracy": 0.1973141074180603,
      "num_tokens": 71261841.0,
      "step": 41080
    },
    {
      "entropy": 5.448593807220459,
      "epoch": 3.1965376385917135,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039892274076125503,
      "loss": 4.8976,
      "mean_token_accuracy": 0.20215146839618683,
      "num_tokens": 71270057.0,
      "step": 41085
    },
    {
      "entropy": 5.371054172515869,
      "epoch": 3.196926667963431,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003988996067688753,
      "loss": 4.8997,
      "mean_token_accuracy": 0.20316412001848222,
      "num_tokens": 71279488.0,
      "step": 41090
    },
    {
      "entropy": 5.464171743392944,
      "epoch": 3.197315697335149,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003988764708965163,
      "loss": 5.0156,
      "mean_token_accuracy": 0.19523449689149858,
      "num_tokens": 71288565.0,
      "step": 41095
    },
    {
      "entropy": 5.389809846878052,
      "epoch": 3.1977047267068666,
      "grad_norm": 1.109375,
      "learning_rate": 0.000398853333144529,
      "loss": 4.8562,
      "mean_token_accuracy": 0.21535607874393464,
      "num_tokens": 71296773.0,
      "step": 41100
    },
    {
      "entropy": 5.438268184661865,
      "epoch": 3.198093756078584,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039883019351326447,
      "loss": 4.9223,
      "mean_token_accuracy": 0.20909268856048585,
      "num_tokens": 71305665.0,
      "step": 41105
    },
    {
      "entropy": 5.349180126190186,
      "epoch": 3.1984827854503015,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039880705200307377,
      "loss": 4.8195,
      "mean_token_accuracy": 0.209517602622509,
      "num_tokens": 71314410.0,
      "step": 41110
    },
    {
      "entropy": 5.420874691009521,
      "epoch": 3.198871814822019,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003987839086143083,
      "loss": 4.9589,
      "mean_token_accuracy": 0.2013794630765915,
      "num_tokens": 71324362.0,
      "step": 41115
    },
    {
      "entropy": 5.397889757156372,
      "epoch": 3.1992608441937365,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039876076334731885,
      "loss": 4.9485,
      "mean_token_accuracy": 0.20154876559972762,
      "num_tokens": 71333244.0,
      "step": 41120
    },
    {
      "entropy": 5.355621671676635,
      "epoch": 3.199649873565454,
      "grad_norm": 1.140625,
      "learning_rate": 0.000398737616202457,
      "loss": 4.8393,
      "mean_token_accuracy": 0.20976633429527283,
      "num_tokens": 71342191.0,
      "step": 41125
    },
    {
      "entropy": 5.403627538681031,
      "epoch": 3.200038902937172,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039871446718007364,
      "loss": 4.9463,
      "mean_token_accuracy": 0.1990462139248848,
      "num_tokens": 71350955.0,
      "step": 41130
    },
    {
      "entropy": 5.405143976211548,
      "epoch": 3.200427932308889,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003986913162805201,
      "loss": 4.912,
      "mean_token_accuracy": 0.20801666527986526,
      "num_tokens": 71359930.0,
      "step": 41135
    },
    {
      "entropy": 5.371746397018432,
      "epoch": 3.200816961680607,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039866816350414786,
      "loss": 4.8696,
      "mean_token_accuracy": 0.20665326863527297,
      "num_tokens": 71368695.0,
      "step": 41140
    },
    {
      "entropy": 5.327304267883301,
      "epoch": 3.2012059910523245,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000398645008851308,
      "loss": 4.8721,
      "mean_token_accuracy": 0.21070967316627504,
      "num_tokens": 71377360.0,
      "step": 41145
    },
    {
      "entropy": 5.421457433700562,
      "epoch": 3.201595020424042,
      "grad_norm": 1.125,
      "learning_rate": 0.00039862185232235196,
      "loss": 4.9197,
      "mean_token_accuracy": 0.2041022762656212,
      "num_tokens": 71386393.0,
      "step": 41150
    },
    {
      "entropy": 5.367809534072876,
      "epoch": 3.2019840497957595,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003985986939176311,
      "loss": 4.85,
      "mean_token_accuracy": 0.2122493118047714,
      "num_tokens": 71394909.0,
      "step": 41155
    },
    {
      "entropy": 5.37042932510376,
      "epoch": 3.202373079167477,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039857553363749674,
      "loss": 4.9439,
      "mean_token_accuracy": 0.20139406323432923,
      "num_tokens": 71403375.0,
      "step": 41160
    },
    {
      "entropy": 5.428268766403198,
      "epoch": 3.202762108539195,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039855237148230026,
      "loss": 4.8724,
      "mean_token_accuracy": 0.2037455677986145,
      "num_tokens": 71412114.0,
      "step": 41165
    },
    {
      "entropy": 5.4241517066955565,
      "epoch": 3.203151137910912,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039852920745239343,
      "loss": 4.9064,
      "mean_token_accuracy": 0.20287071615457536,
      "num_tokens": 71420795.0,
      "step": 41170
    },
    {
      "entropy": 5.297631406784058,
      "epoch": 3.20354016728263,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039850604154812727,
      "loss": 4.8242,
      "mean_token_accuracy": 0.21561457216739655,
      "num_tokens": 71429035.0,
      "step": 41175
    },
    {
      "entropy": 5.287792682647705,
      "epoch": 3.2039291966543475,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003984828737698536,
      "loss": 4.8012,
      "mean_token_accuracy": 0.213507042825222,
      "num_tokens": 71437322.0,
      "step": 41180
    },
    {
      "entropy": 5.3966796875,
      "epoch": 3.204318226026065,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039845970411792384,
      "loss": 4.9188,
      "mean_token_accuracy": 0.2054111674427986,
      "num_tokens": 71445713.0,
      "step": 41185
    },
    {
      "entropy": 5.453556728363037,
      "epoch": 3.2047072553977825,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003984365325926896,
      "loss": 4.8818,
      "mean_token_accuracy": 0.20796265453100204,
      "num_tokens": 71453913.0,
      "step": 41190
    },
    {
      "entropy": 5.466651582717896,
      "epoch": 3.2050962847695,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039841335919450245,
      "loss": 4.8959,
      "mean_token_accuracy": 0.20893990397453308,
      "num_tokens": 71462036.0,
      "step": 41195
    },
    {
      "entropy": 5.348736238479614,
      "epoch": 3.205485314141218,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039839018392371397,
      "loss": 4.893,
      "mean_token_accuracy": 0.20294382721185683,
      "num_tokens": 71470760.0,
      "step": 41200
    },
    {
      "entropy": 5.382997322082519,
      "epoch": 3.205874343512935,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003983670067806759,
      "loss": 4.9249,
      "mean_token_accuracy": 0.19844181090593338,
      "num_tokens": 71479612.0,
      "step": 41205
    },
    {
      "entropy": 5.439393806457519,
      "epoch": 3.206263372884653,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003983438277657398,
      "loss": 4.8992,
      "mean_token_accuracy": 0.2106337293982506,
      "num_tokens": 71487868.0,
      "step": 41210
    },
    {
      "entropy": 5.365131616592407,
      "epoch": 3.2066524022563705,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003983206468792575,
      "loss": 4.8713,
      "mean_token_accuracy": 0.20805854648351668,
      "num_tokens": 71496296.0,
      "step": 41215
    },
    {
      "entropy": 5.271234178543091,
      "epoch": 3.2070414316280877,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003982974641215806,
      "loss": 4.8433,
      "mean_token_accuracy": 0.21512570679187776,
      "num_tokens": 71505076.0,
      "step": 41220
    },
    {
      "entropy": 5.359581184387207,
      "epoch": 3.2074304609998054,
      "grad_norm": 1.15625,
      "learning_rate": 0.000398274279493061,
      "loss": 4.8102,
      "mean_token_accuracy": 0.21830783188343048,
      "num_tokens": 71513585.0,
      "step": 41225
    },
    {
      "entropy": 5.490751934051514,
      "epoch": 3.207819490371523,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039825109299405036,
      "loss": 4.9563,
      "mean_token_accuracy": 0.1992805629968643,
      "num_tokens": 71521519.0,
      "step": 41230
    },
    {
      "entropy": 5.261100196838379,
      "epoch": 3.2082085197432404,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003982279046249006,
      "loss": 4.8074,
      "mean_token_accuracy": 0.21393427550792693,
      "num_tokens": 71529910.0,
      "step": 41235
    },
    {
      "entropy": 5.270900869369507,
      "epoch": 3.208597549114958,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039820471438596346,
      "loss": 4.8562,
      "mean_token_accuracy": 0.21103986352682114,
      "num_tokens": 71538716.0,
      "step": 41240
    },
    {
      "entropy": 5.433122253417968,
      "epoch": 3.208986578486676,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039818152227759097,
      "loss": 5.0068,
      "mean_token_accuracy": 0.2008223533630371,
      "num_tokens": 71548737.0,
      "step": 41245
    },
    {
      "entropy": 5.4912757396698,
      "epoch": 3.2093756078583935,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003981583283001349,
      "loss": 4.9266,
      "mean_token_accuracy": 0.2066113606095314,
      "num_tokens": 71557356.0,
      "step": 41250
    },
    {
      "entropy": 5.400700283050537,
      "epoch": 3.2097646372301107,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003981351324539472,
      "loss": 4.9267,
      "mean_token_accuracy": 0.20826032906770706,
      "num_tokens": 71566858.0,
      "step": 41255
    },
    {
      "entropy": 5.3413153171539305,
      "epoch": 3.2101536666018284,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003981119347393799,
      "loss": 4.8111,
      "mean_token_accuracy": 0.2137700065970421,
      "num_tokens": 71575060.0,
      "step": 41260
    },
    {
      "entropy": 5.4415168285369875,
      "epoch": 3.210542695973546,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039808873515678495,
      "loss": 4.9097,
      "mean_token_accuracy": 0.2108290284872055,
      "num_tokens": 71583987.0,
      "step": 41265
    },
    {
      "entropy": 5.293552875518799,
      "epoch": 3.2109317253452634,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003980655337065144,
      "loss": 4.9868,
      "mean_token_accuracy": 0.1928752303123474,
      "num_tokens": 71593433.0,
      "step": 41270
    },
    {
      "entropy": 5.482369661331177,
      "epoch": 3.211320754716981,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003980423303889201,
      "loss": 4.9034,
      "mean_token_accuracy": 0.20759985744953155,
      "num_tokens": 71601795.0,
      "step": 41275
    },
    {
      "entropy": 5.402105903625488,
      "epoch": 3.211709784088699,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039801912520435437,
      "loss": 4.8162,
      "mean_token_accuracy": 0.21766954511404038,
      "num_tokens": 71611566.0,
      "step": 41280
    },
    {
      "entropy": 5.362250566482544,
      "epoch": 3.2120988134604165,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003979959181531692,
      "loss": 4.8156,
      "mean_token_accuracy": 0.21424843817949296,
      "num_tokens": 71620157.0,
      "step": 41285
    },
    {
      "entropy": 5.432988548278809,
      "epoch": 3.2124878428321337,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039797270923571683,
      "loss": 4.9586,
      "mean_token_accuracy": 0.2011295884847641,
      "num_tokens": 71629342.0,
      "step": 41290
    },
    {
      "entropy": 5.356537246704102,
      "epoch": 3.2128768722038514,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039794949845234925,
      "loss": 4.8317,
      "mean_token_accuracy": 0.2125357583165169,
      "num_tokens": 71638118.0,
      "step": 41295
    },
    {
      "entropy": 5.367728853225708,
      "epoch": 3.213265901575569,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039792628580341874,
      "loss": 4.8632,
      "mean_token_accuracy": 0.2125856414437294,
      "num_tokens": 71647261.0,
      "step": 41300
    },
    {
      "entropy": 5.337952756881714,
      "epoch": 3.2136549309472864,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039790307128927745,
      "loss": 4.9067,
      "mean_token_accuracy": 0.20959130227565764,
      "num_tokens": 71655513.0,
      "step": 41305
    },
    {
      "entropy": 5.372780799865723,
      "epoch": 3.214043960319004,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039787985491027773,
      "loss": 4.928,
      "mean_token_accuracy": 0.20572420060634614,
      "num_tokens": 71663696.0,
      "step": 41310
    },
    {
      "entropy": 5.388763809204102,
      "epoch": 3.2144329896907218,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039785663666677175,
      "loss": 4.9228,
      "mean_token_accuracy": 0.20438790917396546,
      "num_tokens": 71672860.0,
      "step": 41315
    },
    {
      "entropy": 5.357812929153442,
      "epoch": 3.214822019062439,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039783341655911185,
      "loss": 4.8676,
      "mean_token_accuracy": 0.20539714992046357,
      "num_tokens": 71681343.0,
      "step": 41320
    },
    {
      "entropy": 5.371541690826416,
      "epoch": 3.2152110484341567,
      "grad_norm": 1.125,
      "learning_rate": 0.0003978101945876504,
      "loss": 4.8657,
      "mean_token_accuracy": 0.2110227406024933,
      "num_tokens": 71689769.0,
      "step": 41325
    },
    {
      "entropy": 5.380926895141601,
      "epoch": 3.2156000778058744,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039778697075273977,
      "loss": 4.8962,
      "mean_token_accuracy": 0.21422243416309356,
      "num_tokens": 71697782.0,
      "step": 41330
    },
    {
      "entropy": 5.4299054622650145,
      "epoch": 3.2159891071775917,
      "grad_norm": 1.25,
      "learning_rate": 0.00039776374505473224,
      "loss": 4.9256,
      "mean_token_accuracy": 0.2012175813317299,
      "num_tokens": 71706050.0,
      "step": 41335
    },
    {
      "entropy": 5.3100344181060795,
      "epoch": 3.2163781365493094,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003977405174939803,
      "loss": 4.7861,
      "mean_token_accuracy": 0.21079102754592896,
      "num_tokens": 71714623.0,
      "step": 41340
    },
    {
      "entropy": 5.294760274887085,
      "epoch": 3.216767165921027,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039771728807083635,
      "loss": 4.8329,
      "mean_token_accuracy": 0.2149375006556511,
      "num_tokens": 71723450.0,
      "step": 41345
    },
    {
      "entropy": 5.3777018070220945,
      "epoch": 3.2171561952927448,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039769405678565287,
      "loss": 4.8743,
      "mean_token_accuracy": 0.20640100687742233,
      "num_tokens": 71731986.0,
      "step": 41350
    },
    {
      "entropy": 5.4244485855102536,
      "epoch": 3.217545224664462,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039767082363878237,
      "loss": 4.9893,
      "mean_token_accuracy": 0.1947380244731903,
      "num_tokens": 71740650.0,
      "step": 41355
    },
    {
      "entropy": 5.388730573654175,
      "epoch": 3.2179342540361797,
      "grad_norm": 1.125,
      "learning_rate": 0.0003976475886305774,
      "loss": 4.9021,
      "mean_token_accuracy": 0.19871594458818437,
      "num_tokens": 71749358.0,
      "step": 41360
    },
    {
      "entropy": 5.310841608047485,
      "epoch": 3.2183232834078974,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003976243517613904,
      "loss": 4.8534,
      "mean_token_accuracy": 0.20978207141160965,
      "num_tokens": 71757862.0,
      "step": 41365
    },
    {
      "entropy": 5.390683555603028,
      "epoch": 3.2187123127796147,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039760111303157415,
      "loss": 4.9471,
      "mean_token_accuracy": 0.19820580780506133,
      "num_tokens": 71766321.0,
      "step": 41370
    },
    {
      "entropy": 5.389574003219605,
      "epoch": 3.2191013421513324,
      "grad_norm": 1.0625,
      "learning_rate": 0.000397577872441481,
      "loss": 4.9134,
      "mean_token_accuracy": 0.20398980975151063,
      "num_tokens": 71775726.0,
      "step": 41375
    },
    {
      "entropy": 5.4229738235473635,
      "epoch": 3.21949037152305,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003975546299914639,
      "loss": 4.9866,
      "mean_token_accuracy": 0.2040518894791603,
      "num_tokens": 71785227.0,
      "step": 41380
    },
    {
      "entropy": 5.402367353439331,
      "epoch": 3.2198794008947678,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003975313856818753,
      "loss": 4.9302,
      "mean_token_accuracy": 0.20796280056238176,
      "num_tokens": 71794226.0,
      "step": 41385
    },
    {
      "entropy": 5.428212690353393,
      "epoch": 3.220268430266485,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003975081395130679,
      "loss": 4.892,
      "mean_token_accuracy": 0.20707811713218688,
      "num_tokens": 71802758.0,
      "step": 41390
    },
    {
      "entropy": 5.379517841339111,
      "epoch": 3.2206574596382027,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003974848914853946,
      "loss": 4.906,
      "mean_token_accuracy": 0.2102273628115654,
      "num_tokens": 71811156.0,
      "step": 41395
    },
    {
      "entropy": 5.259658670425415,
      "epoch": 3.2210464890099204,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039746164159920795,
      "loss": 4.711,
      "mean_token_accuracy": 0.21655880510807038,
      "num_tokens": 71819243.0,
      "step": 41400
    },
    {
      "entropy": 5.349394655227661,
      "epoch": 3.2214355183816377,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039743838985486075,
      "loss": 4.9212,
      "mean_token_accuracy": 0.200860396027565,
      "num_tokens": 71827551.0,
      "step": 41405
    },
    {
      "entropy": 5.3924932956695555,
      "epoch": 3.2218245477533554,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003974151362527059,
      "loss": 4.8982,
      "mean_token_accuracy": 0.20760398507118225,
      "num_tokens": 71835548.0,
      "step": 41410
    },
    {
      "entropy": 5.457680130004883,
      "epoch": 3.222213577125073,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039739188079309626,
      "loss": 4.9248,
      "mean_token_accuracy": 0.20420821905136108,
      "num_tokens": 71844117.0,
      "step": 41415
    },
    {
      "entropy": 5.438068199157715,
      "epoch": 3.2226026064967903,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003973686234763846,
      "loss": 4.9021,
      "mean_token_accuracy": 0.20057043731212615,
      "num_tokens": 71852665.0,
      "step": 41420
    },
    {
      "entropy": 5.332819890975952,
      "epoch": 3.222991635868508,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003973453643029239,
      "loss": 4.9127,
      "mean_token_accuracy": 0.2046664074063301,
      "num_tokens": 71861162.0,
      "step": 41425
    },
    {
      "entropy": 5.408700704574585,
      "epoch": 3.2233806652402257,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003973221032730669,
      "loss": 4.9206,
      "mean_token_accuracy": 0.20650875121355056,
      "num_tokens": 71870764.0,
      "step": 41430
    },
    {
      "entropy": 5.410676956176758,
      "epoch": 3.2237696946119434,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003972988403871668,
      "loss": 4.8792,
      "mean_token_accuracy": 0.21093602925539018,
      "num_tokens": 71879291.0,
      "step": 41435
    },
    {
      "entropy": 5.3718212127685545,
      "epoch": 3.2241587239836607,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003972755756455764,
      "loss": 4.9215,
      "mean_token_accuracy": 0.20503434985876084,
      "num_tokens": 71888070.0,
      "step": 41440
    },
    {
      "entropy": 5.424950313568115,
      "epoch": 3.2245477533553784,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003972523090486487,
      "loss": 4.9767,
      "mean_token_accuracy": 0.20637868195772172,
      "num_tokens": 71896314.0,
      "step": 41445
    },
    {
      "entropy": 5.289296960830688,
      "epoch": 3.224936782727096,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003972290405967369,
      "loss": 4.7337,
      "mean_token_accuracy": 0.2218379110097885,
      "num_tokens": 71905786.0,
      "step": 41450
    },
    {
      "entropy": 5.359384775161743,
      "epoch": 3.2253258120988133,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003972057702901939,
      "loss": 4.8961,
      "mean_token_accuracy": 0.2042383372783661,
      "num_tokens": 71914329.0,
      "step": 41455
    },
    {
      "entropy": 5.411332988739014,
      "epoch": 3.225714841470531,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003971824981293729,
      "loss": 4.9747,
      "mean_token_accuracy": 0.206637379527092,
      "num_tokens": 71923615.0,
      "step": 41460
    },
    {
      "entropy": 5.4521955966949465,
      "epoch": 3.2261038708422487,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003971592241146269,
      "loss": 4.8803,
      "mean_token_accuracy": 0.20825263112783432,
      "num_tokens": 71931666.0,
      "step": 41465
    },
    {
      "entropy": 5.48477258682251,
      "epoch": 3.226492900213966,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003971359482463092,
      "loss": 4.9738,
      "mean_token_accuracy": 0.20763561129570007,
      "num_tokens": 71940770.0,
      "step": 41470
    },
    {
      "entropy": 5.290102148056031,
      "epoch": 3.2268819295856837,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003971126705247727,
      "loss": 4.8192,
      "mean_token_accuracy": 0.20992695093154906,
      "num_tokens": 71949626.0,
      "step": 41475
    },
    {
      "entropy": 5.350097608566284,
      "epoch": 3.2272709589574013,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003970893909503709,
      "loss": 4.8605,
      "mean_token_accuracy": 0.2097383812069893,
      "num_tokens": 71958055.0,
      "step": 41480
    },
    {
      "entropy": 5.380182361602783,
      "epoch": 3.227659988329119,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000397066109523457,
      "loss": 4.7817,
      "mean_token_accuracy": 0.2213051900267601,
      "num_tokens": 71966353.0,
      "step": 41485
    },
    {
      "entropy": 5.39834246635437,
      "epoch": 3.2280490177008363,
      "grad_norm": 1.125,
      "learning_rate": 0.0003970428262443842,
      "loss": 4.9812,
      "mean_token_accuracy": 0.20385613292455673,
      "num_tokens": 71975479.0,
      "step": 41490
    },
    {
      "entropy": 5.335905694961548,
      "epoch": 3.228438047072554,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039701954111350565,
      "loss": 4.8077,
      "mean_token_accuracy": 0.21200876981019973,
      "num_tokens": 71983859.0,
      "step": 41495
    },
    {
      "entropy": 5.245261573791504,
      "epoch": 3.2288270764442717,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003969962541311748,
      "loss": 4.737,
      "mean_token_accuracy": 0.22028515487909317,
      "num_tokens": 71992804.0,
      "step": 41500
    },
    {
      "entropy": 5.467371034622192,
      "epoch": 3.229216105815989,
      "grad_norm": 1.125,
      "learning_rate": 0.00039697296529774494,
      "loss": 5.0008,
      "mean_token_accuracy": 0.19807840883731842,
      "num_tokens": 72000942.0,
      "step": 41505
    },
    {
      "entropy": 5.390512084960937,
      "epoch": 3.2296051351877066,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039694967461356944,
      "loss": 4.8403,
      "mean_token_accuracy": 0.21506406515836715,
      "num_tokens": 72009275.0,
      "step": 41510
    },
    {
      "entropy": 5.339192867279053,
      "epoch": 3.2299941645594243,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039692638207900177,
      "loss": 4.9186,
      "mean_token_accuracy": 0.2012126177549362,
      "num_tokens": 72018441.0,
      "step": 41515
    },
    {
      "entropy": 5.416414880752564,
      "epoch": 3.2303831939311416,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003969030876943954,
      "loss": 4.8828,
      "mean_token_accuracy": 0.20821574032306672,
      "num_tokens": 72026657.0,
      "step": 41520
    },
    {
      "entropy": 5.344050502777099,
      "epoch": 3.2307722233028593,
      "grad_norm": 1.125,
      "learning_rate": 0.00039687979146010367,
      "loss": 4.7823,
      "mean_token_accuracy": 0.21662601977586746,
      "num_tokens": 72034893.0,
      "step": 41525
    },
    {
      "entropy": 5.427075481414795,
      "epoch": 3.231161252674577,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039685649337648,
      "loss": 4.9696,
      "mean_token_accuracy": 0.20820416659116744,
      "num_tokens": 72044880.0,
      "step": 41530
    },
    {
      "entropy": 5.42249641418457,
      "epoch": 3.2315502820462947,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039683319344387813,
      "loss": 4.8933,
      "mean_token_accuracy": 0.20334199219942092,
      "num_tokens": 72053407.0,
      "step": 41535
    },
    {
      "entropy": 5.343121433258057,
      "epoch": 3.231939311418012,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039680989166265135,
      "loss": 4.8637,
      "mean_token_accuracy": 0.21781446039676666,
      "num_tokens": 72062042.0,
      "step": 41540
    },
    {
      "entropy": 5.338514423370361,
      "epoch": 3.2323283407897296,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039678658803315336,
      "loss": 4.8799,
      "mean_token_accuracy": 0.20290609449148178,
      "num_tokens": 72070120.0,
      "step": 41545
    },
    {
      "entropy": 5.368254089355469,
      "epoch": 3.2327173701614473,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039676328255573777,
      "loss": 4.8856,
      "mean_token_accuracy": 0.20546307563781738,
      "num_tokens": 72079271.0,
      "step": 41550
    },
    {
      "entropy": 5.441349601745605,
      "epoch": 3.2331063995331646,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039673997523075825,
      "loss": 4.9343,
      "mean_token_accuracy": 0.2076259508728981,
      "num_tokens": 72088103.0,
      "step": 41555
    },
    {
      "entropy": 5.46873984336853,
      "epoch": 3.2334954289048823,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039671666605856825,
      "loss": 4.9892,
      "mean_token_accuracy": 0.20788469463586806,
      "num_tokens": 72097244.0,
      "step": 41560
    },
    {
      "entropy": 5.3720427513122555,
      "epoch": 3.2338844582766,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039669335503952165,
      "loss": 4.9478,
      "mean_token_accuracy": 0.20182398557662964,
      "num_tokens": 72105778.0,
      "step": 41565
    },
    {
      "entropy": 5.41860065460205,
      "epoch": 3.2342734876483172,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039667004217397206,
      "loss": 4.9079,
      "mean_token_accuracy": 0.21269229352474212,
      "num_tokens": 72113750.0,
      "step": 41570
    },
    {
      "entropy": 5.3342078685760494,
      "epoch": 3.234662517020035,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039664672746227326,
      "loss": 4.8597,
      "mean_token_accuracy": 0.20644067823886872,
      "num_tokens": 72122181.0,
      "step": 41575
    },
    {
      "entropy": 5.439698886871338,
      "epoch": 3.2350515463917526,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003966234109047789,
      "loss": 4.9158,
      "mean_token_accuracy": 0.2072242945432663,
      "num_tokens": 72131364.0,
      "step": 41580
    },
    {
      "entropy": 5.4215654850006105,
      "epoch": 3.2354405757634703,
      "grad_norm": 1.1875,
      "learning_rate": 0.000396600092501843,
      "loss": 4.9612,
      "mean_token_accuracy": 0.2020516276359558,
      "num_tokens": 72139580.0,
      "step": 41585
    },
    {
      "entropy": 5.340943288803101,
      "epoch": 3.2358296051351876,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039657677225381915,
      "loss": 4.8641,
      "mean_token_accuracy": 0.20832016319036484,
      "num_tokens": 72148414.0,
      "step": 41590
    },
    {
      "entropy": 5.28799033164978,
      "epoch": 3.2362186345069053,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003965534501610613,
      "loss": 4.7584,
      "mean_token_accuracy": 0.21945344805717468,
      "num_tokens": 72156612.0,
      "step": 41595
    },
    {
      "entropy": 5.3809651851654055,
      "epoch": 3.236607663878623,
      "grad_norm": 1.125,
      "learning_rate": 0.0003965301262239234,
      "loss": 4.9442,
      "mean_token_accuracy": 0.20541845560073851,
      "num_tokens": 72164711.0,
      "step": 41600
    },
    {
      "entropy": 5.420930671691894,
      "epoch": 3.2369966932503402,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003965068004427591,
      "loss": 5.0191,
      "mean_token_accuracy": 0.19502892196178437,
      "num_tokens": 72172991.0,
      "step": 41605
    },
    {
      "entropy": 5.442810344696045,
      "epoch": 3.237385722622058,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003964834728179226,
      "loss": 4.919,
      "mean_token_accuracy": 0.20342083722352983,
      "num_tokens": 72181464.0,
      "step": 41610
    },
    {
      "entropy": 5.448131275177002,
      "epoch": 3.2377747519937756,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039646014334976783,
      "loss": 4.8894,
      "mean_token_accuracy": 0.21077728420495986,
      "num_tokens": 72189984.0,
      "step": 41615
    },
    {
      "entropy": 5.234420108795166,
      "epoch": 3.2381637813654933,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039643681203864863,
      "loss": 4.7321,
      "mean_token_accuracy": 0.21412567496299745,
      "num_tokens": 72197845.0,
      "step": 41620
    },
    {
      "entropy": 5.343894147872925,
      "epoch": 3.2385528107372106,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039641347888491905,
      "loss": 4.9238,
      "mean_token_accuracy": 0.20534812957048415,
      "num_tokens": 72206307.0,
      "step": 41625
    },
    {
      "entropy": 5.378876972198486,
      "epoch": 3.2389418401089283,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039639014388893337,
      "loss": 4.9178,
      "mean_token_accuracy": 0.20125647783279418,
      "num_tokens": 72214540.0,
      "step": 41630
    },
    {
      "entropy": 5.379921627044678,
      "epoch": 3.239330869480646,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003963668070510453,
      "loss": 4.9205,
      "mean_token_accuracy": 0.2128364324569702,
      "num_tokens": 72223287.0,
      "step": 41635
    },
    {
      "entropy": 5.379098892211914,
      "epoch": 3.2397198988523632,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003963434683716091,
      "loss": 4.8608,
      "mean_token_accuracy": 0.20771485716104507,
      "num_tokens": 72232314.0,
      "step": 41640
    },
    {
      "entropy": 5.4644852638244625,
      "epoch": 3.240108928224081,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003963201278509789,
      "loss": 4.9845,
      "mean_token_accuracy": 0.19656602144241334,
      "num_tokens": 72241424.0,
      "step": 41645
    },
    {
      "entropy": 5.343140554428101,
      "epoch": 3.2404979575957986,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003962967854895089,
      "loss": 4.8209,
      "mean_token_accuracy": 0.21585943698883056,
      "num_tokens": 72249329.0,
      "step": 41650
    },
    {
      "entropy": 5.389833927154541,
      "epoch": 3.240886986967516,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003962734412875533,
      "loss": 4.964,
      "mean_token_accuracy": 0.20332630276679992,
      "num_tokens": 72258484.0,
      "step": 41655
    },
    {
      "entropy": 5.37431902885437,
      "epoch": 3.2412760163392336,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003962500952454662,
      "loss": 4.9036,
      "mean_token_accuracy": 0.21161109805107117,
      "num_tokens": 72267280.0,
      "step": 41660
    },
    {
      "entropy": 5.3249749660491945,
      "epoch": 3.2416650457109513,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039622674736360195,
      "loss": 4.9233,
      "mean_token_accuracy": 0.20424025058746337,
      "num_tokens": 72275399.0,
      "step": 41665
    },
    {
      "entropy": 5.340392398834228,
      "epoch": 3.2420540750826685,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039620339764231467,
      "loss": 4.8754,
      "mean_token_accuracy": 0.21412305831909179,
      "num_tokens": 72285202.0,
      "step": 41670
    },
    {
      "entropy": 5.329820013046264,
      "epoch": 3.242443104454386,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039618004608195877,
      "loss": 4.7901,
      "mean_token_accuracy": 0.21443211138248444,
      "num_tokens": 72293713.0,
      "step": 41675
    },
    {
      "entropy": 5.354944324493408,
      "epoch": 3.242832133826104,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039615669268288863,
      "loss": 4.8265,
      "mean_token_accuracy": 0.21387371718883513,
      "num_tokens": 72302591.0,
      "step": 41680
    },
    {
      "entropy": 5.36503119468689,
      "epoch": 3.2432211631978216,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039613333744545836,
      "loss": 4.8621,
      "mean_token_accuracy": 0.21321283131837845,
      "num_tokens": 72310981.0,
      "step": 41685
    },
    {
      "entropy": 5.273787784576416,
      "epoch": 3.243610192569539,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003961099803700226,
      "loss": 4.8019,
      "mean_token_accuracy": 0.2164582669734955,
      "num_tokens": 72319797.0,
      "step": 41690
    },
    {
      "entropy": 5.384508848190308,
      "epoch": 3.2439992219412566,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003960866214569357,
      "loss": 4.8992,
      "mean_token_accuracy": 0.20790557712316513,
      "num_tokens": 72329122.0,
      "step": 41695
    },
    {
      "entropy": 5.320718765258789,
      "epoch": 3.2443882513129743,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039606326070655204,
      "loss": 4.8812,
      "mean_token_accuracy": 0.21045181900262833,
      "num_tokens": 72337601.0,
      "step": 41700
    },
    {
      "entropy": 5.46179986000061,
      "epoch": 3.2447772806846915,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039603989811922596,
      "loss": 4.9663,
      "mean_token_accuracy": 0.20764378756284713,
      "num_tokens": 72346484.0,
      "step": 41705
    },
    {
      "entropy": 5.3948663711547855,
      "epoch": 3.245166310056409,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003960165336953122,
      "loss": 4.848,
      "mean_token_accuracy": 0.20332710742950438,
      "num_tokens": 72354989.0,
      "step": 41710
    },
    {
      "entropy": 5.374741840362549,
      "epoch": 3.245555339428127,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003959931674351651,
      "loss": 4.9001,
      "mean_token_accuracy": 0.2020726591348648,
      "num_tokens": 72363216.0,
      "step": 41715
    },
    {
      "entropy": 5.338291501998901,
      "epoch": 3.2459443687998446,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003959697993391393,
      "loss": 4.7937,
      "mean_token_accuracy": 0.21319809854030608,
      "num_tokens": 72371994.0,
      "step": 41720
    },
    {
      "entropy": 5.2396196842193605,
      "epoch": 3.246333398171562,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039594642940758934,
      "loss": 4.7821,
      "mean_token_accuracy": 0.21820472031831742,
      "num_tokens": 72381467.0,
      "step": 41725
    },
    {
      "entropy": 5.337137651443482,
      "epoch": 3.2467224275432796,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039592305764086984,
      "loss": 4.7491,
      "mean_token_accuracy": 0.2187451794743538,
      "num_tokens": 72390087.0,
      "step": 41730
    },
    {
      "entropy": 5.393392419815063,
      "epoch": 3.2471114569149973,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039589968403933533,
      "loss": 4.8407,
      "mean_token_accuracy": 0.20656883865594863,
      "num_tokens": 72399525.0,
      "step": 41735
    },
    {
      "entropy": 5.2897227764129635,
      "epoch": 3.2475004862867145,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003958763086033406,
      "loss": 4.8327,
      "mean_token_accuracy": 0.22144639641046523,
      "num_tokens": 72407487.0,
      "step": 41740
    },
    {
      "entropy": 5.2737571716308596,
      "epoch": 3.247889515658432,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003958529313332403,
      "loss": 4.8158,
      "mean_token_accuracy": 0.2136649891734123,
      "num_tokens": 72416118.0,
      "step": 41745
    },
    {
      "entropy": 5.357871723175049,
      "epoch": 3.24827854503015,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003958295522293891,
      "loss": 4.8127,
      "mean_token_accuracy": 0.21547832041978837,
      "num_tokens": 72424946.0,
      "step": 41750
    },
    {
      "entropy": 5.3265913963317875,
      "epoch": 3.248667574401867,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003958061712921417,
      "loss": 4.8789,
      "mean_token_accuracy": 0.2070385366678238,
      "num_tokens": 72435588.0,
      "step": 41755
    },
    {
      "entropy": 5.2481786727905275,
      "epoch": 3.249056603773585,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000395782788521853,
      "loss": 4.8346,
      "mean_token_accuracy": 0.2093169704079628,
      "num_tokens": 72443728.0,
      "step": 41760
    },
    {
      "entropy": 5.334076929092407,
      "epoch": 3.2494456331453025,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003957594039188778,
      "loss": 4.8031,
      "mean_token_accuracy": 0.21667481660842897,
      "num_tokens": 72452163.0,
      "step": 41765
    },
    {
      "entropy": 5.402799034118653,
      "epoch": 3.24983466251702,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003957360174835708,
      "loss": 4.9694,
      "mean_token_accuracy": 0.20120735019445418,
      "num_tokens": 72461369.0,
      "step": 41770
    },
    {
      "entropy": 5.332149839401245,
      "epoch": 3.2502236918887375,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003957126292162868,
      "loss": 4.7519,
      "mean_token_accuracy": 0.21207719147205353,
      "num_tokens": 72470370.0,
      "step": 41775
    },
    {
      "entropy": 5.305440139770508,
      "epoch": 3.250612721260455,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003956892391173809,
      "loss": 4.7658,
      "mean_token_accuracy": 0.2052665024995804,
      "num_tokens": 72478440.0,
      "step": 41780
    },
    {
      "entropy": 5.397833013534546,
      "epoch": 3.251001750632173,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039566584718720793,
      "loss": 5.0151,
      "mean_token_accuracy": 0.20426901131868364,
      "num_tokens": 72487556.0,
      "step": 41785
    },
    {
      "entropy": 5.313311386108398,
      "epoch": 3.25139078000389,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003956424534261227,
      "loss": 4.8832,
      "mean_token_accuracy": 0.20929237604141235,
      "num_tokens": 72495546.0,
      "step": 41790
    },
    {
      "entropy": 5.398889970779419,
      "epoch": 3.251779809375608,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039561905783448025,
      "loss": 4.8981,
      "mean_token_accuracy": 0.19964513480663298,
      "num_tokens": 72504479.0,
      "step": 41795
    },
    {
      "entropy": 5.361911201477051,
      "epoch": 3.2521688387473255,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039559566041263565,
      "loss": 4.8997,
      "mean_token_accuracy": 0.2103080153465271,
      "num_tokens": 72512758.0,
      "step": 41800
    },
    {
      "entropy": 5.331159782409668,
      "epoch": 3.252557868119043,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003955722611609438,
      "loss": 4.8482,
      "mean_token_accuracy": 0.2054233193397522,
      "num_tokens": 72520920.0,
      "step": 41805
    },
    {
      "entropy": 5.354918193817139,
      "epoch": 3.2529468974907605,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039554886007975976,
      "loss": 4.8573,
      "mean_token_accuracy": 0.20749630331993102,
      "num_tokens": 72529485.0,
      "step": 41810
    },
    {
      "entropy": 5.3143871307373045,
      "epoch": 3.253335926862478,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039552545716943873,
      "loss": 4.8121,
      "mean_token_accuracy": 0.20750755518674852,
      "num_tokens": 72538267.0,
      "step": 41815
    },
    {
      "entropy": 5.299352693557739,
      "epoch": 3.253724956234196,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003955020524303358,
      "loss": 4.8024,
      "mean_token_accuracy": 0.21294431239366532,
      "num_tokens": 72546806.0,
      "step": 41820
    },
    {
      "entropy": 5.335839700698853,
      "epoch": 3.254113985605913,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039547864586280583,
      "loss": 4.9036,
      "mean_token_accuracy": 0.20338983684778214,
      "num_tokens": 72554988.0,
      "step": 41825
    },
    {
      "entropy": 5.355981254577637,
      "epoch": 3.254503014977631,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003954552374672042,
      "loss": 4.8709,
      "mean_token_accuracy": 0.2094633847475052,
      "num_tokens": 72562760.0,
      "step": 41830
    },
    {
      "entropy": 5.287191247940063,
      "epoch": 3.2548920443493485,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003954318272438862,
      "loss": 4.8476,
      "mean_token_accuracy": 0.2076736271381378,
      "num_tokens": 72571908.0,
      "step": 41835
    },
    {
      "entropy": 5.483546733856201,
      "epoch": 3.255281073721066,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003954084151932067,
      "loss": 4.9345,
      "mean_token_accuracy": 0.20738704055547713,
      "num_tokens": 72579840.0,
      "step": 41840
    },
    {
      "entropy": 5.338909721374511,
      "epoch": 3.2556701030927835,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003953850013155214,
      "loss": 4.8053,
      "mean_token_accuracy": 0.20920224189758302,
      "num_tokens": 72588313.0,
      "step": 41845
    },
    {
      "entropy": 5.372808218002319,
      "epoch": 3.256059132464501,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003953615856111852,
      "loss": 4.914,
      "mean_token_accuracy": 0.20202042013406754,
      "num_tokens": 72597634.0,
      "step": 41850
    },
    {
      "entropy": 5.391716718673706,
      "epoch": 3.256448161836219,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039533816808055345,
      "loss": 4.8689,
      "mean_token_accuracy": 0.2087606206536293,
      "num_tokens": 72606092.0,
      "step": 41855
    },
    {
      "entropy": 5.38955249786377,
      "epoch": 3.256837191207936,
      "grad_norm": 1.125,
      "learning_rate": 0.00039531474872398165,
      "loss": 4.9251,
      "mean_token_accuracy": 0.20530149340629578,
      "num_tokens": 72614753.0,
      "step": 41860
    },
    {
      "entropy": 5.381896018981934,
      "epoch": 3.257226220579654,
      "grad_norm": 1.15625,
      "learning_rate": 0.000395291327541825,
      "loss": 4.9351,
      "mean_token_accuracy": 0.2010135069489479,
      "num_tokens": 72624412.0,
      "step": 41865
    },
    {
      "entropy": 5.333378410339355,
      "epoch": 3.257615249951371,
      "grad_norm": 1.125,
      "learning_rate": 0.00039526790453443887,
      "loss": 4.8509,
      "mean_token_accuracy": 0.2023446276783943,
      "num_tokens": 72633680.0,
      "step": 41870
    },
    {
      "entropy": 5.318945264816284,
      "epoch": 3.258004279323089,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039524447970217875,
      "loss": 4.8599,
      "mean_token_accuracy": 0.20652349591255187,
      "num_tokens": 72642077.0,
      "step": 41875
    },
    {
      "entropy": 5.441044044494629,
      "epoch": 3.2583933086948065,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039522105304540007,
      "loss": 4.9445,
      "mean_token_accuracy": 0.20762906074523926,
      "num_tokens": 72650400.0,
      "step": 41880
    },
    {
      "entropy": 5.376451587677002,
      "epoch": 3.258782338066524,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003951976245644582,
      "loss": 4.922,
      "mean_token_accuracy": 0.2064551144838333,
      "num_tokens": 72659164.0,
      "step": 41885
    },
    {
      "entropy": 5.3774515151977536,
      "epoch": 3.2591713674382414,
      "grad_norm": 1.125,
      "learning_rate": 0.0003951741942597087,
      "loss": 4.927,
      "mean_token_accuracy": 0.209536612033844,
      "num_tokens": 72667771.0,
      "step": 41890
    },
    {
      "entropy": 5.395315456390381,
      "epoch": 3.259560396809959,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039515076213150714,
      "loss": 4.9168,
      "mean_token_accuracy": 0.20644933879375457,
      "num_tokens": 72677547.0,
      "step": 41895
    },
    {
      "entropy": 5.311406803131104,
      "epoch": 3.259949426181677,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039512732818020903,
      "loss": 4.84,
      "mean_token_accuracy": 0.21157461702823638,
      "num_tokens": 72686239.0,
      "step": 41900
    },
    {
      "entropy": 5.4303281784057615,
      "epoch": 3.260338455553394,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003951038924061698,
      "loss": 4.9735,
      "mean_token_accuracy": 0.20099610537290574,
      "num_tokens": 72694645.0,
      "step": 41905
    },
    {
      "entropy": 5.501288890838623,
      "epoch": 3.2607274849251118,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039508045480974517,
      "loss": 4.9785,
      "mean_token_accuracy": 0.20128174275159835,
      "num_tokens": 72703166.0,
      "step": 41910
    },
    {
      "entropy": 5.343318843841553,
      "epoch": 3.2611165142968295,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039505701539129087,
      "loss": 4.8122,
      "mean_token_accuracy": 0.21636137068271638,
      "num_tokens": 72711986.0,
      "step": 41915
    },
    {
      "entropy": 5.338136482238769,
      "epoch": 3.261505543668547,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039503357415116236,
      "loss": 4.8513,
      "mean_token_accuracy": 0.21461150646209717,
      "num_tokens": 72720410.0,
      "step": 41920
    },
    {
      "entropy": 5.433432865142822,
      "epoch": 3.2618945730402644,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039501013108971547,
      "loss": 4.9298,
      "mean_token_accuracy": 0.20563365668058395,
      "num_tokens": 72729287.0,
      "step": 41925
    },
    {
      "entropy": 5.408391380310059,
      "epoch": 3.262283602411982,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003949866862073059,
      "loss": 4.9556,
      "mean_token_accuracy": 0.20406564027071,
      "num_tokens": 72737183.0,
      "step": 41930
    },
    {
      "entropy": 5.2857671737670895,
      "epoch": 3.2626726317837,
      "grad_norm": 1.34375,
      "learning_rate": 0.00039496323950428925,
      "loss": 4.9093,
      "mean_token_accuracy": 0.20963615775108338,
      "num_tokens": 72745357.0,
      "step": 41935
    },
    {
      "entropy": 5.334775924682617,
      "epoch": 3.263061661155417,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003949397909810213,
      "loss": 4.8301,
      "mean_token_accuracy": 0.21608931720256805,
      "num_tokens": 72753762.0,
      "step": 41940
    },
    {
      "entropy": 5.376649188995361,
      "epoch": 3.2634506905271348,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039491634063785813,
      "loss": 4.7948,
      "mean_token_accuracy": 0.21431507617235185,
      "num_tokens": 72762259.0,
      "step": 41945
    },
    {
      "entropy": 5.342026853561402,
      "epoch": 3.2638397198988525,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039489288847515525,
      "loss": 4.8968,
      "mean_token_accuracy": 0.20659036189317703,
      "num_tokens": 72770831.0,
      "step": 41950
    },
    {
      "entropy": 5.347036409378052,
      "epoch": 3.26422874927057,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003948694344932687,
      "loss": 4.8618,
      "mean_token_accuracy": 0.2098207637667656,
      "num_tokens": 72779337.0,
      "step": 41955
    },
    {
      "entropy": 5.328391456604004,
      "epoch": 3.2646177786422874,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003948459786925542,
      "loss": 4.7777,
      "mean_token_accuracy": 0.21717894524335862,
      "num_tokens": 72788144.0,
      "step": 41960
    },
    {
      "entropy": 5.365144968032837,
      "epoch": 3.265006808014005,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003948225210733677,
      "loss": 4.8886,
      "mean_token_accuracy": 0.20544988363981248,
      "num_tokens": 72796371.0,
      "step": 41965
    },
    {
      "entropy": 5.317060232162476,
      "epoch": 3.265395837385723,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003947990616360652,
      "loss": 4.7999,
      "mean_token_accuracy": 0.2131517767906189,
      "num_tokens": 72804778.0,
      "step": 41970
    },
    {
      "entropy": 5.397791337966919,
      "epoch": 3.26578486675744,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003947756003810026,
      "loss": 4.9271,
      "mean_token_accuracy": 0.21177922636270524,
      "num_tokens": 72813507.0,
      "step": 41975
    },
    {
      "entropy": 5.450518083572388,
      "epoch": 3.2661738961291578,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003947521373085359,
      "loss": 4.9094,
      "mean_token_accuracy": 0.20602784156799317,
      "num_tokens": 72822772.0,
      "step": 41980
    },
    {
      "entropy": 5.3720207691192625,
      "epoch": 3.2665629255008755,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003947286724190212,
      "loss": 4.9319,
      "mean_token_accuracy": 0.20277654528617858,
      "num_tokens": 72831573.0,
      "step": 41985
    },
    {
      "entropy": 5.367356061935425,
      "epoch": 3.2669519548725927,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039470520571281443,
      "loss": 4.8559,
      "mean_token_accuracy": 0.21823744773864745,
      "num_tokens": 72840461.0,
      "step": 41990
    },
    {
      "entropy": 5.332068634033203,
      "epoch": 3.2673409842443104,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039468173719027163,
      "loss": 4.8963,
      "mean_token_accuracy": 0.20825523436069487,
      "num_tokens": 72848834.0,
      "step": 41995
    },
    {
      "entropy": 5.41693081855774,
      "epoch": 3.267730013616028,
      "grad_norm": 1.25,
      "learning_rate": 0.0003946582668517491,
      "loss": 4.8863,
      "mean_token_accuracy": 0.19776221811771394,
      "num_tokens": 72857679.0,
      "step": 42000
    },
    {
      "epoch": 3.267730013616028,
      "eval_entropy": 5.167160672655005,
      "eval_loss": 5.0328240394592285,
      "eval_mean_token_accuracy": 0.20931800638430387,
      "eval_num_tokens": 72857679.0,
      "eval_runtime": 28.789,
      "eval_samples_per_second": 1443.539,
      "eval_steps_per_second": 180.451,
      "step": 42000
    },
    {
      "entropy": 5.428711366653443,
      "epoch": 3.2681190429877454,
      "grad_norm": 1.125,
      "learning_rate": 0.0003946347946976026,
      "loss": 4.9897,
      "mean_token_accuracy": 0.20459525287151337,
      "num_tokens": 72866664.0,
      "step": 42005
    },
    {
      "entropy": 5.321830940246582,
      "epoch": 3.268508072359463,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003946113207281887,
      "loss": 4.8656,
      "mean_token_accuracy": 0.20998022258281707,
      "num_tokens": 72875520.0,
      "step": 42010
    },
    {
      "entropy": 5.3317183494567875,
      "epoch": 3.2688971017311808,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039458784494386337,
      "loss": 4.8153,
      "mean_token_accuracy": 0.21449126303195953,
      "num_tokens": 72884251.0,
      "step": 42015
    },
    {
      "entropy": 5.452831172943116,
      "epoch": 3.2692861311028985,
      "grad_norm": 1.125,
      "learning_rate": 0.00039456436734498275,
      "loss": 4.9695,
      "mean_token_accuracy": 0.2007792830467224,
      "num_tokens": 72893317.0,
      "step": 42020
    },
    {
      "entropy": 5.3890361309051515,
      "epoch": 3.2696751604746157,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039454088793190316,
      "loss": 4.8561,
      "mean_token_accuracy": 0.2112349271774292,
      "num_tokens": 72902118.0,
      "step": 42025
    },
    {
      "entropy": 5.339156150817871,
      "epoch": 3.2700641898463334,
      "grad_norm": 1.125,
      "learning_rate": 0.00039451740670498094,
      "loss": 4.9121,
      "mean_token_accuracy": 0.20765883922576905,
      "num_tokens": 72911165.0,
      "step": 42030
    },
    {
      "entropy": 5.39759259223938,
      "epoch": 3.270453219218051,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003944939236645723,
      "loss": 4.9832,
      "mean_token_accuracy": 0.1984483554959297,
      "num_tokens": 72920348.0,
      "step": 42035
    },
    {
      "entropy": 5.451859664916992,
      "epoch": 3.2708422485897684,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003944704388110335,
      "loss": 4.8835,
      "mean_token_accuracy": 0.20356089770793914,
      "num_tokens": 72929568.0,
      "step": 42040
    },
    {
      "entropy": 5.362993383407593,
      "epoch": 3.271231277961486,
      "grad_norm": 1.21875,
      "learning_rate": 0.000394446952144721,
      "loss": 4.8928,
      "mean_token_accuracy": 0.20983300507068633,
      "num_tokens": 72937972.0,
      "step": 42045
    },
    {
      "entropy": 5.340504312515259,
      "epoch": 3.2716203073332037,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003944234636659911,
      "loss": 4.8832,
      "mean_token_accuracy": 0.21147648990154266,
      "num_tokens": 72946186.0,
      "step": 42050
    },
    {
      "entropy": 5.3148456573486325,
      "epoch": 3.2720093367049214,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003943999733752003,
      "loss": 4.809,
      "mean_token_accuracy": 0.21389032304286956,
      "num_tokens": 72954521.0,
      "step": 42055
    },
    {
      "entropy": 5.408980178833008,
      "epoch": 3.2723983660766387,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003943764812727048,
      "loss": 4.8386,
      "mean_token_accuracy": 0.21521628946065902,
      "num_tokens": 72962420.0,
      "step": 42060
    },
    {
      "entropy": 5.391130161285401,
      "epoch": 3.2727873954483564,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039435298735886133,
      "loss": 4.8341,
      "mean_token_accuracy": 0.21330428272485732,
      "num_tokens": 72970528.0,
      "step": 42065
    },
    {
      "entropy": 5.315982389450073,
      "epoch": 3.273176424820074,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003943294916340261,
      "loss": 4.8507,
      "mean_token_accuracy": 0.2110617235302925,
      "num_tokens": 72979618.0,
      "step": 42070
    },
    {
      "entropy": 5.281054306030273,
      "epoch": 3.2735654541917913,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003943059940985559,
      "loss": 4.7858,
      "mean_token_accuracy": 0.21885846555233002,
      "num_tokens": 72988133.0,
      "step": 42075
    },
    {
      "entropy": 5.368542098999024,
      "epoch": 3.273954483563509,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039428249475280706,
      "loss": 4.8456,
      "mean_token_accuracy": 0.21754485070705415,
      "num_tokens": 72996872.0,
      "step": 42080
    },
    {
      "entropy": 5.328502798080445,
      "epoch": 3.2743435129352267,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003942589935971363,
      "loss": 4.8719,
      "mean_token_accuracy": 0.22009753584861755,
      "num_tokens": 73005494.0,
      "step": 42085
    },
    {
      "entropy": 5.4510582447052,
      "epoch": 3.274732542306944,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003942354906319001,
      "loss": 4.9656,
      "mean_token_accuracy": 0.1984850749373436,
      "num_tokens": 73013729.0,
      "step": 42090
    },
    {
      "entropy": 5.432101106643676,
      "epoch": 3.2751215716786617,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003942119858574551,
      "loss": 4.8924,
      "mean_token_accuracy": 0.21381503194570542,
      "num_tokens": 73022337.0,
      "step": 42095
    },
    {
      "entropy": 5.319196939468384,
      "epoch": 3.2755106010503794,
      "grad_norm": 1.109375,
      "learning_rate": 0.000394188479274158,
      "loss": 4.9356,
      "mean_token_accuracy": 0.20488589406013488,
      "num_tokens": 73031758.0,
      "step": 42100
    },
    {
      "entropy": 5.368047428131104,
      "epoch": 3.2758996304220966,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039416497088236535,
      "loss": 4.8762,
      "mean_token_accuracy": 0.20543460547924042,
      "num_tokens": 73040046.0,
      "step": 42105
    },
    {
      "entropy": 5.337841320037842,
      "epoch": 3.2762886597938143,
      "grad_norm": 1.1875,
      "learning_rate": 0.000394141460682434,
      "loss": 4.806,
      "mean_token_accuracy": 0.21679086685180665,
      "num_tokens": 73048930.0,
      "step": 42110
    },
    {
      "entropy": 5.328960990905761,
      "epoch": 3.276677689165532,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003941179486747206,
      "loss": 4.8501,
      "mean_token_accuracy": 0.21960316300392152,
      "num_tokens": 73056952.0,
      "step": 42115
    },
    {
      "entropy": 5.3400720119476315,
      "epoch": 3.2770667185372497,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003940944348595819,
      "loss": 4.9071,
      "mean_token_accuracy": 0.20288148522377014,
      "num_tokens": 73066289.0,
      "step": 42120
    },
    {
      "entropy": 5.3371506690979,
      "epoch": 3.277455747908967,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039407091923737475,
      "loss": 4.828,
      "mean_token_accuracy": 0.21449760645627974,
      "num_tokens": 73074484.0,
      "step": 42125
    },
    {
      "entropy": 5.342686128616333,
      "epoch": 3.2778447772806847,
      "grad_norm": 1.46875,
      "learning_rate": 0.00039404740180845594,
      "loss": 4.8764,
      "mean_token_accuracy": 0.2031072735786438,
      "num_tokens": 73082781.0,
      "step": 42130
    },
    {
      "entropy": 5.434997510910034,
      "epoch": 3.2782338066524024,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039402388257318236,
      "loss": 4.8904,
      "mean_token_accuracy": 0.20807009041309357,
      "num_tokens": 73091180.0,
      "step": 42135
    },
    {
      "entropy": 5.336933374404907,
      "epoch": 3.2786228360241196,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003940003615319106,
      "loss": 4.8389,
      "mean_token_accuracy": 0.20721183866262435,
      "num_tokens": 73100328.0,
      "step": 42140
    },
    {
      "entropy": 5.389538097381592,
      "epoch": 3.2790118653958373,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039397683868499784,
      "loss": 4.8576,
      "mean_token_accuracy": 0.22202855497598648,
      "num_tokens": 73108912.0,
      "step": 42145
    },
    {
      "entropy": 5.43189435005188,
      "epoch": 3.279400894767555,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000393953314032801,
      "loss": 4.9359,
      "mean_token_accuracy": 0.20510950386524202,
      "num_tokens": 73118136.0,
      "step": 42150
    },
    {
      "entropy": 5.337279891967773,
      "epoch": 3.2797899241392727,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003939297875756769,
      "loss": 4.8446,
      "mean_token_accuracy": 0.21122337728738785,
      "num_tokens": 73126958.0,
      "step": 42155
    },
    {
      "entropy": 5.3868138790130615,
      "epoch": 3.28017895351099,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003939062593139825,
      "loss": 4.86,
      "mean_token_accuracy": 0.2001390725374222,
      "num_tokens": 73135456.0,
      "step": 42160
    },
    {
      "entropy": 5.28518443107605,
      "epoch": 3.2805679828827077,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000393882729248075,
      "loss": 4.8178,
      "mean_token_accuracy": 0.21503061801195145,
      "num_tokens": 73143190.0,
      "step": 42165
    },
    {
      "entropy": 5.264685487747192,
      "epoch": 3.2809570122544254,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039385919737831127,
      "loss": 4.8361,
      "mean_token_accuracy": 0.21586661636829377,
      "num_tokens": 73152275.0,
      "step": 42170
    },
    {
      "entropy": 5.333975696563721,
      "epoch": 3.2813460416261426,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039383566370504837,
      "loss": 4.8585,
      "mean_token_accuracy": 0.21329369395971298,
      "num_tokens": 73160687.0,
      "step": 42175
    },
    {
      "entropy": 5.357380819320679,
      "epoch": 3.2817350709978603,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003938121282286434,
      "loss": 5.0067,
      "mean_token_accuracy": 0.19609091132879258,
      "num_tokens": 73168843.0,
      "step": 42180
    },
    {
      "entropy": 5.400076913833618,
      "epoch": 3.282124100369578,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039378859094945366,
      "loss": 4.885,
      "mean_token_accuracy": 0.20581024885177612,
      "num_tokens": 73177121.0,
      "step": 42185
    },
    {
      "entropy": 5.406248235702515,
      "epoch": 3.2825131297412953,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003937650518678359,
      "loss": 4.8916,
      "mean_token_accuracy": 0.20771600753068925,
      "num_tokens": 73184952.0,
      "step": 42190
    },
    {
      "entropy": 5.358709764480591,
      "epoch": 3.282902159113013,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003937415109841476,
      "loss": 4.8658,
      "mean_token_accuracy": 0.20596439242362977,
      "num_tokens": 73193509.0,
      "step": 42195
    },
    {
      "entropy": 5.332316923141479,
      "epoch": 3.2832911884847307,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003937179682987459,
      "loss": 4.808,
      "mean_token_accuracy": 0.21530717760324478,
      "num_tokens": 73201083.0,
      "step": 42200
    },
    {
      "entropy": 5.31600341796875,
      "epoch": 3.283680217856448,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039369442381198793,
      "loss": 4.8119,
      "mean_token_accuracy": 0.21223850548267365,
      "num_tokens": 73209852.0,
      "step": 42205
    },
    {
      "entropy": 5.311897802352905,
      "epoch": 3.2840692472281656,
      "grad_norm": 1.140625,
      "learning_rate": 0.000393670877524231,
      "loss": 4.8497,
      "mean_token_accuracy": 0.2147618815302849,
      "num_tokens": 73218338.0,
      "step": 42210
    },
    {
      "entropy": 5.200770902633667,
      "epoch": 3.2844582765998833,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003936473294358325,
      "loss": 4.7511,
      "mean_token_accuracy": 0.2260725423693657,
      "num_tokens": 73226820.0,
      "step": 42215
    },
    {
      "entropy": 5.358584880828857,
      "epoch": 3.284847305971601,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003936237795471495,
      "loss": 4.8964,
      "mean_token_accuracy": 0.21270556449890138,
      "num_tokens": 73235019.0,
      "step": 42220
    },
    {
      "entropy": 5.33240966796875,
      "epoch": 3.2852363353433183,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039360022785853957,
      "loss": 4.8748,
      "mean_token_accuracy": 0.2124168947339058,
      "num_tokens": 73243994.0,
      "step": 42225
    },
    {
      "entropy": 5.355202674865723,
      "epoch": 3.285625364715036,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003935766743703599,
      "loss": 4.8846,
      "mean_token_accuracy": 0.2086198791861534,
      "num_tokens": 73252259.0,
      "step": 42230
    },
    {
      "entropy": 5.397878122329712,
      "epoch": 3.2860143940867537,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039355311908296787,
      "loss": 4.9542,
      "mean_token_accuracy": 0.19856929183006286,
      "num_tokens": 73261559.0,
      "step": 42235
    },
    {
      "entropy": 5.3442024230957035,
      "epoch": 3.286403423458471,
      "grad_norm": 1.1875,
      "learning_rate": 0.000393529561996721,
      "loss": 4.8446,
      "mean_token_accuracy": 0.2225371092557907,
      "num_tokens": 73269534.0,
      "step": 42240
    },
    {
      "entropy": 5.322185897827149,
      "epoch": 3.2867924528301886,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039350600311197667,
      "loss": 4.8222,
      "mean_token_accuracy": 0.2073875054717064,
      "num_tokens": 73277992.0,
      "step": 42245
    },
    {
      "entropy": 5.411247777938843,
      "epoch": 3.2871814822019063,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039348244242909234,
      "loss": 4.9138,
      "mean_token_accuracy": 0.21043745130300523,
      "num_tokens": 73286551.0,
      "step": 42250
    },
    {
      "entropy": 5.4102812767028805,
      "epoch": 3.287570511573624,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039345887994842566,
      "loss": 4.8585,
      "mean_token_accuracy": 0.20302973836660385,
      "num_tokens": 73294945.0,
      "step": 42255
    },
    {
      "entropy": 5.48405499458313,
      "epoch": 3.2879595409453413,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003934353156703339,
      "loss": 4.9548,
      "mean_token_accuracy": 0.20783177465200425,
      "num_tokens": 73303657.0,
      "step": 42260
    },
    {
      "entropy": 5.3461285591125485,
      "epoch": 3.288348570317059,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003934117495951748,
      "loss": 4.9367,
      "mean_token_accuracy": 0.20636518001556398,
      "num_tokens": 73312435.0,
      "step": 42265
    },
    {
      "entropy": 5.415168619155883,
      "epoch": 3.2887375996887767,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003933881817233058,
      "loss": 4.8953,
      "mean_token_accuracy": 0.2120101898908615,
      "num_tokens": 73321268.0,
      "step": 42270
    },
    {
      "entropy": 5.408770990371704,
      "epoch": 3.289126629060494,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039336461205508456,
      "loss": 4.8803,
      "mean_token_accuracy": 0.20962822884321214,
      "num_tokens": 73329608.0,
      "step": 42275
    },
    {
      "entropy": 5.347607803344727,
      "epoch": 3.2895156584322116,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003933410405908689,
      "loss": 4.8269,
      "mean_token_accuracy": 0.21456782668828964,
      "num_tokens": 73338270.0,
      "step": 42280
    },
    {
      "entropy": 5.35476393699646,
      "epoch": 3.2899046878039293,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039331746733101613,
      "loss": 4.8345,
      "mean_token_accuracy": 0.21274319291114807,
      "num_tokens": 73346906.0,
      "step": 42285
    },
    {
      "entropy": 5.35824613571167,
      "epoch": 3.290293717175647,
      "grad_norm": 1.125,
      "learning_rate": 0.00039329389227588416,
      "loss": 4.9028,
      "mean_token_accuracy": 0.21392219066619872,
      "num_tokens": 73355323.0,
      "step": 42290
    },
    {
      "entropy": 5.33562068939209,
      "epoch": 3.2906827465473643,
      "grad_norm": 1.125,
      "learning_rate": 0.00039327031542583067,
      "loss": 4.8919,
      "mean_token_accuracy": 0.20039534717798232,
      "num_tokens": 73364072.0,
      "step": 42295
    },
    {
      "entropy": 5.397718715667724,
      "epoch": 3.291071775919082,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039324673678121344,
      "loss": 4.9557,
      "mean_token_accuracy": 0.20919403433799744,
      "num_tokens": 73372593.0,
      "step": 42300
    },
    {
      "entropy": 5.438922166824341,
      "epoch": 3.291460805290799,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003932231563423901,
      "loss": 5.0064,
      "mean_token_accuracy": 0.19822882860898972,
      "num_tokens": 73381411.0,
      "step": 42305
    },
    {
      "entropy": 5.330200624465943,
      "epoch": 3.291849834662517,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003931995741097186,
      "loss": 4.8155,
      "mean_token_accuracy": 0.21042453050613402,
      "num_tokens": 73389867.0,
      "step": 42310
    },
    {
      "entropy": 5.335262155532837,
      "epoch": 3.2922388640342346,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039317599008355665,
      "loss": 4.8498,
      "mean_token_accuracy": 0.20994527339935304,
      "num_tokens": 73398298.0,
      "step": 42315
    },
    {
      "entropy": 5.282837200164795,
      "epoch": 3.2926278934059523,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003931524042642622,
      "loss": 4.8004,
      "mean_token_accuracy": 0.21810285449028016,
      "num_tokens": 73406566.0,
      "step": 42320
    },
    {
      "entropy": 5.331423759460449,
      "epoch": 3.2930169227776696,
      "grad_norm": 1.125,
      "learning_rate": 0.0003931288166521931,
      "loss": 4.8193,
      "mean_token_accuracy": 0.20740996897220612,
      "num_tokens": 73415390.0,
      "step": 42325
    },
    {
      "entropy": 5.4215796947479244,
      "epoch": 3.2934059521493873,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039310522724770713,
      "loss": 4.926,
      "mean_token_accuracy": 0.20326556414365768,
      "num_tokens": 73424400.0,
      "step": 42330
    },
    {
      "entropy": 5.397515201568604,
      "epoch": 3.293794981521105,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039308163605116235,
      "loss": 4.9787,
      "mean_token_accuracy": 0.2045031249523163,
      "num_tokens": 73433359.0,
      "step": 42335
    },
    {
      "entropy": 5.294189453125,
      "epoch": 3.294184010892822,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003930580430629167,
      "loss": 4.7844,
      "mean_token_accuracy": 0.21304414570331573,
      "num_tokens": 73442353.0,
      "step": 42340
    },
    {
      "entropy": 5.394235467910766,
      "epoch": 3.29457304026454,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003930344482833281,
      "loss": 4.8604,
      "mean_token_accuracy": 0.21535392105579376,
      "num_tokens": 73451122.0,
      "step": 42345
    },
    {
      "entropy": 5.353601503372192,
      "epoch": 3.2949620696362576,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039301085171275475,
      "loss": 4.8932,
      "mean_token_accuracy": 0.20417425334453582,
      "num_tokens": 73459733.0,
      "step": 42350
    },
    {
      "entropy": 5.367926406860351,
      "epoch": 3.2953510990079753,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003929872533515545,
      "loss": 4.8206,
      "mean_token_accuracy": 0.21330493837594985,
      "num_tokens": 73468454.0,
      "step": 42355
    },
    {
      "entropy": 5.387057876586914,
      "epoch": 3.2957401283796925,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039296365320008546,
      "loss": 4.9501,
      "mean_token_accuracy": 0.203630530834198,
      "num_tokens": 73477541.0,
      "step": 42360
    },
    {
      "entropy": 5.335952854156494,
      "epoch": 3.2961291577514102,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039294005125870576,
      "loss": 4.9403,
      "mean_token_accuracy": 0.2026025727391243,
      "num_tokens": 73485625.0,
      "step": 42365
    },
    {
      "entropy": 5.3484447479248045,
      "epoch": 3.296518187123128,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039291644752777346,
      "loss": 4.8751,
      "mean_token_accuracy": 0.21213537603616714,
      "num_tokens": 73494157.0,
      "step": 42370
    },
    {
      "entropy": 5.353247785568238,
      "epoch": 3.296907216494845,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039289284200764676,
      "loss": 4.8791,
      "mean_token_accuracy": 0.2118928000330925,
      "num_tokens": 73503322.0,
      "step": 42375
    },
    {
      "entropy": 5.3896036624908445,
      "epoch": 3.297296245866563,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003928692346986839,
      "loss": 4.9363,
      "mean_token_accuracy": 0.19805720895528794,
      "num_tokens": 73513009.0,
      "step": 42380
    },
    {
      "entropy": 5.410406827926636,
      "epoch": 3.2976852752382806,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039284562560124297,
      "loss": 4.8555,
      "mean_token_accuracy": 0.208360156416893,
      "num_tokens": 73520911.0,
      "step": 42385
    },
    {
      "entropy": 5.344490051269531,
      "epoch": 3.2980743046099983,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003928220147156822,
      "loss": 4.8466,
      "mean_token_accuracy": 0.2149355784058571,
      "num_tokens": 73529810.0,
      "step": 42390
    },
    {
      "entropy": 5.426161670684815,
      "epoch": 3.2984633339817155,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039279840204235994,
      "loss": 4.9948,
      "mean_token_accuracy": 0.20574368089437484,
      "num_tokens": 73539305.0,
      "step": 42395
    },
    {
      "entropy": 5.407580995559693,
      "epoch": 3.2988523633534332,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039277478758163443,
      "loss": 4.9016,
      "mean_token_accuracy": 0.20561892092227935,
      "num_tokens": 73548073.0,
      "step": 42400
    },
    {
      "entropy": 5.368300247192383,
      "epoch": 3.299241392725151,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000392751171333864,
      "loss": 4.8963,
      "mean_token_accuracy": 0.2087974801659584,
      "num_tokens": 73556825.0,
      "step": 42405
    },
    {
      "entropy": 5.337278604507446,
      "epoch": 3.299630422096868,
      "grad_norm": 1.125,
      "learning_rate": 0.00039272755329940683,
      "loss": 4.8705,
      "mean_token_accuracy": 0.20931815952062607,
      "num_tokens": 73565966.0,
      "step": 42410
    },
    {
      "entropy": 5.364250564575196,
      "epoch": 3.300019451468586,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039270393347862155,
      "loss": 4.9158,
      "mean_token_accuracy": 0.21052158027887344,
      "num_tokens": 73574653.0,
      "step": 42415
    },
    {
      "entropy": 5.410108661651611,
      "epoch": 3.3004084808403036,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039268031187186635,
      "loss": 4.9031,
      "mean_token_accuracy": 0.20821256637573243,
      "num_tokens": 73583608.0,
      "step": 42420
    },
    {
      "entropy": 5.292123031616211,
      "epoch": 3.300797510212021,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039265668847949977,
      "loss": 4.8325,
      "mean_token_accuracy": 0.21808480769395827,
      "num_tokens": 73592431.0,
      "step": 42425
    },
    {
      "entropy": 5.355046319961548,
      "epoch": 3.3011865395837385,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003926330633018802,
      "loss": 4.8791,
      "mean_token_accuracy": 0.20998267978429794,
      "num_tokens": 73600842.0,
      "step": 42430
    },
    {
      "entropy": 5.404012632369995,
      "epoch": 3.3015755689554562,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039260943633936615,
      "loss": 4.883,
      "mean_token_accuracy": 0.20850574672222139,
      "num_tokens": 73609725.0,
      "step": 42435
    },
    {
      "entropy": 5.283267831802368,
      "epoch": 3.3019645983271735,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039258580759231603,
      "loss": 4.7608,
      "mean_token_accuracy": 0.21916815638542175,
      "num_tokens": 73619429.0,
      "step": 42440
    },
    {
      "entropy": 5.369707775115967,
      "epoch": 3.302353627698891,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003925621770610885,
      "loss": 4.9364,
      "mean_token_accuracy": 0.19841749370098113,
      "num_tokens": 73628370.0,
      "step": 42445
    },
    {
      "entropy": 5.377126598358155,
      "epoch": 3.302742657070609,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000392538544746042,
      "loss": 4.8687,
      "mean_token_accuracy": 0.21084951609373093,
      "num_tokens": 73637500.0,
      "step": 42450
    },
    {
      "entropy": 5.348564577102661,
      "epoch": 3.3031316864423266,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039251491064753525,
      "loss": 4.8504,
      "mean_token_accuracy": 0.20758171677589415,
      "num_tokens": 73646597.0,
      "step": 42455
    },
    {
      "entropy": 5.31620717048645,
      "epoch": 3.303520715814044,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003924912747659267,
      "loss": 4.7759,
      "mean_token_accuracy": 0.21321991533041001,
      "num_tokens": 73654611.0,
      "step": 42460
    },
    {
      "entropy": 5.344032096862793,
      "epoch": 3.3039097451857615,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039246763710157513,
      "loss": 4.9334,
      "mean_token_accuracy": 0.2014251559972763,
      "num_tokens": 73663011.0,
      "step": 42465
    },
    {
      "entropy": 5.367544651031494,
      "epoch": 3.3042987745574792,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003924439976548391,
      "loss": 4.8608,
      "mean_token_accuracy": 0.21135739833116532,
      "num_tokens": 73670920.0,
      "step": 42470
    },
    {
      "entropy": 5.299671411514282,
      "epoch": 3.3046878039291965,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039242035642607733,
      "loss": 4.8267,
      "mean_token_accuracy": 0.211912901699543,
      "num_tokens": 73680124.0,
      "step": 42475
    },
    {
      "entropy": 5.310601711273193,
      "epoch": 3.305076833300914,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003923967134156486,
      "loss": 4.8445,
      "mean_token_accuracy": 0.2077726572751999,
      "num_tokens": 73688573.0,
      "step": 42480
    },
    {
      "entropy": 5.354006719589234,
      "epoch": 3.305465862672632,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003923730686239115,
      "loss": 4.9103,
      "mean_token_accuracy": 0.2047051101922989,
      "num_tokens": 73697617.0,
      "step": 42485
    },
    {
      "entropy": 5.347281789779663,
      "epoch": 3.3058548920443496,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003923494220512249,
      "loss": 4.7918,
      "mean_token_accuracy": 0.21237691938877107,
      "num_tokens": 73706284.0,
      "step": 42490
    },
    {
      "entropy": 5.325585985183716,
      "epoch": 3.306243921416067,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003923257736979476,
      "loss": 4.8301,
      "mean_token_accuracy": 0.2116803526878357,
      "num_tokens": 73714820.0,
      "step": 42495
    },
    {
      "entropy": 5.3002242088317875,
      "epoch": 3.3066329507877845,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039230212356443855,
      "loss": 4.8605,
      "mean_token_accuracy": 0.20864193588495256,
      "num_tokens": 73723292.0,
      "step": 42500
    },
    {
      "entropy": 5.393737745285034,
      "epoch": 3.307021980159502,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039227847165105635,
      "loss": 4.9093,
      "mean_token_accuracy": 0.19720976501703263,
      "num_tokens": 73732063.0,
      "step": 42505
    },
    {
      "entropy": 5.365620374679565,
      "epoch": 3.3074110095312195,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039225481795816016,
      "loss": 4.9108,
      "mean_token_accuracy": 0.20839255303144455,
      "num_tokens": 73741072.0,
      "step": 42510
    },
    {
      "entropy": 5.2999982833862305,
      "epoch": 3.307800038902937,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003922311624861086,
      "loss": 4.821,
      "mean_token_accuracy": 0.21689532697200775,
      "num_tokens": 73749612.0,
      "step": 42515
    },
    {
      "entropy": 5.335235357284546,
      "epoch": 3.308189068274655,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003922075052352608,
      "loss": 4.8705,
      "mean_token_accuracy": 0.2102089822292328,
      "num_tokens": 73757829.0,
      "step": 42520
    },
    {
      "entropy": 5.333981084823608,
      "epoch": 3.308578097646372,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003921838462059756,
      "loss": 4.8528,
      "mean_token_accuracy": 0.21077800244092942,
      "num_tokens": 73766760.0,
      "step": 42525
    },
    {
      "entropy": 5.397437047958374,
      "epoch": 3.30896712701809,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039216018539861217,
      "loss": 5.0068,
      "mean_token_accuracy": 0.197414068877697,
      "num_tokens": 73775851.0,
      "step": 42530
    },
    {
      "entropy": 5.344344997406006,
      "epoch": 3.3093561563898075,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003921365228135293,
      "loss": 4.7593,
      "mean_token_accuracy": 0.21201589554548264,
      "num_tokens": 73784090.0,
      "step": 42535
    },
    {
      "entropy": 5.335421705245972,
      "epoch": 3.3097451857615248,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039211285845108617,
      "loss": 4.8778,
      "mean_token_accuracy": 0.20496286749839782,
      "num_tokens": 73792709.0,
      "step": 42540
    },
    {
      "entropy": 5.3643499374389645,
      "epoch": 3.3101342151332425,
      "grad_norm": 1.125,
      "learning_rate": 0.0003920891923116418,
      "loss": 4.8951,
      "mean_token_accuracy": 0.21068963557481765,
      "num_tokens": 73801017.0,
      "step": 42545
    },
    {
      "entropy": 5.305859184265136,
      "epoch": 3.31052324450496,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039206552439555544,
      "loss": 4.8372,
      "mean_token_accuracy": 0.21494196504354476,
      "num_tokens": 73809701.0,
      "step": 42550
    },
    {
      "entropy": 5.4098351955413815,
      "epoch": 3.310912273876678,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003920418547031859,
      "loss": 4.9628,
      "mean_token_accuracy": 0.21018091291189195,
      "num_tokens": 73817935.0,
      "step": 42555
    },
    {
      "entropy": 5.281893348693847,
      "epoch": 3.311301303248395,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003920181832348926,
      "loss": 4.7658,
      "mean_token_accuracy": 0.21881279945373536,
      "num_tokens": 73827212.0,
      "step": 42560
    },
    {
      "entropy": 5.347813272476197,
      "epoch": 3.311690332620113,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003919945099910346,
      "loss": 4.9571,
      "mean_token_accuracy": 0.19738508611917496,
      "num_tokens": 73835114.0,
      "step": 42565
    },
    {
      "entropy": 5.37152361869812,
      "epoch": 3.3120793619918305,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039197083497197107,
      "loss": 4.8955,
      "mean_token_accuracy": 0.2067392274737358,
      "num_tokens": 73844137.0,
      "step": 42570
    },
    {
      "entropy": 5.386372470855713,
      "epoch": 3.3124683913635478,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003919471581780613,
      "loss": 4.8638,
      "mean_token_accuracy": 0.2174961358308792,
      "num_tokens": 73852949.0,
      "step": 42575
    },
    {
      "entropy": 5.294262456893921,
      "epoch": 3.3128574207352655,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003919234796096646,
      "loss": 4.7779,
      "mean_token_accuracy": 0.2084813505411148,
      "num_tokens": 73860980.0,
      "step": 42580
    },
    {
      "entropy": 5.358675193786621,
      "epoch": 3.313246450106983,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003918997992671402,
      "loss": 4.9183,
      "mean_token_accuracy": 0.20189463049173356,
      "num_tokens": 73869895.0,
      "step": 42585
    },
    {
      "entropy": 5.305631256103515,
      "epoch": 3.313635479478701,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003918761171508475,
      "loss": 4.7976,
      "mean_token_accuracy": 0.21584342271089554,
      "num_tokens": 73879297.0,
      "step": 42590
    },
    {
      "entropy": 5.362803506851196,
      "epoch": 3.314024508850418,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003918524332611456,
      "loss": 4.8596,
      "mean_token_accuracy": 0.2136973559856415,
      "num_tokens": 73887317.0,
      "step": 42595
    },
    {
      "entropy": 5.302360486984253,
      "epoch": 3.314413538222136,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003918287475983941,
      "loss": 4.84,
      "mean_token_accuracy": 0.20589075684547425,
      "num_tokens": 73896744.0,
      "step": 42600
    },
    {
      "entropy": 5.392413854598999,
      "epoch": 3.3148025675938535,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039180506016295226,
      "loss": 4.8704,
      "mean_token_accuracy": 0.20439125895500182,
      "num_tokens": 73905084.0,
      "step": 42605
    },
    {
      "entropy": 5.431194257736206,
      "epoch": 3.3151915969655708,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039178137095517956,
      "loss": 4.9361,
      "mean_token_accuracy": 0.2032051905989647,
      "num_tokens": 73914342.0,
      "step": 42610
    },
    {
      "entropy": 5.319161319732666,
      "epoch": 3.3155806263372885,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003917576799754354,
      "loss": 4.8568,
      "mean_token_accuracy": 0.21430832892656326,
      "num_tokens": 73923241.0,
      "step": 42615
    },
    {
      "entropy": 5.400479459762574,
      "epoch": 3.315969655709006,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003917339872240793,
      "loss": 4.8673,
      "mean_token_accuracy": 0.20486333221197128,
      "num_tokens": 73932142.0,
      "step": 42620
    },
    {
      "entropy": 5.326025438308716,
      "epoch": 3.3163586850807234,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003917102927014708,
      "loss": 4.802,
      "mean_token_accuracy": 0.21244189739227295,
      "num_tokens": 73941163.0,
      "step": 42625
    },
    {
      "entropy": 5.329640531539917,
      "epoch": 3.316747714452441,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003916865964079694,
      "loss": 4.8182,
      "mean_token_accuracy": 0.21841903775930405,
      "num_tokens": 73949766.0,
      "step": 42630
    },
    {
      "entropy": 5.424621486663819,
      "epoch": 3.317136743824159,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003916628983439345,
      "loss": 4.9156,
      "mean_token_accuracy": 0.2007525384426117,
      "num_tokens": 73958387.0,
      "step": 42635
    },
    {
      "entropy": 5.268437337875366,
      "epoch": 3.317525773195876,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039163919850972585,
      "loss": 4.8285,
      "mean_token_accuracy": 0.20940502285957335,
      "num_tokens": 73967068.0,
      "step": 42640
    },
    {
      "entropy": 5.362547254562378,
      "epoch": 3.3179148025675937,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003916154969057031,
      "loss": 4.9229,
      "mean_token_accuracy": 0.20407520234584808,
      "num_tokens": 73976325.0,
      "step": 42645
    },
    {
      "entropy": 5.407585477828979,
      "epoch": 3.3183038319393114,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039159179353222566,
      "loss": 4.833,
      "mean_token_accuracy": 0.21050372570753098,
      "num_tokens": 73985403.0,
      "step": 42650
    },
    {
      "entropy": 5.4544507503509525,
      "epoch": 3.318692861311029,
      "grad_norm": 1.125,
      "learning_rate": 0.00039156808838965336,
      "loss": 4.9448,
      "mean_token_accuracy": 0.2030564248561859,
      "num_tokens": 73993832.0,
      "step": 42655
    },
    {
      "entropy": 5.34275484085083,
      "epoch": 3.3190818906827464,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039154438147834595,
      "loss": 4.8752,
      "mean_token_accuracy": 0.21381334960460663,
      "num_tokens": 74002374.0,
      "step": 42660
    },
    {
      "entropy": 5.357646846771241,
      "epoch": 3.319470920054464,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039152067279866294,
      "loss": 4.8461,
      "mean_token_accuracy": 0.21271135210990905,
      "num_tokens": 74011655.0,
      "step": 42665
    },
    {
      "entropy": 5.32284026145935,
      "epoch": 3.319859949426182,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003914969623509643,
      "loss": 4.8793,
      "mean_token_accuracy": 0.2072302982211113,
      "num_tokens": 74020760.0,
      "step": 42670
    },
    {
      "entropy": 5.31123833656311,
      "epoch": 3.320248978797899,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003914732501356096,
      "loss": 4.8487,
      "mean_token_accuracy": 0.21172153055667878,
      "num_tokens": 74029315.0,
      "step": 42675
    },
    {
      "entropy": 5.3893169403076175,
      "epoch": 3.3206380081696167,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039144953615295877,
      "loss": 4.8834,
      "mean_token_accuracy": 0.21309739351272583,
      "num_tokens": 74038675.0,
      "step": 42680
    },
    {
      "entropy": 5.426571893692016,
      "epoch": 3.3210270375413344,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039142582040337147,
      "loss": 4.8796,
      "mean_token_accuracy": 0.21346623003482817,
      "num_tokens": 74046836.0,
      "step": 42685
    },
    {
      "entropy": 5.342963409423828,
      "epoch": 3.321416066913052,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003914021028872077,
      "loss": 4.8465,
      "mean_token_accuracy": 0.21121364682912827,
      "num_tokens": 74055903.0,
      "step": 42690
    },
    {
      "entropy": 5.339608573913575,
      "epoch": 3.3218050962847694,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003913783836048274,
      "loss": 4.8642,
      "mean_token_accuracy": 0.21063688695430755,
      "num_tokens": 74064770.0,
      "step": 42695
    },
    {
      "entropy": 5.375512886047363,
      "epoch": 3.322194125656487,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003913546625565902,
      "loss": 4.9775,
      "mean_token_accuracy": 0.20126980692148208,
      "num_tokens": 74072879.0,
      "step": 42700
    },
    {
      "entropy": 5.351089000701904,
      "epoch": 3.322583155028205,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039133093974285635,
      "loss": 4.9022,
      "mean_token_accuracy": 0.20668511986732482,
      "num_tokens": 74081696.0,
      "step": 42705
    },
    {
      "entropy": 5.398468017578125,
      "epoch": 3.322972184399922,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039130721516398556,
      "loss": 4.9021,
      "mean_token_accuracy": 0.20968734920024873,
      "num_tokens": 74090336.0,
      "step": 42710
    },
    {
      "entropy": 5.456526231765747,
      "epoch": 3.3233612137716397,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003912834888203379,
      "loss": 4.9924,
      "mean_token_accuracy": 0.19691791981458664,
      "num_tokens": 74098720.0,
      "step": 42715
    },
    {
      "entropy": 5.396238374710083,
      "epoch": 3.3237502431433574,
      "grad_norm": 1.125,
      "learning_rate": 0.0003912597607122734,
      "loss": 4.9002,
      "mean_token_accuracy": 0.20502217262983322,
      "num_tokens": 74107441.0,
      "step": 42720
    },
    {
      "entropy": 5.316054344177246,
      "epoch": 3.324139272515075,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039123603084015204,
      "loss": 4.8925,
      "mean_token_accuracy": 0.20953400433063507,
      "num_tokens": 74116200.0,
      "step": 42725
    },
    {
      "entropy": 5.327217483520508,
      "epoch": 3.3245283018867924,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039121229920433394,
      "loss": 4.8714,
      "mean_token_accuracy": 0.2094476714730263,
      "num_tokens": 74124886.0,
      "step": 42730
    },
    {
      "entropy": 5.426711130142212,
      "epoch": 3.32491733125851,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003911885658051792,
      "loss": 4.9065,
      "mean_token_accuracy": 0.20472918301820756,
      "num_tokens": 74133204.0,
      "step": 42735
    },
    {
      "entropy": 5.36160249710083,
      "epoch": 3.3253063606302278,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039116483064304794,
      "loss": 4.8586,
      "mean_token_accuracy": 0.20660397112369538,
      "num_tokens": 74141813.0,
      "step": 42740
    },
    {
      "entropy": 5.328420543670655,
      "epoch": 3.325695390001945,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003911410937183002,
      "loss": 4.9055,
      "mean_token_accuracy": 0.20508962571620942,
      "num_tokens": 74150170.0,
      "step": 42745
    },
    {
      "entropy": 5.31853756904602,
      "epoch": 3.3260844193736627,
      "grad_norm": 1.125,
      "learning_rate": 0.00039111735503129626,
      "loss": 4.8419,
      "mean_token_accuracy": 0.2069627583026886,
      "num_tokens": 74158935.0,
      "step": 42750
    },
    {
      "entropy": 5.385170030593872,
      "epoch": 3.3264734487453804,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003910936145823962,
      "loss": 4.8889,
      "mean_token_accuracy": 0.20804361402988433,
      "num_tokens": 74167467.0,
      "step": 42755
    },
    {
      "entropy": 5.315180063247681,
      "epoch": 3.3268624781170977,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003910698723719604,
      "loss": 4.7679,
      "mean_token_accuracy": 0.21956311613321305,
      "num_tokens": 74175513.0,
      "step": 42760
    },
    {
      "entropy": 5.396733665466309,
      "epoch": 3.3272515074888154,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039104612840034904,
      "loss": 4.9215,
      "mean_token_accuracy": 0.2123357132077217,
      "num_tokens": 74183617.0,
      "step": 42765
    },
    {
      "entropy": 5.449248456954956,
      "epoch": 3.327640536860533,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039102238266792234,
      "loss": 4.9188,
      "mean_token_accuracy": 0.20871601998806,
      "num_tokens": 74191662.0,
      "step": 42770
    },
    {
      "entropy": 5.335971403121948,
      "epoch": 3.3280295662322503,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039099863517504064,
      "loss": 4.8224,
      "mean_token_accuracy": 0.21704926788806916,
      "num_tokens": 74199610.0,
      "step": 42775
    },
    {
      "entropy": 5.29502124786377,
      "epoch": 3.328418595603968,
      "grad_norm": 1.25,
      "learning_rate": 0.0003909748859220644,
      "loss": 4.8867,
      "mean_token_accuracy": 0.21051667481660843,
      "num_tokens": 74207579.0,
      "step": 42780
    },
    {
      "entropy": 5.40522575378418,
      "epoch": 3.3288076249756857,
      "grad_norm": 1.25,
      "learning_rate": 0.00039095113490935377,
      "loss": 4.8938,
      "mean_token_accuracy": 0.2075913742184639,
      "num_tokens": 74215428.0,
      "step": 42785
    },
    {
      "entropy": 5.421006536483764,
      "epoch": 3.3291966543474034,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003909273821372692,
      "loss": 4.9369,
      "mean_token_accuracy": 0.20262244790792466,
      "num_tokens": 74223694.0,
      "step": 42790
    },
    {
      "entropy": 5.370386552810669,
      "epoch": 3.3295856837191207,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039090362760617114,
      "loss": 4.9443,
      "mean_token_accuracy": 0.2107130154967308,
      "num_tokens": 74232791.0,
      "step": 42795
    },
    {
      "entropy": 5.422810173034668,
      "epoch": 3.3299747130908384,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039087987131642,
      "loss": 4.9499,
      "mean_token_accuracy": 0.20726177096366882,
      "num_tokens": 74241581.0,
      "step": 42800
    },
    {
      "entropy": 5.303895616531372,
      "epoch": 3.330363742462556,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003908561132683762,
      "loss": 4.8288,
      "mean_token_accuracy": 0.21727501451969147,
      "num_tokens": 74250868.0,
      "step": 42805
    },
    {
      "entropy": 5.3522600650787355,
      "epoch": 3.3307527718342733,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039083235346240044,
      "loss": 4.867,
      "mean_token_accuracy": 0.20561354607343674,
      "num_tokens": 74259992.0,
      "step": 42810
    },
    {
      "entropy": 5.396994304656983,
      "epoch": 3.331141801205991,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000390808591898853,
      "loss": 4.9761,
      "mean_token_accuracy": 0.20532349050045012,
      "num_tokens": 74267900.0,
      "step": 42815
    },
    {
      "entropy": 5.418791627883911,
      "epoch": 3.3315308305777087,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003907848285780945,
      "loss": 4.8323,
      "mean_token_accuracy": 0.20979541540145874,
      "num_tokens": 74276448.0,
      "step": 42820
    },
    {
      "entropy": 5.360913562774658,
      "epoch": 3.3319198599494264,
      "grad_norm": 1.25,
      "learning_rate": 0.00039076106350048557,
      "loss": 4.8886,
      "mean_token_accuracy": 0.20720884054899216,
      "num_tokens": 74284878.0,
      "step": 42825
    },
    {
      "entropy": 5.343692922592163,
      "epoch": 3.3323088893211437,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003907372966663867,
      "loss": 4.8186,
      "mean_token_accuracy": 0.20895814448595046,
      "num_tokens": 74293647.0,
      "step": 42830
    },
    {
      "entropy": 5.37368426322937,
      "epoch": 3.3326979186928614,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003907135280761585,
      "loss": 4.9366,
      "mean_token_accuracy": 0.20624768882989883,
      "num_tokens": 74302122.0,
      "step": 42835
    },
    {
      "entropy": 5.496716737747192,
      "epoch": 3.333086948064579,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003906897577301618,
      "loss": 4.9235,
      "mean_token_accuracy": 0.2008320301771164,
      "num_tokens": 74311729.0,
      "step": 42840
    },
    {
      "entropy": 5.389238595962524,
      "epoch": 3.3334759774362963,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003906659856287571,
      "loss": 4.857,
      "mean_token_accuracy": 0.2144181251525879,
      "num_tokens": 74320694.0,
      "step": 42845
    },
    {
      "entropy": 5.265493679046631,
      "epoch": 3.333865006808014,
      "grad_norm": 1.125,
      "learning_rate": 0.00039064221177230517,
      "loss": 4.7499,
      "mean_token_accuracy": 0.22271642088890076,
      "num_tokens": 74328810.0,
      "step": 42850
    },
    {
      "entropy": 5.280264377593994,
      "epoch": 3.3342540361797317,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039061843616116683,
      "loss": 4.8169,
      "mean_token_accuracy": 0.20777574330568313,
      "num_tokens": 74337005.0,
      "step": 42855
    },
    {
      "entropy": 5.356982564926147,
      "epoch": 3.334643065551449,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039059465879570264,
      "loss": 4.8714,
      "mean_token_accuracy": 0.21355022490024567,
      "num_tokens": 74346353.0,
      "step": 42860
    },
    {
      "entropy": 5.421571063995361,
      "epoch": 3.3350320949231667,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003905708796762736,
      "loss": 5.037,
      "mean_token_accuracy": 0.20000346153974533,
      "num_tokens": 74355698.0,
      "step": 42865
    },
    {
      "entropy": 5.438991928100586,
      "epoch": 3.3354211242948844,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003905470988032403,
      "loss": 4.9946,
      "mean_token_accuracy": 0.20593009889125824,
      "num_tokens": 74365503.0,
      "step": 42870
    },
    {
      "entropy": 5.409659147262573,
      "epoch": 3.3358101536666016,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003905233161769637,
      "loss": 4.9072,
      "mean_token_accuracy": 0.20346214175224303,
      "num_tokens": 74374788.0,
      "step": 42875
    },
    {
      "entropy": 5.3195923328399655,
      "epoch": 3.3361991830383193,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003904995317978048,
      "loss": 4.7488,
      "mean_token_accuracy": 0.2231346294283867,
      "num_tokens": 74384026.0,
      "step": 42880
    },
    {
      "entropy": 5.429484272003174,
      "epoch": 3.336588212410037,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039047574566612416,
      "loss": 4.9564,
      "mean_token_accuracy": 0.2031549870967865,
      "num_tokens": 74392454.0,
      "step": 42885
    },
    {
      "entropy": 5.3413434505462645,
      "epoch": 3.3369772417817547,
      "grad_norm": 1.25,
      "learning_rate": 0.00039045195778228295,
      "loss": 4.8303,
      "mean_token_accuracy": 0.2143862634897232,
      "num_tokens": 74401301.0,
      "step": 42890
    },
    {
      "entropy": 5.291250276565552,
      "epoch": 3.337366271153472,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003904281681466421,
      "loss": 4.849,
      "mean_token_accuracy": 0.21049708276987075,
      "num_tokens": 74409883.0,
      "step": 42895
    },
    {
      "entropy": 5.3549333095550535,
      "epoch": 3.3377553005251897,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003904043767595624,
      "loss": 4.8922,
      "mean_token_accuracy": 0.2025994837284088,
      "num_tokens": 74418999.0,
      "step": 42900
    },
    {
      "entropy": 5.364375829696655,
      "epoch": 3.3381443298969073,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039038058362140507,
      "loss": 4.8067,
      "mean_token_accuracy": 0.2162918508052826,
      "num_tokens": 74427078.0,
      "step": 42905
    },
    {
      "entropy": 5.401328802108765,
      "epoch": 3.3385333592686246,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039035678873253096,
      "loss": 4.8839,
      "mean_token_accuracy": 0.20220344066619872,
      "num_tokens": 74435537.0,
      "step": 42910
    },
    {
      "entropy": 5.323138570785522,
      "epoch": 3.3389223886403423,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003903329920933012,
      "loss": 4.8196,
      "mean_token_accuracy": 0.21777136027812957,
      "num_tokens": 74444941.0,
      "step": 42915
    },
    {
      "entropy": 5.271149206161499,
      "epoch": 3.33931141801206,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003903091937040769,
      "loss": 4.8159,
      "mean_token_accuracy": 0.20615795701742173,
      "num_tokens": 74453772.0,
      "step": 42920
    },
    {
      "entropy": 5.398161792755127,
      "epoch": 3.3397004473837777,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003902853935652191,
      "loss": 4.8249,
      "mean_token_accuracy": 0.20950998812913896,
      "num_tokens": 74462548.0,
      "step": 42925
    },
    {
      "entropy": 5.409592771530152,
      "epoch": 3.340089476755495,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003902615916770889,
      "loss": 4.9048,
      "mean_token_accuracy": 0.20714317858219147,
      "num_tokens": 74471532.0,
      "step": 42930
    },
    {
      "entropy": 5.255369043350219,
      "epoch": 3.3404785061272126,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039023778804004753,
      "loss": 4.7568,
      "mean_token_accuracy": 0.22305561751127242,
      "num_tokens": 74480754.0,
      "step": 42935
    },
    {
      "entropy": 5.36554708480835,
      "epoch": 3.3408675354989303,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039021398265445616,
      "loss": 4.8866,
      "mean_token_accuracy": 0.20617039054632186,
      "num_tokens": 74489597.0,
      "step": 42940
    },
    {
      "entropy": 5.471898412704467,
      "epoch": 3.3412565648706476,
      "grad_norm": 1.1875,
      "learning_rate": 0.000390190175520676,
      "loss": 5.0402,
      "mean_token_accuracy": 0.20137904584407806,
      "num_tokens": 74498871.0,
      "step": 42945
    },
    {
      "entropy": 5.330620861053466,
      "epoch": 3.3416455942423653,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003901663666390683,
      "loss": 4.8643,
      "mean_token_accuracy": 0.2068793371319771,
      "num_tokens": 74507613.0,
      "step": 42950
    },
    {
      "entropy": 5.365389442443847,
      "epoch": 3.342034623614083,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003901425560099943,
      "loss": 4.93,
      "mean_token_accuracy": 0.20374042987823487,
      "num_tokens": 74516072.0,
      "step": 42955
    },
    {
      "entropy": 5.354096984863281,
      "epoch": 3.3424236529858002,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039011874363381523,
      "loss": 4.791,
      "mean_token_accuracy": 0.21314190626144408,
      "num_tokens": 74524947.0,
      "step": 42960
    },
    {
      "entropy": 5.3526490211486815,
      "epoch": 3.342812682357518,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039009492951089236,
      "loss": 4.9155,
      "mean_token_accuracy": 0.20947614312171936,
      "num_tokens": 74533654.0,
      "step": 42965
    },
    {
      "entropy": 5.3304253101348875,
      "epoch": 3.3432017117292356,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003900711136415873,
      "loss": 4.8515,
      "mean_token_accuracy": 0.2180296555161476,
      "num_tokens": 74541850.0,
      "step": 42970
    },
    {
      "entropy": 5.3708991527557375,
      "epoch": 3.343590741100953,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003900472960262611,
      "loss": 4.905,
      "mean_token_accuracy": 0.2054011344909668,
      "num_tokens": 74550037.0,
      "step": 42975
    },
    {
      "entropy": 5.439033937454224,
      "epoch": 3.3439797704726706,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039002347666527537,
      "loss": 4.9602,
      "mean_token_accuracy": 0.20841556191444396,
      "num_tokens": 74558680.0,
      "step": 42980
    },
    {
      "entropy": 5.326992607116699,
      "epoch": 3.3443687998443883,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038999965555899145,
      "loss": 4.7805,
      "mean_token_accuracy": 0.21964341402053833,
      "num_tokens": 74566696.0,
      "step": 42985
    },
    {
      "entropy": 5.299019432067871,
      "epoch": 3.344757829216106,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003899758327077709,
      "loss": 4.8509,
      "mean_token_accuracy": 0.20871831476688385,
      "num_tokens": 74575621.0,
      "step": 42990
    },
    {
      "entropy": 5.384990787506103,
      "epoch": 3.3451468585878232,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038995200811197496,
      "loss": 4.9917,
      "mean_token_accuracy": 0.20930714756250382,
      "num_tokens": 74583587.0,
      "step": 42995
    },
    {
      "entropy": 5.419119596481323,
      "epoch": 3.345535887959541,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038992818177196526,
      "loss": 4.9357,
      "mean_token_accuracy": 0.20468255430459975,
      "num_tokens": 74591530.0,
      "step": 43000
    },
    {
      "entropy": 5.323709297180176,
      "epoch": 3.3459249173312586,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003899043536881034,
      "loss": 4.9173,
      "mean_token_accuracy": 0.19933694005012512,
      "num_tokens": 74599502.0,
      "step": 43005
    },
    {
      "entropy": 5.335243797302246,
      "epoch": 3.346313946702976,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038988052386075084,
      "loss": 4.8674,
      "mean_token_accuracy": 0.21189167499542236,
      "num_tokens": 74608800.0,
      "step": 43010
    },
    {
      "entropy": 5.342799377441406,
      "epoch": 3.3467029760746936,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038985669229026917,
      "loss": 4.8155,
      "mean_token_accuracy": 0.20843221843242646,
      "num_tokens": 74616801.0,
      "step": 43015
    },
    {
      "entropy": 5.334310102462768,
      "epoch": 3.3470920054464113,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038983285897702005,
      "loss": 4.8242,
      "mean_token_accuracy": 0.21673105359077455,
      "num_tokens": 74625605.0,
      "step": 43020
    },
    {
      "entropy": 5.3214117050170895,
      "epoch": 3.347481034818129,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038980902392136496,
      "loss": 4.837,
      "mean_token_accuracy": 0.211628258228302,
      "num_tokens": 74634149.0,
      "step": 43025
    },
    {
      "entropy": 5.371107959747315,
      "epoch": 3.3478700641898462,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003897851871236657,
      "loss": 4.878,
      "mean_token_accuracy": 0.2153734967112541,
      "num_tokens": 74642408.0,
      "step": 43030
    },
    {
      "entropy": 5.378838968276978,
      "epoch": 3.348259093561564,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038976134858428397,
      "loss": 4.9076,
      "mean_token_accuracy": 0.20543947368860244,
      "num_tokens": 74651003.0,
      "step": 43035
    },
    {
      "entropy": 5.441165113449097,
      "epoch": 3.3486481229332816,
      "grad_norm": 1.34375,
      "learning_rate": 0.00038973750830358144,
      "loss": 4.9312,
      "mean_token_accuracy": 0.2076731339097023,
      "num_tokens": 74659282.0,
      "step": 43040
    },
    {
      "entropy": 5.330021858215332,
      "epoch": 3.349037152304999,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038971366628191986,
      "loss": 4.819,
      "mean_token_accuracy": 0.21209264397621155,
      "num_tokens": 74667739.0,
      "step": 43045
    },
    {
      "entropy": 5.384643697738648,
      "epoch": 3.3494261816767166,
      "grad_norm": 1.203125,
      "learning_rate": 0.000389689822519661,
      "loss": 4.9039,
      "mean_token_accuracy": 0.2035349577665329,
      "num_tokens": 74675937.0,
      "step": 43050
    },
    {
      "entropy": 5.332985067367554,
      "epoch": 3.3498152110484343,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038966597701716653,
      "loss": 4.7839,
      "mean_token_accuracy": 0.2245734304189682,
      "num_tokens": 74684529.0,
      "step": 43055
    },
    {
      "entropy": 5.327898263931274,
      "epoch": 3.350204240420152,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003896421297747985,
      "loss": 4.8612,
      "mean_token_accuracy": 0.21872642040252685,
      "num_tokens": 74692814.0,
      "step": 43060
    },
    {
      "entropy": 5.386593914031982,
      "epoch": 3.350593269791869,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003896182807929185,
      "loss": 4.9225,
      "mean_token_accuracy": 0.1999619722366333,
      "num_tokens": 74701477.0,
      "step": 43065
    },
    {
      "entropy": 5.350518417358399,
      "epoch": 3.350982299163587,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003895944300718885,
      "loss": 4.8612,
      "mean_token_accuracy": 0.20919594317674636,
      "num_tokens": 74710158.0,
      "step": 43070
    },
    {
      "entropy": 5.379569625854492,
      "epoch": 3.351371328535304,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038957057761207053,
      "loss": 4.8906,
      "mean_token_accuracy": 0.20774369537830353,
      "num_tokens": 74718269.0,
      "step": 43075
    },
    {
      "entropy": 5.356180572509766,
      "epoch": 3.351760357907022,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038954672341382635,
      "loss": 4.878,
      "mean_token_accuracy": 0.21265193074941635,
      "num_tokens": 74726636.0,
      "step": 43080
    },
    {
      "entropy": 5.373827171325684,
      "epoch": 3.3521493872787396,
      "grad_norm": 1.15625,
      "learning_rate": 0.000389522867477518,
      "loss": 4.8409,
      "mean_token_accuracy": 0.21258383095264435,
      "num_tokens": 74735618.0,
      "step": 43085
    },
    {
      "entropy": 5.388303327560425,
      "epoch": 3.3525384166504573,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003894990098035074,
      "loss": 4.9924,
      "mean_token_accuracy": 0.20318834781646727,
      "num_tokens": 74745161.0,
      "step": 43090
    },
    {
      "entropy": 5.421233224868774,
      "epoch": 3.3529274460221745,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003894751503921566,
      "loss": 4.8669,
      "mean_token_accuracy": 0.20351468175649642,
      "num_tokens": 74753501.0,
      "step": 43095
    },
    {
      "entropy": 5.400782585144043,
      "epoch": 3.353316475393892,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003894512892438275,
      "loss": 4.9293,
      "mean_token_accuracy": 0.2031934529542923,
      "num_tokens": 74761797.0,
      "step": 43100
    },
    {
      "entropy": 5.352028465270996,
      "epoch": 3.35370550476561,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038942742635888236,
      "loss": 4.9424,
      "mean_token_accuracy": 0.20495518147945405,
      "num_tokens": 74771334.0,
      "step": 43105
    },
    {
      "entropy": 5.370048236846924,
      "epoch": 3.354094534137327,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038940356173768303,
      "loss": 4.8876,
      "mean_token_accuracy": 0.2133067950606346,
      "num_tokens": 74780018.0,
      "step": 43110
    },
    {
      "entropy": 5.36496376991272,
      "epoch": 3.354483563509045,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003893796953805919,
      "loss": 4.8203,
      "mean_token_accuracy": 0.21126953959465028,
      "num_tokens": 74788509.0,
      "step": 43115
    },
    {
      "entropy": 5.410099458694458,
      "epoch": 3.3548725928807626,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038935582728797085,
      "loss": 5.0335,
      "mean_token_accuracy": 0.2042275384068489,
      "num_tokens": 74797348.0,
      "step": 43120
    },
    {
      "entropy": 5.408620929718017,
      "epoch": 3.3552616222524803,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038933195746018213,
      "loss": 4.8669,
      "mean_token_accuracy": 0.2087838351726532,
      "num_tokens": 74805815.0,
      "step": 43125
    },
    {
      "entropy": 5.436784553527832,
      "epoch": 3.3556506516241975,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038930808589758797,
      "loss": 4.9311,
      "mean_token_accuracy": 0.20467423498630524,
      "num_tokens": 74814577.0,
      "step": 43130
    },
    {
      "entropy": 5.3527976989746096,
      "epoch": 3.356039680995915,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003892842126005506,
      "loss": 4.7758,
      "mean_token_accuracy": 0.2157517120242119,
      "num_tokens": 74822488.0,
      "step": 43135
    },
    {
      "entropy": 5.366860723495483,
      "epoch": 3.356428710367633,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003892603375694321,
      "loss": 4.9141,
      "mean_token_accuracy": 0.2060529738664627,
      "num_tokens": 74831888.0,
      "step": 43140
    },
    {
      "entropy": 5.388736248016357,
      "epoch": 3.35681773973935,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038923646080459484,
      "loss": 4.91,
      "mean_token_accuracy": 0.2094779133796692,
      "num_tokens": 74840883.0,
      "step": 43145
    },
    {
      "entropy": 5.3407563209533695,
      "epoch": 3.357206769111068,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038921258230640123,
      "loss": 4.7432,
      "mean_token_accuracy": 0.21366207003593446,
      "num_tokens": 74848517.0,
      "step": 43150
    },
    {
      "entropy": 5.279930686950683,
      "epoch": 3.3575957984827856,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038918870207521334,
      "loss": 4.8568,
      "mean_token_accuracy": 0.2116076961159706,
      "num_tokens": 74857189.0,
      "step": 43155
    },
    {
      "entropy": 5.361717605590821,
      "epoch": 3.3579848278545033,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038916482011139373,
      "loss": 4.9738,
      "mean_token_accuracy": 0.2036323994398117,
      "num_tokens": 74866192.0,
      "step": 43160
    },
    {
      "entropy": 5.417784595489502,
      "epoch": 3.3583738572262205,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003891409364153047,
      "loss": 4.8578,
      "mean_token_accuracy": 0.21087003648281097,
      "num_tokens": 74875062.0,
      "step": 43165
    },
    {
      "entropy": 5.373315763473511,
      "epoch": 3.358762886597938,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003891170509873086,
      "loss": 4.9124,
      "mean_token_accuracy": 0.20539368838071823,
      "num_tokens": 74884471.0,
      "step": 43170
    },
    {
      "entropy": 5.2984155178070065,
      "epoch": 3.359151915969656,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038909316382776784,
      "loss": 4.8748,
      "mean_token_accuracy": 0.21019565165042878,
      "num_tokens": 74892945.0,
      "step": 43175
    },
    {
      "entropy": 5.40103325843811,
      "epoch": 3.359540945341373,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003890692749370449,
      "loss": 4.8664,
      "mean_token_accuracy": 0.21363862305879594,
      "num_tokens": 74901944.0,
      "step": 43180
    },
    {
      "entropy": 5.330128908157349,
      "epoch": 3.359929974713091,
      "grad_norm": 1.125,
      "learning_rate": 0.00038904538431550235,
      "loss": 4.8615,
      "mean_token_accuracy": 0.20742748230695723,
      "num_tokens": 74910387.0,
      "step": 43185
    },
    {
      "entropy": 5.399722671508789,
      "epoch": 3.3603190040848085,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003890214919635026,
      "loss": 4.8812,
      "mean_token_accuracy": 0.20547912567853927,
      "num_tokens": 74918570.0,
      "step": 43190
    },
    {
      "entropy": 5.304295539855957,
      "epoch": 3.360708033456526,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003889975978814082,
      "loss": 4.7534,
      "mean_token_accuracy": 0.2156881496310234,
      "num_tokens": 74927009.0,
      "step": 43195
    },
    {
      "entropy": 5.372062349319458,
      "epoch": 3.3610970628282435,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003889737020695817,
      "loss": 4.9179,
      "mean_token_accuracy": 0.20980607569217682,
      "num_tokens": 74935131.0,
      "step": 43200
    },
    {
      "entropy": 5.350213527679443,
      "epoch": 3.361486092199961,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038894980452838573,
      "loss": 4.9075,
      "mean_token_accuracy": 0.20401963889598845,
      "num_tokens": 74944711.0,
      "step": 43205
    },
    {
      "entropy": 5.504417896270752,
      "epoch": 3.3618751215716784,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003889259052581828,
      "loss": 4.9588,
      "mean_token_accuracy": 0.19970207959413527,
      "num_tokens": 74953721.0,
      "step": 43210
    },
    {
      "entropy": 5.337421941757202,
      "epoch": 3.362264150943396,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038890200425933545,
      "loss": 4.8621,
      "mean_token_accuracy": 0.2100758969783783,
      "num_tokens": 74962250.0,
      "step": 43215
    },
    {
      "entropy": 5.389607048034668,
      "epoch": 3.362653180315114,
      "grad_norm": 1.125,
      "learning_rate": 0.0003888781015322067,
      "loss": 4.8956,
      "mean_token_accuracy": 0.21571268886327744,
      "num_tokens": 74971279.0,
      "step": 43220
    },
    {
      "entropy": 5.336884260177612,
      "epoch": 3.3630422096868315,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038885419707715885,
      "loss": 4.9049,
      "mean_token_accuracy": 0.20179797112941741,
      "num_tokens": 74980167.0,
      "step": 43225
    },
    {
      "entropy": 5.412335586547852,
      "epoch": 3.363431239058549,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038883029089455487,
      "loss": 4.9677,
      "mean_token_accuracy": 0.20492493510246276,
      "num_tokens": 74989014.0,
      "step": 43230
    },
    {
      "entropy": 5.360791254043579,
      "epoch": 3.3638202684302665,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003888063829847574,
      "loss": 4.8139,
      "mean_token_accuracy": 0.20741328001022338,
      "num_tokens": 74997395.0,
      "step": 43235
    },
    {
      "entropy": 5.291745710372925,
      "epoch": 3.364209297801984,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003887824733481292,
      "loss": 4.8454,
      "mean_token_accuracy": 0.20952364355325698,
      "num_tokens": 75005962.0,
      "step": 43240
    },
    {
      "entropy": 5.2857666015625,
      "epoch": 3.3645983271737014,
      "grad_norm": 1.125,
      "learning_rate": 0.00038875856198503306,
      "loss": 4.7717,
      "mean_token_accuracy": 0.21336501985788345,
      "num_tokens": 75014201.0,
      "step": 43245
    },
    {
      "entropy": 5.3973448276519775,
      "epoch": 3.364987356545419,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038873464889583177,
      "loss": 4.8823,
      "mean_token_accuracy": 0.20233920961618423,
      "num_tokens": 75022702.0,
      "step": 43250
    },
    {
      "entropy": 5.362800121307373,
      "epoch": 3.365376385917137,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038871073408088826,
      "loss": 4.844,
      "mean_token_accuracy": 0.21447531431913375,
      "num_tokens": 75031488.0,
      "step": 43255
    },
    {
      "entropy": 5.4675404071807865,
      "epoch": 3.3657654152888545,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003886868175405653,
      "loss": 5.0217,
      "mean_token_accuracy": 0.19828703999519348,
      "num_tokens": 75039511.0,
      "step": 43260
    },
    {
      "entropy": 5.395234823226929,
      "epoch": 3.366154444660572,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003886628992752259,
      "loss": 4.913,
      "mean_token_accuracy": 0.20536805391311647,
      "num_tokens": 75048989.0,
      "step": 43265
    },
    {
      "entropy": 5.4423919200897215,
      "epoch": 3.3665434740322895,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003886389792852329,
      "loss": 4.9845,
      "mean_token_accuracy": 0.200530344247818,
      "num_tokens": 75057924.0,
      "step": 43270
    },
    {
      "entropy": 5.331408166885376,
      "epoch": 3.366932503404007,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038861505757094923,
      "loss": 4.8001,
      "mean_token_accuracy": 0.21004365980625153,
      "num_tokens": 75066743.0,
      "step": 43275
    },
    {
      "entropy": 5.324078321456909,
      "epoch": 3.3673215327757244,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003885911341327379,
      "loss": 4.9174,
      "mean_token_accuracy": 0.19951648712158204,
      "num_tokens": 75075784.0,
      "step": 43280
    },
    {
      "entropy": 5.317600727081299,
      "epoch": 3.367710562147442,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000388567208970962,
      "loss": 4.9168,
      "mean_token_accuracy": 0.212921079993248,
      "num_tokens": 75084031.0,
      "step": 43285
    },
    {
      "entropy": 5.296626138687134,
      "epoch": 3.36809959151916,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003885432820859844,
      "loss": 4.8693,
      "mean_token_accuracy": 0.2112334340810776,
      "num_tokens": 75093029.0,
      "step": 43290
    },
    {
      "entropy": 5.3327374935150145,
      "epoch": 3.368488620890877,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038851935347816827,
      "loss": 4.7704,
      "mean_token_accuracy": 0.21931397467851638,
      "num_tokens": 75101620.0,
      "step": 43295
    },
    {
      "entropy": 5.346506357192993,
      "epoch": 3.368877650262595,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003884954231478764,
      "loss": 4.8339,
      "mean_token_accuracy": 0.21202548146247863,
      "num_tokens": 75110179.0,
      "step": 43300
    },
    {
      "entropy": 5.38269453048706,
      "epoch": 3.3692666796343125,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003884714910954724,
      "loss": 4.8976,
      "mean_token_accuracy": 0.20520093142986298,
      "num_tokens": 75119254.0,
      "step": 43305
    },
    {
      "entropy": 5.394597482681275,
      "epoch": 3.3696557090060297,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000388447557321319,
      "loss": 4.8245,
      "mean_token_accuracy": 0.2072989448904991,
      "num_tokens": 75127630.0,
      "step": 43310
    },
    {
      "entropy": 5.385781335830688,
      "epoch": 3.3700447383777474,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003884236218257795,
      "loss": 4.8181,
      "mean_token_accuracy": 0.21806117743253708,
      "num_tokens": 75136866.0,
      "step": 43315
    },
    {
      "entropy": 5.327859926223755,
      "epoch": 3.370433767749465,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003883996846092171,
      "loss": 4.8264,
      "mean_token_accuracy": 0.2168169990181923,
      "num_tokens": 75145570.0,
      "step": 43320
    },
    {
      "entropy": 5.340885257720947,
      "epoch": 3.370822797121183,
      "grad_norm": 1.125,
      "learning_rate": 0.000388375745671995,
      "loss": 4.9403,
      "mean_token_accuracy": 0.2121698960661888,
      "num_tokens": 75154994.0,
      "step": 43325
    },
    {
      "entropy": 5.3572306632995605,
      "epoch": 3.3712118264929,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038835180501447646,
      "loss": 4.8082,
      "mean_token_accuracy": 0.21717213094234467,
      "num_tokens": 75163018.0,
      "step": 43330
    },
    {
      "entropy": 5.293763780593872,
      "epoch": 3.3716008558646178,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003883278626370245,
      "loss": 4.7916,
      "mean_token_accuracy": 0.2100897952914238,
      "num_tokens": 75172234.0,
      "step": 43335
    },
    {
      "entropy": 5.395349311828613,
      "epoch": 3.3719898852363355,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038830391854000285,
      "loss": 4.933,
      "mean_token_accuracy": 0.21201498657464982,
      "num_tokens": 75180816.0,
      "step": 43340
    },
    {
      "entropy": 5.3947898864746096,
      "epoch": 3.3723789146080527,
      "grad_norm": 1.125,
      "learning_rate": 0.00038827997272377447,
      "loss": 4.9095,
      "mean_token_accuracy": 0.2069484546780586,
      "num_tokens": 75189676.0,
      "step": 43345
    },
    {
      "entropy": 5.401403951644897,
      "epoch": 3.3727679439797704,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038825602518870276,
      "loss": 4.8821,
      "mean_token_accuracy": 0.20872806310653685,
      "num_tokens": 75198729.0,
      "step": 43350
    },
    {
      "entropy": 5.327141761779785,
      "epoch": 3.373156973351488,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003882320759351512,
      "loss": 4.8444,
      "mean_token_accuracy": 0.21183761060237885,
      "num_tokens": 75207414.0,
      "step": 43355
    },
    {
      "entropy": 5.306655740737915,
      "epoch": 3.373546002723206,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038820812496348305,
      "loss": 4.7942,
      "mean_token_accuracy": 0.21091445982456208,
      "num_tokens": 75215454.0,
      "step": 43360
    },
    {
      "entropy": 5.35180697441101,
      "epoch": 3.373935032094923,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038818417227406185,
      "loss": 4.8921,
      "mean_token_accuracy": 0.21308453679084777,
      "num_tokens": 75223445.0,
      "step": 43365
    },
    {
      "entropy": 5.41229100227356,
      "epoch": 3.3743240614666408,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038816021786725096,
      "loss": 4.9026,
      "mean_token_accuracy": 0.20590450912714003,
      "num_tokens": 75232120.0,
      "step": 43370
    },
    {
      "entropy": 5.386498355865479,
      "epoch": 3.3747130908383585,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038813626174341386,
      "loss": 4.8587,
      "mean_token_accuracy": 0.20460774451494218,
      "num_tokens": 75240666.0,
      "step": 43375
    },
    {
      "entropy": 5.419239044189453,
      "epoch": 3.3751021202100757,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003881123039029141,
      "loss": 4.8878,
      "mean_token_accuracy": 0.20134558230638505,
      "num_tokens": 75249523.0,
      "step": 43380
    },
    {
      "entropy": 5.384564638137817,
      "epoch": 3.3754911495817934,
      "grad_norm": 1.125,
      "learning_rate": 0.0003880883443461152,
      "loss": 4.9685,
      "mean_token_accuracy": 0.20469918996095657,
      "num_tokens": 75258720.0,
      "step": 43385
    },
    {
      "entropy": 5.3958922863006595,
      "epoch": 3.375880178953511,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003880643830733806,
      "loss": 4.9462,
      "mean_token_accuracy": 0.20746778547763825,
      "num_tokens": 75267424.0,
      "step": 43390
    },
    {
      "entropy": 5.4532005310058596,
      "epoch": 3.3762692083252284,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038804042008507396,
      "loss": 4.898,
      "mean_token_accuracy": 0.2032387763261795,
      "num_tokens": 75276475.0,
      "step": 43395
    },
    {
      "entropy": 5.348258590698242,
      "epoch": 3.376658237696946,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038801645538155883,
      "loss": 4.8371,
      "mean_token_accuracy": 0.21033633798360823,
      "num_tokens": 75284648.0,
      "step": 43400
    },
    {
      "entropy": 5.37525372505188,
      "epoch": 3.3770472670686638,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038799248896319896,
      "loss": 4.9402,
      "mean_token_accuracy": 0.20632299780845642,
      "num_tokens": 75293708.0,
      "step": 43405
    },
    {
      "entropy": 5.329643106460571,
      "epoch": 3.377436296440381,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038796852083035794,
      "loss": 4.8494,
      "mean_token_accuracy": 0.21234897375106812,
      "num_tokens": 75302484.0,
      "step": 43410
    },
    {
      "entropy": 5.440329122543335,
      "epoch": 3.3778253258120987,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003879445509833993,
      "loss": 4.8382,
      "mean_token_accuracy": 0.21537380069494247,
      "num_tokens": 75311112.0,
      "step": 43415
    },
    {
      "entropy": 5.406229066848755,
      "epoch": 3.3782143551838164,
      "grad_norm": 1.234375,
      "learning_rate": 0.000387920579422687,
      "loss": 4.932,
      "mean_token_accuracy": 0.20527267307043076,
      "num_tokens": 75319126.0,
      "step": 43420
    },
    {
      "entropy": 5.320613622665405,
      "epoch": 3.378603384555534,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003878966061485845,
      "loss": 4.8679,
      "mean_token_accuracy": 0.20649921000003815,
      "num_tokens": 75328296.0,
      "step": 43425
    },
    {
      "entropy": 5.3227332592010494,
      "epoch": 3.3789924139272514,
      "grad_norm": 1.125,
      "learning_rate": 0.00038787263116145577,
      "loss": 4.8261,
      "mean_token_accuracy": 0.21385467648506165,
      "num_tokens": 75337036.0,
      "step": 43430
    },
    {
      "entropy": 5.347154569625855,
      "epoch": 3.379381443298969,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038784865446166454,
      "loss": 4.8999,
      "mean_token_accuracy": 0.2154158115386963,
      "num_tokens": 75345691.0,
      "step": 43435
    },
    {
      "entropy": 5.381141281127929,
      "epoch": 3.3797704726706868,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038782467604957455,
      "loss": 4.8626,
      "mean_token_accuracy": 0.2044912353157997,
      "num_tokens": 75354469.0,
      "step": 43440
    },
    {
      "entropy": 5.392892169952392,
      "epoch": 3.380159502042404,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038780069592554977,
      "loss": 4.8826,
      "mean_token_accuracy": 0.2129475474357605,
      "num_tokens": 75363369.0,
      "step": 43445
    },
    {
      "entropy": 5.300584697723389,
      "epoch": 3.3805485314141217,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003877767140899539,
      "loss": 4.8272,
      "mean_token_accuracy": 0.2216135859489441,
      "num_tokens": 75371882.0,
      "step": 43450
    },
    {
      "entropy": 5.295615482330322,
      "epoch": 3.3809375607858394,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038775273054315087,
      "loss": 4.8475,
      "mean_token_accuracy": 0.20961885005235673,
      "num_tokens": 75379801.0,
      "step": 43455
    },
    {
      "entropy": 5.387754631042481,
      "epoch": 3.381326590157557,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003877287452855046,
      "loss": 4.9187,
      "mean_token_accuracy": 0.20855819880962373,
      "num_tokens": 75389588.0,
      "step": 43460
    },
    {
      "entropy": 5.385463285446167,
      "epoch": 3.3817156195292744,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038770475831737905,
      "loss": 4.8998,
      "mean_token_accuracy": 0.20854541957378386,
      "num_tokens": 75398055.0,
      "step": 43465
    },
    {
      "entropy": 5.385175800323486,
      "epoch": 3.382104648900992,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038768076963913823,
      "loss": 4.9026,
      "mean_token_accuracy": 0.20402120500802995,
      "num_tokens": 75406742.0,
      "step": 43470
    },
    {
      "entropy": 5.300346612930298,
      "epoch": 3.3824936782727097,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038765677925114614,
      "loss": 4.7778,
      "mean_token_accuracy": 0.2219518706202507,
      "num_tokens": 75415447.0,
      "step": 43475
    },
    {
      "entropy": 5.363443660736084,
      "epoch": 3.382882707644427,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003876327871537667,
      "loss": 4.9102,
      "mean_token_accuracy": 0.2027765095233917,
      "num_tokens": 75424688.0,
      "step": 43480
    },
    {
      "entropy": 5.452423048019409,
      "epoch": 3.3832717370161447,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038760879334736394,
      "loss": 4.9149,
      "mean_token_accuracy": 0.20905605107545852,
      "num_tokens": 75433334.0,
      "step": 43485
    },
    {
      "entropy": 5.456679010391236,
      "epoch": 3.3836607663878624,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038758479783230197,
      "loss": 4.9595,
      "mean_token_accuracy": 0.1982625588774681,
      "num_tokens": 75442597.0,
      "step": 43490
    },
    {
      "entropy": 5.302169418334961,
      "epoch": 3.38404979575958,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003875608006089449,
      "loss": 4.8721,
      "mean_token_accuracy": 0.21103187650442123,
      "num_tokens": 75450547.0,
      "step": 43495
    },
    {
      "entropy": 5.425356101989746,
      "epoch": 3.3844388251312973,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003875368016776569,
      "loss": 4.9792,
      "mean_token_accuracy": 0.19147971719503404,
      "num_tokens": 75460254.0,
      "step": 43500
    },
    {
      "entropy": 5.448147630691528,
      "epoch": 3.384827854503015,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000387512801038802,
      "loss": 4.9231,
      "mean_token_accuracy": 0.19829928427934645,
      "num_tokens": 75469070.0,
      "step": 43505
    },
    {
      "entropy": 5.413862991333008,
      "epoch": 3.3852168838747323,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003874887986927444,
      "loss": 4.9123,
      "mean_token_accuracy": 0.20581189393997193,
      "num_tokens": 75477610.0,
      "step": 43510
    },
    {
      "entropy": 5.301967287063599,
      "epoch": 3.38560591324645,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003874647946398484,
      "loss": 4.8637,
      "mean_token_accuracy": 0.21003012210130692,
      "num_tokens": 75486395.0,
      "step": 43515
    },
    {
      "entropy": 5.439209508895874,
      "epoch": 3.3859949426181677,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003874407888804781,
      "loss": 5.0188,
      "mean_token_accuracy": 0.20199840962886811,
      "num_tokens": 75495987.0,
      "step": 43520
    },
    {
      "entropy": 5.40192756652832,
      "epoch": 3.3863839719898854,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038741678141499783,
      "loss": 4.8929,
      "mean_token_accuracy": 0.20770883560180664,
      "num_tokens": 75504407.0,
      "step": 43525
    },
    {
      "entropy": 5.418005132675171,
      "epoch": 3.3867730013616026,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003873927722437718,
      "loss": 4.9298,
      "mean_token_accuracy": 0.20730438232421874,
      "num_tokens": 75512953.0,
      "step": 43530
    },
    {
      "entropy": 5.358303022384644,
      "epoch": 3.3871620307333203,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003873687613671643,
      "loss": 4.9221,
      "mean_token_accuracy": 0.20563029050827025,
      "num_tokens": 75521349.0,
      "step": 43535
    },
    {
      "entropy": 5.415304803848267,
      "epoch": 3.387551060105038,
      "grad_norm": 1.25,
      "learning_rate": 0.0003873447487855398,
      "loss": 4.9303,
      "mean_token_accuracy": 0.2028991773724556,
      "num_tokens": 75530178.0,
      "step": 43540
    },
    {
      "entropy": 5.400946998596192,
      "epoch": 3.3879400894767553,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038732073449926244,
      "loss": 4.7939,
      "mean_token_accuracy": 0.21503149569034577,
      "num_tokens": 75539033.0,
      "step": 43545
    },
    {
      "entropy": 5.257432842254639,
      "epoch": 3.388329118848473,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003872967185086968,
      "loss": 4.7008,
      "mean_token_accuracy": 0.22249925285577773,
      "num_tokens": 75547198.0,
      "step": 43550
    },
    {
      "entropy": 5.279355573654175,
      "epoch": 3.3887181482201907,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003872727008142072,
      "loss": 4.8074,
      "mean_token_accuracy": 0.2129985898733139,
      "num_tokens": 75555493.0,
      "step": 43555
    },
    {
      "entropy": 5.353801345825195,
      "epoch": 3.3891071775919084,
      "grad_norm": 1.125,
      "learning_rate": 0.00038724868141615807,
      "loss": 4.8806,
      "mean_token_accuracy": 0.20349131524562836,
      "num_tokens": 75565109.0,
      "step": 43560
    },
    {
      "entropy": 5.417181873321534,
      "epoch": 3.3894962069636256,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003872246603149138,
      "loss": 4.8767,
      "mean_token_accuracy": 0.20838109105825425,
      "num_tokens": 75574156.0,
      "step": 43565
    },
    {
      "entropy": 5.334564208984375,
      "epoch": 3.3898852363353433,
      "grad_norm": 1.125,
      "learning_rate": 0.0003872006375108389,
      "loss": 4.8613,
      "mean_token_accuracy": 0.21686425060033798,
      "num_tokens": 75583160.0,
      "step": 43570
    },
    {
      "entropy": 5.325445938110351,
      "epoch": 3.390274265707061,
      "grad_norm": 1.15625,
      "learning_rate": 0.000387176613004298,
      "loss": 4.8467,
      "mean_token_accuracy": 0.20426160991191863,
      "num_tokens": 75592545.0,
      "step": 43575
    },
    {
      "entropy": 5.3731285572052006,
      "epoch": 3.3906632950787783,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003871525867956555,
      "loss": 4.909,
      "mean_token_accuracy": 0.2015579402446747,
      "num_tokens": 75600999.0,
      "step": 43580
    },
    {
      "entropy": 5.322982215881348,
      "epoch": 3.391052324450496,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038712855888527604,
      "loss": 4.9138,
      "mean_token_accuracy": 0.20270217955112457,
      "num_tokens": 75609807.0,
      "step": 43585
    },
    {
      "entropy": 5.310770559310913,
      "epoch": 3.3914413538222137,
      "grad_norm": 1.125,
      "learning_rate": 0.00038710452927352414,
      "loss": 4.831,
      "mean_token_accuracy": 0.22107521146535875,
      "num_tokens": 75618209.0,
      "step": 43590
    },
    {
      "entropy": 5.394522953033447,
      "epoch": 3.3918303831939314,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003870804979607645,
      "loss": 4.8836,
      "mean_token_accuracy": 0.21430575996637344,
      "num_tokens": 75626547.0,
      "step": 43595
    },
    {
      "entropy": 5.434898948669433,
      "epoch": 3.3922194125656486,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038705646494736163,
      "loss": 4.9204,
      "mean_token_accuracy": 0.21737412214279175,
      "num_tokens": 75634064.0,
      "step": 43600
    },
    {
      "entropy": 5.31025390625,
      "epoch": 3.3926084419373663,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003870324302336802,
      "loss": 4.8392,
      "mean_token_accuracy": 0.2131431743502617,
      "num_tokens": 75642745.0,
      "step": 43605
    },
    {
      "entropy": 5.374652624130249,
      "epoch": 3.392997471309084,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003870083938200851,
      "loss": 4.95,
      "mean_token_accuracy": 0.2056036338210106,
      "num_tokens": 75651581.0,
      "step": 43610
    },
    {
      "entropy": 5.364953994750977,
      "epoch": 3.3933865006808013,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038698435570694084,
      "loss": 4.8835,
      "mean_token_accuracy": 0.20501182675361634,
      "num_tokens": 75661031.0,
      "step": 43615
    },
    {
      "entropy": 5.400718641281128,
      "epoch": 3.393775530052519,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003869603158946122,
      "loss": 4.8208,
      "mean_token_accuracy": 0.21555140614509583,
      "num_tokens": 75669150.0,
      "step": 43620
    },
    {
      "entropy": 5.398316144943237,
      "epoch": 3.3941645594242367,
      "grad_norm": 1.1875,
      "learning_rate": 0.000386936274383464,
      "loss": 4.8861,
      "mean_token_accuracy": 0.20907258838415146,
      "num_tokens": 75677194.0,
      "step": 43625
    },
    {
      "entropy": 5.205297756195068,
      "epoch": 3.394553588795954,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000386912231173861,
      "loss": 4.7945,
      "mean_token_accuracy": 0.21576634496450425,
      "num_tokens": 75685640.0,
      "step": 43630
    },
    {
      "entropy": 5.401810550689698,
      "epoch": 3.3949426181676716,
      "grad_norm": 1.203125,
      "learning_rate": 0.000386888186266168,
      "loss": 4.9246,
      "mean_token_accuracy": 0.20529228895902635,
      "num_tokens": 75693743.0,
      "step": 43635
    },
    {
      "entropy": 5.362131595611572,
      "epoch": 3.3953316475393893,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003868641396607499,
      "loss": 4.8004,
      "mean_token_accuracy": 0.21718397587537766,
      "num_tokens": 75701951.0,
      "step": 43640
    },
    {
      "entropy": 5.336853694915772,
      "epoch": 3.3957206769111066,
      "grad_norm": 1.125,
      "learning_rate": 0.0003868400913579715,
      "loss": 4.9009,
      "mean_token_accuracy": 0.20887548625469207,
      "num_tokens": 75711038.0,
      "step": 43645
    },
    {
      "entropy": 5.404508018493653,
      "epoch": 3.3961097062828243,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038681604135819774,
      "loss": 4.9254,
      "mean_token_accuracy": 0.20015495717525483,
      "num_tokens": 75719649.0,
      "step": 43650
    },
    {
      "entropy": 5.412271451950073,
      "epoch": 3.396498735654542,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038679198966179353,
      "loss": 4.8944,
      "mean_token_accuracy": 0.208041712641716,
      "num_tokens": 75727880.0,
      "step": 43655
    },
    {
      "entropy": 5.4065046310424805,
      "epoch": 3.3968877650262597,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038676793626912375,
      "loss": 4.9336,
      "mean_token_accuracy": 0.20958970189094545,
      "num_tokens": 75736885.0,
      "step": 43660
    },
    {
      "entropy": 5.387508916854858,
      "epoch": 3.397276794397977,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038674388118055354,
      "loss": 4.9093,
      "mean_token_accuracy": 0.21266703307628632,
      "num_tokens": 75745922.0,
      "step": 43665
    },
    {
      "entropy": 5.340303182601929,
      "epoch": 3.3976658237696946,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003867198243964477,
      "loss": 4.7948,
      "mean_token_accuracy": 0.21244871318340303,
      "num_tokens": 75754526.0,
      "step": 43670
    },
    {
      "entropy": 5.391309404373169,
      "epoch": 3.3980548531414123,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003866957659171714,
      "loss": 4.9293,
      "mean_token_accuracy": 0.20475511103868485,
      "num_tokens": 75763154.0,
      "step": 43675
    },
    {
      "entropy": 5.3852311134338375,
      "epoch": 3.3984438825131296,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003866717057430896,
      "loss": 4.8895,
      "mean_token_accuracy": 0.21061914116144181,
      "num_tokens": 75772201.0,
      "step": 43680
    },
    {
      "entropy": 5.4077893733978275,
      "epoch": 3.3988329118848473,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003866476438745675,
      "loss": 4.886,
      "mean_token_accuracy": 0.20980178117752074,
      "num_tokens": 75780351.0,
      "step": 43685
    },
    {
      "entropy": 5.269053936004639,
      "epoch": 3.399221941256565,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038662358031197,
      "loss": 4.7327,
      "mean_token_accuracy": 0.22073321789503098,
      "num_tokens": 75788942.0,
      "step": 43690
    },
    {
      "entropy": 5.37186975479126,
      "epoch": 3.3996109706282827,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003865995150556624,
      "loss": 4.9684,
      "mean_token_accuracy": 0.20149261951446534,
      "num_tokens": 75797362.0,
      "step": 43695
    },
    {
      "entropy": 5.403136825561523,
      "epoch": 3.4,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038657544810600966,
      "loss": 4.8546,
      "mean_token_accuracy": 0.20576683580875396,
      "num_tokens": 75805839.0,
      "step": 43700
    },
    {
      "entropy": 5.277457571029663,
      "epoch": 3.4003890293717176,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003865513794633773,
      "loss": 4.7773,
      "mean_token_accuracy": 0.21069447696208954,
      "num_tokens": 75814544.0,
      "step": 43705
    },
    {
      "entropy": 5.276537799835205,
      "epoch": 3.4007780587434353,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003865273091281301,
      "loss": 4.7471,
      "mean_token_accuracy": 0.21929188519716264,
      "num_tokens": 75822906.0,
      "step": 43710
    },
    {
      "entropy": 5.386422061920166,
      "epoch": 3.4011670881151526,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003865032371006336,
      "loss": 4.8646,
      "mean_token_accuracy": 0.20433398187160492,
      "num_tokens": 75832115.0,
      "step": 43715
    },
    {
      "entropy": 5.313847303390503,
      "epoch": 3.4015561174868703,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000386479163381253,
      "loss": 4.7516,
      "mean_token_accuracy": 0.21593912690877914,
      "num_tokens": 75840920.0,
      "step": 43720
    },
    {
      "entropy": 5.350656795501709,
      "epoch": 3.401945146858588,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038645508797035345,
      "loss": 4.8931,
      "mean_token_accuracy": 0.2093177020549774,
      "num_tokens": 75849654.0,
      "step": 43725
    },
    {
      "entropy": 5.3759928226470945,
      "epoch": 3.402334176230305,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038643101086830036,
      "loss": 4.8666,
      "mean_token_accuracy": 0.20057980567216874,
      "num_tokens": 75857864.0,
      "step": 43730
    },
    {
      "entropy": 5.322567796707153,
      "epoch": 3.402723205602023,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000386406932075459,
      "loss": 4.8121,
      "mean_token_accuracy": 0.22517990320920944,
      "num_tokens": 75866426.0,
      "step": 43735
    },
    {
      "entropy": 5.385067892074585,
      "epoch": 3.4031122349737406,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003863828515921948,
      "loss": 4.905,
      "mean_token_accuracy": 0.21112716645002366,
      "num_tokens": 75875261.0,
      "step": 43740
    },
    {
      "entropy": 5.339459371566773,
      "epoch": 3.403501264345458,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003863587694188732,
      "loss": 4.8134,
      "mean_token_accuracy": 0.21746813952922822,
      "num_tokens": 75883675.0,
      "step": 43745
    },
    {
      "entropy": 5.2879527568817135,
      "epoch": 3.4038902937171756,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038633468555585943,
      "loss": 4.8155,
      "mean_token_accuracy": 0.20890948325395584,
      "num_tokens": 75892209.0,
      "step": 43750
    },
    {
      "entropy": 5.334087944030761,
      "epoch": 3.4042793230888932,
      "grad_norm": 1.125,
      "learning_rate": 0.0003863106000035191,
      "loss": 4.8923,
      "mean_token_accuracy": 0.2129145532846451,
      "num_tokens": 75899870.0,
      "step": 43755
    },
    {
      "entropy": 5.301849699020385,
      "epoch": 3.404668352460611,
      "grad_norm": 1.125,
      "learning_rate": 0.0003862865127622175,
      "loss": 4.8792,
      "mean_token_accuracy": 0.21175415366888045,
      "num_tokens": 75908616.0,
      "step": 43760
    },
    {
      "entropy": 5.361241436004638,
      "epoch": 3.405057381832328,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003862624238323201,
      "loss": 4.8844,
      "mean_token_accuracy": 0.20827001929283143,
      "num_tokens": 75917065.0,
      "step": 43765
    },
    {
      "entropy": 5.290961790084839,
      "epoch": 3.405446411204046,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003862383332141927,
      "loss": 4.786,
      "mean_token_accuracy": 0.22062335312366485,
      "num_tokens": 75925691.0,
      "step": 43770
    },
    {
      "entropy": 5.376216554641724,
      "epoch": 3.4058354405757636,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003862142409082006,
      "loss": 4.8993,
      "mean_token_accuracy": 0.2098408341407776,
      "num_tokens": 75934285.0,
      "step": 43775
    },
    {
      "entropy": 5.444793367385865,
      "epoch": 3.406224469947481,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003861901469147093,
      "loss": 4.9829,
      "mean_token_accuracy": 0.20457764565944672,
      "num_tokens": 75942861.0,
      "step": 43780
    },
    {
      "entropy": 5.381565856933594,
      "epoch": 3.4066134993191985,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038616605123408464,
      "loss": 4.9058,
      "mean_token_accuracy": 0.21265919208526612,
      "num_tokens": 75952172.0,
      "step": 43785
    },
    {
      "entropy": 5.302419662475586,
      "epoch": 3.4070025286909162,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038614195386669207,
      "loss": 4.8878,
      "mean_token_accuracy": 0.2124156951904297,
      "num_tokens": 75960973.0,
      "step": 43790
    },
    {
      "entropy": 5.303713417053222,
      "epoch": 3.407391558062634,
      "grad_norm": 1.125,
      "learning_rate": 0.0003861178548128972,
      "loss": 4.7643,
      "mean_token_accuracy": 0.2182506799697876,
      "num_tokens": 75969220.0,
      "step": 43795
    },
    {
      "entropy": 5.426542186737061,
      "epoch": 3.407780587434351,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038609375407306566,
      "loss": 4.9056,
      "mean_token_accuracy": 0.20692589282989501,
      "num_tokens": 75977345.0,
      "step": 43800
    },
    {
      "entropy": 5.3514509201049805,
      "epoch": 3.408169616806069,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038606965164756334,
      "loss": 4.8914,
      "mean_token_accuracy": 0.2058811902999878,
      "num_tokens": 75986500.0,
      "step": 43805
    },
    {
      "entropy": 5.332065677642822,
      "epoch": 3.4085586461777866,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038604554753675585,
      "loss": 4.8582,
      "mean_token_accuracy": 0.210355681180954,
      "num_tokens": 75995164.0,
      "step": 43810
    },
    {
      "entropy": 5.414210796356201,
      "epoch": 3.408947675549504,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003860214417410089,
      "loss": 4.9423,
      "mean_token_accuracy": 0.20678454637527466,
      "num_tokens": 76003927.0,
      "step": 43815
    },
    {
      "entropy": 5.408791589736938,
      "epoch": 3.4093367049212215,
      "grad_norm": 1.125,
      "learning_rate": 0.00038599733426068826,
      "loss": 4.9316,
      "mean_token_accuracy": 0.20415865927934645,
      "num_tokens": 76012703.0,
      "step": 43820
    },
    {
      "entropy": 5.419067621231079,
      "epoch": 3.4097257342929392,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038597322509615977,
      "loss": 4.8966,
      "mean_token_accuracy": 0.20355138629674913,
      "num_tokens": 76021218.0,
      "step": 43825
    },
    {
      "entropy": 5.3717231273651125,
      "epoch": 3.4101147636646565,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038594911424778925,
      "loss": 4.9255,
      "mean_token_accuracy": 0.20123552829027175,
      "num_tokens": 76030430.0,
      "step": 43830
    },
    {
      "entropy": 5.393952465057373,
      "epoch": 3.410503793036374,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038592500171594245,
      "loss": 4.9442,
      "mean_token_accuracy": 0.20689460635185242,
      "num_tokens": 76039201.0,
      "step": 43835
    },
    {
      "entropy": 5.391186618804932,
      "epoch": 3.410892822408092,
      "grad_norm": 1.125,
      "learning_rate": 0.00038590088750098536,
      "loss": 4.8806,
      "mean_token_accuracy": 0.21480803489685057,
      "num_tokens": 76048250.0,
      "step": 43840
    },
    {
      "entropy": 5.414554214477539,
      "epoch": 3.411281851779809,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038587677160328377,
      "loss": 4.9254,
      "mean_token_accuracy": 0.20182383805513382,
      "num_tokens": 76057056.0,
      "step": 43845
    },
    {
      "entropy": 5.379528617858886,
      "epoch": 3.411670881151527,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003858526540232037,
      "loss": 4.8457,
      "mean_token_accuracy": 0.2111973911523819,
      "num_tokens": 76065602.0,
      "step": 43850
    },
    {
      "entropy": 5.406962156295776,
      "epoch": 3.4120599105232445,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038582853476111094,
      "loss": 4.9914,
      "mean_token_accuracy": 0.20101846158504486,
      "num_tokens": 76073818.0,
      "step": 43855
    },
    {
      "entropy": 5.407097387313843,
      "epoch": 3.4124489398949622,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003858044138173717,
      "loss": 4.9389,
      "mean_token_accuracy": 0.20374141484498978,
      "num_tokens": 76082343.0,
      "step": 43860
    },
    {
      "entropy": 5.360495662689209,
      "epoch": 3.4128379692666795,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038578029119235176,
      "loss": 4.8442,
      "mean_token_accuracy": 0.21291319727897645,
      "num_tokens": 76090478.0,
      "step": 43865
    },
    {
      "entropy": 5.308180332183838,
      "epoch": 3.413226998638397,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038575616688641724,
      "loss": 4.8303,
      "mean_token_accuracy": 0.21565528064966202,
      "num_tokens": 76098901.0,
      "step": 43870
    },
    {
      "entropy": 5.31659722328186,
      "epoch": 3.413616028010115,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003857320408999342,
      "loss": 4.828,
      "mean_token_accuracy": 0.2142665281891823,
      "num_tokens": 76107281.0,
      "step": 43875
    },
    {
      "entropy": 5.341084384918213,
      "epoch": 3.414005057381832,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003857079132332687,
      "loss": 4.8661,
      "mean_token_accuracy": 0.2094491645693779,
      "num_tokens": 76116277.0,
      "step": 43880
    },
    {
      "entropy": 5.403334283828736,
      "epoch": 3.41439408675355,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038568378388678675,
      "loss": 4.9533,
      "mean_token_accuracy": 0.20402031391859055,
      "num_tokens": 76125412.0,
      "step": 43885
    },
    {
      "entropy": 5.34952335357666,
      "epoch": 3.4147831161252675,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003856596528608546,
      "loss": 4.8192,
      "mean_token_accuracy": 0.21202885657548903,
      "num_tokens": 76134259.0,
      "step": 43890
    },
    {
      "entropy": 5.263854837417602,
      "epoch": 3.4151721454969852,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003856355201558384,
      "loss": 4.7563,
      "mean_token_accuracy": 0.22410248816013337,
      "num_tokens": 76142557.0,
      "step": 43895
    },
    {
      "entropy": 5.36772985458374,
      "epoch": 3.4155611748687025,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038561138577210426,
      "loss": 4.905,
      "mean_token_accuracy": 0.21243343949317933,
      "num_tokens": 76150495.0,
      "step": 43900
    },
    {
      "entropy": 5.32286319732666,
      "epoch": 3.41595020424042,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038558724971001835,
      "loss": 4.7848,
      "mean_token_accuracy": 0.21454565972089767,
      "num_tokens": 76158729.0,
      "step": 43905
    },
    {
      "entropy": 5.322009038925171,
      "epoch": 3.416339233612138,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000385563111969947,
      "loss": 4.9004,
      "mean_token_accuracy": 0.2015802413225174,
      "num_tokens": 76167917.0,
      "step": 43910
    },
    {
      "entropy": 5.336364221572876,
      "epoch": 3.416728262983855,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038553897255225633,
      "loss": 4.8001,
      "mean_token_accuracy": 0.2147582694888115,
      "num_tokens": 76176645.0,
      "step": 43915
    },
    {
      "entropy": 5.282109975814819,
      "epoch": 3.417117292355573,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038551483145731277,
      "loss": 4.9173,
      "mean_token_accuracy": 0.210238716006279,
      "num_tokens": 76185344.0,
      "step": 43920
    },
    {
      "entropy": 5.38662633895874,
      "epoch": 3.4175063217272905,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003854906886854825,
      "loss": 4.8893,
      "mean_token_accuracy": 0.2091282159090042,
      "num_tokens": 76193303.0,
      "step": 43925
    },
    {
      "entropy": 5.412346315383911,
      "epoch": 3.417895351099008,
      "grad_norm": 1.125,
      "learning_rate": 0.00038546654423713187,
      "loss": 4.9523,
      "mean_token_accuracy": 0.1990896373987198,
      "num_tokens": 76202872.0,
      "step": 43930
    },
    {
      "entropy": 5.4019097805023195,
      "epoch": 3.4182843804707255,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003854423981126273,
      "loss": 4.9485,
      "mean_token_accuracy": 0.20305564999580383,
      "num_tokens": 76212763.0,
      "step": 43935
    },
    {
      "entropy": 5.415450572967529,
      "epoch": 3.418673409842443,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003854182503123351,
      "loss": 4.9399,
      "mean_token_accuracy": 0.20852164328098297,
      "num_tokens": 76221336.0,
      "step": 43940
    },
    {
      "entropy": 5.390273332595825,
      "epoch": 3.4190624392141604,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038539410083662166,
      "loss": 4.8513,
      "mean_token_accuracy": 0.21169913411140442,
      "num_tokens": 76229658.0,
      "step": 43945
    },
    {
      "entropy": 5.441069459915161,
      "epoch": 3.419451468585878,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038536994968585355,
      "loss": 5.0374,
      "mean_token_accuracy": 0.1966761603951454,
      "num_tokens": 76238745.0,
      "step": 43950
    },
    {
      "entropy": 5.433959007263184,
      "epoch": 3.419840497957596,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038534579686039705,
      "loss": 4.9045,
      "mean_token_accuracy": 0.20257362276315688,
      "num_tokens": 76247463.0,
      "step": 43955
    },
    {
      "entropy": 5.367196750640869,
      "epoch": 3.4202295273293135,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038532164236061874,
      "loss": 4.8625,
      "mean_token_accuracy": 0.21195569187402724,
      "num_tokens": 76256050.0,
      "step": 43960
    },
    {
      "entropy": 5.422271347045898,
      "epoch": 3.4206185567010308,
      "grad_norm": 1.1875,
      "learning_rate": 0.000385297486186885,
      "loss": 4.8558,
      "mean_token_accuracy": 0.21538924425840378,
      "num_tokens": 76264229.0,
      "step": 43965
    },
    {
      "entropy": 5.256823873519897,
      "epoch": 3.4210075860727485,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003852733283395626,
      "loss": 4.8458,
      "mean_token_accuracy": 0.2113883003592491,
      "num_tokens": 76273273.0,
      "step": 43970
    },
    {
      "entropy": 5.3312907218933105,
      "epoch": 3.421396615444466,
      "grad_norm": 1.078125,
      "learning_rate": 0.000385249168819018,
      "loss": 4.7292,
      "mean_token_accuracy": 0.21911280751228332,
      "num_tokens": 76281887.0,
      "step": 43975
    },
    {
      "entropy": 5.30733470916748,
      "epoch": 3.4217856448161834,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003852250076256177,
      "loss": 4.7626,
      "mean_token_accuracy": 0.21440667361021043,
      "num_tokens": 76290317.0,
      "step": 43980
    },
    {
      "entropy": 5.297487735748291,
      "epoch": 3.422174674187901,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003852008447597283,
      "loss": 4.8428,
      "mean_token_accuracy": 0.2138625428080559,
      "num_tokens": 76299016.0,
      "step": 43985
    },
    {
      "entropy": 5.339577913284302,
      "epoch": 3.422563703559619,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038517668022171656,
      "loss": 4.873,
      "mean_token_accuracy": 0.21290654987096785,
      "num_tokens": 76307723.0,
      "step": 43990
    },
    {
      "entropy": 5.408030319213867,
      "epoch": 3.4229527329313365,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003851525140119489,
      "loss": 4.9662,
      "mean_token_accuracy": 0.20986070483922958,
      "num_tokens": 76316615.0,
      "step": 43995
    },
    {
      "entropy": 5.407819032669067,
      "epoch": 3.4233417623030538,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038512834613079226,
      "loss": 4.8894,
      "mean_token_accuracy": 0.214308699965477,
      "num_tokens": 76325920.0,
      "step": 44000
    },
    {
      "entropy": 5.412186145782471,
      "epoch": 3.4237307916747715,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003851041765786133,
      "loss": 4.952,
      "mean_token_accuracy": 0.20249849408864976,
      "num_tokens": 76334223.0,
      "step": 44005
    },
    {
      "entropy": 5.386535787582398,
      "epoch": 3.424119821046489,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038508000535577864,
      "loss": 4.8367,
      "mean_token_accuracy": 0.21700337380170823,
      "num_tokens": 76342741.0,
      "step": 44010
    },
    {
      "entropy": 5.371453094482422,
      "epoch": 3.4245088504182064,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003850558324626552,
      "loss": 4.9244,
      "mean_token_accuracy": 0.20780442357063295,
      "num_tokens": 76351494.0,
      "step": 44015
    },
    {
      "entropy": 5.383586978912353,
      "epoch": 3.424897879789924,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003850316578996096,
      "loss": 4.9097,
      "mean_token_accuracy": 0.20596901029348375,
      "num_tokens": 76361005.0,
      "step": 44020
    },
    {
      "entropy": 5.447936153411865,
      "epoch": 3.425286909161642,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003850074816670087,
      "loss": 4.9616,
      "mean_token_accuracy": 0.20765435099601745,
      "num_tokens": 76369552.0,
      "step": 44025
    },
    {
      "entropy": 5.409248399734497,
      "epoch": 3.4256759385333595,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038498330376521945,
      "loss": 4.9897,
      "mean_token_accuracy": 0.20015438050031661,
      "num_tokens": 76378248.0,
      "step": 44030
    },
    {
      "entropy": 5.299365282058716,
      "epoch": 3.4260649679050768,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003849591241946085,
      "loss": 4.7968,
      "mean_token_accuracy": 0.21924690306186675,
      "num_tokens": 76386022.0,
      "step": 44035
    },
    {
      "entropy": 5.3207704544067385,
      "epoch": 3.4264539972767944,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000384934942955543,
      "loss": 4.8085,
      "mean_token_accuracy": 0.22133046239614487,
      "num_tokens": 76394278.0,
      "step": 44040
    },
    {
      "entropy": 5.379897737503052,
      "epoch": 3.426843026648512,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038491076004838965,
      "loss": 4.8641,
      "mean_token_accuracy": 0.21763150691986083,
      "num_tokens": 76402450.0,
      "step": 44045
    },
    {
      "entropy": 5.285750246047973,
      "epoch": 3.4272320560202294,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038488657547351544,
      "loss": 4.8063,
      "mean_token_accuracy": 0.2224019929766655,
      "num_tokens": 76410384.0,
      "step": 44050
    },
    {
      "entropy": 5.305353116989136,
      "epoch": 3.427621085391947,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003848623892312874,
      "loss": 4.8595,
      "mean_token_accuracy": 0.2090450718998909,
      "num_tokens": 76418739.0,
      "step": 44055
    },
    {
      "entropy": 5.3538408279418945,
      "epoch": 3.428010114763665,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003848382013220725,
      "loss": 4.8812,
      "mean_token_accuracy": 0.21959384828805922,
      "num_tokens": 76427704.0,
      "step": 44060
    },
    {
      "entropy": 5.367846870422364,
      "epoch": 3.428399144135382,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003848140117462376,
      "loss": 4.8321,
      "mean_token_accuracy": 0.2141658693552017,
      "num_tokens": 76436247.0,
      "step": 44065
    },
    {
      "entropy": 5.305167293548584,
      "epoch": 3.4287881735070997,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003847898205041499,
      "loss": 4.9086,
      "mean_token_accuracy": 0.21031559407711028,
      "num_tokens": 76445218.0,
      "step": 44070
    },
    {
      "entropy": 5.354399681091309,
      "epoch": 3.4291772028788174,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038476562759617644,
      "loss": 4.8023,
      "mean_token_accuracy": 0.21128046959638597,
      "num_tokens": 76454042.0,
      "step": 44075
    },
    {
      "entropy": 5.399473667144775,
      "epoch": 3.4295662322505347,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038474143302268425,
      "loss": 4.8985,
      "mean_token_accuracy": 0.20821607410907744,
      "num_tokens": 76462659.0,
      "step": 44080
    },
    {
      "entropy": 5.333794116973877,
      "epoch": 3.4299552616222524,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003847172367840406,
      "loss": 4.9034,
      "mean_token_accuracy": 0.21613748669624328,
      "num_tokens": 76471268.0,
      "step": 44085
    },
    {
      "entropy": 5.439135217666626,
      "epoch": 3.43034429099397,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003846930388806124,
      "loss": 4.9535,
      "mean_token_accuracy": 0.20176736414432525,
      "num_tokens": 76479914.0,
      "step": 44090
    },
    {
      "entropy": 5.45746922492981,
      "epoch": 3.430733320365688,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038466883931276703,
      "loss": 4.9611,
      "mean_token_accuracy": 0.19856687486171723,
      "num_tokens": 76488330.0,
      "step": 44095
    },
    {
      "entropy": 5.339880752563476,
      "epoch": 3.431122349737405,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038464463808087146,
      "loss": 4.8848,
      "mean_token_accuracy": 0.2101561412215233,
      "num_tokens": 76496827.0,
      "step": 44100
    },
    {
      "entropy": 5.304679489135742,
      "epoch": 3.4315113791091227,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003846204351852931,
      "loss": 4.8245,
      "mean_token_accuracy": 0.20712484717369078,
      "num_tokens": 76505220.0,
      "step": 44105
    },
    {
      "entropy": 5.27211709022522,
      "epoch": 3.4319004084808404,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038459623062639907,
      "loss": 4.7367,
      "mean_token_accuracy": 0.2240686759352684,
      "num_tokens": 76513823.0,
      "step": 44110
    },
    {
      "entropy": 5.402150440216064,
      "epoch": 3.4322894378525577,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038457202440455673,
      "loss": 4.8868,
      "mean_token_accuracy": 0.21330241560935975,
      "num_tokens": 76522648.0,
      "step": 44115
    },
    {
      "entropy": 5.353517293930054,
      "epoch": 3.4326784672242754,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003845478165201333,
      "loss": 4.8763,
      "mean_token_accuracy": 0.20407778918743133,
      "num_tokens": 76530288.0,
      "step": 44120
    },
    {
      "entropy": 5.281690454483032,
      "epoch": 3.433067496595993,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038452360697349615,
      "loss": 4.7612,
      "mean_token_accuracy": 0.21284128874540328,
      "num_tokens": 76538428.0,
      "step": 44125
    },
    {
      "entropy": 5.378219079971314,
      "epoch": 3.433456525967711,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003844993957650125,
      "loss": 4.8374,
      "mean_token_accuracy": 0.2128179579973221,
      "num_tokens": 76547396.0,
      "step": 44130
    },
    {
      "entropy": 5.3518470287322994,
      "epoch": 3.433845555339428,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003844751828950499,
      "loss": 4.8759,
      "mean_token_accuracy": 0.20738088190555573,
      "num_tokens": 76556163.0,
      "step": 44135
    },
    {
      "entropy": 5.358539819717407,
      "epoch": 3.4342345847111457,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038445096836397564,
      "loss": 4.8766,
      "mean_token_accuracy": 0.20768878906965255,
      "num_tokens": 76564717.0,
      "step": 44140
    },
    {
      "entropy": 5.324561309814453,
      "epoch": 3.4346236140828634,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038442675217215713,
      "loss": 4.8132,
      "mean_token_accuracy": 0.21306930780410765,
      "num_tokens": 76573049.0,
      "step": 44145
    },
    {
      "entropy": 5.371606016159058,
      "epoch": 3.4350126434545807,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038440253431996175,
      "loss": 4.8559,
      "mean_token_accuracy": 0.20912517309188844,
      "num_tokens": 76581897.0,
      "step": 44150
    },
    {
      "entropy": 5.355568313598633,
      "epoch": 3.4354016728262984,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038437831480775717,
      "loss": 4.9052,
      "mean_token_accuracy": 0.20379181653261186,
      "num_tokens": 76590443.0,
      "step": 44155
    },
    {
      "entropy": 5.285912561416626,
      "epoch": 3.435790702198016,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038435409363591066,
      "loss": 4.7599,
      "mean_token_accuracy": 0.2189452275633812,
      "num_tokens": 76598979.0,
      "step": 44160
    },
    {
      "entropy": 5.3162689208984375,
      "epoch": 3.4361797315697333,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038432987080478986,
      "loss": 4.9254,
      "mean_token_accuracy": 0.206170853972435,
      "num_tokens": 76607658.0,
      "step": 44165
    },
    {
      "entropy": 5.4633502006530765,
      "epoch": 3.436568760941451,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003843056463147623,
      "loss": 4.9604,
      "mean_token_accuracy": 0.2060238838195801,
      "num_tokens": 76615952.0,
      "step": 44170
    },
    {
      "entropy": 5.328668117523193,
      "epoch": 3.4369577903131687,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038428142016619555,
      "loss": 4.8353,
      "mean_token_accuracy": 0.2159118726849556,
      "num_tokens": 76624016.0,
      "step": 44175
    },
    {
      "entropy": 5.404518508911133,
      "epoch": 3.437346819684886,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003842571923594572,
      "loss": 4.9417,
      "mean_token_accuracy": 0.20634030252695085,
      "num_tokens": 76633394.0,
      "step": 44180
    },
    {
      "entropy": 5.361873197555542,
      "epoch": 3.4377358490566037,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038423296289491477,
      "loss": 4.7824,
      "mean_token_accuracy": 0.223851278424263,
      "num_tokens": 76641518.0,
      "step": 44185
    },
    {
      "entropy": 5.34814076423645,
      "epoch": 3.4381248784283214,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038420873177293606,
      "loss": 4.8992,
      "mean_token_accuracy": 0.20483697056770325,
      "num_tokens": 76649823.0,
      "step": 44190
    },
    {
      "entropy": 5.342047643661499,
      "epoch": 3.438513907800039,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003841844989938886,
      "loss": 4.9488,
      "mean_token_accuracy": 0.2027640163898468,
      "num_tokens": 76659704.0,
      "step": 44195
    },
    {
      "entropy": 5.448216295242309,
      "epoch": 3.4389029371717563,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003841602645581402,
      "loss": 4.8688,
      "mean_token_accuracy": 0.20437523275613784,
      "num_tokens": 76668205.0,
      "step": 44200
    },
    {
      "entropy": 5.440160512924194,
      "epoch": 3.439291966543474,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038413602846605854,
      "loss": 5.0047,
      "mean_token_accuracy": 0.20420601218938828,
      "num_tokens": 76677163.0,
      "step": 44205
    },
    {
      "entropy": 5.316384744644165,
      "epoch": 3.4396809959151917,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003841117907180114,
      "loss": 4.8804,
      "mean_token_accuracy": 0.20425205677747726,
      "num_tokens": 76685437.0,
      "step": 44210
    },
    {
      "entropy": 5.470270824432373,
      "epoch": 3.440070025286909,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038408755131436647,
      "loss": 4.9737,
      "mean_token_accuracy": 0.20573582947254182,
      "num_tokens": 76693648.0,
      "step": 44215
    },
    {
      "entropy": 5.347452783584595,
      "epoch": 3.4404590546586267,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038406331025549157,
      "loss": 4.7914,
      "mean_token_accuracy": 0.21655315309762954,
      "num_tokens": 76701790.0,
      "step": 44220
    },
    {
      "entropy": 5.431823062896728,
      "epoch": 3.4408480840303444,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003840390675417545,
      "loss": 5.0024,
      "mean_token_accuracy": 0.20095865428447723,
      "num_tokens": 76710696.0,
      "step": 44225
    },
    {
      "entropy": 5.414216136932373,
      "epoch": 3.441237113402062,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003840148231735231,
      "loss": 4.9461,
      "mean_token_accuracy": 0.21327468752861023,
      "num_tokens": 76718731.0,
      "step": 44230
    },
    {
      "entropy": 5.395946025848389,
      "epoch": 3.4416261427737793,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038399057715116536,
      "loss": 4.9879,
      "mean_token_accuracy": 0.1980130359530449,
      "num_tokens": 76727678.0,
      "step": 44235
    },
    {
      "entropy": 5.357743930816651,
      "epoch": 3.442015172145497,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038396632947504904,
      "loss": 4.8085,
      "mean_token_accuracy": 0.21749017238616944,
      "num_tokens": 76736605.0,
      "step": 44240
    },
    {
      "entropy": 5.414746570587158,
      "epoch": 3.4424042015172147,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003839420801455422,
      "loss": 4.9918,
      "mean_token_accuracy": 0.20089211612939833,
      "num_tokens": 76746050.0,
      "step": 44245
    },
    {
      "entropy": 5.398993110656738,
      "epoch": 3.442793230888932,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003839178291630126,
      "loss": 4.8678,
      "mean_token_accuracy": 0.21333519369363785,
      "num_tokens": 76754825.0,
      "step": 44250
    },
    {
      "entropy": 5.346442222595215,
      "epoch": 3.4431822602606497,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038389357652782836,
      "loss": 4.843,
      "mean_token_accuracy": 0.21309790015220642,
      "num_tokens": 76763761.0,
      "step": 44255
    },
    {
      "entropy": 5.351872348785401,
      "epoch": 3.4435712896323674,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038386932224035734,
      "loss": 4.8855,
      "mean_token_accuracy": 0.20804891735315323,
      "num_tokens": 76772187.0,
      "step": 44260
    },
    {
      "entropy": 5.331553506851196,
      "epoch": 3.443960319004085,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038384506630096765,
      "loss": 4.7915,
      "mean_token_accuracy": 0.21334637552499772,
      "num_tokens": 76780722.0,
      "step": 44265
    },
    {
      "entropy": 5.360098361968994,
      "epoch": 3.4443493483758023,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003838208087100274,
      "loss": 4.8726,
      "mean_token_accuracy": 0.2041327401995659,
      "num_tokens": 76789752.0,
      "step": 44270
    },
    {
      "entropy": 5.434886169433594,
      "epoch": 3.44473837774752,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003837965494679045,
      "loss": 4.8631,
      "mean_token_accuracy": 0.21674614250659943,
      "num_tokens": 76798052.0,
      "step": 44275
    },
    {
      "entropy": 5.326298332214355,
      "epoch": 3.4451274071192373,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003837722885749672,
      "loss": 4.7856,
      "mean_token_accuracy": 0.2141245499253273,
      "num_tokens": 76806543.0,
      "step": 44280
    },
    {
      "entropy": 5.3487707614898685,
      "epoch": 3.445516436490955,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003837480260315835,
      "loss": 4.8736,
      "mean_token_accuracy": 0.21056022197008134,
      "num_tokens": 76815316.0,
      "step": 44285
    },
    {
      "entropy": 5.264605617523193,
      "epoch": 3.4459054658626727,
      "grad_norm": 1.125,
      "learning_rate": 0.00038372376183812165,
      "loss": 4.7139,
      "mean_token_accuracy": 0.21785422265529633,
      "num_tokens": 76823826.0,
      "step": 44290
    },
    {
      "entropy": 5.356287622451783,
      "epoch": 3.4462944952343904,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003836994959949497,
      "loss": 4.9122,
      "mean_token_accuracy": 0.20722005069255828,
      "num_tokens": 76832225.0,
      "step": 44295
    },
    {
      "entropy": 5.395603227615356,
      "epoch": 3.4466835246061076,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038367522850243593,
      "loss": 4.9581,
      "mean_token_accuracy": 0.20258038938045503,
      "num_tokens": 76839732.0,
      "step": 44300
    },
    {
      "entropy": 5.275016450881958,
      "epoch": 3.4470725539778253,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038365095936094857,
      "loss": 4.7782,
      "mean_token_accuracy": 0.21891596615314485,
      "num_tokens": 76848041.0,
      "step": 44305
    },
    {
      "entropy": 5.375964069366455,
      "epoch": 3.447461583349543,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003836266885708558,
      "loss": 4.8697,
      "mean_token_accuracy": 0.2147322416305542,
      "num_tokens": 76856869.0,
      "step": 44310
    },
    {
      "entropy": 5.427698564529419,
      "epoch": 3.4478506127212603,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038360241613252593,
      "loss": 4.9496,
      "mean_token_accuracy": 0.20214297920465468,
      "num_tokens": 76865309.0,
      "step": 44315
    },
    {
      "entropy": 5.340597915649414,
      "epoch": 3.448239642092978,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003835781420463273,
      "loss": 4.8438,
      "mean_token_accuracy": 0.21674845218658448,
      "num_tokens": 76873300.0,
      "step": 44320
    },
    {
      "entropy": 5.372597551345825,
      "epoch": 3.4486286714646956,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003835538663126282,
      "loss": 4.8529,
      "mean_token_accuracy": 0.2088318571448326,
      "num_tokens": 76881662.0,
      "step": 44325
    },
    {
      "entropy": 5.371818161010742,
      "epoch": 3.4490177008364133,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003835295889317969,
      "loss": 4.8236,
      "mean_token_accuracy": 0.2105466514825821,
      "num_tokens": 76890537.0,
      "step": 44330
    },
    {
      "entropy": 5.3207779884338375,
      "epoch": 3.4494067302081306,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038350530990420177,
      "loss": 4.8683,
      "mean_token_accuracy": 0.2131390690803528,
      "num_tokens": 76898976.0,
      "step": 44335
    },
    {
      "entropy": 5.351888847351074,
      "epoch": 3.4497957595798483,
      "grad_norm": 1.25,
      "learning_rate": 0.0003834810292302114,
      "loss": 4.8409,
      "mean_token_accuracy": 0.21886058896780014,
      "num_tokens": 76907094.0,
      "step": 44340
    },
    {
      "entropy": 5.370271825790406,
      "epoch": 3.450184788951566,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000383456746910194,
      "loss": 4.8993,
      "mean_token_accuracy": 0.21064492911100388,
      "num_tokens": 76916324.0,
      "step": 44345
    },
    {
      "entropy": 5.300068616867065,
      "epoch": 3.4505738183232832,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003834324629445182,
      "loss": 4.8059,
      "mean_token_accuracy": 0.2198035180568695,
      "num_tokens": 76925165.0,
      "step": 44350
    },
    {
      "entropy": 5.3317633152008055,
      "epoch": 3.450962847695001,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003834081773335522,
      "loss": 4.858,
      "mean_token_accuracy": 0.2146083101630211,
      "num_tokens": 76933317.0,
      "step": 44355
    },
    {
      "entropy": 5.369013261795044,
      "epoch": 3.4513518770667186,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038338389007766473,
      "loss": 4.883,
      "mean_token_accuracy": 0.20833025723695756,
      "num_tokens": 76942353.0,
      "step": 44360
    },
    {
      "entropy": 5.314143800735474,
      "epoch": 3.4517409064384363,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003833596011772242,
      "loss": 4.8558,
      "mean_token_accuracy": 0.2158580645918846,
      "num_tokens": 76950844.0,
      "step": 44365
    },
    {
      "entropy": 5.30028977394104,
      "epoch": 3.4521299358101536,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003833353106325993,
      "loss": 4.8608,
      "mean_token_accuracy": 0.2134937345981598,
      "num_tokens": 76959159.0,
      "step": 44370
    },
    {
      "entropy": 5.253014707565308,
      "epoch": 3.4525189651818713,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003833110184441584,
      "loss": 4.8389,
      "mean_token_accuracy": 0.21132627725601197,
      "num_tokens": 76967287.0,
      "step": 44375
    },
    {
      "entropy": 5.350424480438233,
      "epoch": 3.452907994553589,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038328672461227024,
      "loss": 4.9009,
      "mean_token_accuracy": 0.21104328334331512,
      "num_tokens": 76975984.0,
      "step": 44380
    },
    {
      "entropy": 5.389069938659668,
      "epoch": 3.4532970239253062,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003832624291373034,
      "loss": 4.9153,
      "mean_token_accuracy": 0.20550612211227418,
      "num_tokens": 76985045.0,
      "step": 44385
    },
    {
      "entropy": 5.4156256198883055,
      "epoch": 3.453686053297024,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038323813201962645,
      "loss": 4.9022,
      "mean_token_accuracy": 0.20268590599298478,
      "num_tokens": 76993221.0,
      "step": 44390
    },
    {
      "entropy": 5.367113351821899,
      "epoch": 3.4540750826687416,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038321383325960816,
      "loss": 4.9018,
      "mean_token_accuracy": 0.2048519417643547,
      "num_tokens": 77002053.0,
      "step": 44395
    },
    {
      "entropy": 5.329811763763428,
      "epoch": 3.454464112040459,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003831895328576172,
      "loss": 4.8226,
      "mean_token_accuracy": 0.20851605981588364,
      "num_tokens": 77011113.0,
      "step": 44400
    },
    {
      "entropy": 5.355357074737549,
      "epoch": 3.4548531414121766,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003831652308140223,
      "loss": 4.8814,
      "mean_token_accuracy": 0.20771587789058685,
      "num_tokens": 77019975.0,
      "step": 44405
    },
    {
      "entropy": 5.285380220413208,
      "epoch": 3.4552421707838943,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003831409271291921,
      "loss": 4.8217,
      "mean_token_accuracy": 0.21427868753671647,
      "num_tokens": 77027854.0,
      "step": 44410
    },
    {
      "entropy": 5.214561939239502,
      "epoch": 3.4556312001556115,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038311662180349554,
      "loss": 4.7628,
      "mean_token_accuracy": 0.22265122532844545,
      "num_tokens": 77036342.0,
      "step": 44415
    },
    {
      "entropy": 5.290084266662598,
      "epoch": 3.4560202295273292,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038309231483730133,
      "loss": 4.7542,
      "mean_token_accuracy": 0.21736222356557847,
      "num_tokens": 77044403.0,
      "step": 44420
    },
    {
      "entropy": 5.414895391464233,
      "epoch": 3.456409258899047,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003830680062309782,
      "loss": 4.9067,
      "mean_token_accuracy": 0.21002959012985228,
      "num_tokens": 77053096.0,
      "step": 44425
    },
    {
      "entropy": 5.354868459701538,
      "epoch": 3.4567982882707646,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038304369598489516,
      "loss": 4.849,
      "mean_token_accuracy": 0.2119813472032547,
      "num_tokens": 77061902.0,
      "step": 44430
    },
    {
      "entropy": 5.378109312057495,
      "epoch": 3.457187317642482,
      "grad_norm": 1.15625,
      "learning_rate": 0.000383019384099421,
      "loss": 4.864,
      "mean_token_accuracy": 0.2117719903588295,
      "num_tokens": 77070414.0,
      "step": 44435
    },
    {
      "entropy": 5.375939512252808,
      "epoch": 3.4575763470141996,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003829950705749246,
      "loss": 4.9414,
      "mean_token_accuracy": 0.20815983414649963,
      "num_tokens": 77078799.0,
      "step": 44440
    },
    {
      "entropy": 5.439076328277588,
      "epoch": 3.4579653763859173,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038297075541177496,
      "loss": 4.9312,
      "mean_token_accuracy": 0.20850176811218263,
      "num_tokens": 77088143.0,
      "step": 44445
    },
    {
      "entropy": 5.363088464736938,
      "epoch": 3.4583544057576345,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038294643861034087,
      "loss": 4.8744,
      "mean_token_accuracy": 0.20764935463666917,
      "num_tokens": 77097566.0,
      "step": 44450
    },
    {
      "entropy": 5.349290752410889,
      "epoch": 3.4587434351293522,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003829221201709914,
      "loss": 4.7997,
      "mean_token_accuracy": 0.21626420468091964,
      "num_tokens": 77105846.0,
      "step": 44455
    },
    {
      "entropy": 5.2779652118682865,
      "epoch": 3.45913246450107,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003828978000940955,
      "loss": 4.8362,
      "mean_token_accuracy": 0.21589682698249818,
      "num_tokens": 77114661.0,
      "step": 44460
    },
    {
      "entropy": 5.3341529846191404,
      "epoch": 3.4595214938727876,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003828734783800223,
      "loss": 4.8332,
      "mean_token_accuracy": 0.21051027327775956,
      "num_tokens": 77122918.0,
      "step": 44465
    },
    {
      "entropy": 5.406696510314942,
      "epoch": 3.459910523244505,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003828491550291408,
      "loss": 4.8567,
      "mean_token_accuracy": 0.2100216567516327,
      "num_tokens": 77132480.0,
      "step": 44470
    },
    {
      "entropy": 5.360545206069946,
      "epoch": 3.4602995526162226,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003828248300418199,
      "loss": 4.8638,
      "mean_token_accuracy": 0.20658172070980071,
      "num_tokens": 77141277.0,
      "step": 44475
    },
    {
      "entropy": 5.416366147994995,
      "epoch": 3.4606885819879403,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038280050341842894,
      "loss": 4.9239,
      "mean_token_accuracy": 0.20729179829359054,
      "num_tokens": 77149818.0,
      "step": 44480
    },
    {
      "entropy": 5.400912618637085,
      "epoch": 3.4610776113596575,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003827761751593369,
      "loss": 4.8589,
      "mean_token_accuracy": 0.20934441834688186,
      "num_tokens": 77158368.0,
      "step": 44485
    },
    {
      "entropy": 5.368451452255249,
      "epoch": 3.461466640731375,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003827518452649129,
      "loss": 4.9491,
      "mean_token_accuracy": 0.20282781571149827,
      "num_tokens": 77167538.0,
      "step": 44490
    },
    {
      "entropy": 5.379152631759643,
      "epoch": 3.461855670103093,
      "grad_norm": 1.125,
      "learning_rate": 0.0003827275137355261,
      "loss": 4.8472,
      "mean_token_accuracy": 0.2111928105354309,
      "num_tokens": 77175920.0,
      "step": 44495
    },
    {
      "entropy": 5.409280109405517,
      "epoch": 3.46224469947481,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003827031805715458,
      "loss": 5.0016,
      "mean_token_accuracy": 0.1978055626153946,
      "num_tokens": 77184297.0,
      "step": 44500
    },
    {
      "entropy": 5.327898597717285,
      "epoch": 3.462633728846528,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038267884577334124,
      "loss": 4.8266,
      "mean_token_accuracy": 0.2100634142756462,
      "num_tokens": 77192651.0,
      "step": 44505
    },
    {
      "entropy": 5.315609502792358,
      "epoch": 3.4630227582182456,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003826545093412815,
      "loss": 4.7488,
      "mean_token_accuracy": 0.21980190873146058,
      "num_tokens": 77200997.0,
      "step": 44510
    },
    {
      "entropy": 5.357059717178345,
      "epoch": 3.463411787589963,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038263017127573596,
      "loss": 4.8287,
      "mean_token_accuracy": 0.2075844958424568,
      "num_tokens": 77209080.0,
      "step": 44515
    },
    {
      "entropy": 5.330939435958863,
      "epoch": 3.4638008169616805,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003826058315770739,
      "loss": 4.8523,
      "mean_token_accuracy": 0.20976097583770753,
      "num_tokens": 77219167.0,
      "step": 44520
    },
    {
      "entropy": 5.296736001968384,
      "epoch": 3.464189846333398,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038258149024566465,
      "loss": 4.821,
      "mean_token_accuracy": 0.20952046811580657,
      "num_tokens": 77228099.0,
      "step": 44525
    },
    {
      "entropy": 5.38392128944397,
      "epoch": 3.464578875705116,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003825571472818774,
      "loss": 4.908,
      "mean_token_accuracy": 0.20585681349039078,
      "num_tokens": 77237384.0,
      "step": 44530
    },
    {
      "entropy": 5.344997787475586,
      "epoch": 3.464967905076833,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003825328026860817,
      "loss": 4.7637,
      "mean_token_accuracy": 0.21872926950454713,
      "num_tokens": 77245443.0,
      "step": 44535
    },
    {
      "entropy": 5.286933660507202,
      "epoch": 3.465356934448551,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038250845645864686,
      "loss": 4.7619,
      "mean_token_accuracy": 0.22390384078025818,
      "num_tokens": 77253193.0,
      "step": 44540
    },
    {
      "entropy": 5.326523733139038,
      "epoch": 3.4657459638202686,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003824841085999423,
      "loss": 4.9206,
      "mean_token_accuracy": 0.20267895013093948,
      "num_tokens": 77263358.0,
      "step": 44545
    },
    {
      "entropy": 5.364959955215454,
      "epoch": 3.466134993191986,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003824597591103375,
      "loss": 4.9328,
      "mean_token_accuracy": 0.20369053930044173,
      "num_tokens": 77271996.0,
      "step": 44550
    },
    {
      "entropy": 5.3641682147979735,
      "epoch": 3.4665240225637035,
      "grad_norm": 1.125,
      "learning_rate": 0.00038243540799020193,
      "loss": 4.7485,
      "mean_token_accuracy": 0.22252302765846252,
      "num_tokens": 77280795.0,
      "step": 44555
    },
    {
      "entropy": 5.345756340026855,
      "epoch": 3.466913051935421,
      "grad_norm": 1.125,
      "learning_rate": 0.00038241105523990497,
      "loss": 4.8566,
      "mean_token_accuracy": 0.21098969131708145,
      "num_tokens": 77289191.0,
      "step": 44560
    },
    {
      "entropy": 5.346543312072754,
      "epoch": 3.467302081307139,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003823867008598163,
      "loss": 4.9015,
      "mean_token_accuracy": 0.20031464397907256,
      "num_tokens": 77298022.0,
      "step": 44565
    },
    {
      "entropy": 5.335916519165039,
      "epoch": 3.467691110678856,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038236234485030524,
      "loss": 4.8803,
      "mean_token_accuracy": 0.20301207304000854,
      "num_tokens": 77307145.0,
      "step": 44570
    },
    {
      "entropy": 5.340648365020752,
      "epoch": 3.468080140050574,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038233798721174156,
      "loss": 4.8755,
      "mean_token_accuracy": 0.21623898148536683,
      "num_tokens": 77316694.0,
      "step": 44575
    },
    {
      "entropy": 5.415460872650146,
      "epoch": 3.4684691694222916,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003823136279444948,
      "loss": 4.9778,
      "mean_token_accuracy": 0.20616236329078674,
      "num_tokens": 77325072.0,
      "step": 44580
    },
    {
      "entropy": 5.3838294506072994,
      "epoch": 3.468858198794009,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003822892670489345,
      "loss": 4.9072,
      "mean_token_accuracy": 0.21152201294898987,
      "num_tokens": 77333823.0,
      "step": 44585
    },
    {
      "entropy": 5.400976610183716,
      "epoch": 3.4692472281657265,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038226490452543036,
      "loss": 4.9007,
      "mean_token_accuracy": 0.20008691847324372,
      "num_tokens": 77342753.0,
      "step": 44590
    },
    {
      "entropy": 5.312636470794677,
      "epoch": 3.469636257537444,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000382240540374352,
      "loss": 4.8396,
      "mean_token_accuracy": 0.21337733566761016,
      "num_tokens": 77351345.0,
      "step": 44595
    },
    {
      "entropy": 5.44354944229126,
      "epoch": 3.4700252869091615,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003822161745960691,
      "loss": 4.9677,
      "mean_token_accuracy": 0.2047753304243088,
      "num_tokens": 77360144.0,
      "step": 44600
    },
    {
      "entropy": 5.4501060962677,
      "epoch": 3.470414316280879,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003821918071909515,
      "loss": 4.9629,
      "mean_token_accuracy": 0.20604571998119353,
      "num_tokens": 77368821.0,
      "step": 44605
    },
    {
      "entropy": 5.381224298477173,
      "epoch": 3.470803345652597,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003821674381593687,
      "loss": 4.8847,
      "mean_token_accuracy": 0.21045186817646028,
      "num_tokens": 77377904.0,
      "step": 44610
    },
    {
      "entropy": 5.426573944091797,
      "epoch": 3.471192375024314,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003821430675016907,
      "loss": 4.9995,
      "mean_token_accuracy": 0.198626746237278,
      "num_tokens": 77387335.0,
      "step": 44615
    },
    {
      "entropy": 5.353687334060669,
      "epoch": 3.471581404396032,
      "grad_norm": 1.125,
      "learning_rate": 0.00038211869521828716,
      "loss": 4.9188,
      "mean_token_accuracy": 0.20529052317142488,
      "num_tokens": 77396783.0,
      "step": 44620
    },
    {
      "entropy": 5.4907050132751465,
      "epoch": 3.4719704337677495,
      "grad_norm": 1.140625,
      "learning_rate": 0.000382094321309528,
      "loss": 5.034,
      "mean_token_accuracy": 0.20266800224781037,
      "num_tokens": 77406035.0,
      "step": 44625
    },
    {
      "entropy": 5.433233261108398,
      "epoch": 3.472359463139467,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003820699457757829,
      "loss": 4.8509,
      "mean_token_accuracy": 0.2108508139848709,
      "num_tokens": 77415064.0,
      "step": 44630
    },
    {
      "entropy": 5.355993461608887,
      "epoch": 3.4727484925111844,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003820455686174218,
      "loss": 4.8194,
      "mean_token_accuracy": 0.2112145647406578,
      "num_tokens": 77423001.0,
      "step": 44635
    },
    {
      "entropy": 5.415927124023438,
      "epoch": 3.473137521882902,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038202118983481456,
      "loss": 4.9437,
      "mean_token_accuracy": 0.20275495052337647,
      "num_tokens": 77431520.0,
      "step": 44640
    },
    {
      "entropy": 5.359023952484131,
      "epoch": 3.47352655125462,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003819968094283311,
      "loss": 4.925,
      "mean_token_accuracy": 0.20460680425167083,
      "num_tokens": 77439476.0,
      "step": 44645
    },
    {
      "entropy": 5.477322769165039,
      "epoch": 3.473915580626337,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003819724273983414,
      "loss": 4.9597,
      "mean_token_accuracy": 0.20322203636169434,
      "num_tokens": 77448466.0,
      "step": 44650
    },
    {
      "entropy": 5.3285393714904785,
      "epoch": 3.474304609998055,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038194804374521536,
      "loss": 4.8369,
      "mean_token_accuracy": 0.2219265356659889,
      "num_tokens": 77457858.0,
      "step": 44655
    },
    {
      "entropy": 5.342967319488525,
      "epoch": 3.4746936393697725,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003819236584693229,
      "loss": 4.8732,
      "mean_token_accuracy": 0.20543833523988725,
      "num_tokens": 77467308.0,
      "step": 44660
    },
    {
      "entropy": 5.305013704299927,
      "epoch": 3.47508266874149,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038189927157103427,
      "loss": 4.8238,
      "mean_token_accuracy": 0.20706083476543427,
      "num_tokens": 77476227.0,
      "step": 44665
    },
    {
      "entropy": 5.364732694625855,
      "epoch": 3.4754716981132074,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003818748830507192,
      "loss": 4.8111,
      "mean_token_accuracy": 0.21035706549882888,
      "num_tokens": 77484915.0,
      "step": 44670
    },
    {
      "entropy": 5.4304746150970455,
      "epoch": 3.475860727484925,
      "grad_norm": 1.109375,
      "learning_rate": 0.000381850492908748,
      "loss": 4.8901,
      "mean_token_accuracy": 0.2073024183511734,
      "num_tokens": 77493638.0,
      "step": 44675
    },
    {
      "entropy": 5.394555473327637,
      "epoch": 3.476249756856643,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038182610114549057,
      "loss": 4.8972,
      "mean_token_accuracy": 0.20854175835847855,
      "num_tokens": 77502834.0,
      "step": 44680
    },
    {
      "entropy": 5.366734266281128,
      "epoch": 3.47663878622836,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038180170776131714,
      "loss": 4.8438,
      "mean_token_accuracy": 0.2094848319888115,
      "num_tokens": 77511193.0,
      "step": 44685
    },
    {
      "entropy": 5.289023208618164,
      "epoch": 3.477027815600078,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003817773127565977,
      "loss": 4.7382,
      "mean_token_accuracy": 0.22577516585588456,
      "num_tokens": 77519839.0,
      "step": 44690
    },
    {
      "entropy": 5.35586109161377,
      "epoch": 3.4774168449717955,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038175291613170247,
      "loss": 4.8915,
      "mean_token_accuracy": 0.20742276906967164,
      "num_tokens": 77528581.0,
      "step": 44695
    },
    {
      "entropy": 5.394116163253784,
      "epoch": 3.477805874343513,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038172851788700174,
      "loss": 4.936,
      "mean_token_accuracy": 0.20792502015829087,
      "num_tokens": 77536426.0,
      "step": 44700
    },
    {
      "entropy": 5.369876050949097,
      "epoch": 3.4781949037152304,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003817041180228657,
      "loss": 4.8822,
      "mean_token_accuracy": 0.20722484439611435,
      "num_tokens": 77544785.0,
      "step": 44705
    },
    {
      "entropy": 5.384251356124878,
      "epoch": 3.478583933086948,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003816797165396643,
      "loss": 4.8831,
      "mean_token_accuracy": 0.21001382619142533,
      "num_tokens": 77554731.0,
      "step": 44710
    },
    {
      "entropy": 5.346391487121582,
      "epoch": 3.4789729624586654,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038165531343776815,
      "loss": 4.7432,
      "mean_token_accuracy": 0.21466246545314788,
      "num_tokens": 77562953.0,
      "step": 44715
    },
    {
      "entropy": 5.369436120986938,
      "epoch": 3.479361991830383,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003816309087175474,
      "loss": 4.8756,
      "mean_token_accuracy": 0.21494832187891005,
      "num_tokens": 77571740.0,
      "step": 44720
    },
    {
      "entropy": 5.3778684616088865,
      "epoch": 3.479751021202101,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038160650237937225,
      "loss": 4.9137,
      "mean_token_accuracy": 0.21087800562381745,
      "num_tokens": 77580160.0,
      "step": 44725
    },
    {
      "entropy": 5.338549613952637,
      "epoch": 3.4801400505738185,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003815820944236131,
      "loss": 4.8461,
      "mean_token_accuracy": 0.21498069763183594,
      "num_tokens": 77588562.0,
      "step": 44730
    },
    {
      "entropy": 5.30530276298523,
      "epoch": 3.4805290799455357,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038155768485064037,
      "loss": 4.7766,
      "mean_token_accuracy": 0.2216700568795204,
      "num_tokens": 77596865.0,
      "step": 44735
    },
    {
      "entropy": 5.374427461624146,
      "epoch": 3.4809181093172534,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038153327366082434,
      "loss": 4.9341,
      "mean_token_accuracy": 0.2070258677005768,
      "num_tokens": 77604976.0,
      "step": 44740
    },
    {
      "entropy": 5.43536148071289,
      "epoch": 3.481307138688971,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038150886085453546,
      "loss": 4.9497,
      "mean_token_accuracy": 0.2060566321015358,
      "num_tokens": 77614216.0,
      "step": 44745
    },
    {
      "entropy": 5.36468472480774,
      "epoch": 3.4816961680606884,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038148444643214415,
      "loss": 4.879,
      "mean_token_accuracy": 0.21005220264196395,
      "num_tokens": 77622294.0,
      "step": 44750
    },
    {
      "entropy": 5.4450806140899655,
      "epoch": 3.482085197432406,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003814600303940208,
      "loss": 4.9136,
      "mean_token_accuracy": 0.20356812328100204,
      "num_tokens": 77630231.0,
      "step": 44755
    },
    {
      "entropy": 5.31572322845459,
      "epoch": 3.4824742268041238,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000381435612740536,
      "loss": 4.7829,
      "mean_token_accuracy": 0.21263141334056854,
      "num_tokens": 77638841.0,
      "step": 44760
    },
    {
      "entropy": 5.368469476699829,
      "epoch": 3.4828632561758415,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038141119347206016,
      "loss": 4.8853,
      "mean_token_accuracy": 0.21435127407312393,
      "num_tokens": 77647487.0,
      "step": 44765
    },
    {
      "entropy": 5.446864175796509,
      "epoch": 3.4832522855475587,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038138677258896383,
      "loss": 4.8986,
      "mean_token_accuracy": 0.2096485748887062,
      "num_tokens": 77655882.0,
      "step": 44770
    },
    {
      "entropy": 5.295647239685058,
      "epoch": 3.4836413149192764,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003813623500916176,
      "loss": 4.7546,
      "mean_token_accuracy": 0.2160818487405777,
      "num_tokens": 77664919.0,
      "step": 44775
    },
    {
      "entropy": 5.351758670806885,
      "epoch": 3.484030344290994,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038133792598039196,
      "loss": 4.8636,
      "mean_token_accuracy": 0.21550663113594054,
      "num_tokens": 77673848.0,
      "step": 44780
    },
    {
      "entropy": 5.39409441947937,
      "epoch": 3.4844193736627114,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003813135002556575,
      "loss": 4.8873,
      "mean_token_accuracy": 0.20722482949495316,
      "num_tokens": 77682352.0,
      "step": 44785
    },
    {
      "entropy": 5.415047121047974,
      "epoch": 3.484808403034429,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038128907291778497,
      "loss": 4.8774,
      "mean_token_accuracy": 0.20512043982744216,
      "num_tokens": 77690347.0,
      "step": 44790
    },
    {
      "entropy": 5.314858865737915,
      "epoch": 3.4851974324061468,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038126464396714494,
      "loss": 4.8041,
      "mean_token_accuracy": 0.21791022419929504,
      "num_tokens": 77699172.0,
      "step": 44795
    },
    {
      "entropy": 5.427189540863037,
      "epoch": 3.4855864617778645,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038124021340410795,
      "loss": 4.89,
      "mean_token_accuracy": 0.20650639086961747,
      "num_tokens": 77707790.0,
      "step": 44800
    },
    {
      "entropy": 5.382651233673096,
      "epoch": 3.4859754911495817,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038121578122904496,
      "loss": 4.863,
      "mean_token_accuracy": 0.2135154351592064,
      "num_tokens": 77716602.0,
      "step": 44805
    },
    {
      "entropy": 5.355307960510254,
      "epoch": 3.4863645205212994,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003811913474423265,
      "loss": 4.7422,
      "mean_token_accuracy": 0.22911380976438522,
      "num_tokens": 77725999.0,
      "step": 44810
    },
    {
      "entropy": 5.420509958267212,
      "epoch": 3.486753549893017,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003811669120443234,
      "loss": 4.9,
      "mean_token_accuracy": 0.20644643157720566,
      "num_tokens": 77733829.0,
      "step": 44815
    },
    {
      "entropy": 5.409259414672851,
      "epoch": 3.4871425792647344,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003811424750354063,
      "loss": 4.856,
      "mean_token_accuracy": 0.21237395852804183,
      "num_tokens": 77742481.0,
      "step": 44820
    },
    {
      "entropy": 5.311704301834107,
      "epoch": 3.487531608636452,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003811180364159462,
      "loss": 4.7631,
      "mean_token_accuracy": 0.21622443050146103,
      "num_tokens": 77751368.0,
      "step": 44825
    },
    {
      "entropy": 5.321013069152832,
      "epoch": 3.4879206380081698,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038109359618631374,
      "loss": 4.8164,
      "mean_token_accuracy": 0.22115306556224823,
      "num_tokens": 77759134.0,
      "step": 44830
    },
    {
      "entropy": 5.3124878883361815,
      "epoch": 3.488309667379887,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038106915434687983,
      "loss": 4.8117,
      "mean_token_accuracy": 0.2034165754914284,
      "num_tokens": 77767337.0,
      "step": 44835
    },
    {
      "entropy": 5.316043996810913,
      "epoch": 3.4886986967516047,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003810447108980153,
      "loss": 4.8167,
      "mean_token_accuracy": 0.2115349844098091,
      "num_tokens": 77776296.0,
      "step": 44840
    },
    {
      "entropy": 5.410290813446045,
      "epoch": 3.4890877261233224,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003810202658400911,
      "loss": 4.9642,
      "mean_token_accuracy": 0.1983846366405487,
      "num_tokens": 77785252.0,
      "step": 44845
    },
    {
      "entropy": 5.4877434253692625,
      "epoch": 3.4894767554950397,
      "grad_norm": 1.125,
      "learning_rate": 0.0003809958191734781,
      "loss": 4.9434,
      "mean_token_accuracy": 0.2040013000369072,
      "num_tokens": 77793971.0,
      "step": 44850
    },
    {
      "entropy": 5.3847955703735355,
      "epoch": 3.4898657848667574,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038097137089854725,
      "loss": 4.8244,
      "mean_token_accuracy": 0.20978716164827346,
      "num_tokens": 77803060.0,
      "step": 44855
    },
    {
      "entropy": 5.278874158859253,
      "epoch": 3.490254814238475,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038094692101566955,
      "loss": 4.8236,
      "mean_token_accuracy": 0.22411368042230606,
      "num_tokens": 77811046.0,
      "step": 44860
    },
    {
      "entropy": 5.326783895492554,
      "epoch": 3.4906438436101928,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003809224695252159,
      "loss": 4.8372,
      "mean_token_accuracy": 0.21200261861085892,
      "num_tokens": 77819725.0,
      "step": 44865
    },
    {
      "entropy": 5.451242351531983,
      "epoch": 3.49103287298191,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038089801642755746,
      "loss": 4.9019,
      "mean_token_accuracy": 0.21296709775924683,
      "num_tokens": 77828306.0,
      "step": 44870
    },
    {
      "entropy": 5.330370855331421,
      "epoch": 3.4914219023536277,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038087356172306514,
      "loss": 4.8748,
      "mean_token_accuracy": 0.21634988933801652,
      "num_tokens": 77836273.0,
      "step": 44875
    },
    {
      "entropy": 5.367626523971557,
      "epoch": 3.4918109317253454,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038084910541211,
      "loss": 4.9255,
      "mean_token_accuracy": 0.20553493052721022,
      "num_tokens": 77846032.0,
      "step": 44880
    },
    {
      "entropy": 5.464908933639526,
      "epoch": 3.4921999610970627,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038082464749506314,
      "loss": 5.0436,
      "mean_token_accuracy": 0.20175820142030715,
      "num_tokens": 77854943.0,
      "step": 44885
    },
    {
      "entropy": 5.327736282348633,
      "epoch": 3.4925889904687804,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038080018797229576,
      "loss": 4.8838,
      "mean_token_accuracy": 0.2106252208352089,
      "num_tokens": 77864058.0,
      "step": 44890
    },
    {
      "entropy": 5.3849998950958256,
      "epoch": 3.492978019840498,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003807757268441789,
      "loss": 4.8083,
      "mean_token_accuracy": 0.20872959792613982,
      "num_tokens": 77872661.0,
      "step": 44895
    },
    {
      "entropy": 5.457427310943603,
      "epoch": 3.4933670492122157,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038075126411108367,
      "loss": 4.8898,
      "mean_token_accuracy": 0.20960034877061845,
      "num_tokens": 77882141.0,
      "step": 44900
    },
    {
      "entropy": 5.411701202392578,
      "epoch": 3.493756078583933,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038072679977338143,
      "loss": 4.9352,
      "mean_token_accuracy": 0.20882305949926377,
      "num_tokens": 77891131.0,
      "step": 44905
    },
    {
      "entropy": 5.324038028717041,
      "epoch": 3.4941451079556507,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038070233383144324,
      "loss": 4.8213,
      "mean_token_accuracy": 0.20613412111997603,
      "num_tokens": 77899444.0,
      "step": 44910
    },
    {
      "entropy": 5.345835208892822,
      "epoch": 3.4945341373273684,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003806778662856404,
      "loss": 4.8748,
      "mean_token_accuracy": 0.20726423859596252,
      "num_tokens": 77907909.0,
      "step": 44915
    },
    {
      "entropy": 5.33070216178894,
      "epoch": 3.4949231666990856,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003806533971363441,
      "loss": 4.8261,
      "mean_token_accuracy": 0.2166428104043007,
      "num_tokens": 77915559.0,
      "step": 44920
    },
    {
      "entropy": 5.29870719909668,
      "epoch": 3.4953121960708033,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003806289263839257,
      "loss": 4.7677,
      "mean_token_accuracy": 0.22023807018995284,
      "num_tokens": 77924692.0,
      "step": 44925
    },
    {
      "entropy": 5.3765613555908205,
      "epoch": 3.495701225442521,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003806044540287564,
      "loss": 4.9085,
      "mean_token_accuracy": 0.21011395007371902,
      "num_tokens": 77933306.0,
      "step": 44930
    },
    {
      "entropy": 5.323753833770752,
      "epoch": 3.4960902548142383,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038057998007120776,
      "loss": 4.8151,
      "mean_token_accuracy": 0.21106274724006652,
      "num_tokens": 77941537.0,
      "step": 44935
    },
    {
      "entropy": 5.36200213432312,
      "epoch": 3.496479284185956,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038055550451165086,
      "loss": 4.9358,
      "mean_token_accuracy": 0.20609404742717743,
      "num_tokens": 77950706.0,
      "step": 44940
    },
    {
      "entropy": 5.503410005569458,
      "epoch": 3.4968683135576737,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003805310273504572,
      "loss": 4.9852,
      "mean_token_accuracy": 0.2002805605530739,
      "num_tokens": 77959711.0,
      "step": 44945
    },
    {
      "entropy": 5.485320281982422,
      "epoch": 3.497257342929391,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003805065485879982,
      "loss": 4.9928,
      "mean_token_accuracy": 0.19864141643047334,
      "num_tokens": 77968534.0,
      "step": 44950
    },
    {
      "entropy": 5.369960832595825,
      "epoch": 3.4976463723011086,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038048206822464514,
      "loss": 4.8821,
      "mean_token_accuracy": 0.21188917607069016,
      "num_tokens": 77976752.0,
      "step": 44955
    },
    {
      "entropy": 5.41899676322937,
      "epoch": 3.4980354016728263,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038045758626076965,
      "loss": 4.8974,
      "mean_token_accuracy": 0.21283506751060485,
      "num_tokens": 77985310.0,
      "step": 44960
    },
    {
      "entropy": 5.440156698226929,
      "epoch": 3.498424431044544,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003804331026967432,
      "loss": 4.8367,
      "mean_token_accuracy": 0.2186812564730644,
      "num_tokens": 77994096.0,
      "step": 44965
    },
    {
      "entropy": 5.403603887557983,
      "epoch": 3.4988134604162613,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003804086175329372,
      "loss": 4.8703,
      "mean_token_accuracy": 0.20670704841613768,
      "num_tokens": 78003341.0,
      "step": 44970
    },
    {
      "entropy": 5.435726833343506,
      "epoch": 3.499202489787979,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003803841307697232,
      "loss": 4.9366,
      "mean_token_accuracy": 0.20646716356277467,
      "num_tokens": 78012102.0,
      "step": 44975
    },
    {
      "entropy": 5.3952288150787355,
      "epoch": 3.4995915191596967,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038035964240747274,
      "loss": 4.8718,
      "mean_token_accuracy": 0.2115003690123558,
      "num_tokens": 78020338.0,
      "step": 44980
    },
    {
      "entropy": 5.389092302322387,
      "epoch": 3.499980548531414,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003803351524465574,
      "loss": 4.9125,
      "mean_token_accuracy": 0.20957560390233992,
      "num_tokens": 78027762.0,
      "step": 44985
    },
    {
      "entropy": 5.438582515716552,
      "epoch": 3.5003695779031316,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038031066088734883,
      "loss": 4.9647,
      "mean_token_accuracy": 0.20572825372219086,
      "num_tokens": 78037186.0,
      "step": 44990
    },
    {
      "entropy": 5.387633752822876,
      "epoch": 3.5007586072748493,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003802861677302185,
      "loss": 4.8957,
      "mean_token_accuracy": 0.2114165022969246,
      "num_tokens": 78046214.0,
      "step": 44995
    },
    {
      "entropy": 5.365544176101684,
      "epoch": 3.501147636646567,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038026167297553826,
      "loss": 4.8859,
      "mean_token_accuracy": 0.2094174638390541,
      "num_tokens": 78054411.0,
      "step": 45000
    },
    {
      "epoch": 3.501147636646567,
      "eval_entropy": 5.134741972867288,
      "eval_loss": 5.003249168395996,
      "eval_mean_token_accuracy": 0.213059131343624,
      "eval_num_tokens": 78054411.0,
      "eval_runtime": 28.5259,
      "eval_samples_per_second": 1456.85,
      "eval_steps_per_second": 182.115,
      "step": 45000
    },
    {
      "entropy": 5.361317920684814,
      "epoch": 3.5015366660182843,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038023717662367956,
      "loss": 4.9076,
      "mean_token_accuracy": 0.20549869686365127,
      "num_tokens": 78063346.0,
      "step": 45005
    },
    {
      "entropy": 5.350410652160645,
      "epoch": 3.501925695390002,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003802126786750143,
      "loss": 4.8329,
      "mean_token_accuracy": 0.20535302758216858,
      "num_tokens": 78072018.0,
      "step": 45010
    },
    {
      "entropy": 5.411782693862915,
      "epoch": 3.5023147247617192,
      "grad_norm": 1.125,
      "learning_rate": 0.00038018817912991413,
      "loss": 4.94,
      "mean_token_accuracy": 0.20410131067037582,
      "num_tokens": 78080161.0,
      "step": 45015
    },
    {
      "entropy": 5.406914710998535,
      "epoch": 3.502703754133437,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038016367798875074,
      "loss": 4.8483,
      "mean_token_accuracy": 0.21136441826820374,
      "num_tokens": 78089543.0,
      "step": 45020
    },
    {
      "entropy": 5.426284646987915,
      "epoch": 3.5030927835051546,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003801391752518959,
      "loss": 4.94,
      "mean_token_accuracy": 0.2026752144098282,
      "num_tokens": 78098221.0,
      "step": 45025
    },
    {
      "entropy": 5.3190408706665036,
      "epoch": 3.5034818128768723,
      "grad_norm": 1.328125,
      "learning_rate": 0.00038011467091972134,
      "loss": 4.7997,
      "mean_token_accuracy": 0.212889364361763,
      "num_tokens": 78106775.0,
      "step": 45030
    },
    {
      "entropy": 5.36604323387146,
      "epoch": 3.50387084224859,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003800901649925991,
      "loss": 4.8707,
      "mean_token_accuracy": 0.2158042922616005,
      "num_tokens": 78115096.0,
      "step": 45035
    },
    {
      "entropy": 5.326368618011474,
      "epoch": 3.5042598716203073,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038006565747090076,
      "loss": 4.7968,
      "mean_token_accuracy": 0.21513827443122863,
      "num_tokens": 78123739.0,
      "step": 45040
    },
    {
      "entropy": 5.436754608154297,
      "epoch": 3.504648900992025,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038004114835499833,
      "loss": 4.9488,
      "mean_token_accuracy": 0.20664715617895127,
      "num_tokens": 78132765.0,
      "step": 45045
    },
    {
      "entropy": 5.416302967071533,
      "epoch": 3.5050379303637422,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038001663764526374,
      "loss": 4.8779,
      "mean_token_accuracy": 0.20784488320350647,
      "num_tokens": 78141152.0,
      "step": 45050
    },
    {
      "entropy": 5.466329336166382,
      "epoch": 3.50542695973546,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037999212534206876,
      "loss": 4.8987,
      "mean_token_accuracy": 0.20754835307598113,
      "num_tokens": 78150396.0,
      "step": 45055
    },
    {
      "entropy": 5.3058679580688475,
      "epoch": 3.5058159891071776,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003799676114457853,
      "loss": 4.7979,
      "mean_token_accuracy": 0.2152017593383789,
      "num_tokens": 78158678.0,
      "step": 45060
    },
    {
      "entropy": 5.398160123825074,
      "epoch": 3.5062050184788953,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037994309595678553,
      "loss": 4.9782,
      "mean_token_accuracy": 0.1999985620379448,
      "num_tokens": 78166140.0,
      "step": 45065
    },
    {
      "entropy": 5.446952629089355,
      "epoch": 3.5065940478506126,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003799185788754413,
      "loss": 4.9546,
      "mean_token_accuracy": 0.2083387091755867,
      "num_tokens": 78174321.0,
      "step": 45070
    },
    {
      "entropy": 5.415827846527099,
      "epoch": 3.5069830772223303,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003798940602021247,
      "loss": 4.8981,
      "mean_token_accuracy": 0.2054516300559044,
      "num_tokens": 78183437.0,
      "step": 45075
    },
    {
      "entropy": 5.355153608322143,
      "epoch": 3.507372106594048,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003798695399372076,
      "loss": 4.8415,
      "mean_token_accuracy": 0.21385342478752137,
      "num_tokens": 78191619.0,
      "step": 45080
    },
    {
      "entropy": 5.291917896270752,
      "epoch": 3.507761135965765,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003798450180810621,
      "loss": 4.8022,
      "mean_token_accuracy": 0.21129710525274276,
      "num_tokens": 78200178.0,
      "step": 45085
    },
    {
      "entropy": 5.355296182632446,
      "epoch": 3.508150165337483,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037982049463406044,
      "loss": 4.8345,
      "mean_token_accuracy": 0.21269507855176925,
      "num_tokens": 78209036.0,
      "step": 45090
    },
    {
      "entropy": 5.374589586257935,
      "epoch": 3.5085391947092006,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003797959695965745,
      "loss": 4.8886,
      "mean_token_accuracy": 0.2091663047671318,
      "num_tokens": 78218668.0,
      "step": 45095
    },
    {
      "entropy": 5.407313680648803,
      "epoch": 3.5089282240809183,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037977144296897665,
      "loss": 4.8269,
      "mean_token_accuracy": 0.2064720943570137,
      "num_tokens": 78226590.0,
      "step": 45100
    },
    {
      "entropy": 5.381232595443725,
      "epoch": 3.5093172534526356,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003797469147516388,
      "loss": 4.9012,
      "mean_token_accuracy": 0.2075979307293892,
      "num_tokens": 78235714.0,
      "step": 45105
    },
    {
      "entropy": 5.317760705947876,
      "epoch": 3.5097062828243533,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037972238494493333,
      "loss": 4.7758,
      "mean_token_accuracy": 0.2148730233311653,
      "num_tokens": 78244428.0,
      "step": 45110
    },
    {
      "entropy": 5.197449827194214,
      "epoch": 3.510095312196071,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003796978535492323,
      "loss": 4.6999,
      "mean_token_accuracy": 0.22743071168661116,
      "num_tokens": 78253594.0,
      "step": 45115
    },
    {
      "entropy": 5.355169773101807,
      "epoch": 3.510484341567788,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000379673320564908,
      "loss": 4.8077,
      "mean_token_accuracy": 0.21441177874803544,
      "num_tokens": 78262483.0,
      "step": 45120
    },
    {
      "entropy": 5.456123733520508,
      "epoch": 3.510873370939506,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037964878599233264,
      "loss": 4.9337,
      "mean_token_accuracy": 0.20043265521526338,
      "num_tokens": 78271659.0,
      "step": 45125
    },
    {
      "entropy": 5.290518712997437,
      "epoch": 3.5112624003112236,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037962424983187856,
      "loss": 4.8064,
      "mean_token_accuracy": 0.2148437276482582,
      "num_tokens": 78280211.0,
      "step": 45130
    },
    {
      "entropy": 5.262149953842163,
      "epoch": 3.5116514296829413,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000379599712083918,
      "loss": 4.8263,
      "mean_token_accuracy": 0.21430227011442185,
      "num_tokens": 78288834.0,
      "step": 45135
    },
    {
      "entropy": 5.315751218795777,
      "epoch": 3.5120404590546586,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003795751727488233,
      "loss": 4.7394,
      "mean_token_accuracy": 0.2179761603474617,
      "num_tokens": 78298058.0,
      "step": 45140
    },
    {
      "entropy": 5.415918922424316,
      "epoch": 3.5124294884263763,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037955063182696676,
      "loss": 4.8427,
      "mean_token_accuracy": 0.21158117651939393,
      "num_tokens": 78307074.0,
      "step": 45145
    },
    {
      "entropy": 5.352178859710693,
      "epoch": 3.5128185177980935,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037952608931872086,
      "loss": 4.8638,
      "mean_token_accuracy": 0.21605750322341918,
      "num_tokens": 78315760.0,
      "step": 45150
    },
    {
      "entropy": 5.349720859527588,
      "epoch": 3.513207547169811,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037950154522445784,
      "loss": 4.8109,
      "mean_token_accuracy": 0.21984634846448897,
      "num_tokens": 78323273.0,
      "step": 45155
    },
    {
      "entropy": 5.4327592849731445,
      "epoch": 3.513596576541529,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003794769995445502,
      "loss": 4.9341,
      "mean_token_accuracy": 0.20415549874305725,
      "num_tokens": 78332561.0,
      "step": 45160
    },
    {
      "entropy": 5.417402553558349,
      "epoch": 3.5139856059132466,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003794524522793705,
      "loss": 4.8886,
      "mean_token_accuracy": 0.20726952701807022,
      "num_tokens": 78340988.0,
      "step": 45165
    },
    {
      "entropy": 5.300640201568603,
      "epoch": 3.514374635284964,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037942790342929106,
      "loss": 4.7554,
      "mean_token_accuracy": 0.22118674218654633,
      "num_tokens": 78348910.0,
      "step": 45170
    },
    {
      "entropy": 5.383945369720459,
      "epoch": 3.5147636646566816,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037940335299468437,
      "loss": 4.8671,
      "mean_token_accuracy": 0.20781001448631287,
      "num_tokens": 78357354.0,
      "step": 45175
    },
    {
      "entropy": 5.393444442749024,
      "epoch": 3.5151526940283992,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037937880097592294,
      "loss": 4.9783,
      "mean_token_accuracy": 0.19870034158229827,
      "num_tokens": 78366974.0,
      "step": 45180
    },
    {
      "entropy": 5.388302278518677,
      "epoch": 3.5155417234001165,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037935424737337936,
      "loss": 4.8477,
      "mean_token_accuracy": 0.21021169573068618,
      "num_tokens": 78375384.0,
      "step": 45185
    },
    {
      "entropy": 5.285903263092041,
      "epoch": 3.515930752771834,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037932969218742616,
      "loss": 4.7941,
      "mean_token_accuracy": 0.21153487414121627,
      "num_tokens": 78383450.0,
      "step": 45190
    },
    {
      "entropy": 5.276757526397705,
      "epoch": 3.516319782143552,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037930513541843593,
      "loss": 4.6774,
      "mean_token_accuracy": 0.2282014861702919,
      "num_tokens": 78392676.0,
      "step": 45195
    },
    {
      "entropy": 5.378009271621704,
      "epoch": 3.5167088115152696,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037928057706678136,
      "loss": 4.8702,
      "mean_token_accuracy": 0.20442668944597245,
      "num_tokens": 78401193.0,
      "step": 45200
    },
    {
      "entropy": 5.378287267684937,
      "epoch": 3.517097840886987,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037925601713283504,
      "loss": 4.8243,
      "mean_token_accuracy": 0.21854268014431,
      "num_tokens": 78409399.0,
      "step": 45205
    },
    {
      "entropy": 5.346860408782959,
      "epoch": 3.5174868702587045,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003792314556169695,
      "loss": 4.8466,
      "mean_token_accuracy": 0.2080584019422531,
      "num_tokens": 78417973.0,
      "step": 45210
    },
    {
      "entropy": 5.302037000656128,
      "epoch": 3.5178758996304222,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003792068925195576,
      "loss": 4.921,
      "mean_token_accuracy": 0.20893283039331437,
      "num_tokens": 78426638.0,
      "step": 45215
    },
    {
      "entropy": 5.36264967918396,
      "epoch": 3.5182649290021395,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003791823278409719,
      "loss": 4.8964,
      "mean_token_accuracy": 0.20582932829856873,
      "num_tokens": 78434394.0,
      "step": 45220
    },
    {
      "entropy": 5.325146341323853,
      "epoch": 3.518653958373857,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037915776158158515,
      "loss": 4.853,
      "mean_token_accuracy": 0.20980437994003295,
      "num_tokens": 78443204.0,
      "step": 45225
    },
    {
      "entropy": 5.389709758758545,
      "epoch": 3.519042987745575,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003791331937417703,
      "loss": 4.8163,
      "mean_token_accuracy": 0.217941777408123,
      "num_tokens": 78451612.0,
      "step": 45230
    },
    {
      "entropy": 5.3801624298095705,
      "epoch": 3.5194320171172926,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037910862432189986,
      "loss": 4.8982,
      "mean_token_accuracy": 0.20434323102235794,
      "num_tokens": 78459456.0,
      "step": 45235
    },
    {
      "entropy": 5.308131885528565,
      "epoch": 3.51982104648901,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003790840533223467,
      "loss": 4.8405,
      "mean_token_accuracy": 0.20770837962627411,
      "num_tokens": 78468587.0,
      "step": 45240
    },
    {
      "entropy": 5.363768196105957,
      "epoch": 3.5202100758607275,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037905948074348387,
      "loss": 4.8504,
      "mean_token_accuracy": 0.2083433210849762,
      "num_tokens": 78476229.0,
      "step": 45245
    },
    {
      "entropy": 5.412431812286377,
      "epoch": 3.520599105232445,
      "grad_norm": 1.09375,
      "learning_rate": 0.000379034906585684,
      "loss": 4.9896,
      "mean_token_accuracy": 0.2004853218793869,
      "num_tokens": 78484996.0,
      "step": 45250
    },
    {
      "entropy": 5.424630880355835,
      "epoch": 3.5209881346041625,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037901033084932,
      "loss": 4.9842,
      "mean_token_accuracy": 0.19791247248649596,
      "num_tokens": 78493597.0,
      "step": 45255
    },
    {
      "entropy": 5.366647672653198,
      "epoch": 3.52137716397588,
      "grad_norm": 1.125,
      "learning_rate": 0.0003789857535347648,
      "loss": 4.8113,
      "mean_token_accuracy": 0.2119579493999481,
      "num_tokens": 78502269.0,
      "step": 45260
    },
    {
      "entropy": 5.312234210968017,
      "epoch": 3.521766193347598,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003789611746423912,
      "loss": 4.8327,
      "mean_token_accuracy": 0.21394023001194,
      "num_tokens": 78510676.0,
      "step": 45265
    },
    {
      "entropy": 5.3032890319824215,
      "epoch": 3.5221552227193156,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037893659417257233,
      "loss": 4.8365,
      "mean_token_accuracy": 0.21103252917528154,
      "num_tokens": 78518879.0,
      "step": 45270
    },
    {
      "entropy": 5.381455612182617,
      "epoch": 3.522544252091033,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037891201212568107,
      "loss": 4.8751,
      "mean_token_accuracy": 0.21435323655605315,
      "num_tokens": 78527704.0,
      "step": 45275
    },
    {
      "entropy": 5.462736463546753,
      "epoch": 3.5229332814627505,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037888742850209037,
      "loss": 4.8357,
      "mean_token_accuracy": 0.21086146980524062,
      "num_tokens": 78534893.0,
      "step": 45280
    },
    {
      "entropy": 5.350700855255127,
      "epoch": 3.523322310834468,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003788628433021733,
      "loss": 4.8492,
      "mean_token_accuracy": 0.21139488369226456,
      "num_tokens": 78543269.0,
      "step": 45285
    },
    {
      "entropy": 5.311590385437012,
      "epoch": 3.5237113402061855,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000378838256526303,
      "loss": 4.8144,
      "mean_token_accuracy": 0.21182577461004257,
      "num_tokens": 78551508.0,
      "step": 45290
    },
    {
      "entropy": 5.34890718460083,
      "epoch": 3.524100369577903,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003788136681748523,
      "loss": 4.8379,
      "mean_token_accuracy": 0.21465999335050584,
      "num_tokens": 78560466.0,
      "step": 45295
    },
    {
      "entropy": 5.466533756256103,
      "epoch": 3.524489398949621,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003787890782481945,
      "loss": 4.8973,
      "mean_token_accuracy": 0.20386092513799667,
      "num_tokens": 78569174.0,
      "step": 45300
    },
    {
      "entropy": 5.338572931289673,
      "epoch": 3.524878428321338,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003787644867467026,
      "loss": 4.9301,
      "mean_token_accuracy": 0.2042231887578964,
      "num_tokens": 78577955.0,
      "step": 45305
    },
    {
      "entropy": 5.2951642036437985,
      "epoch": 3.525267457693056,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037873989367074977,
      "loss": 4.7524,
      "mean_token_accuracy": 0.2157907873392105,
      "num_tokens": 78586570.0,
      "step": 45310
    },
    {
      "entropy": 5.387935304641724,
      "epoch": 3.5256564870647735,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037871529902070924,
      "loss": 4.8478,
      "mean_token_accuracy": 0.21014175862073897,
      "num_tokens": 78594714.0,
      "step": 45315
    },
    {
      "entropy": 5.408081436157227,
      "epoch": 3.526045516436491,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037869070279695404,
      "loss": 5.0113,
      "mean_token_accuracy": 0.1987893134355545,
      "num_tokens": 78604010.0,
      "step": 45320
    },
    {
      "entropy": 5.369876337051392,
      "epoch": 3.5264345458082085,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037866610499985755,
      "loss": 4.7933,
      "mean_token_accuracy": 0.21567542254924774,
      "num_tokens": 78611948.0,
      "step": 45325
    },
    {
      "entropy": 5.3476968765258786,
      "epoch": 3.526823575179926,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003786415056297929,
      "loss": 4.8758,
      "mean_token_accuracy": 0.21306609958410264,
      "num_tokens": 78621047.0,
      "step": 45330
    },
    {
      "entropy": 5.311571073532105,
      "epoch": 3.527212604551644,
      "grad_norm": 1.125,
      "learning_rate": 0.0003786169046871333,
      "loss": 4.8997,
      "mean_token_accuracy": 0.2105301395058632,
      "num_tokens": 78629739.0,
      "step": 45335
    },
    {
      "entropy": 5.32279634475708,
      "epoch": 3.527601633923361,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003785923021722521,
      "loss": 4.8086,
      "mean_token_accuracy": 0.21228309571743012,
      "num_tokens": 78638527.0,
      "step": 45340
    },
    {
      "entropy": 5.3915722370147705,
      "epoch": 3.527990663295079,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037856769808552267,
      "loss": 4.9443,
      "mean_token_accuracy": 0.20455940663814545,
      "num_tokens": 78648141.0,
      "step": 45345
    },
    {
      "entropy": 5.386763286590576,
      "epoch": 3.528379692666796,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003785430924273182,
      "loss": 4.8775,
      "mean_token_accuracy": 0.2176896646618843,
      "num_tokens": 78657164.0,
      "step": 45350
    },
    {
      "entropy": 5.333935213088989,
      "epoch": 3.5287687220385138,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037851848519801214,
      "loss": 4.7848,
      "mean_token_accuracy": 0.21336877048015596,
      "num_tokens": 78665267.0,
      "step": 45355
    },
    {
      "entropy": 5.284735155105591,
      "epoch": 3.5291577514102315,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003784938763979778,
      "loss": 4.8789,
      "mean_token_accuracy": 0.21539074182510376,
      "num_tokens": 78673423.0,
      "step": 45360
    },
    {
      "entropy": 5.3493256092071535,
      "epoch": 3.529546780781949,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003784692660275886,
      "loss": 4.8792,
      "mean_token_accuracy": 0.20247115194797516,
      "num_tokens": 78682593.0,
      "step": 45365
    },
    {
      "entropy": 5.304942083358765,
      "epoch": 3.529935810153667,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003784446540872181,
      "loss": 4.8245,
      "mean_token_accuracy": 0.20344803631305694,
      "num_tokens": 78690885.0,
      "step": 45370
    },
    {
      "entropy": 5.400669908523559,
      "epoch": 3.530324839525384,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037842004057723957,
      "loss": 4.8787,
      "mean_token_accuracy": 0.21496908813714982,
      "num_tokens": 78699970.0,
      "step": 45375
    },
    {
      "entropy": 5.413967084884644,
      "epoch": 3.530713868897102,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003783954254980265,
      "loss": 4.8617,
      "mean_token_accuracy": 0.2125125288963318,
      "num_tokens": 78709002.0,
      "step": 45380
    },
    {
      "entropy": 5.3131176948547365,
      "epoch": 3.531102898268819,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003783708088499524,
      "loss": 4.8419,
      "mean_token_accuracy": 0.215519018471241,
      "num_tokens": 78718062.0,
      "step": 45385
    },
    {
      "entropy": 5.285049819946289,
      "epoch": 3.5314919276405368,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037834619063339085,
      "loss": 4.8056,
      "mean_token_accuracy": 0.21343080848455429,
      "num_tokens": 78726558.0,
      "step": 45390
    },
    {
      "entropy": 5.3680634021759035,
      "epoch": 3.5318809570122545,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003783215708487153,
      "loss": 4.7946,
      "mean_token_accuracy": 0.21887348592281342,
      "num_tokens": 78734939.0,
      "step": 45395
    },
    {
      "entropy": 5.384794902801514,
      "epoch": 3.532269986383972,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003782969494962994,
      "loss": 4.9207,
      "mean_token_accuracy": 0.20882296860218047,
      "num_tokens": 78744418.0,
      "step": 45400
    },
    {
      "entropy": 5.331861305236816,
      "epoch": 3.5326590157556894,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003782723265765167,
      "loss": 4.8576,
      "mean_token_accuracy": 0.21029502600431443,
      "num_tokens": 78753021.0,
      "step": 45405
    },
    {
      "entropy": 5.334423398971557,
      "epoch": 3.533048045127407,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037824770208974085,
      "loss": 4.7432,
      "mean_token_accuracy": 0.21413806527853013,
      "num_tokens": 78761107.0,
      "step": 45410
    },
    {
      "entropy": 5.288662052154541,
      "epoch": 3.533437074499125,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003782230760363454,
      "loss": 4.838,
      "mean_token_accuracy": 0.2075060024857521,
      "num_tokens": 78770143.0,
      "step": 45415
    },
    {
      "entropy": 5.335245847702026,
      "epoch": 3.533826103870842,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037819844841670416,
      "loss": 4.8185,
      "mean_token_accuracy": 0.2041589230298996,
      "num_tokens": 78778458.0,
      "step": 45420
    },
    {
      "entropy": 5.360023403167725,
      "epoch": 3.5342151332425598,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003781738192311906,
      "loss": 4.9204,
      "mean_token_accuracy": 0.21567057371139525,
      "num_tokens": 78787510.0,
      "step": 45425
    },
    {
      "entropy": 5.413481521606445,
      "epoch": 3.5346041626142775,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037814918848017855,
      "loss": 4.956,
      "mean_token_accuracy": 0.20616626590490342,
      "num_tokens": 78796331.0,
      "step": 45430
    },
    {
      "entropy": 5.383185815811157,
      "epoch": 3.534993191985995,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003781245561640418,
      "loss": 4.9638,
      "mean_token_accuracy": 0.2106638789176941,
      "num_tokens": 78805606.0,
      "step": 45435
    },
    {
      "entropy": 5.45123987197876,
      "epoch": 3.5353822213577124,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003780999222831541,
      "loss": 4.8824,
      "mean_token_accuracy": 0.20432367771863938,
      "num_tokens": 78814435.0,
      "step": 45440
    },
    {
      "entropy": 5.47799825668335,
      "epoch": 3.53577125072943,
      "grad_norm": 1.25,
      "learning_rate": 0.0003780752868378892,
      "loss": 4.9765,
      "mean_token_accuracy": 0.20623561143875122,
      "num_tokens": 78822361.0,
      "step": 45445
    },
    {
      "entropy": 5.3703714370727536,
      "epoch": 3.536160280101148,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037805064982862076,
      "loss": 4.8776,
      "mean_token_accuracy": 0.211610908806324,
      "num_tokens": 78831000.0,
      "step": 45450
    },
    {
      "entropy": 5.425612115859986,
      "epoch": 3.536549309472865,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003780260112557228,
      "loss": 4.9479,
      "mean_token_accuracy": 0.20029371082782746,
      "num_tokens": 78840386.0,
      "step": 45455
    },
    {
      "entropy": 5.3691956996917725,
      "epoch": 3.5369383388445828,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00037800137111956904,
      "loss": 4.9294,
      "mean_token_accuracy": 0.20000745356082916,
      "num_tokens": 78849399.0,
      "step": 45460
    },
    {
      "entropy": 5.372079849243164,
      "epoch": 3.5373273682163004,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037797672942053344,
      "loss": 4.8001,
      "mean_token_accuracy": 0.21666875034570693,
      "num_tokens": 78857116.0,
      "step": 45465
    },
    {
      "entropy": 5.3707356452941895,
      "epoch": 3.537716397588018,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037795208615898997,
      "loss": 4.9351,
      "mean_token_accuracy": 0.20827598124742508,
      "num_tokens": 78865733.0,
      "step": 45470
    },
    {
      "entropy": 5.339529609680175,
      "epoch": 3.5381054269597354,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003779274413353124,
      "loss": 4.8821,
      "mean_token_accuracy": 0.20467738062143326,
      "num_tokens": 78873870.0,
      "step": 45475
    },
    {
      "entropy": 5.412511205673217,
      "epoch": 3.538494456331453,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037790279494987475,
      "loss": 4.9235,
      "mean_token_accuracy": 0.21310148239135743,
      "num_tokens": 78883314.0,
      "step": 45480
    },
    {
      "entropy": 5.348595190048218,
      "epoch": 3.5388834857031704,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037787814700305096,
      "loss": 4.8121,
      "mean_token_accuracy": 0.21185794472694397,
      "num_tokens": 78892161.0,
      "step": 45485
    },
    {
      "entropy": 5.317692232131958,
      "epoch": 3.539272515074888,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000377853497495215,
      "loss": 4.763,
      "mean_token_accuracy": 0.2190377339720726,
      "num_tokens": 78900432.0,
      "step": 45490
    },
    {
      "entropy": 5.334459733963013,
      "epoch": 3.5396615444466057,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037782884642674085,
      "loss": 4.8476,
      "mean_token_accuracy": 0.2072419673204422,
      "num_tokens": 78910091.0,
      "step": 45495
    },
    {
      "entropy": 5.379205274581909,
      "epoch": 3.5400505738183234,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003778041937980028,
      "loss": 4.9303,
      "mean_token_accuracy": 0.20614203959703445,
      "num_tokens": 78918863.0,
      "step": 45500
    },
    {
      "entropy": 5.366581916809082,
      "epoch": 3.5404396031900407,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037777953960937467,
      "loss": 4.8578,
      "mean_token_accuracy": 0.20734538733959199,
      "num_tokens": 78927380.0,
      "step": 45505
    },
    {
      "entropy": 5.357504177093506,
      "epoch": 3.5408286325617584,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037775488386123057,
      "loss": 4.8215,
      "mean_token_accuracy": 0.21592286974191666,
      "num_tokens": 78935847.0,
      "step": 45510
    },
    {
      "entropy": 5.366195440292358,
      "epoch": 3.541217661933476,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037773022655394466,
      "loss": 4.912,
      "mean_token_accuracy": 0.21204430609941483,
      "num_tokens": 78943941.0,
      "step": 45515
    },
    {
      "entropy": 5.4155675888061525,
      "epoch": 3.5416066913051933,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037770556768789115,
      "loss": 4.905,
      "mean_token_accuracy": 0.20575576275587082,
      "num_tokens": 78952966.0,
      "step": 45520
    },
    {
      "entropy": 5.283182859420776,
      "epoch": 3.541995720676911,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037768090726344414,
      "loss": 4.7407,
      "mean_token_accuracy": 0.21573355197906494,
      "num_tokens": 78961502.0,
      "step": 45525
    },
    {
      "entropy": 5.375920963287354,
      "epoch": 3.5423847500486287,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003776562452809777,
      "loss": 4.8559,
      "mean_token_accuracy": 0.21139290034770966,
      "num_tokens": 78969995.0,
      "step": 45530
    },
    {
      "entropy": 5.4716754913330075,
      "epoch": 3.5427737794203464,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037763158174086616,
      "loss": 5.0376,
      "mean_token_accuracy": 0.19798943549394607,
      "num_tokens": 78978385.0,
      "step": 45535
    },
    {
      "entropy": 5.367848062515259,
      "epoch": 3.5431628087920637,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037760691664348367,
      "loss": 4.8174,
      "mean_token_accuracy": 0.20784605741500856,
      "num_tokens": 78986985.0,
      "step": 45540
    },
    {
      "entropy": 5.2880395412445065,
      "epoch": 3.5435518381637814,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003775822499892046,
      "loss": 4.8028,
      "mean_token_accuracy": 0.21552790105342864,
      "num_tokens": 78995466.0,
      "step": 45545
    },
    {
      "entropy": 5.356474494934082,
      "epoch": 3.543940867535499,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003775575817784032,
      "loss": 4.9137,
      "mean_token_accuracy": 0.2094582661986351,
      "num_tokens": 79003809.0,
      "step": 45550
    },
    {
      "entropy": 5.3609109878540036,
      "epoch": 3.5443298969072163,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037753291201145375,
      "loss": 4.8223,
      "mean_token_accuracy": 0.21162704676389693,
      "num_tokens": 79011876.0,
      "step": 45555
    },
    {
      "entropy": 5.295887565612793,
      "epoch": 3.544718926278934,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003775082406887305,
      "loss": 4.8163,
      "mean_token_accuracy": 0.216938553750515,
      "num_tokens": 79021379.0,
      "step": 45560
    },
    {
      "entropy": 5.412968921661377,
      "epoch": 3.5451079556506517,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037748356781060785,
      "loss": 4.9688,
      "mean_token_accuracy": 0.20165472328662873,
      "num_tokens": 79031006.0,
      "step": 45565
    },
    {
      "entropy": 5.401702260971069,
      "epoch": 3.5454969850223694,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003774588933774602,
      "loss": 4.9049,
      "mean_token_accuracy": 0.21031848937273026,
      "num_tokens": 79039676.0,
      "step": 45570
    },
    {
      "entropy": 5.360059833526611,
      "epoch": 3.5458860143940867,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003774342173896619,
      "loss": 4.826,
      "mean_token_accuracy": 0.21433681696653367,
      "num_tokens": 79048111.0,
      "step": 45575
    },
    {
      "entropy": 5.369013118743896,
      "epoch": 3.5462750437658044,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037740953984758736,
      "loss": 4.9274,
      "mean_token_accuracy": 0.20701586455106735,
      "num_tokens": 79057447.0,
      "step": 45580
    },
    {
      "entropy": 5.337816858291626,
      "epoch": 3.5466640731375216,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003773848607516111,
      "loss": 4.8723,
      "mean_token_accuracy": 0.21563268899917604,
      "num_tokens": 79066120.0,
      "step": 45585
    },
    {
      "entropy": 5.434275531768799,
      "epoch": 3.5470531025092393,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003773601801021075,
      "loss": 4.8429,
      "mean_token_accuracy": 0.2121105670928955,
      "num_tokens": 79074732.0,
      "step": 45590
    },
    {
      "entropy": 5.373475122451782,
      "epoch": 3.547442131880957,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003773354978994512,
      "loss": 4.9258,
      "mean_token_accuracy": 0.20615026652812957,
      "num_tokens": 79084914.0,
      "step": 45595
    },
    {
      "entropy": 5.251870346069336,
      "epoch": 3.5478311612526747,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037731081414401653,
      "loss": 4.8076,
      "mean_token_accuracy": 0.21933836191892625,
      "num_tokens": 79093467.0,
      "step": 45600
    },
    {
      "entropy": 5.33868145942688,
      "epoch": 3.548220190624392,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037728612883617806,
      "loss": 4.8711,
      "mean_token_accuracy": 0.21208764910697936,
      "num_tokens": 79102710.0,
      "step": 45605
    },
    {
      "entropy": 5.4210809707641605,
      "epoch": 3.5486092199961097,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003772614419763104,
      "loss": 4.9591,
      "mean_token_accuracy": 0.20685671716928483,
      "num_tokens": 79111130.0,
      "step": 45610
    },
    {
      "entropy": 5.338911771774292,
      "epoch": 3.5489982493678274,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003772367535647882,
      "loss": 4.8097,
      "mean_token_accuracy": 0.21386290192604065,
      "num_tokens": 79119775.0,
      "step": 45615
    },
    {
      "entropy": 5.304570627212525,
      "epoch": 3.5493872787395446,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037721206360198596,
      "loss": 4.8,
      "mean_token_accuracy": 0.21682381331920625,
      "num_tokens": 79128543.0,
      "step": 45620
    },
    {
      "entropy": 5.3001853942871096,
      "epoch": 3.5497763081112623,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003771873720882783,
      "loss": 4.8641,
      "mean_token_accuracy": 0.2156204730272293,
      "num_tokens": 79136372.0,
      "step": 45625
    },
    {
      "entropy": 5.363806962966919,
      "epoch": 3.55016533748298,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037716267902403996,
      "loss": 4.8791,
      "mean_token_accuracy": 0.21082372069358826,
      "num_tokens": 79144638.0,
      "step": 45630
    },
    {
      "entropy": 5.382849597930909,
      "epoch": 3.5505543668546977,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003771379844096456,
      "loss": 4.951,
      "mean_token_accuracy": 0.20679997354745866,
      "num_tokens": 79152626.0,
      "step": 45635
    },
    {
      "entropy": 5.445909643173218,
      "epoch": 3.550943396226415,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003771132882454698,
      "loss": 4.906,
      "mean_token_accuracy": 0.20968775898218156,
      "num_tokens": 79161127.0,
      "step": 45640
    },
    {
      "entropy": 5.330301237106323,
      "epoch": 3.5513324255981327,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003770885905318874,
      "loss": 4.8076,
      "mean_token_accuracy": 0.22205820977687835,
      "num_tokens": 79169453.0,
      "step": 45645
    },
    {
      "entropy": 5.371573066711425,
      "epoch": 3.5517214549698504,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003770638912692732,
      "loss": 4.8703,
      "mean_token_accuracy": 0.20799188911914826,
      "num_tokens": 79177591.0,
      "step": 45650
    },
    {
      "entropy": 5.351689577102661,
      "epoch": 3.5521104843415676,
      "grad_norm": 1.125,
      "learning_rate": 0.0003770391904580019,
      "loss": 4.8957,
      "mean_token_accuracy": 0.20742850005626678,
      "num_tokens": 79185987.0,
      "step": 45655
    },
    {
      "entropy": 5.46686577796936,
      "epoch": 3.5524995137132853,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003770144880984482,
      "loss": 4.9934,
      "mean_token_accuracy": 0.20019074380397797,
      "num_tokens": 79194005.0,
      "step": 45660
    },
    {
      "entropy": 5.292559862136841,
      "epoch": 3.552888543085003,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037698978419098704,
      "loss": 4.7386,
      "mean_token_accuracy": 0.22500164061784744,
      "num_tokens": 79202314.0,
      "step": 45665
    },
    {
      "entropy": 5.363996315002441,
      "epoch": 3.5532775724567207,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037696507873599336,
      "loss": 4.9527,
      "mean_token_accuracy": 0.20358259826898575,
      "num_tokens": 79211213.0,
      "step": 45670
    },
    {
      "entropy": 5.313408613204956,
      "epoch": 3.553666601828438,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037694037173384177,
      "loss": 4.7459,
      "mean_token_accuracy": 0.22371175587177278,
      "num_tokens": 79219489.0,
      "step": 45675
    },
    {
      "entropy": 5.349441289901733,
      "epoch": 3.5540556312001557,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003769156631849074,
      "loss": 4.8382,
      "mean_token_accuracy": 0.21773654073476792,
      "num_tokens": 79228846.0,
      "step": 45680
    },
    {
      "entropy": 5.339849758148193,
      "epoch": 3.554444660571873,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000376890953089565,
      "loss": 4.8765,
      "mean_token_accuracy": 0.21128596365451813,
      "num_tokens": 79236924.0,
      "step": 45685
    },
    {
      "entropy": 5.35886025428772,
      "epoch": 3.5548336899435906,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003768662414481895,
      "loss": 4.9299,
      "mean_token_accuracy": 0.21177061647176743,
      "num_tokens": 79244649.0,
      "step": 45690
    },
    {
      "entropy": 5.4179017543792725,
      "epoch": 3.5552227193153083,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037684152826115595,
      "loss": 4.9461,
      "mean_token_accuracy": 0.21095108538866042,
      "num_tokens": 79252972.0,
      "step": 45695
    },
    {
      "entropy": 5.375231313705444,
      "epoch": 3.555611748687026,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003768168135288394,
      "loss": 4.8592,
      "mean_token_accuracy": 0.20862558037042617,
      "num_tokens": 79261497.0,
      "step": 45700
    },
    {
      "entropy": 5.354634141921997,
      "epoch": 3.5560007780587437,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003767920972516147,
      "loss": 4.8919,
      "mean_token_accuracy": 0.21257970333099366,
      "num_tokens": 79270297.0,
      "step": 45705
    },
    {
      "entropy": 5.403325748443604,
      "epoch": 3.556389807430461,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037676737942985697,
      "loss": 4.8968,
      "mean_token_accuracy": 0.20933594554662704,
      "num_tokens": 79278635.0,
      "step": 45710
    },
    {
      "entropy": 5.46719765663147,
      "epoch": 3.5567788368021787,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037674266006394123,
      "loss": 4.9233,
      "mean_token_accuracy": 0.20261537581682204,
      "num_tokens": 79287816.0,
      "step": 45715
    },
    {
      "entropy": 5.306892347335816,
      "epoch": 3.557167866173896,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037671793915424254,
      "loss": 4.799,
      "mean_token_accuracy": 0.2111009478569031,
      "num_tokens": 79296837.0,
      "step": 45720
    },
    {
      "entropy": 5.4563555240631105,
      "epoch": 3.5575568955456136,
      "grad_norm": 1.09375,
      "learning_rate": 0.000376693216701136,
      "loss": 4.9797,
      "mean_token_accuracy": 0.2013792023062706,
      "num_tokens": 79306442.0,
      "step": 45725
    },
    {
      "entropy": 5.529389142990112,
      "epoch": 3.5579459249173313,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003766684927049968,
      "loss": 4.9993,
      "mean_token_accuracy": 0.20627006143331528,
      "num_tokens": 79316483.0,
      "step": 45730
    },
    {
      "entropy": 5.3456484317779545,
      "epoch": 3.558334954289049,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003766437671662,
      "loss": 4.7,
      "mean_token_accuracy": 0.22627594023942948,
      "num_tokens": 79324858.0,
      "step": 45735
    },
    {
      "entropy": 5.339090728759766,
      "epoch": 3.5587239836607663,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003766190400851209,
      "loss": 4.9266,
      "mean_token_accuracy": 0.21103347837924957,
      "num_tokens": 79333744.0,
      "step": 45740
    },
    {
      "entropy": 5.385576009750366,
      "epoch": 3.559113013032484,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003765943114621345,
      "loss": 4.8149,
      "mean_token_accuracy": 0.2129213333129883,
      "num_tokens": 79342589.0,
      "step": 45745
    },
    {
      "entropy": 5.315264940261841,
      "epoch": 3.5595020424042016,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037656958129761626,
      "loss": 4.808,
      "mean_token_accuracy": 0.216843618452549,
      "num_tokens": 79350869.0,
      "step": 45750
    },
    {
      "entropy": 5.3487025737762455,
      "epoch": 3.559891071775919,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037654484959194126,
      "loss": 4.8775,
      "mean_token_accuracy": 0.21131789684295654,
      "num_tokens": 79359407.0,
      "step": 45755
    },
    {
      "entropy": 5.393037176132202,
      "epoch": 3.5602801011476366,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037652011634548466,
      "loss": 4.898,
      "mean_token_accuracy": 0.21466627717018127,
      "num_tokens": 79367564.0,
      "step": 45760
    },
    {
      "entropy": 5.3143572330474855,
      "epoch": 3.5606691305193543,
      "grad_norm": 1.171875,
      "learning_rate": 0.000376495381558622,
      "loss": 4.781,
      "mean_token_accuracy": 0.21719335168600082,
      "num_tokens": 79376300.0,
      "step": 45765
    },
    {
      "entropy": 5.402415657043457,
      "epoch": 3.561058159891072,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003764706452317284,
      "loss": 4.9138,
      "mean_token_accuracy": 0.21030608862638472,
      "num_tokens": 79385484.0,
      "step": 45770
    },
    {
      "entropy": 5.381895875930786,
      "epoch": 3.5614471892627892,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037644590736517935,
      "loss": 4.8724,
      "mean_token_accuracy": 0.215175761282444,
      "num_tokens": 79393971.0,
      "step": 45775
    },
    {
      "entropy": 5.473497343063355,
      "epoch": 3.561836218634507,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003764211679593501,
      "loss": 5.0082,
      "mean_token_accuracy": 0.20043233931064605,
      "num_tokens": 79402315.0,
      "step": 45780
    },
    {
      "entropy": 5.409636545181274,
      "epoch": 3.562225248006224,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037639642701461603,
      "loss": 4.8505,
      "mean_token_accuracy": 0.21595456898212434,
      "num_tokens": 79411071.0,
      "step": 45785
    },
    {
      "entropy": 5.311245536804199,
      "epoch": 3.562614277377942,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003763716845313526,
      "loss": 4.8294,
      "mean_token_accuracy": 0.21633348166942595,
      "num_tokens": 79418896.0,
      "step": 45790
    },
    {
      "entropy": 5.3709173679351805,
      "epoch": 3.5630033067496596,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003763469405099352,
      "loss": 4.8562,
      "mean_token_accuracy": 0.20685584396123885,
      "num_tokens": 79427223.0,
      "step": 45795
    },
    {
      "entropy": 5.416921854019165,
      "epoch": 3.5633923361213773,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003763221949507392,
      "loss": 4.8637,
      "mean_token_accuracy": 0.21078304052352906,
      "num_tokens": 79435895.0,
      "step": 45800
    },
    {
      "entropy": 5.419690418243408,
      "epoch": 3.563781365493095,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037629744785414027,
      "loss": 4.9203,
      "mean_token_accuracy": 0.20382775962352753,
      "num_tokens": 79443957.0,
      "step": 45805
    },
    {
      "entropy": 5.387167596817017,
      "epoch": 3.5641703948648122,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037627269922051377,
      "loss": 4.8956,
      "mean_token_accuracy": 0.2097598135471344,
      "num_tokens": 79452134.0,
      "step": 45810
    },
    {
      "entropy": 5.342566680908203,
      "epoch": 3.56455942423653,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037624794905023523,
      "loss": 4.8732,
      "mean_token_accuracy": 0.2081510528922081,
      "num_tokens": 79460087.0,
      "step": 45815
    },
    {
      "entropy": 5.374317836761475,
      "epoch": 3.564948453608247,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003762231973436802,
      "loss": 4.8978,
      "mean_token_accuracy": 0.20252387821674347,
      "num_tokens": 79469869.0,
      "step": 45820
    },
    {
      "entropy": 5.420507907867432,
      "epoch": 3.565337482979965,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003761984441012243,
      "loss": 4.8968,
      "mean_token_accuracy": 0.2083616688847542,
      "num_tokens": 79478559.0,
      "step": 45825
    },
    {
      "entropy": 5.39992241859436,
      "epoch": 3.5657265123516826,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003761736893232431,
      "loss": 4.7938,
      "mean_token_accuracy": 0.2227136731147766,
      "num_tokens": 79486945.0,
      "step": 45830
    },
    {
      "entropy": 5.264429521560669,
      "epoch": 3.5661155417234003,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037614893301011223,
      "loss": 4.7805,
      "mean_token_accuracy": 0.21383192241191865,
      "num_tokens": 79496526.0,
      "step": 45835
    },
    {
      "entropy": 5.332279539108276,
      "epoch": 3.5665045710951175,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003761241751622072,
      "loss": 4.8995,
      "mean_token_accuracy": 0.20822951644659043,
      "num_tokens": 79505503.0,
      "step": 45840
    },
    {
      "entropy": 5.396700000762939,
      "epoch": 3.5668936004668352,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003760994157799038,
      "loss": 4.9519,
      "mean_token_accuracy": 0.20017847716808318,
      "num_tokens": 79514561.0,
      "step": 45845
    },
    {
      "entropy": 5.332700395584107,
      "epoch": 3.567282629838553,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003760746548635777,
      "loss": 4.8681,
      "mean_token_accuracy": 0.21444705426692962,
      "num_tokens": 79522973.0,
      "step": 45850
    },
    {
      "entropy": 5.351207208633423,
      "epoch": 3.56767165921027,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003760498924136046,
      "loss": 4.8053,
      "mean_token_accuracy": 0.2152150899171829,
      "num_tokens": 79532130.0,
      "step": 45855
    },
    {
      "entropy": 5.387542819976806,
      "epoch": 3.568060688581988,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037602512843036025,
      "loss": 4.8939,
      "mean_token_accuracy": 0.20648878514766694,
      "num_tokens": 79540581.0,
      "step": 45860
    },
    {
      "entropy": 5.32386884689331,
      "epoch": 3.5684497179537056,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003760003629142203,
      "loss": 4.7718,
      "mean_token_accuracy": 0.2246812179684639,
      "num_tokens": 79548674.0,
      "step": 45865
    },
    {
      "entropy": 5.342483806610107,
      "epoch": 3.5688387473254233,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037597559586556063,
      "loss": 4.8286,
      "mean_token_accuracy": 0.21390326768159867,
      "num_tokens": 79558300.0,
      "step": 45870
    },
    {
      "entropy": 5.371205615997314,
      "epoch": 3.5692277766971405,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000375950827284757,
      "loss": 4.8661,
      "mean_token_accuracy": 0.2090674415230751,
      "num_tokens": 79566687.0,
      "step": 45875
    },
    {
      "entropy": 5.435831356048584,
      "epoch": 3.5696168060688582,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037592605717218523,
      "loss": 4.9039,
      "mean_token_accuracy": 0.21104466766119004,
      "num_tokens": 79575204.0,
      "step": 45880
    },
    {
      "entropy": 5.386499214172363,
      "epoch": 3.570005835440576,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037590128552822123,
      "loss": 4.8331,
      "mean_token_accuracy": 0.21853763610124588,
      "num_tokens": 79583102.0,
      "step": 45885
    },
    {
      "entropy": 5.3587729930877686,
      "epoch": 3.570394864812293,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003758765123532407,
      "loss": 4.858,
      "mean_token_accuracy": 0.20406140834093095,
      "num_tokens": 79591933.0,
      "step": 45890
    },
    {
      "entropy": 5.414136552810669,
      "epoch": 3.570783894184011,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003758517376476198,
      "loss": 4.8766,
      "mean_token_accuracy": 0.20506496727466583,
      "num_tokens": 79600588.0,
      "step": 45895
    },
    {
      "entropy": 5.370579099655151,
      "epoch": 3.5711729235557286,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037582696141173435,
      "loss": 4.8369,
      "mean_token_accuracy": 0.2162756785750389,
      "num_tokens": 79609252.0,
      "step": 45900
    },
    {
      "entropy": 5.371086597442627,
      "epoch": 3.5715619529274463,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037580218364596015,
      "loss": 4.8329,
      "mean_token_accuracy": 0.21743370443582535,
      "num_tokens": 79617191.0,
      "step": 45905
    },
    {
      "entropy": 5.38363995552063,
      "epoch": 3.5719509822991635,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037577740435067327,
      "loss": 4.9411,
      "mean_token_accuracy": 0.20150582045316695,
      "num_tokens": 79626174.0,
      "step": 45910
    },
    {
      "entropy": 5.392297983169556,
      "epoch": 3.572340011670881,
      "grad_norm": 1.125,
      "learning_rate": 0.00037575262352624965,
      "loss": 4.8714,
      "mean_token_accuracy": 0.2171070784330368,
      "num_tokens": 79634760.0,
      "step": 45915
    },
    {
      "entropy": 5.305218267440796,
      "epoch": 3.5727290410425985,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037572784117306535,
      "loss": 4.7519,
      "mean_token_accuracy": 0.22990741729736328,
      "num_tokens": 79643335.0,
      "step": 45920
    },
    {
      "entropy": 5.412169075012207,
      "epoch": 3.573118070414316,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003757030572914963,
      "loss": 4.9576,
      "mean_token_accuracy": 0.20484157055616378,
      "num_tokens": 79651695.0,
      "step": 45925
    },
    {
      "entropy": 5.387224817276001,
      "epoch": 3.573507099786034,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003756782718819188,
      "loss": 4.8715,
      "mean_token_accuracy": 0.21286609172821044,
      "num_tokens": 79659674.0,
      "step": 45930
    },
    {
      "entropy": 5.3691725730896,
      "epoch": 3.5738961291577516,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037565348494470873,
      "loss": 4.7966,
      "mean_token_accuracy": 0.2168641433119774,
      "num_tokens": 79668592.0,
      "step": 45935
    },
    {
      "entropy": 5.288401460647583,
      "epoch": 3.574285158529469,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003756286964802422,
      "loss": 4.7461,
      "mean_token_accuracy": 0.21727731674909592,
      "num_tokens": 79677828.0,
      "step": 45940
    },
    {
      "entropy": 5.38605899810791,
      "epoch": 3.5746741879011865,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003756039064888954,
      "loss": 4.8981,
      "mean_token_accuracy": 0.21261195540428163,
      "num_tokens": 79686660.0,
      "step": 45945
    },
    {
      "entropy": 5.386173582077026,
      "epoch": 3.575063217272904,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003755791149710444,
      "loss": 4.9441,
      "mean_token_accuracy": 0.2114563375711441,
      "num_tokens": 79695263.0,
      "step": 45950
    },
    {
      "entropy": 5.405178785324097,
      "epoch": 3.5754522466446215,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037555432192706547,
      "loss": 4.8446,
      "mean_token_accuracy": 0.21095335334539414,
      "num_tokens": 79703602.0,
      "step": 45955
    },
    {
      "entropy": 5.404139184951783,
      "epoch": 3.575841276016339,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037552952735733466,
      "loss": 4.8813,
      "mean_token_accuracy": 0.214123372733593,
      "num_tokens": 79711783.0,
      "step": 45960
    },
    {
      "entropy": 5.46048903465271,
      "epoch": 3.576230305388057,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003755047312622283,
      "loss": 4.9627,
      "mean_token_accuracy": 0.19972453266382217,
      "num_tokens": 79721461.0,
      "step": 45965
    },
    {
      "entropy": 5.319296932220459,
      "epoch": 3.5766193347597746,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037547993364212267,
      "loss": 4.8037,
      "mean_token_accuracy": 0.21298013627529144,
      "num_tokens": 79729230.0,
      "step": 45970
    },
    {
      "entropy": 5.35458402633667,
      "epoch": 3.577008364131492,
      "grad_norm": 1.15625,
      "learning_rate": 0.000375455134497394,
      "loss": 4.8954,
      "mean_token_accuracy": 0.2101075530052185,
      "num_tokens": 79738165.0,
      "step": 45975
    },
    {
      "entropy": 5.35061502456665,
      "epoch": 3.5773973935032095,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003754303338284186,
      "loss": 4.7844,
      "mean_token_accuracy": 0.21575349420309067,
      "num_tokens": 79746857.0,
      "step": 45980
    },
    {
      "entropy": 5.458946800231933,
      "epoch": 3.577786422874927,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037540553163557263,
      "loss": 4.9483,
      "mean_token_accuracy": 0.2113783538341522,
      "num_tokens": 79755172.0,
      "step": 45985
    },
    {
      "entropy": 5.256872367858887,
      "epoch": 3.5781754522466445,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037538072791923257,
      "loss": 4.776,
      "mean_token_accuracy": 0.21889315843582152,
      "num_tokens": 79763542.0,
      "step": 45990
    },
    {
      "entropy": 5.340472984313965,
      "epoch": 3.578564481618362,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037535592267977467,
      "loss": 4.8454,
      "mean_token_accuracy": 0.20987128615379333,
      "num_tokens": 79771805.0,
      "step": 45995
    },
    {
      "entropy": 5.407039308547974,
      "epoch": 3.57895351099008,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003753311159175754,
      "loss": 4.8866,
      "mean_token_accuracy": 0.21345460712909697,
      "num_tokens": 79780645.0,
      "step": 46000
    },
    {
      "entropy": 5.293634986877441,
      "epoch": 3.5793425403617976,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037530630763301113,
      "loss": 4.7717,
      "mean_token_accuracy": 0.22159164994955063,
      "num_tokens": 79789218.0,
      "step": 46005
    },
    {
      "entropy": 5.39737458229065,
      "epoch": 3.579731569733515,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003752814978264584,
      "loss": 4.9166,
      "mean_token_accuracy": 0.20027800500392914,
      "num_tokens": 79798214.0,
      "step": 46010
    },
    {
      "entropy": 5.352280235290527,
      "epoch": 3.5801205991052325,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003752566864982934,
      "loss": 4.8223,
      "mean_token_accuracy": 0.21190735697746277,
      "num_tokens": 79806698.0,
      "step": 46015
    },
    {
      "entropy": 5.315085792541504,
      "epoch": 3.5805096284769498,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037523187364889293,
      "loss": 4.8874,
      "mean_token_accuracy": 0.20557451099157334,
      "num_tokens": 79814956.0,
      "step": 46020
    },
    {
      "entropy": 5.327392053604126,
      "epoch": 3.5808986578486675,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003752070592786332,
      "loss": 4.8152,
      "mean_token_accuracy": 0.21579249799251557,
      "num_tokens": 79823955.0,
      "step": 46025
    },
    {
      "entropy": 5.323977661132813,
      "epoch": 3.581287687220385,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037518224338789084,
      "loss": 4.728,
      "mean_token_accuracy": 0.21561214923858643,
      "num_tokens": 79832383.0,
      "step": 46030
    },
    {
      "entropy": 5.369806480407715,
      "epoch": 3.581676716592103,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003751574259770424,
      "loss": 4.8448,
      "mean_token_accuracy": 0.20796600729227066,
      "num_tokens": 79841778.0,
      "step": 46035
    },
    {
      "entropy": 5.2708429336547855,
      "epoch": 3.58206574596382,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003751326070464645,
      "loss": 4.7758,
      "mean_token_accuracy": 0.21068284511566163,
      "num_tokens": 79851024.0,
      "step": 46040
    },
    {
      "entropy": 5.364293336868286,
      "epoch": 3.582454775335538,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003751077865965337,
      "loss": 4.8831,
      "mean_token_accuracy": 0.2154039666056633,
      "num_tokens": 79859987.0,
      "step": 46045
    },
    {
      "entropy": 5.298119688034058,
      "epoch": 3.5828438047072555,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037508296462762647,
      "loss": 4.7662,
      "mean_token_accuracy": 0.22345883399248123,
      "num_tokens": 79868389.0,
      "step": 46050
    },
    {
      "entropy": 5.392587900161743,
      "epoch": 3.5832328340789728,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037505814114011956,
      "loss": 4.956,
      "mean_token_accuracy": 0.19964467585086823,
      "num_tokens": 79877784.0,
      "step": 46055
    },
    {
      "entropy": 5.394743347167969,
      "epoch": 3.5836218634506904,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037503331613438965,
      "loss": 4.8606,
      "mean_token_accuracy": 0.21893921196460725,
      "num_tokens": 79886118.0,
      "step": 46060
    },
    {
      "entropy": 5.335175466537476,
      "epoch": 3.584010892822408,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037500848961081334,
      "loss": 4.8138,
      "mean_token_accuracy": 0.22220377027988433,
      "num_tokens": 79894753.0,
      "step": 46065
    },
    {
      "entropy": 5.340113878250122,
      "epoch": 3.584399922194126,
      "grad_norm": 1.359375,
      "learning_rate": 0.00037498366156976746,
      "loss": 4.8354,
      "mean_token_accuracy": 0.21485548913478852,
      "num_tokens": 79903285.0,
      "step": 46070
    },
    {
      "entropy": 5.384540939331055,
      "epoch": 3.584788951565843,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037495883201162854,
      "loss": 4.8507,
      "mean_token_accuracy": 0.2106296271085739,
      "num_tokens": 79912076.0,
      "step": 46075
    },
    {
      "entropy": 5.3170544624328615,
      "epoch": 3.585177980937561,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003749340009367736,
      "loss": 4.8282,
      "mean_token_accuracy": 0.20960128903388978,
      "num_tokens": 79920942.0,
      "step": 46080
    },
    {
      "entropy": 5.365418338775635,
      "epoch": 3.5855670103092785,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037490916834557916,
      "loss": 4.9794,
      "mean_token_accuracy": 0.2040545791387558,
      "num_tokens": 79929818.0,
      "step": 46085
    },
    {
      "entropy": 5.410957098007202,
      "epoch": 3.5859560396809957,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003748843342384221,
      "loss": 4.8804,
      "mean_token_accuracy": 0.21477923393249512,
      "num_tokens": 79939001.0,
      "step": 46090
    },
    {
      "entropy": 5.4663848876953125,
      "epoch": 3.5863450690527134,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037485949861567925,
      "loss": 4.9456,
      "mean_token_accuracy": 0.2039807215332985,
      "num_tokens": 79947311.0,
      "step": 46095
    },
    {
      "entropy": 5.327258014678955,
      "epoch": 3.586734098424431,
      "grad_norm": 1.125,
      "learning_rate": 0.0003748346614777275,
      "loss": 4.8163,
      "mean_token_accuracy": 0.2211000517010689,
      "num_tokens": 79955266.0,
      "step": 46100
    },
    {
      "entropy": 5.396526670455932,
      "epoch": 3.587123127796149,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003748098228249437,
      "loss": 4.8672,
      "mean_token_accuracy": 0.2157622754573822,
      "num_tokens": 79962465.0,
      "step": 46105
    },
    {
      "entropy": 5.28253583908081,
      "epoch": 3.587512157167866,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037478498265770463,
      "loss": 4.8308,
      "mean_token_accuracy": 0.21479502320289612,
      "num_tokens": 79971040.0,
      "step": 46110
    },
    {
      "entropy": 5.368221759796143,
      "epoch": 3.587901186539584,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003747601409763872,
      "loss": 4.8532,
      "mean_token_accuracy": 0.21187446266412735,
      "num_tokens": 79979454.0,
      "step": 46115
    },
    {
      "entropy": 5.375804996490478,
      "epoch": 3.588290215911301,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037473529778136853,
      "loss": 4.8066,
      "mean_token_accuracy": 0.2162273570895195,
      "num_tokens": 79988240.0,
      "step": 46120
    },
    {
      "entropy": 5.3784057140350345,
      "epoch": 3.5886792452830187,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037471045307302544,
      "loss": 4.9702,
      "mean_token_accuracy": 0.21573228836059571,
      "num_tokens": 79997753.0,
      "step": 46125
    },
    {
      "entropy": 5.343861246109009,
      "epoch": 3.5890682746547364,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037468560685173497,
      "loss": 4.8603,
      "mean_token_accuracy": 0.20834775269031525,
      "num_tokens": 80007274.0,
      "step": 46130
    },
    {
      "entropy": 5.368830680847168,
      "epoch": 3.589457304026454,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000374660759117874,
      "loss": 4.8383,
      "mean_token_accuracy": 0.21289100199937822,
      "num_tokens": 80015796.0,
      "step": 46135
    },
    {
      "entropy": 5.331884241104126,
      "epoch": 3.589846333398172,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037463590987181966,
      "loss": 4.8022,
      "mean_token_accuracy": 0.21325447112321855,
      "num_tokens": 80024049.0,
      "step": 46140
    },
    {
      "entropy": 5.223087453842163,
      "epoch": 3.590235362769889,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037461105911394904,
      "loss": 4.7046,
      "mean_token_accuracy": 0.2245904713869095,
      "num_tokens": 80032284.0,
      "step": 46145
    },
    {
      "entropy": 5.315054893493652,
      "epoch": 3.590624392141607,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037458620684463907,
      "loss": 4.8306,
      "mean_token_accuracy": 0.21044742316007614,
      "num_tokens": 80040154.0,
      "step": 46150
    },
    {
      "entropy": 5.395602512359619,
      "epoch": 3.591013421513324,
      "grad_norm": 1.125,
      "learning_rate": 0.00037456135306426693,
      "loss": 4.9123,
      "mean_token_accuracy": 0.21983245015144348,
      "num_tokens": 80048156.0,
      "step": 46155
    },
    {
      "entropy": 5.291496419906617,
      "epoch": 3.5914024508850417,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003745364977732097,
      "loss": 4.8105,
      "mean_token_accuracy": 0.21978435218334197,
      "num_tokens": 80057467.0,
      "step": 46160
    },
    {
      "entropy": 5.364649534225464,
      "epoch": 3.5917914802567594,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003745116409718446,
      "loss": 4.8903,
      "mean_token_accuracy": 0.21300036758184432,
      "num_tokens": 80066864.0,
      "step": 46165
    },
    {
      "entropy": 5.40105791091919,
      "epoch": 3.592180509628477,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003744867826605488,
      "loss": 4.927,
      "mean_token_accuracy": 0.20288191735744476,
      "num_tokens": 80076940.0,
      "step": 46170
    },
    {
      "entropy": 5.382034683227539,
      "epoch": 3.5925695390001944,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003744619228396993,
      "loss": 4.8364,
      "mean_token_accuracy": 0.2175360456109047,
      "num_tokens": 80085994.0,
      "step": 46175
    },
    {
      "entropy": 5.3054131984710695,
      "epoch": 3.592958568371912,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003744370615096734,
      "loss": 4.7961,
      "mean_token_accuracy": 0.21235192120075225,
      "num_tokens": 80093721.0,
      "step": 46180
    },
    {
      "entropy": 5.3134490013122555,
      "epoch": 3.5933475977436298,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003744121986708485,
      "loss": 4.8074,
      "mean_token_accuracy": 0.2174096077680588,
      "num_tokens": 80101833.0,
      "step": 46185
    },
    {
      "entropy": 5.350430202484131,
      "epoch": 3.593736627115347,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037438733432360163,
      "loss": 4.8063,
      "mean_token_accuracy": 0.2237432599067688,
      "num_tokens": 80110337.0,
      "step": 46190
    },
    {
      "entropy": 5.339279747009277,
      "epoch": 3.5941256564870647,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003743624684683102,
      "loss": 4.7986,
      "mean_token_accuracy": 0.21971596032381058,
      "num_tokens": 80118218.0,
      "step": 46195
    },
    {
      "entropy": 5.342680883407593,
      "epoch": 3.5945146858587824,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003743376011053514,
      "loss": 4.8802,
      "mean_token_accuracy": 0.2046196088194847,
      "num_tokens": 80126409.0,
      "step": 46200
    },
    {
      "entropy": 5.309059000015258,
      "epoch": 3.5949037152305,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037431273223510265,
      "loss": 4.8489,
      "mean_token_accuracy": 0.21931012123823165,
      "num_tokens": 80134415.0,
      "step": 46205
    },
    {
      "entropy": 5.3414427757263185,
      "epoch": 3.5952927446022174,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003742878618579413,
      "loss": 4.8754,
      "mean_token_accuracy": 0.2158853754401207,
      "num_tokens": 80142504.0,
      "step": 46210
    },
    {
      "entropy": 5.3322498321533205,
      "epoch": 3.595681773973935,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003742629899742446,
      "loss": 4.8946,
      "mean_token_accuracy": 0.21256586760282517,
      "num_tokens": 80151892.0,
      "step": 46215
    },
    {
      "entropy": 5.391326999664306,
      "epoch": 3.5960708033456523,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037423811658439014,
      "loss": 4.8579,
      "mean_token_accuracy": 0.21121630668640137,
      "num_tokens": 80160194.0,
      "step": 46220
    },
    {
      "entropy": 5.43248872756958,
      "epoch": 3.59645983271737,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037421324168875503,
      "loss": 4.9052,
      "mean_token_accuracy": 0.21015152484178543,
      "num_tokens": 80168397.0,
      "step": 46225
    },
    {
      "entropy": 5.286218500137329,
      "epoch": 3.5968488620890877,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000374188365287717,
      "loss": 4.838,
      "mean_token_accuracy": 0.21344176828861236,
      "num_tokens": 80177142.0,
      "step": 46230
    },
    {
      "entropy": 5.373470258712769,
      "epoch": 3.5972378914608054,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003741634873816534,
      "loss": 4.9029,
      "mean_token_accuracy": 0.20238704681396485,
      "num_tokens": 80186221.0,
      "step": 46235
    },
    {
      "entropy": 5.42016978263855,
      "epoch": 3.597626920832523,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037413860797094175,
      "loss": 4.9192,
      "mean_token_accuracy": 0.20729054212570192,
      "num_tokens": 80194334.0,
      "step": 46240
    },
    {
      "entropy": 5.392420673370362,
      "epoch": 3.5980159502042404,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037411372705595947,
      "loss": 4.9182,
      "mean_token_accuracy": 0.2054744154214859,
      "num_tokens": 80202986.0,
      "step": 46245
    },
    {
      "entropy": 5.4008262157440186,
      "epoch": 3.598404979575958,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003740888446370842,
      "loss": 4.8833,
      "mean_token_accuracy": 0.21755042672157288,
      "num_tokens": 80211392.0,
      "step": 46250
    },
    {
      "entropy": 5.365060615539551,
      "epoch": 3.5987940089476753,
      "grad_norm": 1.125,
      "learning_rate": 0.0003740639607146933,
      "loss": 4.9182,
      "mean_token_accuracy": 0.19629386365413665,
      "num_tokens": 80219572.0,
      "step": 46255
    },
    {
      "entropy": 5.421970796585083,
      "epoch": 3.599183038319393,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037403907528916455,
      "loss": 4.9307,
      "mean_token_accuracy": 0.2074686974287033,
      "num_tokens": 80227847.0,
      "step": 46260
    },
    {
      "entropy": 5.369739246368408,
      "epoch": 3.5995720676911107,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003740141883608755,
      "loss": 4.9365,
      "mean_token_accuracy": 0.2021120235323906,
      "num_tokens": 80237610.0,
      "step": 46265
    },
    {
      "entropy": 5.346039533615112,
      "epoch": 3.5999610970628284,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003739892999302037,
      "loss": 4.8021,
      "mean_token_accuracy": 0.21924768686294555,
      "num_tokens": 80245948.0,
      "step": 46270
    },
    {
      "entropy": 5.354439210891724,
      "epoch": 3.6003501264345457,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003739644099975269,
      "loss": 4.9045,
      "mean_token_accuracy": 0.21224594116210938,
      "num_tokens": 80254457.0,
      "step": 46275
    },
    {
      "entropy": 5.4103845119476315,
      "epoch": 3.6007391558062634,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037393951856322263,
      "loss": 4.9228,
      "mean_token_accuracy": 0.20880051851272582,
      "num_tokens": 80264204.0,
      "step": 46280
    },
    {
      "entropy": 5.45153398513794,
      "epoch": 3.601128185177981,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003739146256276686,
      "loss": 4.9017,
      "mean_token_accuracy": 0.21286383122205735,
      "num_tokens": 80272188.0,
      "step": 46285
    },
    {
      "entropy": 5.361737203598023,
      "epoch": 3.6015172145496983,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037388973119124267,
      "loss": 4.8936,
      "mean_token_accuracy": 0.2037728786468506,
      "num_tokens": 80282009.0,
      "step": 46290
    },
    {
      "entropy": 5.4231476306915285,
      "epoch": 3.601906243921416,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003738648352543224,
      "loss": 4.8356,
      "mean_token_accuracy": 0.21708330363035203,
      "num_tokens": 80290268.0,
      "step": 46295
    },
    {
      "entropy": 5.385340976715088,
      "epoch": 3.6022952732931337,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003738399378172856,
      "loss": 4.8425,
      "mean_token_accuracy": 0.20730377584695817,
      "num_tokens": 80298919.0,
      "step": 46300
    },
    {
      "entropy": 5.3667865753173825,
      "epoch": 3.6026843026648514,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003738150388805101,
      "loss": 4.8945,
      "mean_token_accuracy": 0.2113905981183052,
      "num_tokens": 80308117.0,
      "step": 46305
    },
    {
      "entropy": 5.297799825668335,
      "epoch": 3.6030733320365687,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037379013844437373,
      "loss": 4.875,
      "mean_token_accuracy": 0.21415880918502808,
      "num_tokens": 80316840.0,
      "step": 46310
    },
    {
      "entropy": 5.370544004440307,
      "epoch": 3.6034623614082864,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037376523650925415,
      "loss": 4.788,
      "mean_token_accuracy": 0.2231057271361351,
      "num_tokens": 80326044.0,
      "step": 46315
    },
    {
      "entropy": 5.370165252685547,
      "epoch": 3.603851390780004,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003737403330755293,
      "loss": 4.852,
      "mean_token_accuracy": 0.20823064744472503,
      "num_tokens": 80334297.0,
      "step": 46320
    },
    {
      "entropy": 5.311166095733642,
      "epoch": 3.6042404201517213,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003737154281435771,
      "loss": 4.8498,
      "mean_token_accuracy": 0.21300268471240996,
      "num_tokens": 80342522.0,
      "step": 46325
    },
    {
      "entropy": 5.396177816390991,
      "epoch": 3.604629449523439,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037369052171377535,
      "loss": 4.9279,
      "mean_token_accuracy": 0.20780676752328872,
      "num_tokens": 80351055.0,
      "step": 46330
    },
    {
      "entropy": 5.35433783531189,
      "epoch": 3.6050184788951567,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003736656137865021,
      "loss": 4.8649,
      "mean_token_accuracy": 0.20731803476810456,
      "num_tokens": 80359627.0,
      "step": 46335
    },
    {
      "entropy": 5.372165393829346,
      "epoch": 3.6054075082668744,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037364070436213514,
      "loss": 4.9294,
      "mean_token_accuracy": 0.20639152973890304,
      "num_tokens": 80368080.0,
      "step": 46340
    },
    {
      "entropy": 5.349333477020264,
      "epoch": 3.6057965376385916,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003736157934410525,
      "loss": 4.787,
      "mean_token_accuracy": 0.21628152430057526,
      "num_tokens": 80376814.0,
      "step": 46345
    },
    {
      "entropy": 5.3038170337677,
      "epoch": 3.6061855670103093,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003735908810236322,
      "loss": 4.8714,
      "mean_token_accuracy": 0.212046779692173,
      "num_tokens": 80385364.0,
      "step": 46350
    },
    {
      "entropy": 5.376226758956909,
      "epoch": 3.6065745963820266,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037356596711025214,
      "loss": 4.844,
      "mean_token_accuracy": 0.21661100536584854,
      "num_tokens": 80393674.0,
      "step": 46355
    },
    {
      "entropy": 5.371115636825562,
      "epoch": 3.6069636257537443,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003735410517012905,
      "loss": 4.8703,
      "mean_token_accuracy": 0.21429896354675293,
      "num_tokens": 80401981.0,
      "step": 46360
    },
    {
      "entropy": 5.327208566665649,
      "epoch": 3.607352655125462,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003735161347971252,
      "loss": 4.8551,
      "mean_token_accuracy": 0.21539910435676574,
      "num_tokens": 80410550.0,
      "step": 46365
    },
    {
      "entropy": 5.39033317565918,
      "epoch": 3.6077416844971797,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003734912163981344,
      "loss": 4.87,
      "mean_token_accuracy": 0.20593751221895218,
      "num_tokens": 80420310.0,
      "step": 46370
    },
    {
      "entropy": 5.297357177734375,
      "epoch": 3.608130713868897,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037346629650469605,
      "loss": 4.8278,
      "mean_token_accuracy": 0.21171630322933196,
      "num_tokens": 80429167.0,
      "step": 46375
    },
    {
      "entropy": 5.2734943389892575,
      "epoch": 3.6085197432406146,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037344137511718847,
      "loss": 4.7573,
      "mean_token_accuracy": 0.2259500354528427,
      "num_tokens": 80438162.0,
      "step": 46380
    },
    {
      "entropy": 5.372167253494263,
      "epoch": 3.6089087726123323,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037341645223598965,
      "loss": 4.8916,
      "mean_token_accuracy": 0.21464264243841172,
      "num_tokens": 80446989.0,
      "step": 46385
    },
    {
      "entropy": 5.424570369720459,
      "epoch": 3.6092978019840496,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037339152786147777,
      "loss": 4.9332,
      "mean_token_accuracy": 0.2038123369216919,
      "num_tokens": 80455696.0,
      "step": 46390
    },
    {
      "entropy": 5.470143032073975,
      "epoch": 3.6096868313557673,
      "grad_norm": 1.125,
      "learning_rate": 0.00037336660199403115,
      "loss": 4.9826,
      "mean_token_accuracy": 0.20275453180074693,
      "num_tokens": 80465029.0,
      "step": 46395
    },
    {
      "entropy": 5.43672285079956,
      "epoch": 3.610075860727485,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003733416746340279,
      "loss": 4.9101,
      "mean_token_accuracy": 0.21647294014692306,
      "num_tokens": 80472834.0,
      "step": 46400
    },
    {
      "entropy": 5.39040060043335,
      "epoch": 3.6104648900992027,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003733167457818463,
      "loss": 4.8316,
      "mean_token_accuracy": 0.20891461670398712,
      "num_tokens": 80481689.0,
      "step": 46405
    },
    {
      "entropy": 5.377660512924194,
      "epoch": 3.61085391947092,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003732918154378645,
      "loss": 4.9276,
      "mean_token_accuracy": 0.20984197556972503,
      "num_tokens": 80489277.0,
      "step": 46410
    },
    {
      "entropy": 5.306615686416626,
      "epoch": 3.6112429488426376,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037326688360246087,
      "loss": 4.847,
      "mean_token_accuracy": 0.22103274017572402,
      "num_tokens": 80497567.0,
      "step": 46415
    },
    {
      "entropy": 5.406009292602539,
      "epoch": 3.6116319782143553,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003732419502760137,
      "loss": 4.9671,
      "mean_token_accuracy": 0.2047761559486389,
      "num_tokens": 80506554.0,
      "step": 46420
    },
    {
      "entropy": 5.3683021545410154,
      "epoch": 3.6120210075860726,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037321701545890126,
      "loss": 4.8523,
      "mean_token_accuracy": 0.21365817040205,
      "num_tokens": 80515648.0,
      "step": 46425
    },
    {
      "entropy": 5.417666721343994,
      "epoch": 3.6124100369577903,
      "grad_norm": 1.125,
      "learning_rate": 0.0003731920791515019,
      "loss": 4.9271,
      "mean_token_accuracy": 0.20564096868038179,
      "num_tokens": 80524368.0,
      "step": 46430
    },
    {
      "entropy": 5.357018280029297,
      "epoch": 3.612799066329508,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003731671413541942,
      "loss": 4.8115,
      "mean_token_accuracy": 0.21313745379447938,
      "num_tokens": 80533575.0,
      "step": 46435
    },
    {
      "entropy": 5.366616153717041,
      "epoch": 3.6131880957012257,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037314220206735626,
      "loss": 4.8187,
      "mean_token_accuracy": 0.21483396291732787,
      "num_tokens": 80542053.0,
      "step": 46440
    },
    {
      "entropy": 5.462084484100342,
      "epoch": 3.613577125072943,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037311726129136665,
      "loss": 4.9228,
      "mean_token_accuracy": 0.20712907314300538,
      "num_tokens": 80551903.0,
      "step": 46445
    },
    {
      "entropy": 5.381314611434936,
      "epoch": 3.6139661544446606,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003730923190266038,
      "loss": 4.8995,
      "mean_token_accuracy": 0.20371965318918228,
      "num_tokens": 80559260.0,
      "step": 46450
    },
    {
      "entropy": 5.3936299800872805,
      "epoch": 3.614355183816378,
      "grad_norm": 1.140625,
      "learning_rate": 0.000373067375273446,
      "loss": 4.9095,
      "mean_token_accuracy": 0.21196842938661575,
      "num_tokens": 80568689.0,
      "step": 46455
    },
    {
      "entropy": 5.383879470825195,
      "epoch": 3.6147442131880956,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037304243003227197,
      "loss": 4.8974,
      "mean_token_accuracy": 0.20314313620328903,
      "num_tokens": 80577073.0,
      "step": 46460
    },
    {
      "entropy": 5.304295635223388,
      "epoch": 3.6151332425598133,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003730174833034601,
      "loss": 4.7857,
      "mean_token_accuracy": 0.21240022033452988,
      "num_tokens": 80586595.0,
      "step": 46465
    },
    {
      "entropy": 5.343378257751465,
      "epoch": 3.615522271931531,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037299253508738894,
      "loss": 4.84,
      "mean_token_accuracy": 0.2106585130095482,
      "num_tokens": 80595074.0,
      "step": 46470
    },
    {
      "entropy": 5.354985046386719,
      "epoch": 3.6159113013032487,
      "grad_norm": 1.25,
      "learning_rate": 0.00037296758538443703,
      "loss": 4.9096,
      "mean_token_accuracy": 0.20369569212198257,
      "num_tokens": 80603801.0,
      "step": 46475
    },
    {
      "entropy": 5.436268281936646,
      "epoch": 3.616300330674966,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037294263419498295,
      "loss": 4.9359,
      "mean_token_accuracy": 0.2146188110113144,
      "num_tokens": 80612993.0,
      "step": 46480
    },
    {
      "entropy": 5.352521228790283,
      "epoch": 3.6166893600466836,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003729176815194053,
      "loss": 4.7808,
      "mean_token_accuracy": 0.22187235057353974,
      "num_tokens": 80621712.0,
      "step": 46485
    },
    {
      "entropy": 5.391929578781128,
      "epoch": 3.617078389418401,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037289272735808267,
      "loss": 4.8546,
      "mean_token_accuracy": 0.21278309524059297,
      "num_tokens": 80630657.0,
      "step": 46490
    },
    {
      "entropy": 5.378232431411743,
      "epoch": 3.6174674187901186,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003728677717113936,
      "loss": 4.8815,
      "mean_token_accuracy": 0.2113603264093399,
      "num_tokens": 80638967.0,
      "step": 46495
    },
    {
      "entropy": 5.31564884185791,
      "epoch": 3.6178564481618363,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037284281457971707,
      "loss": 4.8226,
      "mean_token_accuracy": 0.21725354790687562,
      "num_tokens": 80647791.0,
      "step": 46500
    },
    {
      "entropy": 5.346744966506958,
      "epoch": 3.618245477533554,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003728178559634315,
      "loss": 4.7869,
      "mean_token_accuracy": 0.22069291770458221,
      "num_tokens": 80656162.0,
      "step": 46505
    },
    {
      "entropy": 5.330867958068848,
      "epoch": 3.618634506905271,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003727928958629156,
      "loss": 4.8702,
      "mean_token_accuracy": 0.2125949129462242,
      "num_tokens": 80665203.0,
      "step": 46510
    },
    {
      "entropy": 5.3909399032592775,
      "epoch": 3.619023536276989,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037276793427854817,
      "loss": 4.8789,
      "mean_token_accuracy": 0.2084796130657196,
      "num_tokens": 80673748.0,
      "step": 46515
    },
    {
      "entropy": 5.324099063873291,
      "epoch": 3.6194125656487066,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037274297121070796,
      "loss": 4.7909,
      "mean_token_accuracy": 0.22222814857959747,
      "num_tokens": 80682166.0,
      "step": 46520
    },
    {
      "entropy": 5.372356986999511,
      "epoch": 3.619801595020424,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003727180066597737,
      "loss": 4.816,
      "mean_token_accuracy": 0.2135859966278076,
      "num_tokens": 80689972.0,
      "step": 46525
    },
    {
      "entropy": 5.336402034759521,
      "epoch": 3.6201906243921416,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003726930406261243,
      "loss": 4.8523,
      "mean_token_accuracy": 0.20832593888044357,
      "num_tokens": 80698193.0,
      "step": 46530
    },
    {
      "entropy": 5.391450262069702,
      "epoch": 3.6205796537638593,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003726680731101385,
      "loss": 4.9192,
      "mean_token_accuracy": 0.20272493064403535,
      "num_tokens": 80707464.0,
      "step": 46535
    },
    {
      "entropy": 5.379547595977783,
      "epoch": 3.620968683135577,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003726431041121951,
      "loss": 4.8162,
      "mean_token_accuracy": 0.21616086959838868,
      "num_tokens": 80715185.0,
      "step": 46540
    },
    {
      "entropy": 5.355844163894654,
      "epoch": 3.621357712507294,
      "grad_norm": 1.203125,
      "learning_rate": 0.000372618133632673,
      "loss": 4.8541,
      "mean_token_accuracy": 0.20428049117326735,
      "num_tokens": 80723499.0,
      "step": 46545
    },
    {
      "entropy": 5.40355544090271,
      "epoch": 3.621746741879012,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003725931616719511,
      "loss": 4.8275,
      "mean_token_accuracy": 0.21575037389993668,
      "num_tokens": 80732539.0,
      "step": 46550
    },
    {
      "entropy": 5.482115745544434,
      "epoch": 3.622135771250729,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003725681882304084,
      "loss": 4.9129,
      "mean_token_accuracy": 0.20281482338905335,
      "num_tokens": 80742332.0,
      "step": 46555
    },
    {
      "entropy": 5.321111583709717,
      "epoch": 3.622524800622447,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003725432133084237,
      "loss": 4.7989,
      "mean_token_accuracy": 0.22250897735357283,
      "num_tokens": 80750553.0,
      "step": 46560
    },
    {
      "entropy": 5.373487377166748,
      "epoch": 3.6229138299941646,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003725182369063759,
      "loss": 4.8442,
      "mean_token_accuracy": 0.214154152572155,
      "num_tokens": 80758721.0,
      "step": 46565
    },
    {
      "entropy": 5.360918617248535,
      "epoch": 3.6233028593658823,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003724932590246441,
      "loss": 4.8852,
      "mean_token_accuracy": 0.21329243183135987,
      "num_tokens": 80768242.0,
      "step": 46570
    },
    {
      "entropy": 5.315581560134888,
      "epoch": 3.6236918887376,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003724682796636073,
      "loss": 4.7549,
      "mean_token_accuracy": 0.21743698120117189,
      "num_tokens": 80776494.0,
      "step": 46575
    },
    {
      "entropy": 5.300078439712524,
      "epoch": 3.624080918109317,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003724432988236445,
      "loss": 4.7864,
      "mean_token_accuracy": 0.21626759320497513,
      "num_tokens": 80785198.0,
      "step": 46580
    },
    {
      "entropy": 5.352545738220215,
      "epoch": 3.624469947481035,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003724183165051347,
      "loss": 4.9047,
      "mean_token_accuracy": 0.21085509061813354,
      "num_tokens": 80794126.0,
      "step": 46585
    },
    {
      "entropy": 5.324484348297119,
      "epoch": 3.624858976852752,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037239333270845714,
      "loss": 4.7743,
      "mean_token_accuracy": 0.21440768539905547,
      "num_tokens": 80802199.0,
      "step": 46590
    },
    {
      "entropy": 5.314590978622436,
      "epoch": 3.62524800622447,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003723683474339907,
      "loss": 4.8137,
      "mean_token_accuracy": 0.21393654197454454,
      "num_tokens": 80810537.0,
      "step": 46595
    },
    {
      "entropy": 5.299376487731934,
      "epoch": 3.6256370355961876,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003723433606821145,
      "loss": 4.8571,
      "mean_token_accuracy": 0.21834296882152557,
      "num_tokens": 80818661.0,
      "step": 46600
    },
    {
      "entropy": 5.28003306388855,
      "epoch": 3.6260260649679052,
      "grad_norm": 1.125,
      "learning_rate": 0.00037231837245320777,
      "loss": 4.7649,
      "mean_token_accuracy": 0.22352368682622908,
      "num_tokens": 80827883.0,
      "step": 46605
    },
    {
      "entropy": 5.406387805938721,
      "epoch": 3.6264150943396225,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037229338274764965,
      "loss": 4.9048,
      "mean_token_accuracy": 0.21257759183645247,
      "num_tokens": 80836692.0,
      "step": 46610
    },
    {
      "entropy": 5.4556131839752195,
      "epoch": 3.62680412371134,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003722683915658193,
      "loss": 4.879,
      "mean_token_accuracy": 0.2154424488544464,
      "num_tokens": 80845342.0,
      "step": 46615
    },
    {
      "entropy": 5.280368709564209,
      "epoch": 3.627193153083058,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037224339890809587,
      "loss": 4.7535,
      "mean_token_accuracy": 0.21835696697235107,
      "num_tokens": 80854393.0,
      "step": 46620
    },
    {
      "entropy": 5.363966655731201,
      "epoch": 3.627582182454775,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003722184047748587,
      "loss": 4.7853,
      "mean_token_accuracy": 0.22336314171552657,
      "num_tokens": 80862712.0,
      "step": 46625
    },
    {
      "entropy": 5.383457040786743,
      "epoch": 3.627971211826493,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037219340916648703,
      "loss": 4.8996,
      "mean_token_accuracy": 0.20361993163824083,
      "num_tokens": 80872455.0,
      "step": 46630
    },
    {
      "entropy": 5.356365919113159,
      "epoch": 3.6283602411982105,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037216841208336,
      "loss": 4.824,
      "mean_token_accuracy": 0.21407736241817474,
      "num_tokens": 80881305.0,
      "step": 46635
    },
    {
      "entropy": 5.388237857818604,
      "epoch": 3.6287492705699282,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037214341352585697,
      "loss": 4.8638,
      "mean_token_accuracy": 0.22094998955726625,
      "num_tokens": 80890242.0,
      "step": 46640
    },
    {
      "entropy": 5.419369411468506,
      "epoch": 3.6291382999416455,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003721184134943573,
      "loss": 4.9486,
      "mean_token_accuracy": 0.21141682118177413,
      "num_tokens": 80899707.0,
      "step": 46645
    },
    {
      "entropy": 5.346998405456543,
      "epoch": 3.629527329313363,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037209341198924024,
      "loss": 4.8254,
      "mean_token_accuracy": 0.21290561705827712,
      "num_tokens": 80908543.0,
      "step": 46650
    },
    {
      "entropy": 5.363423919677734,
      "epoch": 3.6299163586850804,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003720684090108852,
      "loss": 4.9203,
      "mean_token_accuracy": 0.20853622555732726,
      "num_tokens": 80916916.0,
      "step": 46655
    },
    {
      "entropy": 5.426221561431885,
      "epoch": 3.630305388056798,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003720434045596716,
      "loss": 4.9877,
      "mean_token_accuracy": 0.19575837403535842,
      "num_tokens": 80925782.0,
      "step": 46660
    },
    {
      "entropy": 5.358912420272827,
      "epoch": 3.630694417428516,
      "grad_norm": 1.125,
      "learning_rate": 0.0003720183986359788,
      "loss": 4.758,
      "mean_token_accuracy": 0.2226752057671547,
      "num_tokens": 80933902.0,
      "step": 46665
    },
    {
      "entropy": 5.372754669189453,
      "epoch": 3.6310834468002335,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003719933912401862,
      "loss": 4.9117,
      "mean_token_accuracy": 0.2023972138762474,
      "num_tokens": 80941993.0,
      "step": 46670
    },
    {
      "entropy": 5.322053050994873,
      "epoch": 3.6314724761719512,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003719683823726732,
      "loss": 4.766,
      "mean_token_accuracy": 0.2157853960990906,
      "num_tokens": 80950038.0,
      "step": 46675
    },
    {
      "entropy": 5.386645269393921,
      "epoch": 3.6318615055436685,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037194337203381945,
      "loss": 4.8961,
      "mean_token_accuracy": 0.20981784760951996,
      "num_tokens": 80958684.0,
      "step": 46680
    },
    {
      "entropy": 5.393042516708374,
      "epoch": 3.632250534915386,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037191836022400427,
      "loss": 4.8608,
      "mean_token_accuracy": 0.2094475194811821,
      "num_tokens": 80967302.0,
      "step": 46685
    },
    {
      "entropy": 5.322162342071533,
      "epoch": 3.6326395642871034,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037189334694360724,
      "loss": 4.8696,
      "mean_token_accuracy": 0.2123696208000183,
      "num_tokens": 80975691.0,
      "step": 46690
    },
    {
      "entropy": 5.297226762771606,
      "epoch": 3.633028593658821,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003718683321930079,
      "loss": 4.8436,
      "mean_token_accuracy": 0.21797813028097152,
      "num_tokens": 80984282.0,
      "step": 46695
    },
    {
      "entropy": 5.472274875640869,
      "epoch": 3.633417623030539,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003718433159725857,
      "loss": 4.8731,
      "mean_token_accuracy": 0.21265694946050645,
      "num_tokens": 80992991.0,
      "step": 46700
    },
    {
      "entropy": 5.383316993713379,
      "epoch": 3.6338066524022565,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003718182982827205,
      "loss": 4.8763,
      "mean_token_accuracy": 0.20865497142076492,
      "num_tokens": 81001263.0,
      "step": 46705
    },
    {
      "entropy": 5.317346000671387,
      "epoch": 3.634195681773974,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037179327912379166,
      "loss": 4.8195,
      "mean_token_accuracy": 0.21988535374403,
      "num_tokens": 81010680.0,
      "step": 46710
    },
    {
      "entropy": 5.30451455116272,
      "epoch": 3.6345847111456915,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037176825849617886,
      "loss": 4.7953,
      "mean_token_accuracy": 0.21292907297611235,
      "num_tokens": 81019359.0,
      "step": 46715
    },
    {
      "entropy": 5.453900671005249,
      "epoch": 3.634973740517409,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037174323640026173,
      "loss": 4.9784,
      "mean_token_accuracy": 0.2042897954583168,
      "num_tokens": 81028808.0,
      "step": 46720
    },
    {
      "entropy": 5.406426048278808,
      "epoch": 3.6353627698891264,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037171821283642003,
      "loss": 4.8497,
      "mean_token_accuracy": 0.2067391097545624,
      "num_tokens": 81037778.0,
      "step": 46725
    },
    {
      "entropy": 5.413697624206543,
      "epoch": 3.635751799260844,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003716931878050334,
      "loss": 4.8779,
      "mean_token_accuracy": 0.20466281622648239,
      "num_tokens": 81047657.0,
      "step": 46730
    },
    {
      "entropy": 5.294223976135254,
      "epoch": 3.636140828632562,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037166816130648143,
      "loss": 4.7521,
      "mean_token_accuracy": 0.21895309686660766,
      "num_tokens": 81056585.0,
      "step": 46735
    },
    {
      "entropy": 5.3575067043304445,
      "epoch": 3.6365298580042795,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037164313334114407,
      "loss": 4.9044,
      "mean_token_accuracy": 0.2057991772890091,
      "num_tokens": 81064463.0,
      "step": 46740
    },
    {
      "entropy": 5.324848222732544,
      "epoch": 3.636918887375997,
      "grad_norm": 1.140625,
      "learning_rate": 0.000371618103909401,
      "loss": 4.9229,
      "mean_token_accuracy": 0.2141909584403038,
      "num_tokens": 81073210.0,
      "step": 46745
    },
    {
      "entropy": 5.490639495849609,
      "epoch": 3.6373079167477145,
      "grad_norm": 1.21875,
      "learning_rate": 0.000371593073011632,
      "loss": 4.9712,
      "mean_token_accuracy": 0.21098536849021912,
      "num_tokens": 81081991.0,
      "step": 46750
    },
    {
      "entropy": 5.495748853683471,
      "epoch": 3.637696946119432,
      "grad_norm": 1.125,
      "learning_rate": 0.00037156804064821674,
      "loss": 4.9506,
      "mean_token_accuracy": 0.20795520544052123,
      "num_tokens": 81090592.0,
      "step": 46755
    },
    {
      "entropy": 5.36137170791626,
      "epoch": 3.6380859754911494,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037154300681953523,
      "loss": 4.8858,
      "mean_token_accuracy": 0.20915915668010712,
      "num_tokens": 81099278.0,
      "step": 46760
    },
    {
      "entropy": 5.3598490238189695,
      "epoch": 3.638475004862867,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003715179715259673,
      "loss": 4.8456,
      "mean_token_accuracy": 0.21495743840932846,
      "num_tokens": 81108106.0,
      "step": 46765
    },
    {
      "entropy": 5.353936767578125,
      "epoch": 3.638864034234585,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037149293476789277,
      "loss": 4.7862,
      "mean_token_accuracy": 0.2128440871834755,
      "num_tokens": 81116679.0,
      "step": 46770
    },
    {
      "entropy": 5.41002402305603,
      "epoch": 3.6392530636063025,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003714678965456915,
      "loss": 4.9084,
      "mean_token_accuracy": 0.2040422111749649,
      "num_tokens": 81125308.0,
      "step": 46775
    },
    {
      "entropy": 5.437566232681275,
      "epoch": 3.6396420929780198,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037144285685974346,
      "loss": 4.9207,
      "mean_token_accuracy": 0.21176941096782684,
      "num_tokens": 81134092.0,
      "step": 46780
    },
    {
      "entropy": 5.425619983673096,
      "epoch": 3.6400311223497375,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037141781571042854,
      "loss": 4.8821,
      "mean_token_accuracy": 0.2100697159767151,
      "num_tokens": 81142413.0,
      "step": 46785
    },
    {
      "entropy": 5.36117639541626,
      "epoch": 3.6404201517214547,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037139277309812676,
      "loss": 4.876,
      "mean_token_accuracy": 0.2071504056453705,
      "num_tokens": 81150952.0,
      "step": 46790
    },
    {
      "entropy": 5.412106990814209,
      "epoch": 3.6408091810931724,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003713677290232181,
      "loss": 4.952,
      "mean_token_accuracy": 0.20451395511627196,
      "num_tokens": 81161118.0,
      "step": 46795
    },
    {
      "entropy": 5.384509611129761,
      "epoch": 3.64119821046489,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00037134268348608255,
      "loss": 4.8533,
      "mean_token_accuracy": 0.21169256567955017,
      "num_tokens": 81169682.0,
      "step": 46800
    },
    {
      "entropy": 5.42182502746582,
      "epoch": 3.641587239836608,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037131763648710003,
      "loss": 4.8368,
      "mean_token_accuracy": 0.21508550643920898,
      "num_tokens": 81177892.0,
      "step": 46805
    },
    {
      "entropy": 5.361674642562866,
      "epoch": 3.641976269208325,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037129258802665076,
      "loss": 4.8629,
      "mean_token_accuracy": 0.21256581991910933,
      "num_tokens": 81186612.0,
      "step": 46810
    },
    {
      "entropy": 5.304353952407837,
      "epoch": 3.6423652985800428,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003712675381051147,
      "loss": 4.8404,
      "mean_token_accuracy": 0.20409987568855287,
      "num_tokens": 81195518.0,
      "step": 46815
    },
    {
      "entropy": 5.468058300018311,
      "epoch": 3.6427543279517605,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000371242486722872,
      "loss": 5.0134,
      "mean_token_accuracy": 0.20576711893081664,
      "num_tokens": 81205105.0,
      "step": 46820
    },
    {
      "entropy": 5.49623293876648,
      "epoch": 3.6431433573234777,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003712174338803027,
      "loss": 4.8946,
      "mean_token_accuracy": 0.2026061534881592,
      "num_tokens": 81213629.0,
      "step": 46825
    },
    {
      "entropy": 5.395466470718384,
      "epoch": 3.6435323866951954,
      "grad_norm": 1.109375,
      "learning_rate": 0.000371192379577787,
      "loss": 4.8537,
      "mean_token_accuracy": 0.20984504222869874,
      "num_tokens": 81222387.0,
      "step": 46830
    },
    {
      "entropy": 5.219000673294067,
      "epoch": 3.643921416066913,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003711673238157051,
      "loss": 4.7402,
      "mean_token_accuracy": 0.228009395301342,
      "num_tokens": 81231280.0,
      "step": 46835
    },
    {
      "entropy": 5.399661874771118,
      "epoch": 3.644310445438631,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037114226659443704,
      "loss": 4.9069,
      "mean_token_accuracy": 0.20749039947986603,
      "num_tokens": 81239681.0,
      "step": 46840
    },
    {
      "entropy": 5.375925827026367,
      "epoch": 3.644699474810348,
      "grad_norm": 1.25,
      "learning_rate": 0.0003711172079143632,
      "loss": 4.9007,
      "mean_token_accuracy": 0.20477538853883742,
      "num_tokens": 81247909.0,
      "step": 46845
    },
    {
      "entropy": 5.35288372039795,
      "epoch": 3.6450885041820658,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003710921477758637,
      "loss": 4.8418,
      "mean_token_accuracy": 0.2095668375492096,
      "num_tokens": 81256507.0,
      "step": 46850
    },
    {
      "entropy": 5.410383844375611,
      "epoch": 3.6454775335537835,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003710670861793187,
      "loss": 5.028,
      "mean_token_accuracy": 0.2004512906074524,
      "num_tokens": 81265301.0,
      "step": 46855
    },
    {
      "entropy": 5.399573278427124,
      "epoch": 3.6458665629255007,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003710420231251087,
      "loss": 4.8193,
      "mean_token_accuracy": 0.2150818407535553,
      "num_tokens": 81273968.0,
      "step": 46860
    },
    {
      "entropy": 5.39545316696167,
      "epoch": 3.6462555922972184,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037101695861361384,
      "loss": 4.8666,
      "mean_token_accuracy": 0.20519386231899261,
      "num_tokens": 81283511.0,
      "step": 46865
    },
    {
      "entropy": 5.364624309539795,
      "epoch": 3.646644621668936,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037099189264521446,
      "loss": 4.8585,
      "mean_token_accuracy": 0.2102563723921776,
      "num_tokens": 81292399.0,
      "step": 46870
    },
    {
      "entropy": 5.334155416488647,
      "epoch": 3.647033651040654,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003709668252202909,
      "loss": 4.909,
      "mean_token_accuracy": 0.20831002593040465,
      "num_tokens": 81300670.0,
      "step": 46875
    },
    {
      "entropy": 5.37366828918457,
      "epoch": 3.647422680412371,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003709417563392235,
      "loss": 4.8265,
      "mean_token_accuracy": 0.21570503413677217,
      "num_tokens": 81309419.0,
      "step": 46880
    },
    {
      "entropy": 5.398037099838257,
      "epoch": 3.6478117097840888,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003709166860023926,
      "loss": 4.8604,
      "mean_token_accuracy": 0.21377164572477342,
      "num_tokens": 81318349.0,
      "step": 46885
    },
    {
      "entropy": 5.356763887405395,
      "epoch": 3.648200739155806,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037089161421017874,
      "loss": 4.8619,
      "mean_token_accuracy": 0.21135079711675644,
      "num_tokens": 81326715.0,
      "step": 46890
    },
    {
      "entropy": 5.285213422775269,
      "epoch": 3.6485897685275237,
      "grad_norm": 1.125,
      "learning_rate": 0.00037086654096296217,
      "loss": 4.7673,
      "mean_token_accuracy": 0.22384574711322786,
      "num_tokens": 81335113.0,
      "step": 46895
    },
    {
      "entropy": 5.453222179412842,
      "epoch": 3.6489787978992414,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003708414662611235,
      "loss": 4.9635,
      "mean_token_accuracy": 0.20365542322397232,
      "num_tokens": 81343928.0,
      "step": 46900
    },
    {
      "entropy": 5.377603530883789,
      "epoch": 3.649367827270959,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003708163901050432,
      "loss": 4.8164,
      "mean_token_accuracy": 0.21860815435647965,
      "num_tokens": 81352244.0,
      "step": 46905
    },
    {
      "entropy": 5.354126405715943,
      "epoch": 3.649756856642677,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037079131249510155,
      "loss": 4.8796,
      "mean_token_accuracy": 0.21360485553741454,
      "num_tokens": 81361769.0,
      "step": 46910
    },
    {
      "entropy": 5.382349157333374,
      "epoch": 3.650145886014394,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003707662334316793,
      "loss": 4.816,
      "mean_token_accuracy": 0.21640096008777618,
      "num_tokens": 81369936.0,
      "step": 46915
    },
    {
      "entropy": 5.368713855743408,
      "epoch": 3.6505349153861117,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003707411529151568,
      "loss": 4.8638,
      "mean_token_accuracy": 0.21717785000801088,
      "num_tokens": 81379846.0,
      "step": 46920
    },
    {
      "entropy": 5.362018394470215,
      "epoch": 3.650923944757829,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003707160709459147,
      "loss": 4.8188,
      "mean_token_accuracy": 0.21400633752346038,
      "num_tokens": 81388123.0,
      "step": 46925
    },
    {
      "entropy": 5.44135799407959,
      "epoch": 3.6513129741295467,
      "grad_norm": 1.125,
      "learning_rate": 0.0003706909875243336,
      "loss": 4.9276,
      "mean_token_accuracy": 0.2067818373441696,
      "num_tokens": 81397000.0,
      "step": 46930
    },
    {
      "entropy": 5.38765025138855,
      "epoch": 3.6517020035012644,
      "grad_norm": 1.109375,
      "learning_rate": 0.000370665902650794,
      "loss": 4.9191,
      "mean_token_accuracy": 0.2051563248038292,
      "num_tokens": 81406218.0,
      "step": 46935
    },
    {
      "entropy": 5.389194583892822,
      "epoch": 3.652091032872982,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037064081632567675,
      "loss": 4.7837,
      "mean_token_accuracy": 0.21756212711334227,
      "num_tokens": 81414934.0,
      "step": 46940
    },
    {
      "entropy": 5.3309876918792725,
      "epoch": 3.6524800622446993,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037061572854936225,
      "loss": 4.769,
      "mean_token_accuracy": 0.2158322274684906,
      "num_tokens": 81423345.0,
      "step": 46945
    },
    {
      "entropy": 5.344428634643554,
      "epoch": 3.652869091616417,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003705906393222313,
      "loss": 4.8029,
      "mean_token_accuracy": 0.21828072369098664,
      "num_tokens": 81431858.0,
      "step": 46950
    },
    {
      "entropy": 5.438388013839722,
      "epoch": 3.6532581209881347,
      "grad_norm": 1.125,
      "learning_rate": 0.0003705655486446645,
      "loss": 4.9306,
      "mean_token_accuracy": 0.2076655849814415,
      "num_tokens": 81440538.0,
      "step": 46955
    },
    {
      "entropy": 5.480002212524414,
      "epoch": 3.653647150359852,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003705404565170426,
      "loss": 4.9274,
      "mean_token_accuracy": 0.21332752406597139,
      "num_tokens": 81449187.0,
      "step": 46960
    },
    {
      "entropy": 5.356939315795898,
      "epoch": 3.6540361797315697,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003705153629397463,
      "loss": 4.746,
      "mean_token_accuracy": 0.21789834648370743,
      "num_tokens": 81456708.0,
      "step": 46965
    },
    {
      "entropy": 5.361954927444458,
      "epoch": 3.6544252091032874,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037049026791315646,
      "loss": 4.8838,
      "mean_token_accuracy": 0.20844581425189973,
      "num_tokens": 81464933.0,
      "step": 46970
    },
    {
      "entropy": 5.3753955364227295,
      "epoch": 3.654814238475005,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003704651714376538,
      "loss": 4.9295,
      "mean_token_accuracy": 0.21061077266931533,
      "num_tokens": 81473846.0,
      "step": 46975
    },
    {
      "entropy": 5.362191104888916,
      "epoch": 3.6552032678467223,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003704400735136191,
      "loss": 4.7681,
      "mean_token_accuracy": 0.22042788714170455,
      "num_tokens": 81481928.0,
      "step": 46980
    },
    {
      "entropy": 5.427572345733642,
      "epoch": 3.65559229721844,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003704149741414332,
      "loss": 4.9871,
      "mean_token_accuracy": 0.19762064069509505,
      "num_tokens": 81490913.0,
      "step": 46985
    },
    {
      "entropy": 5.463778257369995,
      "epoch": 3.6559813265901573,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003703898733214769,
      "loss": 4.942,
      "mean_token_accuracy": 0.20737839341163636,
      "num_tokens": 81499779.0,
      "step": 46990
    },
    {
      "entropy": 5.324742412567138,
      "epoch": 3.656370355961875,
      "grad_norm": 1.125,
      "learning_rate": 0.0003703647710541311,
      "loss": 4.8302,
      "mean_token_accuracy": 0.21601635068655015,
      "num_tokens": 81507707.0,
      "step": 46995
    },
    {
      "entropy": 5.318167304992675,
      "epoch": 3.6567593853335927,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003703396673397768,
      "loss": 4.848,
      "mean_token_accuracy": 0.21239372193813325,
      "num_tokens": 81516418.0,
      "step": 47000
    },
    {
      "entropy": 5.38788104057312,
      "epoch": 3.6571484147053104,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003703145621787947,
      "loss": 4.8326,
      "mean_token_accuracy": 0.21541489958763121,
      "num_tokens": 81525218.0,
      "step": 47005
    },
    {
      "entropy": 5.28867917060852,
      "epoch": 3.657537444077028,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037028945557156587,
      "loss": 4.7487,
      "mean_token_accuracy": 0.22531513571739198,
      "num_tokens": 81533739.0,
      "step": 47010
    },
    {
      "entropy": 5.315992832183838,
      "epoch": 3.6579264734487453,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037026434751847117,
      "loss": 4.8126,
      "mean_token_accuracy": 0.21803421080112456,
      "num_tokens": 81541967.0,
      "step": 47015
    },
    {
      "entropy": 5.38592209815979,
      "epoch": 3.658315502820463,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037023923801989163,
      "loss": 4.8622,
      "mean_token_accuracy": 0.20761808902025222,
      "num_tokens": 81549892.0,
      "step": 47020
    },
    {
      "entropy": 5.346289110183716,
      "epoch": 3.6587045321921803,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003702141270762083,
      "loss": 4.8172,
      "mean_token_accuracy": 0.2110361784696579,
      "num_tokens": 81558441.0,
      "step": 47025
    },
    {
      "entropy": 5.4026021480560305,
      "epoch": 3.659093561563898,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037018901468780207,
      "loss": 4.9364,
      "mean_token_accuracy": 0.2145633652806282,
      "num_tokens": 81566753.0,
      "step": 47030
    },
    {
      "entropy": 5.471596193313599,
      "epoch": 3.6594825909356157,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003701639008550541,
      "loss": 4.9614,
      "mean_token_accuracy": 0.20758402794599534,
      "num_tokens": 81575038.0,
      "step": 47035
    },
    {
      "entropy": 5.376137733459473,
      "epoch": 3.6598716203073334,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003701387855783454,
      "loss": 4.8239,
      "mean_token_accuracy": 0.21505968719720842,
      "num_tokens": 81584016.0,
      "step": 47040
    },
    {
      "entropy": 5.425729417800904,
      "epoch": 3.6602606496790506,
      "grad_norm": 1.1875,
      "learning_rate": 0.000370113668858057,
      "loss": 4.8672,
      "mean_token_accuracy": 0.2121990904211998,
      "num_tokens": 81592757.0,
      "step": 47045
    },
    {
      "entropy": 5.3454344272613525,
      "epoch": 3.6606496790507683,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003700885506945701,
      "loss": 4.857,
      "mean_token_accuracy": 0.2155336007475853,
      "num_tokens": 81601225.0,
      "step": 47050
    },
    {
      "entropy": 5.391754627227783,
      "epoch": 3.661038708422486,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003700634310882658,
      "loss": 4.9223,
      "mean_token_accuracy": 0.2033282607793808,
      "num_tokens": 81610043.0,
      "step": 47055
    },
    {
      "entropy": 5.4308788776397705,
      "epoch": 3.6614277377942033,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037003831003952525,
      "loss": 4.845,
      "mean_token_accuracy": 0.2185583606362343,
      "num_tokens": 81618540.0,
      "step": 47060
    },
    {
      "entropy": 5.3553272724151615,
      "epoch": 3.661816767165921,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037001318754872957,
      "loss": 4.8034,
      "mean_token_accuracy": 0.22531151175498962,
      "num_tokens": 81626682.0,
      "step": 47065
    },
    {
      "entropy": 5.260670137405396,
      "epoch": 3.6622057965376387,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036998806361626,
      "loss": 4.8731,
      "mean_token_accuracy": 0.2156827598810196,
      "num_tokens": 81635917.0,
      "step": 47070
    },
    {
      "entropy": 5.301696825027466,
      "epoch": 3.6625948259093564,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003699629382424979,
      "loss": 4.8204,
      "mean_token_accuracy": 0.21608836501836776,
      "num_tokens": 81644306.0,
      "step": 47075
    },
    {
      "entropy": 5.430362844467163,
      "epoch": 3.6629838552810736,
      "grad_norm": 1.125,
      "learning_rate": 0.00036993781142782416,
      "loss": 4.8949,
      "mean_token_accuracy": 0.20389288812875747,
      "num_tokens": 81653207.0,
      "step": 47080
    },
    {
      "entropy": 5.429917192459106,
      "epoch": 3.6633728846527913,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036991268317262033,
      "loss": 4.8749,
      "mean_token_accuracy": 0.2133514925837517,
      "num_tokens": 81661126.0,
      "step": 47085
    },
    {
      "entropy": 5.334979772567749,
      "epoch": 3.663761914024509,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036988755347726766,
      "loss": 4.8369,
      "mean_token_accuracy": 0.2151196852326393,
      "num_tokens": 81670721.0,
      "step": 47090
    },
    {
      "entropy": 5.367210721969604,
      "epoch": 3.6641509433962263,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036986242234214736,
      "loss": 4.9304,
      "mean_token_accuracy": 0.20234448909759523,
      "num_tokens": 81680281.0,
      "step": 47095
    },
    {
      "entropy": 5.41204981803894,
      "epoch": 3.664539972767944,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036983728976764074,
      "loss": 4.825,
      "mean_token_accuracy": 0.21472640633583068,
      "num_tokens": 81688727.0,
      "step": 47100
    },
    {
      "entropy": 5.3155848503112795,
      "epoch": 3.6649290021396617,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036981215575412923,
      "loss": 4.8198,
      "mean_token_accuracy": 0.2170701116323471,
      "num_tokens": 81697903.0,
      "step": 47105
    },
    {
      "entropy": 5.31125545501709,
      "epoch": 3.6653180315113794,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003697870203019942,
      "loss": 4.8554,
      "mean_token_accuracy": 0.21162189543247223,
      "num_tokens": 81705712.0,
      "step": 47110
    },
    {
      "entropy": 5.397530841827392,
      "epoch": 3.6657070608830966,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036976188341161694,
      "loss": 4.9079,
      "mean_token_accuracy": 0.21045473366975784,
      "num_tokens": 81714787.0,
      "step": 47115
    },
    {
      "entropy": 5.355281972885132,
      "epoch": 3.6660960902548143,
      "grad_norm": 1.125,
      "learning_rate": 0.0003697367450833789,
      "loss": 4.8537,
      "mean_token_accuracy": 0.2026073768734932,
      "num_tokens": 81723891.0,
      "step": 47120
    },
    {
      "entropy": 5.299651098251343,
      "epoch": 3.6664851196265316,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036971160531766155,
      "loss": 4.7882,
      "mean_token_accuracy": 0.2148401767015457,
      "num_tokens": 81733054.0,
      "step": 47125
    },
    {
      "entropy": 5.342397403717041,
      "epoch": 3.6668741489982493,
      "grad_norm": 1.125,
      "learning_rate": 0.0003696864641148465,
      "loss": 4.8464,
      "mean_token_accuracy": 0.2183964654803276,
      "num_tokens": 81742065.0,
      "step": 47130
    },
    {
      "entropy": 5.325996446609497,
      "epoch": 3.667263178369967,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036966132147531487,
      "loss": 4.8359,
      "mean_token_accuracy": 0.22014235109090804,
      "num_tokens": 81751011.0,
      "step": 47135
    },
    {
      "entropy": 5.430106687545776,
      "epoch": 3.6676522077416847,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036963617739944844,
      "loss": 4.9158,
      "mean_token_accuracy": 0.21342725306749344,
      "num_tokens": 81759864.0,
      "step": 47140
    },
    {
      "entropy": 5.388666200637817,
      "epoch": 3.668041237113402,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036961103188762866,
      "loss": 4.8818,
      "mean_token_accuracy": 0.20579354017972945,
      "num_tokens": 81769420.0,
      "step": 47145
    },
    {
      "entropy": 5.297831201553345,
      "epoch": 3.6684302664851196,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036958588494023705,
      "loss": 4.8027,
      "mean_token_accuracy": 0.21736599057912825,
      "num_tokens": 81777544.0,
      "step": 47150
    },
    {
      "entropy": 5.332208824157715,
      "epoch": 3.6688192958568373,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003695607365576551,
      "loss": 4.7744,
      "mean_token_accuracy": 0.2245257467031479,
      "num_tokens": 81785418.0,
      "step": 47155
    },
    {
      "entropy": 5.32034649848938,
      "epoch": 3.6692083252285546,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003695355867402646,
      "loss": 4.88,
      "mean_token_accuracy": 0.21258388310670853,
      "num_tokens": 81794584.0,
      "step": 47160
    },
    {
      "entropy": 5.322759437561035,
      "epoch": 3.6695973546002723,
      "grad_norm": 1.125,
      "learning_rate": 0.000369510435488447,
      "loss": 4.8047,
      "mean_token_accuracy": 0.21999323666095733,
      "num_tokens": 81802950.0,
      "step": 47165
    },
    {
      "entropy": 5.338032007217407,
      "epoch": 3.66998638397199,
      "grad_norm": 1.125,
      "learning_rate": 0.000369485282802584,
      "loss": 4.7593,
      "mean_token_accuracy": 0.2175430417060852,
      "num_tokens": 81812092.0,
      "step": 47170
    },
    {
      "entropy": 5.417027711868286,
      "epoch": 3.6703754133437076,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036946012868305716,
      "loss": 4.9271,
      "mean_token_accuracy": 0.20586876422166825,
      "num_tokens": 81821437.0,
      "step": 47175
    },
    {
      "entropy": 5.378199195861816,
      "epoch": 3.670764442715425,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003694349731302483,
      "loss": 4.9328,
      "mean_token_accuracy": 0.20728449523448944,
      "num_tokens": 81830207.0,
      "step": 47180
    },
    {
      "entropy": 5.446003437042236,
      "epoch": 3.6711534720871426,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003694098161445389,
      "loss": 4.9787,
      "mean_token_accuracy": 0.20525303930044175,
      "num_tokens": 81838664.0,
      "step": 47185
    },
    {
      "entropy": 5.336422634124756,
      "epoch": 3.6715425014588603,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003693846577263109,
      "loss": 4.8433,
      "mean_token_accuracy": 0.21650296002626418,
      "num_tokens": 81846606.0,
      "step": 47190
    },
    {
      "entropy": 5.380356645584106,
      "epoch": 3.6719315308305776,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003693594978759459,
      "loss": 4.954,
      "mean_token_accuracy": 0.20728806406259537,
      "num_tokens": 81855925.0,
      "step": 47195
    },
    {
      "entropy": 5.358998012542725,
      "epoch": 3.6723205602022952,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036933433659382575,
      "loss": 4.7907,
      "mean_token_accuracy": 0.21966007649898528,
      "num_tokens": 81864113.0,
      "step": 47200
    },
    {
      "entropy": 5.338566541671753,
      "epoch": 3.672709589574013,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003693091738803322,
      "loss": 4.8895,
      "mean_token_accuracy": 0.20016815215349198,
      "num_tokens": 81873325.0,
      "step": 47205
    },
    {
      "entropy": 5.382515907287598,
      "epoch": 3.6730986189457306,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000369284009735847,
      "loss": 4.9019,
      "mean_token_accuracy": 0.21370816379785537,
      "num_tokens": 81881835.0,
      "step": 47210
    },
    {
      "entropy": 5.466736221313477,
      "epoch": 3.673487648317448,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003692588441607521,
      "loss": 4.9885,
      "mean_token_accuracy": 0.2003921702504158,
      "num_tokens": 81890402.0,
      "step": 47215
    },
    {
      "entropy": 5.4390253067016605,
      "epoch": 3.6738766776891656,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036923367715542923,
      "loss": 4.8789,
      "mean_token_accuracy": 0.2064184531569481,
      "num_tokens": 81898742.0,
      "step": 47220
    },
    {
      "entropy": 5.3671056747436525,
      "epoch": 3.674265707060883,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036920850872026025,
      "loss": 4.8298,
      "mean_token_accuracy": 0.21612456291913987,
      "num_tokens": 81906845.0,
      "step": 47225
    },
    {
      "entropy": 5.275747680664063,
      "epoch": 3.6746547364326005,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003691833388556272,
      "loss": 4.6896,
      "mean_token_accuracy": 0.2199362874031067,
      "num_tokens": 81914825.0,
      "step": 47230
    },
    {
      "entropy": 5.36029839515686,
      "epoch": 3.6750437658043182,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003691581675619118,
      "loss": 4.8302,
      "mean_token_accuracy": 0.21252372860908508,
      "num_tokens": 81923626.0,
      "step": 47235
    },
    {
      "entropy": 5.363822889328003,
      "epoch": 3.675432795176036,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003691329948394961,
      "loss": 4.8715,
      "mean_token_accuracy": 0.2050026535987854,
      "num_tokens": 81932515.0,
      "step": 47240
    },
    {
      "entropy": 5.388404083251953,
      "epoch": 3.675821824547753,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036910782068876204,
      "loss": 4.834,
      "mean_token_accuracy": 0.21149262934923171,
      "num_tokens": 81941797.0,
      "step": 47245
    },
    {
      "entropy": 5.359037208557129,
      "epoch": 3.676210853919471,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036908264511009155,
      "loss": 4.8572,
      "mean_token_accuracy": 0.21346010714769365,
      "num_tokens": 81949917.0,
      "step": 47250
    },
    {
      "entropy": 5.408670282363891,
      "epoch": 3.6765998832911886,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003690574681038667,
      "loss": 4.8944,
      "mean_token_accuracy": 0.20366846919059753,
      "num_tokens": 81958901.0,
      "step": 47255
    },
    {
      "entropy": 5.372065258026123,
      "epoch": 3.676988912662906,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036903228967046946,
      "loss": 4.8671,
      "mean_token_accuracy": 0.21343590468168258,
      "num_tokens": 81967491.0,
      "step": 47260
    },
    {
      "entropy": 5.307934999465942,
      "epoch": 3.6773779420346235,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003690071098102819,
      "loss": 4.743,
      "mean_token_accuracy": 0.21796633750200273,
      "num_tokens": 81976046.0,
      "step": 47265
    },
    {
      "entropy": 5.386505222320556,
      "epoch": 3.6777669714063412,
      "grad_norm": 1.125,
      "learning_rate": 0.0003689819285236861,
      "loss": 4.8994,
      "mean_token_accuracy": 0.20982520580291747,
      "num_tokens": 81984573.0,
      "step": 47270
    },
    {
      "entropy": 5.333456802368164,
      "epoch": 3.678156000778059,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036895674581106415,
      "loss": 4.8238,
      "mean_token_accuracy": 0.2167393907904625,
      "num_tokens": 81993992.0,
      "step": 47275
    },
    {
      "entropy": 5.399825572967529,
      "epoch": 3.678545030149776,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036893156167279803,
      "loss": 4.8164,
      "mean_token_accuracy": 0.21511218100786209,
      "num_tokens": 82002715.0,
      "step": 47280
    },
    {
      "entropy": 5.33726806640625,
      "epoch": 3.678934059521494,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003689063761092701,
      "loss": 4.8766,
      "mean_token_accuracy": 0.21589206159114838,
      "num_tokens": 82011692.0,
      "step": 47285
    },
    {
      "entropy": 5.301930952072143,
      "epoch": 3.6793230888932116,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00036888118912086225,
      "loss": 4.8191,
      "mean_token_accuracy": 0.21329231411218644,
      "num_tokens": 82021082.0,
      "step": 47290
    },
    {
      "entropy": 5.325025606155395,
      "epoch": 3.679712118264929,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003688560007079569,
      "loss": 4.8032,
      "mean_token_accuracy": 0.2181471109390259,
      "num_tokens": 82029834.0,
      "step": 47295
    },
    {
      "entropy": 5.380826616287232,
      "epoch": 3.6801011476366465,
      "grad_norm": 1.125,
      "learning_rate": 0.000368830810870936,
      "loss": 4.838,
      "mean_token_accuracy": 0.21597159206867217,
      "num_tokens": 82038486.0,
      "step": 47300
    },
    {
      "entropy": 5.427830410003662,
      "epoch": 3.6804901770083642,
      "grad_norm": 1.140625,
      "learning_rate": 0.000368805619610182,
      "loss": 4.8848,
      "mean_token_accuracy": 0.20869013667106628,
      "num_tokens": 82047034.0,
      "step": 47305
    },
    {
      "entropy": 5.325745534896851,
      "epoch": 3.680879206380082,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036878042692607693,
      "loss": 4.792,
      "mean_token_accuracy": 0.2241353139281273,
      "num_tokens": 82056243.0,
      "step": 47310
    },
    {
      "entropy": 5.366618347167969,
      "epoch": 3.681268235751799,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036875523281900324,
      "loss": 4.8776,
      "mean_token_accuracy": 0.22416869550943375,
      "num_tokens": 82064345.0,
      "step": 47315
    },
    {
      "entropy": 5.260969114303589,
      "epoch": 3.681657265123517,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003687300372893431,
      "loss": 4.7393,
      "mean_token_accuracy": 0.21943260878324508,
      "num_tokens": 82071755.0,
      "step": 47320
    },
    {
      "entropy": 5.274215841293335,
      "epoch": 3.682046294495234,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036870484033747883,
      "loss": 4.8347,
      "mean_token_accuracy": 0.2142368584871292,
      "num_tokens": 82080435.0,
      "step": 47325
    },
    {
      "entropy": 5.334148597717285,
      "epoch": 3.682435323866952,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036867964196379274,
      "loss": 4.9406,
      "mean_token_accuracy": 0.20686953365802765,
      "num_tokens": 82089092.0,
      "step": 47330
    },
    {
      "entropy": 5.335886907577515,
      "epoch": 3.6828243532386695,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003686544421686672,
      "loss": 4.7988,
      "mean_token_accuracy": 0.21934151500463486,
      "num_tokens": 82097431.0,
      "step": 47335
    },
    {
      "entropy": 5.417931938171387,
      "epoch": 3.683213382610387,
      "grad_norm": 1.359375,
      "learning_rate": 0.00036862924095248457,
      "loss": 4.9147,
      "mean_token_accuracy": 0.2179303526878357,
      "num_tokens": 82106785.0,
      "step": 47340
    },
    {
      "entropy": 5.35706582069397,
      "epoch": 3.683602411982105,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036860403831562717,
      "loss": 4.917,
      "mean_token_accuracy": 0.20302112251520157,
      "num_tokens": 82115742.0,
      "step": 47345
    },
    {
      "entropy": 5.320372295379639,
      "epoch": 3.683991441353822,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003685788342584775,
      "loss": 4.8422,
      "mean_token_accuracy": 0.2185914024710655,
      "num_tokens": 82124316.0,
      "step": 47350
    },
    {
      "entropy": 5.373825645446777,
      "epoch": 3.68438047072554,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000368553628781418,
      "loss": 4.8031,
      "mean_token_accuracy": 0.21866165697574616,
      "num_tokens": 82133057.0,
      "step": 47355
    },
    {
      "entropy": 5.355614280700683,
      "epoch": 3.684769500097257,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003685284218848311,
      "loss": 4.8797,
      "mean_token_accuracy": 0.21816698610782623,
      "num_tokens": 82141680.0,
      "step": 47360
    },
    {
      "entropy": 5.294845485687256,
      "epoch": 3.685158529468975,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036850321356909926,
      "loss": 4.84,
      "mean_token_accuracy": 0.21362229138612748,
      "num_tokens": 82150336.0,
      "step": 47365
    },
    {
      "entropy": 5.456057691574097,
      "epoch": 3.6855475588406925,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036847800383460493,
      "loss": 4.9012,
      "mean_token_accuracy": 0.21581654995679855,
      "num_tokens": 82158651.0,
      "step": 47370
    },
    {
      "entropy": 5.380819606781006,
      "epoch": 3.68593658821241,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036845279268173076,
      "loss": 4.8334,
      "mean_token_accuracy": 0.21854581981897353,
      "num_tokens": 82167364.0,
      "step": 47375
    },
    {
      "entropy": 5.3959205627441404,
      "epoch": 3.6863256175841275,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036842758011085914,
      "loss": 4.9125,
      "mean_token_accuracy": 0.2078518897294998,
      "num_tokens": 82176841.0,
      "step": 47380
    },
    {
      "entropy": 5.414372825622559,
      "epoch": 3.686714646955845,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003684023661223727,
      "loss": 4.8923,
      "mean_token_accuracy": 0.20744656175374984,
      "num_tokens": 82185541.0,
      "step": 47385
    },
    {
      "entropy": 5.359154844284058,
      "epoch": 3.687103676327563,
      "grad_norm": 1.125,
      "learning_rate": 0.00036837715071665403,
      "loss": 4.8886,
      "mean_token_accuracy": 0.20280827432870865,
      "num_tokens": 82194406.0,
      "step": 47390
    },
    {
      "entropy": 5.339507293701172,
      "epoch": 3.68749270569928,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003683519338940857,
      "loss": 4.8164,
      "mean_token_accuracy": 0.21961307674646377,
      "num_tokens": 82203372.0,
      "step": 47395
    },
    {
      "entropy": 5.45060977935791,
      "epoch": 3.687881735070998,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036832671565505045,
      "loss": 4.9398,
      "mean_token_accuracy": 0.21270493865013124,
      "num_tokens": 82211467.0,
      "step": 47400
    },
    {
      "entropy": 5.413046550750733,
      "epoch": 3.6882707644427155,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003683014959999307,
      "loss": 4.9226,
      "mean_token_accuracy": 0.2089463546872139,
      "num_tokens": 82220317.0,
      "step": 47405
    },
    {
      "entropy": 5.406777667999267,
      "epoch": 3.688659793814433,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003682762749291094,
      "loss": 4.8302,
      "mean_token_accuracy": 0.21259994059801102,
      "num_tokens": 82228474.0,
      "step": 47410
    },
    {
      "entropy": 5.346289873123169,
      "epoch": 3.6890488231861505,
      "grad_norm": 1.09375,
      "learning_rate": 0.000368251052442969,
      "loss": 4.8298,
      "mean_token_accuracy": 0.21342036575078965,
      "num_tokens": 82237463.0,
      "step": 47415
    },
    {
      "entropy": 5.353835868835449,
      "epoch": 3.689437852557868,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036822582854189234,
      "loss": 4.9171,
      "mean_token_accuracy": 0.20954884588718414,
      "num_tokens": 82245909.0,
      "step": 47420
    },
    {
      "entropy": 5.383306455612183,
      "epoch": 3.6898268819295854,
      "grad_norm": 1.078125,
      "learning_rate": 0.000368200603226262,
      "loss": 4.8304,
      "mean_token_accuracy": 0.2177188888192177,
      "num_tokens": 82254497.0,
      "step": 47425
    },
    {
      "entropy": 5.431554460525513,
      "epoch": 3.690215911301303,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000368175376496461,
      "loss": 4.9673,
      "mean_token_accuracy": 0.200039803981781,
      "num_tokens": 82262250.0,
      "step": 47430
    },
    {
      "entropy": 5.399050188064575,
      "epoch": 3.690604940673021,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003681501483528719,
      "loss": 4.9471,
      "mean_token_accuracy": 0.2116875097155571,
      "num_tokens": 82271645.0,
      "step": 47435
    },
    {
      "entropy": 5.360663986206054,
      "epoch": 3.6909939700447385,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003681249187958776,
      "loss": 4.7818,
      "mean_token_accuracy": 0.21456883400678634,
      "num_tokens": 82280244.0,
      "step": 47440
    },
    {
      "entropy": 5.417734384536743,
      "epoch": 3.691382999416456,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036809968782586087,
      "loss": 4.8733,
      "mean_token_accuracy": 0.2139945611357689,
      "num_tokens": 82288777.0,
      "step": 47445
    },
    {
      "entropy": 5.401992464065552,
      "epoch": 3.6917720287881735,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003680744554432045,
      "loss": 4.9001,
      "mean_token_accuracy": 0.206680503487587,
      "num_tokens": 82297651.0,
      "step": 47450
    },
    {
      "entropy": 5.412092971801758,
      "epoch": 3.692161058159891,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036804922164829153,
      "loss": 4.9212,
      "mean_token_accuracy": 0.20225875824689865,
      "num_tokens": 82306142.0,
      "step": 47455
    },
    {
      "entropy": 5.419519138336182,
      "epoch": 3.6925500875316084,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003680239864415046,
      "loss": 4.9479,
      "mean_token_accuracy": 0.20086863040924072,
      "num_tokens": 82314812.0,
      "step": 47460
    },
    {
      "entropy": 5.359300422668457,
      "epoch": 3.692939116903326,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003679987498232268,
      "loss": 4.7848,
      "mean_token_accuracy": 0.2154761478304863,
      "num_tokens": 82323776.0,
      "step": 47465
    },
    {
      "entropy": 5.390937185287475,
      "epoch": 3.693328146275044,
      "grad_norm": 1.234375,
      "learning_rate": 0.000367973511793841,
      "loss": 4.7839,
      "mean_token_accuracy": 0.223929700255394,
      "num_tokens": 82331846.0,
      "step": 47470
    },
    {
      "entropy": 5.334718132019043,
      "epoch": 3.6937171756467615,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036794827235373006,
      "loss": 4.9106,
      "mean_token_accuracy": 0.20874475836753845,
      "num_tokens": 82340609.0,
      "step": 47475
    },
    {
      "entropy": 5.406247186660766,
      "epoch": 3.6941062050184788,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003679230315032772,
      "loss": 4.864,
      "mean_token_accuracy": 0.21097481697797776,
      "num_tokens": 82350023.0,
      "step": 47480
    },
    {
      "entropy": 5.405655097961426,
      "epoch": 3.6944952343901964,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003678977892428651,
      "loss": 4.8516,
      "mean_token_accuracy": 0.21300053894519805,
      "num_tokens": 82358570.0,
      "step": 47485
    },
    {
      "entropy": 5.354764461517334,
      "epoch": 3.694884263761914,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003678725455728769,
      "loss": 4.8184,
      "mean_token_accuracy": 0.21519537270069122,
      "num_tokens": 82366747.0,
      "step": 47490
    },
    {
      "entropy": 5.447524309158325,
      "epoch": 3.6952732931336314,
      "grad_norm": 1.125,
      "learning_rate": 0.00036784730049369577,
      "loss": 4.9965,
      "mean_token_accuracy": 0.19718868732452394,
      "num_tokens": 82376553.0,
      "step": 47495
    },
    {
      "entropy": 5.348805236816406,
      "epoch": 3.695662322505349,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003678220540057045,
      "loss": 4.7802,
      "mean_token_accuracy": 0.21932324320077895,
      "num_tokens": 82384859.0,
      "step": 47500
    },
    {
      "entropy": 5.346302175521851,
      "epoch": 3.696051351877067,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00036779680610928635,
      "loss": 4.8264,
      "mean_token_accuracy": 0.21634551733732224,
      "num_tokens": 82392636.0,
      "step": 47505
    },
    {
      "entropy": 5.360527086257934,
      "epoch": 3.6964403812487845,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003677715568048244,
      "loss": 4.8742,
      "mean_token_accuracy": 0.21590301245450974,
      "num_tokens": 82401833.0,
      "step": 47510
    },
    {
      "entropy": 5.25359354019165,
      "epoch": 3.6968294106205017,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036774630609270165,
      "loss": 4.683,
      "mean_token_accuracy": 0.22805102467536925,
      "num_tokens": 82410081.0,
      "step": 47515
    },
    {
      "entropy": 5.433086109161377,
      "epoch": 3.6972184399922194,
      "grad_norm": 1.125,
      "learning_rate": 0.00036772105397330147,
      "loss": 4.9338,
      "mean_token_accuracy": 0.20512235313653945,
      "num_tokens": 82418652.0,
      "step": 47520
    },
    {
      "entropy": 5.3614586353302,
      "epoch": 3.697607469363937,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003676958004470067,
      "loss": 4.8149,
      "mean_token_accuracy": 0.21614805459976197,
      "num_tokens": 82426776.0,
      "step": 47525
    },
    {
      "entropy": 5.245257329940796,
      "epoch": 3.6979964987356544,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003676705455142009,
      "loss": 4.7467,
      "mean_token_accuracy": 0.2233837604522705,
      "num_tokens": 82435624.0,
      "step": 47530
    },
    {
      "entropy": 5.384545135498047,
      "epoch": 3.698385528107372,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003676452891752669,
      "loss": 4.9131,
      "mean_token_accuracy": 0.21032561957836152,
      "num_tokens": 82444224.0,
      "step": 47535
    },
    {
      "entropy": 5.447855710983276,
      "epoch": 3.69877455747909,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003676200314305882,
      "loss": 4.9538,
      "mean_token_accuracy": 0.20316437929868697,
      "num_tokens": 82452527.0,
      "step": 47540
    },
    {
      "entropy": 5.411064910888672,
      "epoch": 3.6991635868508075,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003675947722805479,
      "loss": 4.8051,
      "mean_token_accuracy": 0.22199366241693497,
      "num_tokens": 82460180.0,
      "step": 47545
    },
    {
      "entropy": 5.37548828125,
      "epoch": 3.6995526162225247,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003675695117255293,
      "loss": 4.8527,
      "mean_token_accuracy": 0.20894798189401625,
      "num_tokens": 82468725.0,
      "step": 47550
    },
    {
      "entropy": 5.406736755371094,
      "epoch": 3.6999416455942424,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003675442497659157,
      "loss": 4.8867,
      "mean_token_accuracy": 0.20724343955516816,
      "num_tokens": 82477070.0,
      "step": 47555
    },
    {
      "entropy": 5.340423345565796,
      "epoch": 3.7003306749659597,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003675189864020905,
      "loss": 4.8139,
      "mean_token_accuracy": 0.2193831041455269,
      "num_tokens": 82485960.0,
      "step": 47560
    },
    {
      "entropy": 5.268238639831543,
      "epoch": 3.7007197043376774,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003674937216344368,
      "loss": 4.7338,
      "mean_token_accuracy": 0.21971870511770247,
      "num_tokens": 82494498.0,
      "step": 47565
    },
    {
      "entropy": 5.387840986251831,
      "epoch": 3.701108733709395,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003674684554633382,
      "loss": 4.8695,
      "mean_token_accuracy": 0.20704305320978164,
      "num_tokens": 82504505.0,
      "step": 47570
    },
    {
      "entropy": 5.326870632171631,
      "epoch": 3.701497763081113,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003674431878891779,
      "loss": 4.8475,
      "mean_token_accuracy": 0.2131541058421135,
      "num_tokens": 82513747.0,
      "step": 47575
    },
    {
      "entropy": 5.422780752182007,
      "epoch": 3.70188679245283,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003674179189123393,
      "loss": 4.7839,
      "mean_token_accuracy": 0.21281896829605101,
      "num_tokens": 82522621.0,
      "step": 47580
    },
    {
      "entropy": 5.331941938400268,
      "epoch": 3.7022758218245477,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036739264853320606,
      "loss": 4.7559,
      "mean_token_accuracy": 0.22189822643995286,
      "num_tokens": 82530286.0,
      "step": 47585
    },
    {
      "entropy": 5.285174322128296,
      "epoch": 3.7026648511962654,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036736737675216127,
      "loss": 4.8147,
      "mean_token_accuracy": 0.21306973248720168,
      "num_tokens": 82539288.0,
      "step": 47590
    },
    {
      "entropy": 5.396151924133301,
      "epoch": 3.7030538805679827,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003673421035695887,
      "loss": 4.8828,
      "mean_token_accuracy": 0.20856768041849136,
      "num_tokens": 82547271.0,
      "step": 47595
    },
    {
      "entropy": 5.432281112670898,
      "epoch": 3.7034429099397004,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003673168289858716,
      "loss": 4.9296,
      "mean_token_accuracy": 0.2029433473944664,
      "num_tokens": 82555997.0,
      "step": 47600
    },
    {
      "entropy": 5.417050123214722,
      "epoch": 3.703831939311418,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003672915530013936,
      "loss": 4.8405,
      "mean_token_accuracy": 0.21640387624502183,
      "num_tokens": 82565181.0,
      "step": 47605
    },
    {
      "entropy": 5.382005500793457,
      "epoch": 3.7042209686831358,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036726627561653815,
      "loss": 4.879,
      "mean_token_accuracy": 0.21240968406200408,
      "num_tokens": 82573691.0,
      "step": 47610
    },
    {
      "entropy": 5.420676374435425,
      "epoch": 3.704609998054853,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003672409968316888,
      "loss": 4.9355,
      "mean_token_accuracy": 0.20527132153511046,
      "num_tokens": 82582562.0,
      "step": 47615
    },
    {
      "entropy": 5.401203870773315,
      "epoch": 3.7049990274265707,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036721571664722914,
      "loss": 4.9011,
      "mean_token_accuracy": 0.2141050338745117,
      "num_tokens": 82591051.0,
      "step": 47620
    },
    {
      "entropy": 5.438391399383545,
      "epoch": 3.7053880567982884,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003671904350635428,
      "loss": 4.8655,
      "mean_token_accuracy": 0.2139350637793541,
      "num_tokens": 82599385.0,
      "step": 47625
    },
    {
      "entropy": 5.4277503490448,
      "epoch": 3.7057770861700057,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036716515208101334,
      "loss": 4.8127,
      "mean_token_accuracy": 0.21528344452381135,
      "num_tokens": 82607688.0,
      "step": 47630
    },
    {
      "entropy": 5.37467942237854,
      "epoch": 3.7061661155417234,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003671398677000244,
      "loss": 4.8647,
      "mean_token_accuracy": 0.21550339460372925,
      "num_tokens": 82615471.0,
      "step": 47635
    },
    {
      "entropy": 5.302372074127197,
      "epoch": 3.706555144913441,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003671145819209596,
      "loss": 4.7818,
      "mean_token_accuracy": 0.21941887736320495,
      "num_tokens": 82624243.0,
      "step": 47640
    },
    {
      "entropy": 5.390410757064819,
      "epoch": 3.7069441742851588,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003670892947442026,
      "loss": 4.8466,
      "mean_token_accuracy": 0.21348970681428908,
      "num_tokens": 82632821.0,
      "step": 47645
    },
    {
      "entropy": 5.494918346405029,
      "epoch": 3.707333203656876,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003670640061701371,
      "loss": 5.0658,
      "mean_token_accuracy": 0.1995412901043892,
      "num_tokens": 82641225.0,
      "step": 47650
    },
    {
      "entropy": 5.438647270202637,
      "epoch": 3.7077222330285937,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003670387161991469,
      "loss": 4.9116,
      "mean_token_accuracy": 0.21429473012685776,
      "num_tokens": 82649885.0,
      "step": 47655
    },
    {
      "entropy": 5.3290777683258055,
      "epoch": 3.708111262400311,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003670134248316157,
      "loss": 4.7759,
      "mean_token_accuracy": 0.2215218335390091,
      "num_tokens": 82658589.0,
      "step": 47660
    },
    {
      "entropy": 5.399945163726807,
      "epoch": 3.7085002917720287,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036698813206792724,
      "loss": 4.9289,
      "mean_token_accuracy": 0.20571202486753465,
      "num_tokens": 82666676.0,
      "step": 47665
    },
    {
      "entropy": 5.365212106704712,
      "epoch": 3.7088893211437464,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036696283790846524,
      "loss": 4.8026,
      "mean_token_accuracy": 0.2154592514038086,
      "num_tokens": 82675007.0,
      "step": 47670
    },
    {
      "entropy": 5.384629106521606,
      "epoch": 3.709278350515464,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036693754235361364,
      "loss": 4.8697,
      "mean_token_accuracy": 0.21132962107658387,
      "num_tokens": 82684060.0,
      "step": 47675
    },
    {
      "entropy": 5.34520411491394,
      "epoch": 3.7096673798871818,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003669122454037561,
      "loss": 4.9336,
      "mean_token_accuracy": 0.21316074430942536,
      "num_tokens": 82693280.0,
      "step": 47680
    },
    {
      "entropy": 5.347713041305542,
      "epoch": 3.710056409258899,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003668869470592765,
      "loss": 4.7597,
      "mean_token_accuracy": 0.22360343039035796,
      "num_tokens": 82701176.0,
      "step": 47685
    },
    {
      "entropy": 5.327824878692627,
      "epoch": 3.7104454386306167,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036686164732055876,
      "loss": 4.7891,
      "mean_token_accuracy": 0.22305157780647278,
      "num_tokens": 82709410.0,
      "step": 47690
    },
    {
      "entropy": 5.318912315368652,
      "epoch": 3.710834468002334,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036683634618798675,
      "loss": 4.8226,
      "mean_token_accuracy": 0.2101035714149475,
      "num_tokens": 82718535.0,
      "step": 47695
    },
    {
      "entropy": 5.389748430252075,
      "epoch": 3.7112234973740517,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003668110436619443,
      "loss": 4.8932,
      "mean_token_accuracy": 0.21471327543258667,
      "num_tokens": 82727524.0,
      "step": 47700
    },
    {
      "entropy": 5.392421436309815,
      "epoch": 3.7116125267457694,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036678573974281546,
      "loss": 4.9204,
      "mean_token_accuracy": 0.20711326450109482,
      "num_tokens": 82735605.0,
      "step": 47705
    },
    {
      "entropy": 5.39943265914917,
      "epoch": 3.712001556117487,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003667604344309841,
      "loss": 4.9097,
      "mean_token_accuracy": 0.21048038601875305,
      "num_tokens": 82745046.0,
      "step": 47710
    },
    {
      "entropy": 5.482069778442383,
      "epoch": 3.7123905854892043,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036673512772683424,
      "loss": 4.9108,
      "mean_token_accuracy": 0.2126159578561783,
      "num_tokens": 82753961.0,
      "step": 47715
    },
    {
      "entropy": 5.35315580368042,
      "epoch": 3.712779614860922,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003667098196307498,
      "loss": 4.8327,
      "mean_token_accuracy": 0.2193732365965843,
      "num_tokens": 82762684.0,
      "step": 47720
    },
    {
      "entropy": 5.476911211013794,
      "epoch": 3.7131686442326397,
      "grad_norm": 1.125,
      "learning_rate": 0.00036668451014311475,
      "loss": 4.9174,
      "mean_token_accuracy": 0.21720012575387954,
      "num_tokens": 82770958.0,
      "step": 47725
    },
    {
      "entropy": 5.412020587921143,
      "epoch": 3.713557673604357,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036665919926431325,
      "loss": 4.8529,
      "mean_token_accuracy": 0.21815908402204515,
      "num_tokens": 82779891.0,
      "step": 47730
    },
    {
      "entropy": 5.34972071647644,
      "epoch": 3.7139467029760747,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003666338869947293,
      "loss": 4.8692,
      "mean_token_accuracy": 0.20911203622817992,
      "num_tokens": 82788835.0,
      "step": 47735
    },
    {
      "entropy": 5.385988044738769,
      "epoch": 3.7143357323477924,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003666085733347469,
      "loss": 4.8777,
      "mean_token_accuracy": 0.20787182599306106,
      "num_tokens": 82797119.0,
      "step": 47740
    },
    {
      "entropy": 5.413204574584961,
      "epoch": 3.71472476171951,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036658325828475033,
      "loss": 4.7897,
      "mean_token_accuracy": 0.2148541271686554,
      "num_tokens": 82805210.0,
      "step": 47745
    },
    {
      "entropy": 5.381630945205688,
      "epoch": 3.7151137910912273,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003665579418451235,
      "loss": 4.7868,
      "mean_token_accuracy": 0.2153421998023987,
      "num_tokens": 82813916.0,
      "step": 47750
    },
    {
      "entropy": 5.349843072891235,
      "epoch": 3.715502820462945,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003665326240162506,
      "loss": 4.8848,
      "mean_token_accuracy": 0.2132158175110817,
      "num_tokens": 82822938.0,
      "step": 47755
    },
    {
      "entropy": 5.4200750350952145,
      "epoch": 3.7158918498346623,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036650730479851593,
      "loss": 4.9601,
      "mean_token_accuracy": 0.20464060008525847,
      "num_tokens": 82831473.0,
      "step": 47760
    },
    {
      "entropy": 5.356255388259887,
      "epoch": 3.71628087920638,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036648198419230344,
      "loss": 4.9325,
      "mean_token_accuracy": 0.20896043479442597,
      "num_tokens": 82840233.0,
      "step": 47765
    },
    {
      "entropy": 5.397795629501343,
      "epoch": 3.7166699085780976,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036645666219799753,
      "loss": 4.8603,
      "mean_token_accuracy": 0.21931437253952027,
      "num_tokens": 82848982.0,
      "step": 47770
    },
    {
      "entropy": 5.427029418945312,
      "epoch": 3.7170589379498153,
      "grad_norm": 1.125,
      "learning_rate": 0.0003664313388159823,
      "loss": 4.8982,
      "mean_token_accuracy": 0.20943263322114944,
      "num_tokens": 82858029.0,
      "step": 47775
    },
    {
      "entropy": 5.432743501663208,
      "epoch": 3.717447967321533,
      "grad_norm": 1.109375,
      "learning_rate": 0.000366406014046642,
      "loss": 4.8849,
      "mean_token_accuracy": 0.20868797302246095,
      "num_tokens": 82866138.0,
      "step": 47780
    },
    {
      "entropy": 5.464727163314819,
      "epoch": 3.7178369966932503,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000366380687890361,
      "loss": 4.9936,
      "mean_token_accuracy": 0.2028328776359558,
      "num_tokens": 82875330.0,
      "step": 47785
    },
    {
      "entropy": 5.422740030288696,
      "epoch": 3.718226026064968,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003663553603475235,
      "loss": 4.9127,
      "mean_token_accuracy": 0.20891542136669158,
      "num_tokens": 82884456.0,
      "step": 47790
    },
    {
      "entropy": 5.380167436599732,
      "epoch": 3.7186150554366852,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036633003141851373,
      "loss": 4.8278,
      "mean_token_accuracy": 0.22119367271661758,
      "num_tokens": 82893148.0,
      "step": 47795
    },
    {
      "entropy": 5.390344429016113,
      "epoch": 3.719004084808403,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003663047011037162,
      "loss": 4.8481,
      "mean_token_accuracy": 0.21996967494487762,
      "num_tokens": 82901065.0,
      "step": 47800
    },
    {
      "entropy": 5.297053003311158,
      "epoch": 3.7193931141801206,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003662793694035151,
      "loss": 4.8348,
      "mean_token_accuracy": 0.21533189415931703,
      "num_tokens": 82910162.0,
      "step": 47805
    },
    {
      "entropy": 5.39837417602539,
      "epoch": 3.7197821435518383,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003662540363182949,
      "loss": 4.9165,
      "mean_token_accuracy": 0.20835154354572297,
      "num_tokens": 82918393.0,
      "step": 47810
    },
    {
      "entropy": 5.321748542785644,
      "epoch": 3.7201711729235556,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003662287018484398,
      "loss": 4.795,
      "mean_token_accuracy": 0.22680991888046265,
      "num_tokens": 82927607.0,
      "step": 47815
    },
    {
      "entropy": 5.377447700500488,
      "epoch": 3.7205602022952733,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003662033659943345,
      "loss": 4.9198,
      "mean_token_accuracy": 0.21109362244606017,
      "num_tokens": 82936743.0,
      "step": 47820
    },
    {
      "entropy": 5.432659435272217,
      "epoch": 3.720949231666991,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036617802875636333,
      "loss": 4.943,
      "mean_token_accuracy": 0.20714199244976045,
      "num_tokens": 82946027.0,
      "step": 47825
    },
    {
      "entropy": 5.41870083808899,
      "epoch": 3.7213382610387082,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003661526901349106,
      "loss": 4.8232,
      "mean_token_accuracy": 0.21017409563064576,
      "num_tokens": 82954442.0,
      "step": 47830
    },
    {
      "entropy": 5.33777551651001,
      "epoch": 3.721727290410426,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036612735013036085,
      "loss": 4.7641,
      "mean_token_accuracy": 0.2219604194164276,
      "num_tokens": 82962518.0,
      "step": 47835
    },
    {
      "entropy": 5.434994411468506,
      "epoch": 3.7221163197821436,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003661020087430987,
      "loss": 4.9781,
      "mean_token_accuracy": 0.20242183059453964,
      "num_tokens": 82971544.0,
      "step": 47840
    },
    {
      "entropy": 5.401493692398072,
      "epoch": 3.7225053491538613,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003660766659735086,
      "loss": 4.8006,
      "mean_token_accuracy": 0.218464894592762,
      "num_tokens": 82980771.0,
      "step": 47845
    },
    {
      "entropy": 5.390036916732788,
      "epoch": 3.7228943785255786,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003660513218219749,
      "loss": 4.8682,
      "mean_token_accuracy": 0.2119361251592636,
      "num_tokens": 82989525.0,
      "step": 47850
    },
    {
      "entropy": 5.359787559509277,
      "epoch": 3.7232834078972963,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003660259762888824,
      "loss": 4.861,
      "mean_token_accuracy": 0.21262769401073456,
      "num_tokens": 82998135.0,
      "step": 47855
    },
    {
      "entropy": 5.451057481765747,
      "epoch": 3.7236724372690135,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003660006293746156,
      "loss": 4.9243,
      "mean_token_accuracy": 0.20308346748352052,
      "num_tokens": 83007068.0,
      "step": 47860
    },
    {
      "entropy": 5.401442337036133,
      "epoch": 3.7240614666407312,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036597528107955914,
      "loss": 4.8792,
      "mean_token_accuracy": 0.20906789749860763,
      "num_tokens": 83015819.0,
      "step": 47865
    },
    {
      "entropy": 5.478734254837036,
      "epoch": 3.724450496012449,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036594993140409753,
      "loss": 4.936,
      "mean_token_accuracy": 0.20496702790260315,
      "num_tokens": 83025096.0,
      "step": 47870
    },
    {
      "entropy": 5.3364794731140135,
      "epoch": 3.7248395253841666,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003659245803486155,
      "loss": 4.7415,
      "mean_token_accuracy": 0.21970428824424743,
      "num_tokens": 83033181.0,
      "step": 47875
    },
    {
      "entropy": 5.314488697052002,
      "epoch": 3.7252285547558843,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003658992279134977,
      "loss": 4.8577,
      "mean_token_accuracy": 0.21468538343906401,
      "num_tokens": 83042315.0,
      "step": 47880
    },
    {
      "entropy": 5.337536191940307,
      "epoch": 3.7256175841276016,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003658738740991287,
      "loss": 4.8799,
      "mean_token_accuracy": 0.2071785882115364,
      "num_tokens": 83051565.0,
      "step": 47885
    },
    {
      "entropy": 5.413470458984375,
      "epoch": 3.7260066134993193,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036584851890589335,
      "loss": 4.8746,
      "mean_token_accuracy": 0.20534023493528367,
      "num_tokens": 83059889.0,
      "step": 47890
    },
    {
      "entropy": 5.366611289978027,
      "epoch": 3.7263956428710365,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036582316233417636,
      "loss": 4.8636,
      "mean_token_accuracy": 0.214415942132473,
      "num_tokens": 83068764.0,
      "step": 47895
    },
    {
      "entropy": 5.317829656600952,
      "epoch": 3.7267846722427542,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036579780438436246,
      "loss": 4.7795,
      "mean_token_accuracy": 0.2132525622844696,
      "num_tokens": 83077605.0,
      "step": 47900
    },
    {
      "entropy": 5.297066354751587,
      "epoch": 3.727173701614472,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003657724450568363,
      "loss": 4.7894,
      "mean_token_accuracy": 0.21438853442668915,
      "num_tokens": 83086080.0,
      "step": 47905
    },
    {
      "entropy": 5.320501852035522,
      "epoch": 3.7275627309861896,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003657470843519829,
      "loss": 4.7659,
      "mean_token_accuracy": 0.2174546793103218,
      "num_tokens": 83094070.0,
      "step": 47910
    },
    {
      "entropy": 5.299572420120239,
      "epoch": 3.727951760357907,
      "grad_norm": 1.125,
      "learning_rate": 0.0003657217222701868,
      "loss": 4.7579,
      "mean_token_accuracy": 0.2212927371263504,
      "num_tokens": 83102213.0,
      "step": 47915
    },
    {
      "entropy": 5.285571002960205,
      "epoch": 3.7283407897296246,
      "grad_norm": 1.171875,
      "learning_rate": 0.000365696358811833,
      "loss": 4.8004,
      "mean_token_accuracy": 0.22055469453334808,
      "num_tokens": 83110525.0,
      "step": 47920
    },
    {
      "entropy": 5.31011438369751,
      "epoch": 3.7287298191013423,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036567099397730635,
      "loss": 4.8345,
      "mean_token_accuracy": 0.21674588322639465,
      "num_tokens": 83119206.0,
      "step": 47925
    },
    {
      "entropy": 5.392427110671997,
      "epoch": 3.7291188484730595,
      "grad_norm": 1.125,
      "learning_rate": 0.0003656456277669917,
      "loss": 4.924,
      "mean_token_accuracy": 0.20751623511314393,
      "num_tokens": 83127713.0,
      "step": 47930
    },
    {
      "entropy": 5.452603769302368,
      "epoch": 3.729507877844777,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036562026018127386,
      "loss": 4.8977,
      "mean_token_accuracy": 0.21085544675588608,
      "num_tokens": 83135785.0,
      "step": 47935
    },
    {
      "entropy": 5.368570327758789,
      "epoch": 3.729896907216495,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036559489122053786,
      "loss": 4.7982,
      "mean_token_accuracy": 0.21598514765501023,
      "num_tokens": 83144623.0,
      "step": 47940
    },
    {
      "entropy": 5.388584804534912,
      "epoch": 3.7302859365882126,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003655695208851686,
      "loss": 4.8405,
      "mean_token_accuracy": 0.21508225202560424,
      "num_tokens": 83153189.0,
      "step": 47945
    },
    {
      "entropy": 5.3912904262542725,
      "epoch": 3.73067496595993,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036554414917555096,
      "loss": 4.8578,
      "mean_token_accuracy": 0.21852099448442458,
      "num_tokens": 83161821.0,
      "step": 47950
    },
    {
      "entropy": 5.399232339859009,
      "epoch": 3.7310639953316476,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036551877609207,
      "loss": 4.8635,
      "mean_token_accuracy": 0.2144806981086731,
      "num_tokens": 83170527.0,
      "step": 47955
    },
    {
      "entropy": 5.338731288909912,
      "epoch": 3.7314530247033653,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003654934016351107,
      "loss": 4.7715,
      "mean_token_accuracy": 0.22164586782455445,
      "num_tokens": 83178983.0,
      "step": 47960
    },
    {
      "entropy": 5.353824281692505,
      "epoch": 3.7318420540750825,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036546802580505805,
      "loss": 4.8147,
      "mean_token_accuracy": 0.21759581416845322,
      "num_tokens": 83188177.0,
      "step": 47965
    },
    {
      "entropy": 5.3596964359283445,
      "epoch": 3.7322310834468,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003654426486022971,
      "loss": 4.8398,
      "mean_token_accuracy": 0.2099335566163063,
      "num_tokens": 83197090.0,
      "step": 47970
    },
    {
      "entropy": 5.378551244735718,
      "epoch": 3.732620112818518,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003654172700272129,
      "loss": 4.922,
      "mean_token_accuracy": 0.21393698900938035,
      "num_tokens": 83205347.0,
      "step": 47975
    },
    {
      "entropy": 5.375136280059815,
      "epoch": 3.7330091421902356,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003653918900801905,
      "loss": 4.8016,
      "mean_token_accuracy": 0.2193748787045479,
      "num_tokens": 83213378.0,
      "step": 47980
    },
    {
      "entropy": 5.468813800811768,
      "epoch": 3.733398171561953,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036536650876161514,
      "loss": 4.9132,
      "mean_token_accuracy": 0.20753172487020494,
      "num_tokens": 83221898.0,
      "step": 47985
    },
    {
      "entropy": 5.363307666778565,
      "epoch": 3.7337872009336706,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003653411260718718,
      "loss": 4.8035,
      "mean_token_accuracy": 0.21766814291477204,
      "num_tokens": 83231176.0,
      "step": 47990
    },
    {
      "entropy": 5.416937971115113,
      "epoch": 3.734176230305388,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003653157420113457,
      "loss": 4.9577,
      "mean_token_accuracy": 0.2091292291879654,
      "num_tokens": 83239777.0,
      "step": 47995
    },
    {
      "entropy": 5.3985882759094235,
      "epoch": 3.7345652596771055,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003652903565804219,
      "loss": 4.852,
      "mean_token_accuracy": 0.2179276466369629,
      "num_tokens": 83248941.0,
      "step": 48000
    },
    {
      "epoch": 3.7345652596771055,
      "eval_entropy": 5.2014463453136806,
      "eval_loss": 4.981439590454102,
      "eval_mean_token_accuracy": 0.2157636579567131,
      "eval_num_tokens": 83248941.0,
      "eval_runtime": 28.8246,
      "eval_samples_per_second": 1441.754,
      "eval_steps_per_second": 180.228,
      "step": 48000
    },
    {
      "entropy": 5.386441373825074,
      "epoch": 3.734954289048823,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036526496977948574,
      "loss": 4.8592,
      "mean_token_accuracy": 0.21433609127998351,
      "num_tokens": 83257324.0,
      "step": 48005
    },
    {
      "entropy": 5.33705358505249,
      "epoch": 3.735343318420541,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036523958160892227,
      "loss": 4.8724,
      "mean_token_accuracy": 0.21342201828956603,
      "num_tokens": 83265655.0,
      "step": 48010
    },
    {
      "entropy": 5.4089569568634035,
      "epoch": 3.735732347792258,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036521419206911684,
      "loss": 4.8504,
      "mean_token_accuracy": 0.21217601448297502,
      "num_tokens": 83274743.0,
      "step": 48015
    },
    {
      "entropy": 5.4215648651123045,
      "epoch": 3.736121377163976,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003651888011604545,
      "loss": 4.9235,
      "mean_token_accuracy": 0.2090612009167671,
      "num_tokens": 83284344.0,
      "step": 48020
    },
    {
      "entropy": 5.305263328552246,
      "epoch": 3.7365104065356936,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036516340888332076,
      "loss": 4.7926,
      "mean_token_accuracy": 0.2263455718755722,
      "num_tokens": 83292855.0,
      "step": 48025
    },
    {
      "entropy": 5.376588201522827,
      "epoch": 3.736899435907411,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036513801523810075,
      "loss": 4.8574,
      "mean_token_accuracy": 0.20942715257406236,
      "num_tokens": 83301624.0,
      "step": 48030
    },
    {
      "entropy": 5.473322010040283,
      "epoch": 3.7372884652791285,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003651126202251799,
      "loss": 4.9167,
      "mean_token_accuracy": 0.20920263081789017,
      "num_tokens": 83311052.0,
      "step": 48035
    },
    {
      "entropy": 5.432959079742432,
      "epoch": 3.737677494650846,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003650872238449434,
      "loss": 4.838,
      "mean_token_accuracy": 0.218114210665226,
      "num_tokens": 83319382.0,
      "step": 48040
    },
    {
      "entropy": 5.38957781791687,
      "epoch": 3.738066524022564,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003650618260977767,
      "loss": 4.9018,
      "mean_token_accuracy": 0.22011256217956543,
      "num_tokens": 83327548.0,
      "step": 48045
    },
    {
      "entropy": 5.321341180801392,
      "epoch": 3.738455553394281,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003650364269840651,
      "loss": 4.7914,
      "mean_token_accuracy": 0.21491991430521012,
      "num_tokens": 83336463.0,
      "step": 48050
    },
    {
      "entropy": 5.4038755893707275,
      "epoch": 3.738844582765999,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003650110265041941,
      "loss": 4.8268,
      "mean_token_accuracy": 0.22047334164381027,
      "num_tokens": 83344404.0,
      "step": 48055
    },
    {
      "entropy": 5.285020112991333,
      "epoch": 3.7392336121377165,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036498562465854897,
      "loss": 4.748,
      "mean_token_accuracy": 0.2192089632153511,
      "num_tokens": 83352016.0,
      "step": 48060
    },
    {
      "entropy": 5.2938337326049805,
      "epoch": 3.739622641509434,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003649602214475152,
      "loss": 4.8054,
      "mean_token_accuracy": 0.21537040024995804,
      "num_tokens": 83360424.0,
      "step": 48065
    },
    {
      "entropy": 5.392592525482177,
      "epoch": 3.7400116708811515,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036493481687147836,
      "loss": 4.8127,
      "mean_token_accuracy": 0.21050431579351425,
      "num_tokens": 83369413.0,
      "step": 48070
    },
    {
      "entropy": 5.3297539234161375,
      "epoch": 3.740400700252869,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003649094109308237,
      "loss": 4.8379,
      "mean_token_accuracy": 0.20785255134105682,
      "num_tokens": 83378493.0,
      "step": 48075
    },
    {
      "entropy": 5.288585233688354,
      "epoch": 3.740789729624587,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036488400362593695,
      "loss": 4.853,
      "mean_token_accuracy": 0.21077035665512084,
      "num_tokens": 83386668.0,
      "step": 48080
    },
    {
      "entropy": 5.334759664535523,
      "epoch": 3.741178758996304,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036485859495720337,
      "loss": 4.868,
      "mean_token_accuracy": 0.21185035407543182,
      "num_tokens": 83395961.0,
      "step": 48085
    },
    {
      "entropy": 5.347606468200683,
      "epoch": 3.741567788368022,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036483318492500875,
      "loss": 4.8718,
      "mean_token_accuracy": 0.210748353600502,
      "num_tokens": 83405346.0,
      "step": 48090
    },
    {
      "entropy": 5.442207765579224,
      "epoch": 3.741956817739739,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036480777352973856,
      "loss": 4.9081,
      "mean_token_accuracy": 0.21370006948709488,
      "num_tokens": 83413798.0,
      "step": 48095
    },
    {
      "entropy": 5.506014442443847,
      "epoch": 3.742345847111457,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036478236077177825,
      "loss": 4.9759,
      "mean_token_accuracy": 0.20348286032676696,
      "num_tokens": 83421961.0,
      "step": 48100
    },
    {
      "entropy": 5.385786819458008,
      "epoch": 3.7427348764831745,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036475694665151357,
      "loss": 4.9736,
      "mean_token_accuracy": 0.20576438754796983,
      "num_tokens": 83430615.0,
      "step": 48105
    },
    {
      "entropy": 5.397714757919312,
      "epoch": 3.743123905854892,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036473153116933007,
      "loss": 4.8872,
      "mean_token_accuracy": 0.21386412233114244,
      "num_tokens": 83439142.0,
      "step": 48110
    },
    {
      "entropy": 5.34433217048645,
      "epoch": 3.74351293522661,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036470611432561334,
      "loss": 4.7719,
      "mean_token_accuracy": 0.2239024117588997,
      "num_tokens": 83447421.0,
      "step": 48115
    },
    {
      "entropy": 5.3372032165527346,
      "epoch": 3.743901964598327,
      "grad_norm": 1.125,
      "learning_rate": 0.0003646806961207492,
      "loss": 4.8637,
      "mean_token_accuracy": 0.2122753605246544,
      "num_tokens": 83456090.0,
      "step": 48120
    },
    {
      "entropy": 5.378424549102784,
      "epoch": 3.744290993970045,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036465527655512323,
      "loss": 4.8345,
      "mean_token_accuracy": 0.2170809879899025,
      "num_tokens": 83464527.0,
      "step": 48125
    },
    {
      "entropy": 5.366263055801392,
      "epoch": 3.744680023341762,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003646298556291211,
      "loss": 4.8224,
      "mean_token_accuracy": 0.21746802628040313,
      "num_tokens": 83472648.0,
      "step": 48130
    },
    {
      "entropy": 5.293454647064209,
      "epoch": 3.74506905271348,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036460443334312867,
      "loss": 4.7867,
      "mean_token_accuracy": 0.21619207561016082,
      "num_tokens": 83481598.0,
      "step": 48135
    },
    {
      "entropy": 5.370912647247314,
      "epoch": 3.7454580820851975,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003645790096975315,
      "loss": 4.8375,
      "mean_token_accuracy": 0.21149327009916305,
      "num_tokens": 83490153.0,
      "step": 48140
    },
    {
      "entropy": 5.397190952301026,
      "epoch": 3.745847111456915,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00036455358469271546,
      "loss": 4.8637,
      "mean_token_accuracy": 0.2142141968011856,
      "num_tokens": 83499617.0,
      "step": 48145
    },
    {
      "entropy": 5.320184850692749,
      "epoch": 3.7462361408286324,
      "grad_norm": 1.125,
      "learning_rate": 0.00036452815832906625,
      "loss": 4.8564,
      "mean_token_accuracy": 0.20838571190834046,
      "num_tokens": 83507793.0,
      "step": 48150
    },
    {
      "entropy": 5.351131629943848,
      "epoch": 3.74662517020035,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036450273060696984,
      "loss": 4.79,
      "mean_token_accuracy": 0.21538684964179994,
      "num_tokens": 83516490.0,
      "step": 48155
    },
    {
      "entropy": 5.3430657386779785,
      "epoch": 3.747014199572068,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003644773015268118,
      "loss": 4.8888,
      "mean_token_accuracy": 0.21065460443496703,
      "num_tokens": 83524812.0,
      "step": 48160
    },
    {
      "entropy": 5.374954080581665,
      "epoch": 3.747403228943785,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036445187108897825,
      "loss": 4.8542,
      "mean_token_accuracy": 0.21244678497314454,
      "num_tokens": 83533242.0,
      "step": 48165
    },
    {
      "entropy": 5.372977304458618,
      "epoch": 3.747792258315503,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003644264392938549,
      "loss": 4.8011,
      "mean_token_accuracy": 0.22270214259624482,
      "num_tokens": 83541233.0,
      "step": 48170
    },
    {
      "entropy": 5.433784437179566,
      "epoch": 3.7481812876872205,
      "grad_norm": 1.4375,
      "learning_rate": 0.00036440100614182775,
      "loss": 4.9774,
      "mean_token_accuracy": 0.20909656286239625,
      "num_tokens": 83549561.0,
      "step": 48175
    },
    {
      "entropy": 5.419878625869751,
      "epoch": 3.748570317058938,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036437557163328255,
      "loss": 4.9103,
      "mean_token_accuracy": 0.2062024340033531,
      "num_tokens": 83558384.0,
      "step": 48180
    },
    {
      "entropy": 5.375685739517212,
      "epoch": 3.7489593464306554,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003643501357686053,
      "loss": 4.9295,
      "mean_token_accuracy": 0.20656861811876298,
      "num_tokens": 83567573.0,
      "step": 48185
    },
    {
      "entropy": 5.3838108539581295,
      "epoch": 3.749348375802373,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036432469854818194,
      "loss": 4.875,
      "mean_token_accuracy": 0.21278886049985885,
      "num_tokens": 83576632.0,
      "step": 48190
    },
    {
      "entropy": 5.413621616363526,
      "epoch": 3.7497374051740904,
      "grad_norm": 1.125,
      "learning_rate": 0.0003642992599723985,
      "loss": 4.9313,
      "mean_token_accuracy": 0.20403687804937362,
      "num_tokens": 83585995.0,
      "step": 48195
    },
    {
      "entropy": 5.356377124786377,
      "epoch": 3.750126434545808,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003642738200416409,
      "loss": 4.8104,
      "mean_token_accuracy": 0.213023741543293,
      "num_tokens": 83594985.0,
      "step": 48200
    },
    {
      "entropy": 5.366329908370972,
      "epoch": 3.7505154639175258,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036424837875629517,
      "loss": 4.831,
      "mean_token_accuracy": 0.21570476293563842,
      "num_tokens": 83603179.0,
      "step": 48205
    },
    {
      "entropy": 5.43712158203125,
      "epoch": 3.7509044932892435,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036422293611674736,
      "loss": 4.9707,
      "mean_token_accuracy": 0.2096259340643883,
      "num_tokens": 83612890.0,
      "step": 48210
    },
    {
      "entropy": 5.433262348175049,
      "epoch": 3.751293522660961,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036419749212338346,
      "loss": 4.9867,
      "mean_token_accuracy": 0.20604695081710817,
      "num_tokens": 83623249.0,
      "step": 48215
    },
    {
      "entropy": 5.407290935516357,
      "epoch": 3.7516825520326784,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003641720467765897,
      "loss": 4.837,
      "mean_token_accuracy": 0.2184054121375084,
      "num_tokens": 83632899.0,
      "step": 48220
    },
    {
      "entropy": 5.405412673950195,
      "epoch": 3.752071581404396,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000364146600076752,
      "loss": 4.9021,
      "mean_token_accuracy": 0.2075895830988884,
      "num_tokens": 83641773.0,
      "step": 48225
    },
    {
      "entropy": 5.392686557769776,
      "epoch": 3.7524606107761134,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003641211520242565,
      "loss": 4.899,
      "mean_token_accuracy": 0.21425518691539763,
      "num_tokens": 83649740.0,
      "step": 48230
    },
    {
      "entropy": 5.430963706970215,
      "epoch": 3.752849640147831,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036409570261948943,
      "loss": 4.8738,
      "mean_token_accuracy": 0.2098555698990822,
      "num_tokens": 83658764.0,
      "step": 48235
    },
    {
      "entropy": 5.304048824310303,
      "epoch": 3.7532386695195488,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003640702518628369,
      "loss": 4.7986,
      "mean_token_accuracy": 0.21282921582460404,
      "num_tokens": 83667350.0,
      "step": 48240
    },
    {
      "entropy": 5.338675260543823,
      "epoch": 3.7536276988912665,
      "grad_norm": 1.140625,
      "learning_rate": 0.000364044799754685,
      "loss": 4.8152,
      "mean_token_accuracy": 0.2132792592048645,
      "num_tokens": 83676012.0,
      "step": 48245
    },
    {
      "entropy": 5.389057636260986,
      "epoch": 3.7540167282629837,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036401934629542,
      "loss": 4.8684,
      "mean_token_accuracy": 0.22021762728691102,
      "num_tokens": 83684606.0,
      "step": 48250
    },
    {
      "entropy": 5.4090039253234865,
      "epoch": 3.7544057576347014,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036399389148542813,
      "loss": 4.8994,
      "mean_token_accuracy": 0.20691086649894713,
      "num_tokens": 83692735.0,
      "step": 48255
    },
    {
      "entropy": 5.346251630783081,
      "epoch": 3.754794787006419,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036396843532509563,
      "loss": 4.8,
      "mean_token_accuracy": 0.22138732373714448,
      "num_tokens": 83701487.0,
      "step": 48260
    },
    {
      "entropy": 5.379057550430298,
      "epoch": 3.7551838163781364,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003639429778148087,
      "loss": 4.8781,
      "mean_token_accuracy": 0.20976461917161943,
      "num_tokens": 83709660.0,
      "step": 48265
    },
    {
      "entropy": 5.415218305587769,
      "epoch": 3.755572845749854,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003639175189549536,
      "loss": 4.9148,
      "mean_token_accuracy": 0.20973466634750365,
      "num_tokens": 83719267.0,
      "step": 48270
    },
    {
      "entropy": 5.408178329467773,
      "epoch": 3.7559618751215718,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036389205874591676,
      "loss": 4.8969,
      "mean_token_accuracy": 0.2174176901578903,
      "num_tokens": 83727479.0,
      "step": 48275
    },
    {
      "entropy": 5.356724548339844,
      "epoch": 3.7563509044932895,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036386659718808437,
      "loss": 4.7476,
      "mean_token_accuracy": 0.21856663823127748,
      "num_tokens": 83735447.0,
      "step": 48280
    },
    {
      "entropy": 5.399404668807984,
      "epoch": 3.7567399338650067,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036384113428184286,
      "loss": 4.9111,
      "mean_token_accuracy": 0.20864460617303848,
      "num_tokens": 83744147.0,
      "step": 48285
    },
    {
      "entropy": 5.362719535827637,
      "epoch": 3.7571289632367244,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003638156700275784,
      "loss": 4.8915,
      "mean_token_accuracy": 0.2060146600008011,
      "num_tokens": 83752898.0,
      "step": 48290
    },
    {
      "entropy": 5.336288690567017,
      "epoch": 3.757517992608442,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003637902044256777,
      "loss": 4.8195,
      "mean_token_accuracy": 0.2140578180551529,
      "num_tokens": 83762378.0,
      "step": 48295
    },
    {
      "entropy": 5.381037473678589,
      "epoch": 3.7579070219801594,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003637647374765269,
      "loss": 4.8342,
      "mean_token_accuracy": 0.21446406990289688,
      "num_tokens": 83770938.0,
      "step": 48300
    },
    {
      "entropy": 5.365966939926148,
      "epoch": 3.758296051351877,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036373926918051243,
      "loss": 4.8427,
      "mean_token_accuracy": 0.21584258377552032,
      "num_tokens": 83778676.0,
      "step": 48305
    },
    {
      "entropy": 5.353034782409668,
      "epoch": 3.7586850807235948,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036371379953802085,
      "loss": 4.854,
      "mean_token_accuracy": 0.21512504816055297,
      "num_tokens": 83787764.0,
      "step": 48310
    },
    {
      "entropy": 5.351162576675415,
      "epoch": 3.7590741100953124,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036368832854943856,
      "loss": 4.9019,
      "mean_token_accuracy": 0.2080257922410965,
      "num_tokens": 83795913.0,
      "step": 48315
    },
    {
      "entropy": 5.3098742961883545,
      "epoch": 3.7594631394670297,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036366285621515206,
      "loss": 4.6771,
      "mean_token_accuracy": 0.22400540709495545,
      "num_tokens": 83805537.0,
      "step": 48320
    },
    {
      "entropy": 5.420927095413208,
      "epoch": 3.7598521688387474,
      "grad_norm": 1.125,
      "learning_rate": 0.0003636373825355478,
      "loss": 4.8724,
      "mean_token_accuracy": 0.20868360698223115,
      "num_tokens": 83813411.0,
      "step": 48325
    },
    {
      "entropy": 5.32913908958435,
      "epoch": 3.7602411982104647,
      "grad_norm": 1.125,
      "learning_rate": 0.0003636119075110124,
      "loss": 4.9283,
      "mean_token_accuracy": 0.20932894945144653,
      "num_tokens": 83822110.0,
      "step": 48330
    },
    {
      "entropy": 5.33929443359375,
      "epoch": 3.7606302275821823,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003635864311419323,
      "loss": 4.7887,
      "mean_token_accuracy": 0.21437810212373734,
      "num_tokens": 83830302.0,
      "step": 48335
    },
    {
      "entropy": 5.373210334777832,
      "epoch": 3.7610192569539,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036356095342869414,
      "loss": 4.7673,
      "mean_token_accuracy": 0.21564409732818604,
      "num_tokens": 83838595.0,
      "step": 48340
    },
    {
      "entropy": 5.37248969078064,
      "epoch": 3.7614082863256177,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003635354743716844,
      "loss": 4.8652,
      "mean_token_accuracy": 0.21085265129804612,
      "num_tokens": 83847036.0,
      "step": 48345
    },
    {
      "entropy": 5.402022647857666,
      "epoch": 3.761797315697335,
      "grad_norm": 1.125,
      "learning_rate": 0.0003635099939712898,
      "loss": 4.8405,
      "mean_token_accuracy": 0.21138464510440827,
      "num_tokens": 83855256.0,
      "step": 48350
    },
    {
      "entropy": 5.371267318725586,
      "epoch": 3.7621863450690527,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036348451222789694,
      "loss": 4.8173,
      "mean_token_accuracy": 0.22107304781675338,
      "num_tokens": 83864019.0,
      "step": 48355
    },
    {
      "entropy": 5.412528085708618,
      "epoch": 3.7625753744407704,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036345902914189246,
      "loss": 4.8681,
      "mean_token_accuracy": 0.21484641134738922,
      "num_tokens": 83873263.0,
      "step": 48360
    },
    {
      "entropy": 5.4595037460327145,
      "epoch": 3.7629644038124876,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000363433544713663,
      "loss": 4.9628,
      "mean_token_accuracy": 0.19982558637857437,
      "num_tokens": 83882226.0,
      "step": 48365
    },
    {
      "entropy": 5.393287086486817,
      "epoch": 3.7633534331842053,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036340805894359524,
      "loss": 4.7813,
      "mean_token_accuracy": 0.21854738891124725,
      "num_tokens": 83891140.0,
      "step": 48370
    },
    {
      "entropy": 5.32624921798706,
      "epoch": 3.763742462555923,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036338257183207587,
      "loss": 4.8205,
      "mean_token_accuracy": 0.22004197239875795,
      "num_tokens": 83899183.0,
      "step": 48375
    },
    {
      "entropy": 5.366065406799317,
      "epoch": 3.7641314919276407,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036335708337949165,
      "loss": 4.8331,
      "mean_token_accuracy": 0.2206488609313965,
      "num_tokens": 83907631.0,
      "step": 48380
    },
    {
      "entropy": 5.298793840408325,
      "epoch": 3.764520521299358,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003633315935862293,
      "loss": 4.7622,
      "mean_token_accuracy": 0.22630729526281357,
      "num_tokens": 83916201.0,
      "step": 48385
    },
    {
      "entropy": 5.34281358718872,
      "epoch": 3.7649095506710757,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003633061024526757,
      "loss": 4.8078,
      "mean_token_accuracy": 0.2214032381772995,
      "num_tokens": 83924566.0,
      "step": 48390
    },
    {
      "entropy": 5.404642868041992,
      "epoch": 3.7652985800427934,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003632806099792175,
      "loss": 4.8689,
      "mean_token_accuracy": 0.21534093469381332,
      "num_tokens": 83932766.0,
      "step": 48395
    },
    {
      "entropy": 5.386739492416382,
      "epoch": 3.7656876094145106,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036325511616624155,
      "loss": 4.7536,
      "mean_token_accuracy": 0.219247530400753,
      "num_tokens": 83941068.0,
      "step": 48400
    },
    {
      "entropy": 5.4157476902008055,
      "epoch": 3.7660766387862283,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003632296210141348,
      "loss": 4.9263,
      "mean_token_accuracy": 0.20605076253414153,
      "num_tokens": 83950711.0,
      "step": 48405
    },
    {
      "entropy": 5.314728450775147,
      "epoch": 3.766465668157946,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036320412452328383,
      "loss": 4.8563,
      "mean_token_accuracy": 0.20861955285072326,
      "num_tokens": 83959712.0,
      "step": 48410
    },
    {
      "entropy": 5.366595315933227,
      "epoch": 3.7668546975296637,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003631786266940757,
      "loss": 4.7675,
      "mean_token_accuracy": 0.2185141324996948,
      "num_tokens": 83968868.0,
      "step": 48415
    },
    {
      "entropy": 5.362298774719238,
      "epoch": 3.767243726901381,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036315312752689726,
      "loss": 4.7622,
      "mean_token_accuracy": 0.21828568279743193,
      "num_tokens": 83977620.0,
      "step": 48420
    },
    {
      "entropy": 5.3667113304138185,
      "epoch": 3.7676327562730987,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003631276270221354,
      "loss": 4.7894,
      "mean_token_accuracy": 0.21562462598085402,
      "num_tokens": 83987057.0,
      "step": 48425
    },
    {
      "entropy": 5.372176170349121,
      "epoch": 3.768021785644816,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036310212518017717,
      "loss": 4.8007,
      "mean_token_accuracy": 0.2119729071855545,
      "num_tokens": 83995851.0,
      "step": 48430
    },
    {
      "entropy": 5.365168237686158,
      "epoch": 3.7684108150165336,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036307662200140935,
      "loss": 4.8456,
      "mean_token_accuracy": 0.21332936137914657,
      "num_tokens": 84005276.0,
      "step": 48435
    },
    {
      "entropy": 5.3854395866394045,
      "epoch": 3.7687998443882513,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003630511174862189,
      "loss": 4.9195,
      "mean_token_accuracy": 0.21133777499198914,
      "num_tokens": 84014259.0,
      "step": 48440
    },
    {
      "entropy": 5.358669757843018,
      "epoch": 3.769188873759969,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003630256116349931,
      "loss": 4.8915,
      "mean_token_accuracy": 0.21274707615375518,
      "num_tokens": 84023518.0,
      "step": 48445
    },
    {
      "entropy": 5.414945983886719,
      "epoch": 3.7695779031316863,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003630001044481186,
      "loss": 4.9728,
      "mean_token_accuracy": 0.2091632142663002,
      "num_tokens": 84032274.0,
      "step": 48450
    },
    {
      "entropy": 5.469515466690064,
      "epoch": 3.769966932503404,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036297459592598256,
      "loss": 4.8924,
      "mean_token_accuracy": 0.2042225033044815,
      "num_tokens": 84041334.0,
      "step": 48455
    },
    {
      "entropy": 5.348917818069458,
      "epoch": 3.7703559618751217,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003629490860689721,
      "loss": 4.8272,
      "mean_token_accuracy": 0.21530359387397766,
      "num_tokens": 84050443.0,
      "step": 48460
    },
    {
      "entropy": 5.327214479446411,
      "epoch": 3.770744991246839,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036292357487747426,
      "loss": 4.8172,
      "mean_token_accuracy": 0.21303404718637467,
      "num_tokens": 84059130.0,
      "step": 48465
    },
    {
      "entropy": 5.37951626777649,
      "epoch": 3.7711340206185566,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036289806235187606,
      "loss": 4.7766,
      "mean_token_accuracy": 0.22568628787994385,
      "num_tokens": 84068514.0,
      "step": 48470
    },
    {
      "entropy": 5.372968912124634,
      "epoch": 3.7715230499902743,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003628725484925647,
      "loss": 4.7322,
      "mean_token_accuracy": 0.22189069986343385,
      "num_tokens": 84076551.0,
      "step": 48475
    },
    {
      "entropy": 5.417578506469726,
      "epoch": 3.771912079361992,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036284703329992734,
      "loss": 5.0068,
      "mean_token_accuracy": 0.1988322153687477,
      "num_tokens": 84086332.0,
      "step": 48480
    },
    {
      "entropy": 5.351701879501343,
      "epoch": 3.7723011087337093,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000362821516774351,
      "loss": 4.7682,
      "mean_token_accuracy": 0.22622643709182738,
      "num_tokens": 84094538.0,
      "step": 48485
    },
    {
      "entropy": 5.433412504196167,
      "epoch": 3.772690138105427,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036279599891622297,
      "loss": 4.8328,
      "mean_token_accuracy": 0.21462987065315248,
      "num_tokens": 84104034.0,
      "step": 48490
    },
    {
      "entropy": 5.357803106307983,
      "epoch": 3.7730791674771447,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036277047972593036,
      "loss": 4.8977,
      "mean_token_accuracy": 0.2121119201183319,
      "num_tokens": 84112893.0,
      "step": 48495
    },
    {
      "entropy": 5.341241121292114,
      "epoch": 3.773468196848862,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003627449592038604,
      "loss": 4.8684,
      "mean_token_accuracy": 0.21817442327737807,
      "num_tokens": 84120805.0,
      "step": 48500
    },
    {
      "entropy": 5.3794197082519535,
      "epoch": 3.7738572262205796,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003627194373504004,
      "loss": 4.8931,
      "mean_token_accuracy": 0.2160290524363518,
      "num_tokens": 84129355.0,
      "step": 48505
    },
    {
      "entropy": 5.3089488506317135,
      "epoch": 3.7742462555922973,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036269391416593753,
      "loss": 4.7937,
      "mean_token_accuracy": 0.21713491529226303,
      "num_tokens": 84137302.0,
      "step": 48510
    },
    {
      "entropy": 5.396674156188965,
      "epoch": 3.774635284964015,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003626683896508591,
      "loss": 4.9394,
      "mean_token_accuracy": 0.20451409816741944,
      "num_tokens": 84146164.0,
      "step": 48515
    },
    {
      "entropy": 5.344146966934204,
      "epoch": 3.7750243143357323,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003626428638055524,
      "loss": 4.7551,
      "mean_token_accuracy": 0.21822899132966994,
      "num_tokens": 84154495.0,
      "step": 48520
    },
    {
      "entropy": 5.525819921493531,
      "epoch": 3.77541334370745,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036261733663040473,
      "loss": 4.9442,
      "mean_token_accuracy": 0.20247694700956345,
      "num_tokens": 84163239.0,
      "step": 48525
    },
    {
      "entropy": 5.330809545516968,
      "epoch": 3.775802373079167,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036259180812580347,
      "loss": 4.802,
      "mean_token_accuracy": 0.21730223000049592,
      "num_tokens": 84171934.0,
      "step": 48530
    },
    {
      "entropy": 5.309118700027466,
      "epoch": 3.776191402450885,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036256627829213585,
      "loss": 4.8574,
      "mean_token_accuracy": 0.21307612508535384,
      "num_tokens": 84180149.0,
      "step": 48535
    },
    {
      "entropy": 5.393866062164307,
      "epoch": 3.7765804318226026,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036254074712978947,
      "loss": 4.8325,
      "mean_token_accuracy": 0.21931487023830415,
      "num_tokens": 84188290.0,
      "step": 48540
    },
    {
      "entropy": 5.3707677841186525,
      "epoch": 3.7769694611943203,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036251521463915144,
      "loss": 4.8251,
      "mean_token_accuracy": 0.2168906733393669,
      "num_tokens": 84196906.0,
      "step": 48545
    },
    {
      "entropy": 5.245473194122314,
      "epoch": 3.777358490566038,
      "grad_norm": 1.125,
      "learning_rate": 0.0003624896808206093,
      "loss": 4.7744,
      "mean_token_accuracy": 0.212023563683033,
      "num_tokens": 84205416.0,
      "step": 48550
    },
    {
      "entropy": 5.442719602584839,
      "epoch": 3.7777475199377553,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003624641456745507,
      "loss": 4.9986,
      "mean_token_accuracy": 0.198079614341259,
      "num_tokens": 84214279.0,
      "step": 48555
    },
    {
      "entropy": 5.48408088684082,
      "epoch": 3.778136549309473,
      "grad_norm": 1.125,
      "learning_rate": 0.00036243860920136283,
      "loss": 4.911,
      "mean_token_accuracy": 0.208728563785553,
      "num_tokens": 84222976.0,
      "step": 48560
    },
    {
      "entropy": 5.356388235092163,
      "epoch": 3.77852557868119,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036241307140143323,
      "loss": 4.8393,
      "mean_token_accuracy": 0.2090616136789322,
      "num_tokens": 84231315.0,
      "step": 48565
    },
    {
      "entropy": 5.330193424224854,
      "epoch": 3.778914608052908,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036238753227514934,
      "loss": 4.8484,
      "mean_token_accuracy": 0.21083036065101624,
      "num_tokens": 84239691.0,
      "step": 48570
    },
    {
      "entropy": 5.494891214370727,
      "epoch": 3.7793036374246256,
      "grad_norm": 1.25,
      "learning_rate": 0.00036236199182289883,
      "loss": 4.9346,
      "mean_token_accuracy": 0.21492374688386917,
      "num_tokens": 84247542.0,
      "step": 48575
    },
    {
      "entropy": 5.439738988876343,
      "epoch": 3.7796926667963433,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000362336450045069,
      "loss": 5.0324,
      "mean_token_accuracy": 0.20039337277412414,
      "num_tokens": 84255843.0,
      "step": 48580
    },
    {
      "entropy": 5.325422286987305,
      "epoch": 3.7800816961680606,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036231090694204763,
      "loss": 4.7882,
      "mean_token_accuracy": 0.2215324744582176,
      "num_tokens": 84264404.0,
      "step": 48585
    },
    {
      "entropy": 5.403598117828369,
      "epoch": 3.7804707255397783,
      "grad_norm": 1.125,
      "learning_rate": 0.00036228536251422226,
      "loss": 4.8369,
      "mean_token_accuracy": 0.2147928550839424,
      "num_tokens": 84273596.0,
      "step": 48590
    },
    {
      "entropy": 5.363894367218018,
      "epoch": 3.780859754911496,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036225981676198045,
      "loss": 4.7865,
      "mean_token_accuracy": 0.2110724702477455,
      "num_tokens": 84283138.0,
      "step": 48595
    },
    {
      "entropy": 5.316057395935059,
      "epoch": 3.781248784283213,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003622342696857098,
      "loss": 4.7684,
      "mean_token_accuracy": 0.21990798264741898,
      "num_tokens": 84291539.0,
      "step": 48600
    },
    {
      "entropy": 5.311081790924073,
      "epoch": 3.781637813654931,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003622087212857979,
      "loss": 4.8248,
      "mean_token_accuracy": 0.22024541795253755,
      "num_tokens": 84299896.0,
      "step": 48605
    },
    {
      "entropy": 5.383578443527222,
      "epoch": 3.7820268430266486,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003621831715626325,
      "loss": 4.8781,
      "mean_token_accuracy": 0.21606519669294358,
      "num_tokens": 84308195.0,
      "step": 48610
    },
    {
      "entropy": 5.3811619758605955,
      "epoch": 3.7824158723983663,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036215762051660115,
      "loss": 4.8778,
      "mean_token_accuracy": 0.20864836424589156,
      "num_tokens": 84316636.0,
      "step": 48615
    },
    {
      "entropy": 5.346364736557007,
      "epoch": 3.7828049017700835,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036213206814809175,
      "loss": 4.8324,
      "mean_token_accuracy": 0.21636251211166382,
      "num_tokens": 84325634.0,
      "step": 48620
    },
    {
      "entropy": 5.315078926086426,
      "epoch": 3.7831939311418012,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003621065144574919,
      "loss": 4.8053,
      "mean_token_accuracy": 0.22245784252882003,
      "num_tokens": 84334477.0,
      "step": 48625
    },
    {
      "entropy": 5.397788095474243,
      "epoch": 3.7835829605135185,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003620809594451893,
      "loss": 4.931,
      "mean_token_accuracy": 0.21175595819950105,
      "num_tokens": 84342738.0,
      "step": 48630
    },
    {
      "entropy": 5.474747991561889,
      "epoch": 3.783971989885236,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003620554031115718,
      "loss": 4.8742,
      "mean_token_accuracy": 0.20963586419820784,
      "num_tokens": 84351305.0,
      "step": 48635
    },
    {
      "entropy": 5.429792070388794,
      "epoch": 3.784361019256954,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003620298454570271,
      "loss": 4.8604,
      "mean_token_accuracy": 0.20984972268342972,
      "num_tokens": 84360500.0,
      "step": 48640
    },
    {
      "entropy": 5.4192845821380615,
      "epoch": 3.7847500486286716,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003620042864819429,
      "loss": 4.864,
      "mean_token_accuracy": 0.20919908285140992,
      "num_tokens": 84369160.0,
      "step": 48645
    },
    {
      "entropy": 5.458877325057983,
      "epoch": 3.7851390780003893,
      "grad_norm": 1.25,
      "learning_rate": 0.0003619787261867072,
      "loss": 4.9479,
      "mean_token_accuracy": 0.20863496512174606,
      "num_tokens": 84377784.0,
      "step": 48650
    },
    {
      "entropy": 5.42494010925293,
      "epoch": 3.7855281073721065,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003619531645717078,
      "loss": 4.8806,
      "mean_token_accuracy": 0.21045134216547012,
      "num_tokens": 84386502.0,
      "step": 48655
    },
    {
      "entropy": 5.392590427398682,
      "epoch": 3.7859171367438242,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003619276016373325,
      "loss": 4.9012,
      "mean_token_accuracy": 0.2129158928990364,
      "num_tokens": 84396100.0,
      "step": 48660
    },
    {
      "entropy": 5.404166603088379,
      "epoch": 3.7863061661155415,
      "grad_norm": 1.125,
      "learning_rate": 0.0003619020373839693,
      "loss": 4.8858,
      "mean_token_accuracy": 0.20430890023708342,
      "num_tokens": 84404917.0,
      "step": 48665
    },
    {
      "entropy": 5.407828998565674,
      "epoch": 3.786695195487259,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003618764718120059,
      "loss": 4.8449,
      "mean_token_accuracy": 0.21955275982618333,
      "num_tokens": 84413510.0,
      "step": 48670
    },
    {
      "entropy": 5.3997344970703125,
      "epoch": 3.787084224858977,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003618509049218304,
      "loss": 4.8862,
      "mean_token_accuracy": 0.20779266953468323,
      "num_tokens": 84421689.0,
      "step": 48675
    },
    {
      "entropy": 5.417587232589722,
      "epoch": 3.7874732542306946,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036182533671383056,
      "loss": 4.8526,
      "mean_token_accuracy": 0.211449034512043,
      "num_tokens": 84430362.0,
      "step": 48680
    },
    {
      "entropy": 5.291483783721924,
      "epoch": 3.787862283602412,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003617997671883944,
      "loss": 4.7861,
      "mean_token_accuracy": 0.22451868802309036,
      "num_tokens": 84438066.0,
      "step": 48685
    },
    {
      "entropy": 5.31300950050354,
      "epoch": 3.7882513129741295,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036177419634591,
      "loss": 4.7442,
      "mean_token_accuracy": 0.2250799521803856,
      "num_tokens": 84446995.0,
      "step": 48690
    },
    {
      "entropy": 5.3887574672698975,
      "epoch": 3.7886403423458472,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003617486241867654,
      "loss": 4.8761,
      "mean_token_accuracy": 0.21271787136793135,
      "num_tokens": 84455879.0,
      "step": 48695
    },
    {
      "entropy": 5.381903791427613,
      "epoch": 3.7890293717175645,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003617230507113483,
      "loss": 4.7296,
      "mean_token_accuracy": 0.2204664543271065,
      "num_tokens": 84464006.0,
      "step": 48700
    },
    {
      "entropy": 5.36190447807312,
      "epoch": 3.789418401089282,
      "grad_norm": 1.125,
      "learning_rate": 0.00036169747592004704,
      "loss": 4.8442,
      "mean_token_accuracy": 0.2118131309747696,
      "num_tokens": 84472116.0,
      "step": 48705
    },
    {
      "entropy": 5.369759941101075,
      "epoch": 3.789807430461,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036167189981324956,
      "loss": 4.9289,
      "mean_token_accuracy": 0.20844001621007918,
      "num_tokens": 84481303.0,
      "step": 48710
    },
    {
      "entropy": 5.440121841430664,
      "epoch": 3.7901964598327176,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000361646322391344,
      "loss": 4.7804,
      "mean_token_accuracy": 0.21621585190296172,
      "num_tokens": 84490920.0,
      "step": 48715
    },
    {
      "entropy": 5.356743907928466,
      "epoch": 3.790585489204435,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003616207436547183,
      "loss": 4.7909,
      "mean_token_accuracy": 0.22211736738681792,
      "num_tokens": 84498904.0,
      "step": 48720
    },
    {
      "entropy": 5.3964691162109375,
      "epoch": 3.7909745185761525,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036159516360376084,
      "loss": 4.983,
      "mean_token_accuracy": 0.20178569108247757,
      "num_tokens": 84507718.0,
      "step": 48725
    },
    {
      "entropy": 5.356824731826782,
      "epoch": 3.7913635479478702,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036156958223885945,
      "loss": 4.8558,
      "mean_token_accuracy": 0.21116693168878556,
      "num_tokens": 84516859.0,
      "step": 48730
    },
    {
      "entropy": 5.502367544174194,
      "epoch": 3.7917525773195875,
      "grad_norm": 1.125,
      "learning_rate": 0.0003615439995604025,
      "loss": 4.9184,
      "mean_token_accuracy": 0.21280080825090408,
      "num_tokens": 84526554.0,
      "step": 48735
    },
    {
      "entropy": 5.4064232349395756,
      "epoch": 3.792141606691305,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003615184155687781,
      "loss": 4.8248,
      "mean_token_accuracy": 0.21581604778766633,
      "num_tokens": 84535884.0,
      "step": 48740
    },
    {
      "entropy": 5.334324216842651,
      "epoch": 3.792530636063023,
      "grad_norm": 1.125,
      "learning_rate": 0.00036149283026437444,
      "loss": 4.7858,
      "mean_token_accuracy": 0.22141025960445404,
      "num_tokens": 84544022.0,
      "step": 48745
    },
    {
      "entropy": 5.378333616256714,
      "epoch": 3.7929196654347406,
      "grad_norm": 1.125,
      "learning_rate": 0.00036146724364757984,
      "loss": 4.9518,
      "mean_token_accuracy": 0.19941932410001756,
      "num_tokens": 84553408.0,
      "step": 48750
    },
    {
      "entropy": 5.36847152709961,
      "epoch": 3.793308694806458,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036144165571878233,
      "loss": 4.8414,
      "mean_token_accuracy": 0.20652897506952286,
      "num_tokens": 84562794.0,
      "step": 48755
    },
    {
      "entropy": 5.3042103290557865,
      "epoch": 3.7936977241781755,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003614160664783703,
      "loss": 4.7323,
      "mean_token_accuracy": 0.22932633459568025,
      "num_tokens": 84571064.0,
      "step": 48760
    },
    {
      "entropy": 5.419686317443848,
      "epoch": 3.7940867535498928,
      "grad_norm": 1.234375,
      "learning_rate": 0.000361390475926732,
      "loss": 4.7957,
      "mean_token_accuracy": 0.22320399582386016,
      "num_tokens": 84579343.0,
      "step": 48765
    },
    {
      "entropy": 5.394041538238525,
      "epoch": 3.7944757829216105,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003613648840642558,
      "loss": 4.9287,
      "mean_token_accuracy": 0.20517850816249847,
      "num_tokens": 84587726.0,
      "step": 48770
    },
    {
      "entropy": 5.375931024551392,
      "epoch": 3.794864812293328,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003613392908913299,
      "loss": 4.8382,
      "mean_token_accuracy": 0.2109801411628723,
      "num_tokens": 84596293.0,
      "step": 48775
    },
    {
      "entropy": 5.419444274902344,
      "epoch": 3.795253841665046,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003613136964083427,
      "loss": 4.8763,
      "mean_token_accuracy": 0.21804671585559846,
      "num_tokens": 84605102.0,
      "step": 48780
    },
    {
      "entropy": 5.449718284606933,
      "epoch": 3.795642871036763,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003612881006156826,
      "loss": 4.8833,
      "mean_token_accuracy": 0.20645925402641296,
      "num_tokens": 84613696.0,
      "step": 48785
    },
    {
      "entropy": 5.393892288208008,
      "epoch": 3.796031900408481,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036126250351373786,
      "loss": 4.8751,
      "mean_token_accuracy": 0.21512623131275177,
      "num_tokens": 84622439.0,
      "step": 48790
    },
    {
      "entropy": 5.36529688835144,
      "epoch": 3.7964209297801985,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036123690510289684,
      "loss": 4.8683,
      "mean_token_accuracy": 0.21317328065633773,
      "num_tokens": 84631167.0,
      "step": 48795
    },
    {
      "entropy": 5.429307651519776,
      "epoch": 3.7968099591519158,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003612113053835482,
      "loss": 4.9338,
      "mean_token_accuracy": 0.2109585151076317,
      "num_tokens": 84640130.0,
      "step": 48800
    },
    {
      "entropy": 5.507963991165161,
      "epoch": 3.7971989885236335,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003611857043560801,
      "loss": 4.9691,
      "mean_token_accuracy": 0.2054226815700531,
      "num_tokens": 84648719.0,
      "step": 48805
    },
    {
      "entropy": 5.4106368064880375,
      "epoch": 3.797588017895351,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003611601020208812,
      "loss": 4.8614,
      "mean_token_accuracy": 0.21271366328001023,
      "num_tokens": 84656891.0,
      "step": 48810
    },
    {
      "entropy": 5.401108694076538,
      "epoch": 3.797977047267069,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003611344983783398,
      "loss": 4.8663,
      "mean_token_accuracy": 0.21545614302158356,
      "num_tokens": 84665678.0,
      "step": 48815
    },
    {
      "entropy": 5.4180670261383055,
      "epoch": 3.798366076638786,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003611088934288445,
      "loss": 4.8211,
      "mean_token_accuracy": 0.21985283941030503,
      "num_tokens": 84674112.0,
      "step": 48820
    },
    {
      "entropy": 5.329475927352905,
      "epoch": 3.798755106010504,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036108328717278387,
      "loss": 4.8809,
      "mean_token_accuracy": 0.2107171446084976,
      "num_tokens": 84682294.0,
      "step": 48825
    },
    {
      "entropy": 5.305478143692016,
      "epoch": 3.7991441353822215,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036105767961054636,
      "loss": 4.7925,
      "mean_token_accuracy": 0.22056083083152772,
      "num_tokens": 84690399.0,
      "step": 48830
    },
    {
      "entropy": 5.394183206558227,
      "epoch": 3.7995331647539388,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003610320707425205,
      "loss": 4.8466,
      "mean_token_accuracy": 0.2176917463541031,
      "num_tokens": 84698451.0,
      "step": 48835
    },
    {
      "entropy": 5.414718151092529,
      "epoch": 3.7999221941256565,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000361006460569095,
      "loss": 4.8402,
      "mean_token_accuracy": 0.2216223493218422,
      "num_tokens": 84707808.0,
      "step": 48840
    },
    {
      "entropy": 5.413246917724609,
      "epoch": 3.800311223497374,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036098084909065814,
      "loss": 4.8931,
      "mean_token_accuracy": 0.21050593703985215,
      "num_tokens": 84716915.0,
      "step": 48845
    },
    {
      "entropy": 5.34749960899353,
      "epoch": 3.800700252869092,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003609552363075989,
      "loss": 4.8096,
      "mean_token_accuracy": 0.21784162670373916,
      "num_tokens": 84724691.0,
      "step": 48850
    },
    {
      "entropy": 5.364886999130249,
      "epoch": 3.801089282240809,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036092962222030573,
      "loss": 4.8249,
      "mean_token_accuracy": 0.2102698117494583,
      "num_tokens": 84733680.0,
      "step": 48855
    },
    {
      "entropy": 5.358042192459107,
      "epoch": 3.801478311612527,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003609040068291673,
      "loss": 4.8136,
      "mean_token_accuracy": 0.21756737679243088,
      "num_tokens": 84742722.0,
      "step": 48860
    },
    {
      "entropy": 5.330917072296143,
      "epoch": 3.801867340984244,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003608783901345724,
      "loss": 4.8004,
      "mean_token_accuracy": 0.21604165881872178,
      "num_tokens": 84751320.0,
      "step": 48865
    },
    {
      "entropy": 5.368800783157349,
      "epoch": 3.8022563703559618,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036085277213690947,
      "loss": 4.842,
      "mean_token_accuracy": 0.21910916417837142,
      "num_tokens": 84759905.0,
      "step": 48870
    },
    {
      "entropy": 5.378180885314942,
      "epoch": 3.8026453997276795,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003608271528365675,
      "loss": 4.9363,
      "mean_token_accuracy": 0.1978272795677185,
      "num_tokens": 84768821.0,
      "step": 48875
    },
    {
      "entropy": 5.397261619567871,
      "epoch": 3.803034429099397,
      "grad_norm": 1.1875,
      "learning_rate": 0.000360801532233935,
      "loss": 4.784,
      "mean_token_accuracy": 0.21749879866838456,
      "num_tokens": 84776809.0,
      "step": 48880
    },
    {
      "entropy": 5.336229944229126,
      "epoch": 3.8034234584711144,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003607759103294008,
      "loss": 4.8012,
      "mean_token_accuracy": 0.2177709475159645,
      "num_tokens": 84785043.0,
      "step": 48885
    },
    {
      "entropy": 5.297978830337525,
      "epoch": 3.803812487842832,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003607502871233538,
      "loss": 4.825,
      "mean_token_accuracy": 0.2129688337445259,
      "num_tokens": 84793369.0,
      "step": 48890
    },
    {
      "entropy": 5.285146570205688,
      "epoch": 3.80420151721455,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003607246626161826,
      "loss": 4.7341,
      "mean_token_accuracy": 0.21907636374235154,
      "num_tokens": 84801792.0,
      "step": 48895
    },
    {
      "entropy": 5.3475878715515135,
      "epoch": 3.804590546586267,
      "grad_norm": 1.109375,
      "learning_rate": 0.000360699036808276,
      "loss": 4.8143,
      "mean_token_accuracy": 0.21554414182901382,
      "num_tokens": 84810381.0,
      "step": 48900
    },
    {
      "entropy": 5.380750465393066,
      "epoch": 3.8049795759579847,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003606734097000231,
      "loss": 4.7767,
      "mean_token_accuracy": 0.21850337833166122,
      "num_tokens": 84818979.0,
      "step": 48905
    },
    {
      "entropy": 5.333633136749268,
      "epoch": 3.8053686053297024,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003606477812918125,
      "loss": 4.7922,
      "mean_token_accuracy": 0.22206557244062425,
      "num_tokens": 84827844.0,
      "step": 48910
    },
    {
      "entropy": 5.28030366897583,
      "epoch": 3.80575763470142,
      "grad_norm": 1.25,
      "learning_rate": 0.0003606221515840331,
      "loss": 4.8476,
      "mean_token_accuracy": 0.21228626519441604,
      "num_tokens": 84836336.0,
      "step": 48915
    },
    {
      "entropy": 5.3466136932373045,
      "epoch": 3.8061466640731374,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036059652057707394,
      "loss": 4.8395,
      "mean_token_accuracy": 0.22071217596530915,
      "num_tokens": 84844340.0,
      "step": 48920
    },
    {
      "entropy": 5.304082679748535,
      "epoch": 3.806535693444855,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003605708882713237,
      "loss": 4.8062,
      "mean_token_accuracy": 0.2210530400276184,
      "num_tokens": 84853201.0,
      "step": 48925
    },
    {
      "entropy": 5.343185186386108,
      "epoch": 3.806924722816573,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036054525466717145,
      "loss": 4.9108,
      "mean_token_accuracy": 0.2129480168223381,
      "num_tokens": 84861764.0,
      "step": 48930
    },
    {
      "entropy": 5.314773893356323,
      "epoch": 3.80731375218829,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003605196197650062,
      "loss": 4.7982,
      "mean_token_accuracy": 0.212201789021492,
      "num_tokens": 84870536.0,
      "step": 48935
    },
    {
      "entropy": 5.470804834365845,
      "epoch": 3.8077027815600077,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003604939835652167,
      "loss": 4.961,
      "mean_token_accuracy": 0.20273516178131104,
      "num_tokens": 84879258.0,
      "step": 48940
    },
    {
      "entropy": 5.281630563735962,
      "epoch": 3.8080918109317254,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036046834606819224,
      "loss": 4.7424,
      "mean_token_accuracy": 0.22670885175466537,
      "num_tokens": 84888212.0,
      "step": 48945
    },
    {
      "entropy": 5.390174293518067,
      "epoch": 3.808480840303443,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003604427072743216,
      "loss": 4.8683,
      "mean_token_accuracy": 0.21703223139047623,
      "num_tokens": 84896877.0,
      "step": 48950
    },
    {
      "entropy": 5.3506804466247555,
      "epoch": 3.8088698696751604,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036041706718399386,
      "loss": 4.8725,
      "mean_token_accuracy": 0.2105171948671341,
      "num_tokens": 84905594.0,
      "step": 48955
    },
    {
      "entropy": 5.332077598571777,
      "epoch": 3.809258899046878,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000360391425797598,
      "loss": 4.8344,
      "mean_token_accuracy": 0.21445717066526412,
      "num_tokens": 84914658.0,
      "step": 48960
    },
    {
      "entropy": 5.475912094116211,
      "epoch": 3.8096479284185953,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003603657831155233,
      "loss": 4.952,
      "mean_token_accuracy": 0.2087505877017975,
      "num_tokens": 84924053.0,
      "step": 48965
    },
    {
      "entropy": 5.3516120433807375,
      "epoch": 3.810036957790313,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003603401391381587,
      "loss": 4.7599,
      "mean_token_accuracy": 0.2216384917497635,
      "num_tokens": 84932445.0,
      "step": 48970
    },
    {
      "entropy": 5.331463623046875,
      "epoch": 3.8104259871620307,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036031449386589326,
      "loss": 4.8263,
      "mean_token_accuracy": 0.21812264025211334,
      "num_tokens": 84940889.0,
      "step": 48975
    },
    {
      "entropy": 5.377898216247559,
      "epoch": 3.8108150165337484,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003602888472991162,
      "loss": 4.8953,
      "mean_token_accuracy": 0.1996566101908684,
      "num_tokens": 84949314.0,
      "step": 48980
    },
    {
      "entropy": 5.40591025352478,
      "epoch": 3.811204045905466,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036026319943821664,
      "loss": 4.8468,
      "mean_token_accuracy": 0.2203821524977684,
      "num_tokens": 84958313.0,
      "step": 48985
    },
    {
      "entropy": 5.4249590873718265,
      "epoch": 3.8115930752771834,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003602375502835837,
      "loss": 4.903,
      "mean_token_accuracy": 0.21262612491846083,
      "num_tokens": 84967367.0,
      "step": 48990
    },
    {
      "entropy": 5.420601224899292,
      "epoch": 3.811982104648901,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036021189983560667,
      "loss": 4.957,
      "mean_token_accuracy": 0.2099213793873787,
      "num_tokens": 84975890.0,
      "step": 48995
    },
    {
      "entropy": 5.418737602233887,
      "epoch": 3.8123711340206183,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036018624809467466,
      "loss": 4.8721,
      "mean_token_accuracy": 0.20916310101747512,
      "num_tokens": 84985089.0,
      "step": 49000
    },
    {
      "entropy": 5.421949338912964,
      "epoch": 3.812760163392336,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036016059506117687,
      "loss": 4.8619,
      "mean_token_accuracy": 0.21321135759353638,
      "num_tokens": 84993825.0,
      "step": 49005
    },
    {
      "entropy": 5.457699489593506,
      "epoch": 3.8131491927640537,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003601349407355027,
      "loss": 4.9491,
      "mean_token_accuracy": 0.20711114257574081,
      "num_tokens": 85003301.0,
      "step": 49010
    },
    {
      "entropy": 5.392137479782105,
      "epoch": 3.8135382221357714,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003601092851180412,
      "loss": 4.9422,
      "mean_token_accuracy": 0.20204103887081146,
      "num_tokens": 85012082.0,
      "step": 49015
    },
    {
      "entropy": 5.489521598815918,
      "epoch": 3.8139272515074887,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036008362820918186,
      "loss": 4.8921,
      "mean_token_accuracy": 0.20815909504890442,
      "num_tokens": 85021299.0,
      "step": 49020
    },
    {
      "entropy": 5.383515405654907,
      "epoch": 3.8143162808792064,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003600579700093138,
      "loss": 4.8501,
      "mean_token_accuracy": 0.21293762028217317,
      "num_tokens": 85030418.0,
      "step": 49025
    },
    {
      "entropy": 5.35365047454834,
      "epoch": 3.814705310250924,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036003231051882646,
      "loss": 4.8482,
      "mean_token_accuracy": 0.2144520327448845,
      "num_tokens": 85038780.0,
      "step": 49030
    },
    {
      "entropy": 5.39434027671814,
      "epoch": 3.8150943396226413,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036000664973810914,
      "loss": 4.9476,
      "mean_token_accuracy": 0.2117200881242752,
      "num_tokens": 85048456.0,
      "step": 49035
    },
    {
      "entropy": 5.406230068206787,
      "epoch": 3.815483368994359,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003599809876675513,
      "loss": 4.8467,
      "mean_token_accuracy": 0.21495288908481597,
      "num_tokens": 85057058.0,
      "step": 49040
    },
    {
      "entropy": 5.407706642150879,
      "epoch": 3.8158723983660767,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035995532430754205,
      "loss": 4.8304,
      "mean_token_accuracy": 0.21891197562217712,
      "num_tokens": 85066156.0,
      "step": 49045
    },
    {
      "entropy": 5.428645610809326,
      "epoch": 3.8162614277377944,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003599296596584711,
      "loss": 4.8909,
      "mean_token_accuracy": 0.21328189373016357,
      "num_tokens": 85075412.0,
      "step": 49050
    },
    {
      "entropy": 5.457464027404785,
      "epoch": 3.8166504571095117,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003599039937207278,
      "loss": 4.9538,
      "mean_token_accuracy": 0.209725458920002,
      "num_tokens": 85085140.0,
      "step": 49055
    },
    {
      "entropy": 5.345560741424561,
      "epoch": 3.8170394864812294,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003598783264947015,
      "loss": 4.8291,
      "mean_token_accuracy": 0.21992829144001008,
      "num_tokens": 85093707.0,
      "step": 49060
    },
    {
      "entropy": 5.272307920455932,
      "epoch": 3.8174285158529466,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035985265798078165,
      "loss": 4.7416,
      "mean_token_accuracy": 0.22282542884349824,
      "num_tokens": 85102060.0,
      "step": 49065
    },
    {
      "entropy": 5.291296863555909,
      "epoch": 3.8178175452246643,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035982698817935783,
      "loss": 4.7643,
      "mean_token_accuracy": 0.2261415332555771,
      "num_tokens": 85110603.0,
      "step": 49070
    },
    {
      "entropy": 5.438015985488891,
      "epoch": 3.818206574596382,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003598013170908194,
      "loss": 4.8447,
      "mean_token_accuracy": 0.21482376009225845,
      "num_tokens": 85118371.0,
      "step": 49075
    },
    {
      "entropy": 5.3634607791900635,
      "epoch": 3.8185956039680997,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003597756447155559,
      "loss": 4.8731,
      "mean_token_accuracy": 0.2145504340529442,
      "num_tokens": 85127498.0,
      "step": 49080
    },
    {
      "entropy": 5.4380635738372805,
      "epoch": 3.8189846333398174,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035974997105395705,
      "loss": 4.9062,
      "mean_token_accuracy": 0.20289222598075868,
      "num_tokens": 85136321.0,
      "step": 49085
    },
    {
      "entropy": 5.43972749710083,
      "epoch": 3.8193736627115347,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035972429610641224,
      "loss": 4.8671,
      "mean_token_accuracy": 0.20698008090257644,
      "num_tokens": 85144509.0,
      "step": 49090
    },
    {
      "entropy": 5.467578649520874,
      "epoch": 3.8197626920832524,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035969861987331114,
      "loss": 4.9324,
      "mean_token_accuracy": 0.21168333888053895,
      "num_tokens": 85153151.0,
      "step": 49095
    },
    {
      "entropy": 5.373190832138062,
      "epoch": 3.8201517214549696,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003596729423550433,
      "loss": 4.8762,
      "mean_token_accuracy": 0.20938099622726442,
      "num_tokens": 85161481.0,
      "step": 49100
    },
    {
      "entropy": 5.387825536727905,
      "epoch": 3.8205407508266873,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003596472635519983,
      "loss": 4.8378,
      "mean_token_accuracy": 0.21613711565732957,
      "num_tokens": 85170540.0,
      "step": 49105
    },
    {
      "entropy": 5.376368522644043,
      "epoch": 3.820929780198405,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003596215834645657,
      "loss": 4.9133,
      "mean_token_accuracy": 0.21119810342788697,
      "num_tokens": 85179559.0,
      "step": 49110
    },
    {
      "entropy": 5.404464769363403,
      "epoch": 3.8213188095701227,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035959590209313537,
      "loss": 4.7728,
      "mean_token_accuracy": 0.21895685642957688,
      "num_tokens": 85187877.0,
      "step": 49115
    },
    {
      "entropy": 5.443701839447021,
      "epoch": 3.82170783894184,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003595702194380968,
      "loss": 4.847,
      "mean_token_accuracy": 0.21235233098268508,
      "num_tokens": 85196909.0,
      "step": 49120
    },
    {
      "entropy": 5.270828199386597,
      "epoch": 3.8220968683135577,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035954453549983984,
      "loss": 4.7282,
      "mean_token_accuracy": 0.22367016822099686,
      "num_tokens": 85205647.0,
      "step": 49125
    },
    {
      "entropy": 5.378337574005127,
      "epoch": 3.8224858976852754,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035951885027875406,
      "loss": 4.8696,
      "mean_token_accuracy": 0.205944162607193,
      "num_tokens": 85214533.0,
      "step": 49130
    },
    {
      "entropy": 5.45755295753479,
      "epoch": 3.8228749270569926,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035949316377522934,
      "loss": 4.96,
      "mean_token_accuracy": 0.20505741238594055,
      "num_tokens": 85222658.0,
      "step": 49135
    },
    {
      "entropy": 5.449260091781616,
      "epoch": 3.8232639564287103,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003594674759896553,
      "loss": 4.9227,
      "mean_token_accuracy": 0.20826692879199982,
      "num_tokens": 85231547.0,
      "step": 49140
    },
    {
      "entropy": 5.362991619110107,
      "epoch": 3.823652985800428,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003594417869224217,
      "loss": 4.857,
      "mean_token_accuracy": 0.20972586274147034,
      "num_tokens": 85240494.0,
      "step": 49145
    },
    {
      "entropy": 5.345929908752441,
      "epoch": 3.8240420151721457,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035941609657391845,
      "loss": 4.7523,
      "mean_token_accuracy": 0.22582974433898925,
      "num_tokens": 85248887.0,
      "step": 49150
    },
    {
      "entropy": 5.44267692565918,
      "epoch": 3.824431044543863,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003593904049445353,
      "loss": 4.9026,
      "mean_token_accuracy": 0.20788061022758483,
      "num_tokens": 85257472.0,
      "step": 49155
    },
    {
      "entropy": 5.308692359924317,
      "epoch": 3.8248200739155807,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035936471203466206,
      "loss": 4.8236,
      "mean_token_accuracy": 0.21016405075788497,
      "num_tokens": 85265809.0,
      "step": 49160
    },
    {
      "entropy": 5.3967290878295895,
      "epoch": 3.8252091032872984,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003593390178446886,
      "loss": 4.914,
      "mean_token_accuracy": 0.21247025579214096,
      "num_tokens": 85273933.0,
      "step": 49165
    },
    {
      "entropy": 5.409799575805664,
      "epoch": 3.8255981326590156,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003593133223750047,
      "loss": 4.8955,
      "mean_token_accuracy": 0.2122568815946579,
      "num_tokens": 85281916.0,
      "step": 49170
    },
    {
      "entropy": 5.366797256469726,
      "epoch": 3.8259871620307333,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003592876256260004,
      "loss": 4.837,
      "mean_token_accuracy": 0.20937739312648773,
      "num_tokens": 85291613.0,
      "step": 49175
    },
    {
      "entropy": 5.420622777938843,
      "epoch": 3.826376191402451,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003592619275980656,
      "loss": 4.9431,
      "mean_token_accuracy": 0.20032470375299455,
      "num_tokens": 85300027.0,
      "step": 49180
    },
    {
      "entropy": 5.45319013595581,
      "epoch": 3.8267652207741687,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035923622829159014,
      "loss": 4.8739,
      "mean_token_accuracy": 0.21081397384405137,
      "num_tokens": 85309251.0,
      "step": 49185
    },
    {
      "entropy": 5.343557739257813,
      "epoch": 3.827154250145886,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003592105277069639,
      "loss": 4.7673,
      "mean_token_accuracy": 0.21544500738382338,
      "num_tokens": 85316849.0,
      "step": 49190
    },
    {
      "entropy": 5.377697420120239,
      "epoch": 3.8275432795176036,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000359184825844577,
      "loss": 4.8873,
      "mean_token_accuracy": 0.20978275686502457,
      "num_tokens": 85325673.0,
      "step": 49195
    },
    {
      "entropy": 5.370963430404663,
      "epoch": 3.827932308889321,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035915912270481935,
      "loss": 4.8475,
      "mean_token_accuracy": 0.21689437776803971,
      "num_tokens": 85334485.0,
      "step": 49200
    },
    {
      "entropy": 5.352567291259765,
      "epoch": 3.8283213382610386,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000359133418288081,
      "loss": 4.8446,
      "mean_token_accuracy": 0.21981618851423262,
      "num_tokens": 85342459.0,
      "step": 49205
    },
    {
      "entropy": 5.402795934677124,
      "epoch": 3.8287103676327563,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035910771259475187,
      "loss": 4.8633,
      "mean_token_accuracy": 0.21429249346256257,
      "num_tokens": 85351163.0,
      "step": 49210
    },
    {
      "entropy": 5.410806608200073,
      "epoch": 3.829099397004474,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003590820056252222,
      "loss": 4.9107,
      "mean_token_accuracy": 0.20597064048051833,
      "num_tokens": 85359378.0,
      "step": 49215
    },
    {
      "entropy": 5.427314424514771,
      "epoch": 3.8294884263761912,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003590562973798818,
      "loss": 4.9246,
      "mean_token_accuracy": 0.21375837922096252,
      "num_tokens": 85368113.0,
      "step": 49220
    },
    {
      "entropy": 5.354839181900024,
      "epoch": 3.829877455747909,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003590305878591209,
      "loss": 4.8696,
      "mean_token_accuracy": 0.20545235723257066,
      "num_tokens": 85376757.0,
      "step": 49225
    },
    {
      "entropy": 5.452987337112427,
      "epoch": 3.8302664851196266,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035900487706332963,
      "loss": 4.9879,
      "mean_token_accuracy": 0.20366422235965728,
      "num_tokens": 85385382.0,
      "step": 49230
    },
    {
      "entropy": 5.382672929763794,
      "epoch": 3.830655514491344,
      "grad_norm": 1.125,
      "learning_rate": 0.000358979164992898,
      "loss": 4.786,
      "mean_token_accuracy": 0.22284211069345475,
      "num_tokens": 85393947.0,
      "step": 49235
    },
    {
      "entropy": 5.367537784576416,
      "epoch": 3.8310445438630616,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035895345164821624,
      "loss": 4.7522,
      "mean_token_accuracy": 0.22473657727241517,
      "num_tokens": 85402083.0,
      "step": 49240
    },
    {
      "entropy": 5.341214275360107,
      "epoch": 3.8314335732347793,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003589277370296745,
      "loss": 4.8692,
      "mean_token_accuracy": 0.21610562652349471,
      "num_tokens": 85410589.0,
      "step": 49245
    },
    {
      "entropy": 5.448135852813721,
      "epoch": 3.831822602606497,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035890202113766283,
      "loss": 4.8931,
      "mean_token_accuracy": 0.21050004065036773,
      "num_tokens": 85419413.0,
      "step": 49250
    },
    {
      "entropy": 5.46400260925293,
      "epoch": 3.8322116319782142,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035887630397257167,
      "loss": 4.927,
      "mean_token_accuracy": 0.2058223694562912,
      "num_tokens": 85428539.0,
      "step": 49255
    },
    {
      "entropy": 5.347223424911499,
      "epoch": 3.832600661349932,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000358850585534791,
      "loss": 4.858,
      "mean_token_accuracy": 0.21469225287437438,
      "num_tokens": 85437260.0,
      "step": 49260
    },
    {
      "entropy": 5.35832724571228,
      "epoch": 3.8329896907216496,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035882486582471124,
      "loss": 4.8716,
      "mean_token_accuracy": 0.21331861168146132,
      "num_tokens": 85446026.0,
      "step": 49265
    },
    {
      "entropy": 5.397970247268677,
      "epoch": 3.833378720093367,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003587991448427225,
      "loss": 4.9174,
      "mean_token_accuracy": 0.20954670906066894,
      "num_tokens": 85455024.0,
      "step": 49270
    },
    {
      "entropy": 5.313426685333252,
      "epoch": 3.8337677494650846,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035877342258921514,
      "loss": 4.7859,
      "mean_token_accuracy": 0.22052097022533418,
      "num_tokens": 85464089.0,
      "step": 49275
    },
    {
      "entropy": 5.406452751159668,
      "epoch": 3.8341567788368023,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035874769906457936,
      "loss": 4.925,
      "mean_token_accuracy": 0.21464455723762513,
      "num_tokens": 85471909.0,
      "step": 49280
    },
    {
      "entropy": 5.272440576553345,
      "epoch": 3.83454580820852,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035872197426920567,
      "loss": 4.7333,
      "mean_token_accuracy": 0.2214886650443077,
      "num_tokens": 85480291.0,
      "step": 49285
    },
    {
      "entropy": 5.314004230499267,
      "epoch": 3.8349348375802372,
      "grad_norm": 1.25,
      "learning_rate": 0.0003586962482034842,
      "loss": 4.7386,
      "mean_token_accuracy": 0.22440838664770127,
      "num_tokens": 85489155.0,
      "step": 49290
    },
    {
      "entropy": 5.260528039932251,
      "epoch": 3.835323866951955,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035867052086780544,
      "loss": 4.7547,
      "mean_token_accuracy": 0.2201266512274742,
      "num_tokens": 85497610.0,
      "step": 49295
    },
    {
      "entropy": 5.2254434585571286,
      "epoch": 3.835712896323672,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003586447922625596,
      "loss": 4.7554,
      "mean_token_accuracy": 0.2247130960226059,
      "num_tokens": 85506037.0,
      "step": 49300
    },
    {
      "entropy": 5.441733312606812,
      "epoch": 3.83610192569539,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035861906238813726,
      "loss": 4.9222,
      "mean_token_accuracy": 0.20879645198583602,
      "num_tokens": 85515154.0,
      "step": 49305
    },
    {
      "entropy": 5.433365535736084,
      "epoch": 3.8364909550671076,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035859333124492866,
      "loss": 4.8582,
      "mean_token_accuracy": 0.2182568773627281,
      "num_tokens": 85524090.0,
      "step": 49310
    },
    {
      "entropy": 5.443056583404541,
      "epoch": 3.8368799844388253,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003585675988333244,
      "loss": 4.8925,
      "mean_token_accuracy": 0.21472567170858384,
      "num_tokens": 85532946.0,
      "step": 49315
    },
    {
      "entropy": 5.400026273727417,
      "epoch": 3.837269013810543,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003585418651537147,
      "loss": 4.8107,
      "mean_token_accuracy": 0.21380586475133895,
      "num_tokens": 85542720.0,
      "step": 49320
    },
    {
      "entropy": 5.352134037017822,
      "epoch": 3.8376580431822602,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035851613020649036,
      "loss": 4.8377,
      "mean_token_accuracy": 0.22012614458799362,
      "num_tokens": 85551710.0,
      "step": 49325
    },
    {
      "entropy": 5.406709241867065,
      "epoch": 3.838047072553978,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035849039399204157,
      "loss": 4.8768,
      "mean_token_accuracy": 0.21051025837659837,
      "num_tokens": 85560270.0,
      "step": 49330
    },
    {
      "entropy": 5.3659967422485355,
      "epoch": 3.838436101925695,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035846465651075897,
      "loss": 4.8132,
      "mean_token_accuracy": 0.2115047514438629,
      "num_tokens": 85568166.0,
      "step": 49335
    },
    {
      "entropy": 5.357828950881958,
      "epoch": 3.838825131297413,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035843891776303304,
      "loss": 4.8677,
      "mean_token_accuracy": 0.21093716323375702,
      "num_tokens": 85576183.0,
      "step": 49340
    },
    {
      "entropy": 5.3142406940460205,
      "epoch": 3.8392141606691306,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003584131777492543,
      "loss": 4.7841,
      "mean_token_accuracy": 0.21688947975635528,
      "num_tokens": 85584789.0,
      "step": 49345
    },
    {
      "entropy": 5.393529272079467,
      "epoch": 3.8396031900408483,
      "grad_norm": 1.125,
      "learning_rate": 0.0003583874364698134,
      "loss": 4.9237,
      "mean_token_accuracy": 0.21380278915166856,
      "num_tokens": 85593094.0,
      "step": 49350
    },
    {
      "entropy": 5.457193088531494,
      "epoch": 3.8399922194125655,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003583616939251009,
      "loss": 5.0086,
      "mean_token_accuracy": 0.198978029191494,
      "num_tokens": 85601673.0,
      "step": 49355
    },
    {
      "entropy": 5.431853246688843,
      "epoch": 3.840381248784283,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003583359501155073,
      "loss": 4.9163,
      "mean_token_accuracy": 0.2131246730685234,
      "num_tokens": 85609622.0,
      "step": 49360
    },
    {
      "entropy": 5.405243158340454,
      "epoch": 3.840770278156001,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003583102050414234,
      "loss": 4.9109,
      "mean_token_accuracy": 0.20622801780700684,
      "num_tokens": 85619131.0,
      "step": 49365
    },
    {
      "entropy": 5.317914056777954,
      "epoch": 3.841159307527718,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035828445870323967,
      "loss": 4.7956,
      "mean_token_accuracy": 0.21311789900064468,
      "num_tokens": 85626802.0,
      "step": 49370
    },
    {
      "entropy": 5.336965036392212,
      "epoch": 3.841548336899436,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003582587111013469,
      "loss": 4.7895,
      "mean_token_accuracy": 0.2215505674481392,
      "num_tokens": 85635927.0,
      "step": 49375
    },
    {
      "entropy": 5.359397077560425,
      "epoch": 3.8419373662711536,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003582329622361356,
      "loss": 4.7835,
      "mean_token_accuracy": 0.21502152532339097,
      "num_tokens": 85644152.0,
      "step": 49380
    },
    {
      "entropy": 5.377920246124267,
      "epoch": 3.8423263956428713,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003582072121079966,
      "loss": 4.8501,
      "mean_token_accuracy": 0.21364355683326722,
      "num_tokens": 85652436.0,
      "step": 49385
    },
    {
      "entropy": 5.372662019729614,
      "epoch": 3.8427154250145885,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035818146071732064,
      "loss": 4.8996,
      "mean_token_accuracy": 0.20561764985322953,
      "num_tokens": 85661836.0,
      "step": 49390
    },
    {
      "entropy": 5.319490909576416,
      "epoch": 3.843104454386306,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035815570806449845,
      "loss": 4.8571,
      "mean_token_accuracy": 0.20462923496961594,
      "num_tokens": 85671079.0,
      "step": 49395
    },
    {
      "entropy": 5.40166711807251,
      "epoch": 3.8434934837580235,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035812995414992075,
      "loss": 4.855,
      "mean_token_accuracy": 0.21832002997398375,
      "num_tokens": 85680226.0,
      "step": 49400
    },
    {
      "entropy": 5.398334550857544,
      "epoch": 3.843882513129741,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035810419897397823,
      "loss": 4.8103,
      "mean_token_accuracy": 0.21397507935762405,
      "num_tokens": 85688161.0,
      "step": 49405
    },
    {
      "entropy": 5.3294501304626465,
      "epoch": 3.844271542501459,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003580784425370618,
      "loss": 4.7938,
      "mean_token_accuracy": 0.22475031465291978,
      "num_tokens": 85697065.0,
      "step": 49410
    },
    {
      "entropy": 5.406174516677856,
      "epoch": 3.8446605718731766,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003580526848395622,
      "loss": 4.8546,
      "mean_token_accuracy": 0.2149505689740181,
      "num_tokens": 85705863.0,
      "step": 49415
    },
    {
      "entropy": 5.445010805130005,
      "epoch": 3.8450496012448943,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035802692588187036,
      "loss": 4.8976,
      "mean_token_accuracy": 0.20930345058441163,
      "num_tokens": 85714475.0,
      "step": 49420
    },
    {
      "entropy": 5.40729832649231,
      "epoch": 3.8454386306166115,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000358001165664377,
      "loss": 4.8743,
      "mean_token_accuracy": 0.21048289835453032,
      "num_tokens": 85723217.0,
      "step": 49425
    },
    {
      "entropy": 5.438900184631348,
      "epoch": 3.845827659988329,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003579754041874731,
      "loss": 4.8592,
      "mean_token_accuracy": 0.2158377930521965,
      "num_tokens": 85731324.0,
      "step": 49430
    },
    {
      "entropy": 5.426533651351929,
      "epoch": 3.8462166893600465,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003579496414515495,
      "loss": 4.8471,
      "mean_token_accuracy": 0.2072739452123642,
      "num_tokens": 85739407.0,
      "step": 49435
    },
    {
      "entropy": 5.414934825897217,
      "epoch": 3.846605718731764,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035792387745699715,
      "loss": 4.8036,
      "mean_token_accuracy": 0.2222285345196724,
      "num_tokens": 85747749.0,
      "step": 49440
    },
    {
      "entropy": 5.433607149124145,
      "epoch": 3.846994748103482,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003578981122042069,
      "loss": 4.8901,
      "mean_token_accuracy": 0.20729900449514388,
      "num_tokens": 85756008.0,
      "step": 49445
    },
    {
      "entropy": 5.334391975402832,
      "epoch": 3.8473837774751996,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003578723456935698,
      "loss": 4.8862,
      "mean_token_accuracy": 0.21168475598096848,
      "num_tokens": 85764796.0,
      "step": 49450
    },
    {
      "entropy": 5.467380428314209,
      "epoch": 3.847772806846917,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003578465779254768,
      "loss": 4.8482,
      "mean_token_accuracy": 0.2108509510755539,
      "num_tokens": 85773160.0,
      "step": 49455
    },
    {
      "entropy": 5.392522096633911,
      "epoch": 3.8481618362186345,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035782080890031876,
      "loss": 4.84,
      "mean_token_accuracy": 0.21254764795303344,
      "num_tokens": 85781625.0,
      "step": 49460
    },
    {
      "entropy": 5.340482091903686,
      "epoch": 3.848550865590352,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035779503861848677,
      "loss": 4.8146,
      "mean_token_accuracy": 0.2167460709810257,
      "num_tokens": 85790173.0,
      "step": 49465
    },
    {
      "entropy": 5.3352930545806885,
      "epoch": 3.8489398949620695,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003577692670803719,
      "loss": 4.8161,
      "mean_token_accuracy": 0.21474898755550384,
      "num_tokens": 85799965.0,
      "step": 49470
    },
    {
      "entropy": 5.428297138214111,
      "epoch": 3.849328924333787,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003577434942863651,
      "loss": 4.8984,
      "mean_token_accuracy": 0.21247785091400145,
      "num_tokens": 85809293.0,
      "step": 49475
    },
    {
      "entropy": 5.417802476882935,
      "epoch": 3.849717953705505,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003577177202368575,
      "loss": 4.9235,
      "mean_token_accuracy": 0.2178581103682518,
      "num_tokens": 85818321.0,
      "step": 49480
    },
    {
      "entropy": 5.380562973022461,
      "epoch": 3.8501069830772225,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035769194493224014,
      "loss": 4.8311,
      "mean_token_accuracy": 0.2170521527528763,
      "num_tokens": 85826709.0,
      "step": 49485
    },
    {
      "entropy": 5.396248626708984,
      "epoch": 3.85049601244894,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035766616837290413,
      "loss": 4.9298,
      "mean_token_accuracy": 0.20876024961471557,
      "num_tokens": 85835630.0,
      "step": 49490
    },
    {
      "entropy": 5.326535940170288,
      "epoch": 3.8508850418206575,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003576403905592406,
      "loss": 4.7735,
      "mean_token_accuracy": 0.21734911501407622,
      "num_tokens": 85845133.0,
      "step": 49495
    },
    {
      "entropy": 5.40862250328064,
      "epoch": 3.8512740711923747,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035761461149164066,
      "loss": 4.8787,
      "mean_token_accuracy": 0.2097112938761711,
      "num_tokens": 85854589.0,
      "step": 49500
    },
    {
      "entropy": 5.400992393493652,
      "epoch": 3.8516631005640924,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035758883117049543,
      "loss": 4.8162,
      "mean_token_accuracy": 0.2167309269309044,
      "num_tokens": 85863037.0,
      "step": 49505
    },
    {
      "entropy": 5.225657987594604,
      "epoch": 3.85205212993581,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003575630495961962,
      "loss": 4.6946,
      "mean_token_accuracy": 0.22563563734292985,
      "num_tokens": 85871450.0,
      "step": 49510
    },
    {
      "entropy": 5.343592596054077,
      "epoch": 3.852441159307528,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035753726676913413,
      "loss": 4.8916,
      "mean_token_accuracy": 0.21279727518558503,
      "num_tokens": 85879962.0,
      "step": 49515
    },
    {
      "entropy": 5.451533222198487,
      "epoch": 3.8528301886792455,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035751148268970046,
      "loss": 4.9377,
      "mean_token_accuracy": 0.21166047751903533,
      "num_tokens": 85888236.0,
      "step": 49520
    },
    {
      "entropy": 5.384769344329834,
      "epoch": 3.853219218050963,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003574856973582863,
      "loss": 4.806,
      "mean_token_accuracy": 0.21624423414468766,
      "num_tokens": 85897168.0,
      "step": 49525
    },
    {
      "entropy": 5.308226871490478,
      "epoch": 3.8536082474226805,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035745991077528296,
      "loss": 4.7305,
      "mean_token_accuracy": 0.22079455256462097,
      "num_tokens": 85905156.0,
      "step": 49530
    },
    {
      "entropy": 5.283204793930054,
      "epoch": 3.8539972767943977,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035743412294108175,
      "loss": 4.7777,
      "mean_token_accuracy": 0.21717469394207,
      "num_tokens": 85913050.0,
      "step": 49535
    },
    {
      "entropy": 5.225965118408203,
      "epoch": 3.8543863061661154,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003574083338560739,
      "loss": 4.674,
      "mean_token_accuracy": 0.22293821573257447,
      "num_tokens": 85921983.0,
      "step": 49540
    },
    {
      "entropy": 5.3748854637146,
      "epoch": 3.854775335537833,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035738254352065074,
      "loss": 4.7434,
      "mean_token_accuracy": 0.21783807426691054,
      "num_tokens": 85930730.0,
      "step": 49545
    },
    {
      "entropy": 5.398989248275757,
      "epoch": 3.855164364909551,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035735675193520364,
      "loss": 4.9014,
      "mean_token_accuracy": 0.20490865558385848,
      "num_tokens": 85939372.0,
      "step": 49550
    },
    {
      "entropy": 5.3850404739379885,
      "epoch": 3.855553394281268,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003573309591001239,
      "loss": 4.8434,
      "mean_token_accuracy": 0.21465734839439393,
      "num_tokens": 85947701.0,
      "step": 49555
    },
    {
      "entropy": 5.375218629837036,
      "epoch": 3.855942423652986,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000357305165015803,
      "loss": 4.8523,
      "mean_token_accuracy": 0.2174821451306343,
      "num_tokens": 85955314.0,
      "step": 49560
    },
    {
      "entropy": 5.434342241287231,
      "epoch": 3.8563314530247035,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035727936968263206,
      "loss": 4.8534,
      "mean_token_accuracy": 0.21463609039783477,
      "num_tokens": 85964236.0,
      "step": 49565
    },
    {
      "entropy": 5.354101848602295,
      "epoch": 3.8567204823964207,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003572535731010028,
      "loss": 4.7564,
      "mean_token_accuracy": 0.2221148893237114,
      "num_tokens": 85972287.0,
      "step": 49570
    },
    {
      "entropy": 5.295447111129761,
      "epoch": 3.8571095117681384,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003572277752713063,
      "loss": 4.8188,
      "mean_token_accuracy": 0.21898875385522842,
      "num_tokens": 85980661.0,
      "step": 49575
    },
    {
      "entropy": 5.339267778396606,
      "epoch": 3.857498541139856,
      "grad_norm": 1.125,
      "learning_rate": 0.0003572019761939343,
      "loss": 4.8643,
      "mean_token_accuracy": 0.22298433035612106,
      "num_tokens": 85989397.0,
      "step": 49580
    },
    {
      "entropy": 5.363077306747437,
      "epoch": 3.857887570511574,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035717617586927814,
      "loss": 4.8266,
      "mean_token_accuracy": 0.21804153621196748,
      "num_tokens": 85997842.0,
      "step": 49585
    },
    {
      "entropy": 5.375947380065918,
      "epoch": 3.858276599883291,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035715037429772935,
      "loss": 4.8654,
      "mean_token_accuracy": 0.21511486023664475,
      "num_tokens": 86006878.0,
      "step": 49590
    },
    {
      "entropy": 5.4156862735748295,
      "epoch": 3.8586656292550088,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003571245714796793,
      "loss": 4.8804,
      "mean_token_accuracy": 0.21092199981212617,
      "num_tokens": 86015443.0,
      "step": 49595
    },
    {
      "entropy": 5.426168489456177,
      "epoch": 3.8590546586267265,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003570987674155197,
      "loss": 4.9288,
      "mean_token_accuracy": 0.2071852818131447,
      "num_tokens": 86024588.0,
      "step": 49600
    },
    {
      "entropy": 5.442469644546509,
      "epoch": 3.8594436879984437,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003570729621056419,
      "loss": 4.8806,
      "mean_token_accuracy": 0.21255477666854858,
      "num_tokens": 86033175.0,
      "step": 49605
    },
    {
      "entropy": 5.3730326175689695,
      "epoch": 3.8598327173701614,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035704715555043753,
      "loss": 4.811,
      "mean_token_accuracy": 0.21449954062700272,
      "num_tokens": 86041422.0,
      "step": 49610
    },
    {
      "entropy": 5.464237928390503,
      "epoch": 3.860221746741879,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035702134775029813,
      "loss": 5.0295,
      "mean_token_accuracy": 0.19675232619047164,
      "num_tokens": 86050186.0,
      "step": 49615
    },
    {
      "entropy": 5.479752922058106,
      "epoch": 3.860610776113597,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003569955387056154,
      "loss": 4.9069,
      "mean_token_accuracy": 0.2154562696814537,
      "num_tokens": 86057809.0,
      "step": 49620
    },
    {
      "entropy": 5.391401052474976,
      "epoch": 3.860999805485314,
      "grad_norm": 1.125,
      "learning_rate": 0.00035696972841678094,
      "loss": 4.8645,
      "mean_token_accuracy": 0.21277629882097243,
      "num_tokens": 86066127.0,
      "step": 49625
    },
    {
      "entropy": 5.3678525447845455,
      "epoch": 3.8613888348570318,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003569439168841862,
      "loss": 4.8429,
      "mean_token_accuracy": 0.21454396098852158,
      "num_tokens": 86075013.0,
      "step": 49630
    },
    {
      "entropy": 5.410279607772827,
      "epoch": 3.861777864228749,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035691810410822303,
      "loss": 4.8188,
      "mean_token_accuracy": 0.21628714203834534,
      "num_tokens": 86083140.0,
      "step": 49635
    },
    {
      "entropy": 5.376409482955933,
      "epoch": 3.8621668936004667,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035689229008928304,
      "loss": 4.7809,
      "mean_token_accuracy": 0.2223998248577118,
      "num_tokens": 86091882.0,
      "step": 49640
    },
    {
      "entropy": 5.385114049911499,
      "epoch": 3.8625559229721844,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035686647482775774,
      "loss": 4.8486,
      "mean_token_accuracy": 0.22324987798929213,
      "num_tokens": 86100588.0,
      "step": 49645
    },
    {
      "entropy": 5.414225912094116,
      "epoch": 3.862944952343902,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035684065832403903,
      "loss": 4.875,
      "mean_token_accuracy": 0.21872392445802688,
      "num_tokens": 86108859.0,
      "step": 49650
    },
    {
      "entropy": 5.377131652832031,
      "epoch": 3.8633339817156194,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003568148405785187,
      "loss": 4.8259,
      "mean_token_accuracy": 0.21533940434455873,
      "num_tokens": 86117978.0,
      "step": 49655
    },
    {
      "entropy": 5.429769992828369,
      "epoch": 3.863723011087337,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003567890215915883,
      "loss": 4.8971,
      "mean_token_accuracy": 0.20989787578582764,
      "num_tokens": 86127265.0,
      "step": 49660
    },
    {
      "entropy": 5.428150367736817,
      "epoch": 3.8641120404590548,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035676320136363975,
      "loss": 4.8643,
      "mean_token_accuracy": 0.2101878419518471,
      "num_tokens": 86136064.0,
      "step": 49665
    },
    {
      "entropy": 5.495558500289917,
      "epoch": 3.864501069830772,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003567373798950647,
      "loss": 4.9059,
      "mean_token_accuracy": 0.20507902204990386,
      "num_tokens": 86144550.0,
      "step": 49670
    },
    {
      "entropy": 5.458512496948242,
      "epoch": 3.8648900992024897,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035671155718625504,
      "loss": 4.953,
      "mean_token_accuracy": 0.20621129721403123,
      "num_tokens": 86153776.0,
      "step": 49675
    },
    {
      "entropy": 5.439413118362427,
      "epoch": 3.8652791285742074,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003566857332376025,
      "loss": 4.89,
      "mean_token_accuracy": 0.21585310250520706,
      "num_tokens": 86162903.0,
      "step": 49680
    },
    {
      "entropy": 5.437718629837036,
      "epoch": 3.865668157945925,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035665990804949904,
      "loss": 4.9313,
      "mean_token_accuracy": 0.20729794651269912,
      "num_tokens": 86171521.0,
      "step": 49685
    },
    {
      "entropy": 5.370537137985229,
      "epoch": 3.8660571873176424,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003566340816223364,
      "loss": 4.86,
      "mean_token_accuracy": 0.21997511386871338,
      "num_tokens": 86180057.0,
      "step": 49690
    },
    {
      "entropy": 5.480142402648926,
      "epoch": 3.86644621668936,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003566082539565066,
      "loss": 4.8901,
      "mean_token_accuracy": 0.21017359048128129,
      "num_tokens": 86188497.0,
      "step": 49695
    },
    {
      "entropy": 5.390807676315307,
      "epoch": 3.8668352460610778,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003565824250524013,
      "loss": 4.7859,
      "mean_token_accuracy": 0.21848454773426057,
      "num_tokens": 86197454.0,
      "step": 49700
    },
    {
      "entropy": 5.372758960723877,
      "epoch": 3.867224275432795,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035655659491041266,
      "loss": 4.8713,
      "mean_token_accuracy": 0.2149476021528244,
      "num_tokens": 86206680.0,
      "step": 49705
    },
    {
      "entropy": 5.381614446640015,
      "epoch": 3.8676133048045127,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003565307635309325,
      "loss": 4.8167,
      "mean_token_accuracy": 0.22494980543851853,
      "num_tokens": 86215214.0,
      "step": 49710
    },
    {
      "entropy": 5.396752262115479,
      "epoch": 3.8680023341762304,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003565049309143527,
      "loss": 4.8718,
      "mean_token_accuracy": 0.21875571757555007,
      "num_tokens": 86223506.0,
      "step": 49715
    },
    {
      "entropy": 5.380212783813477,
      "epoch": 3.868391363547948,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035647909706106533,
      "loss": 4.8311,
      "mean_token_accuracy": 0.21216350495815278,
      "num_tokens": 86231875.0,
      "step": 49720
    },
    {
      "entropy": 5.438787794113159,
      "epoch": 3.8687803929196654,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003564532619714624,
      "loss": 4.8315,
      "mean_token_accuracy": 0.2131131500005722,
      "num_tokens": 86240920.0,
      "step": 49725
    },
    {
      "entropy": 5.495491170883179,
      "epoch": 3.869169422291383,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003564274256459358,
      "loss": 4.9986,
      "mean_token_accuracy": 0.2044559046626091,
      "num_tokens": 86248815.0,
      "step": 49730
    },
    {
      "entropy": 5.406834125518799,
      "epoch": 3.8695584516631003,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003564015880848777,
      "loss": 4.8525,
      "mean_token_accuracy": 0.2138305574655533,
      "num_tokens": 86257251.0,
      "step": 49735
    },
    {
      "entropy": 5.468204689025879,
      "epoch": 3.869947481034818,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035637574928867996,
      "loss": 4.9404,
      "mean_token_accuracy": 0.2075471594929695,
      "num_tokens": 86266349.0,
      "step": 49740
    },
    {
      "entropy": 5.41004490852356,
      "epoch": 3.8703365104065357,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035634990925773475,
      "loss": 4.8743,
      "mean_token_accuracy": 0.2104436159133911,
      "num_tokens": 86274417.0,
      "step": 49745
    },
    {
      "entropy": 5.3878051280975345,
      "epoch": 3.8707255397782534,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003563240679924342,
      "loss": 4.8831,
      "mean_token_accuracy": 0.2164582446217537,
      "num_tokens": 86283637.0,
      "step": 49750
    },
    {
      "entropy": 5.329652547836304,
      "epoch": 3.871114569149971,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003562982254931704,
      "loss": 4.8158,
      "mean_token_accuracy": 0.21619013398885728,
      "num_tokens": 86292544.0,
      "step": 49755
    },
    {
      "entropy": 5.369157838821411,
      "epoch": 3.8715035985216883,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003562723817603354,
      "loss": 4.8342,
      "mean_token_accuracy": 0.21063119769096375,
      "num_tokens": 86301654.0,
      "step": 49760
    },
    {
      "entropy": 5.42558741569519,
      "epoch": 3.871892627893406,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035624653679432135,
      "loss": 4.8344,
      "mean_token_accuracy": 0.21468476206064224,
      "num_tokens": 86310158.0,
      "step": 49765
    },
    {
      "entropy": 5.424980926513672,
      "epoch": 3.8722816572651233,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035622069059552036,
      "loss": 4.8572,
      "mean_token_accuracy": 0.22494109869003295,
      "num_tokens": 86318233.0,
      "step": 49770
    },
    {
      "entropy": 5.3506293296813965,
      "epoch": 3.872670686636841,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003561948431643247,
      "loss": 4.8921,
      "mean_token_accuracy": 0.21584369391202926,
      "num_tokens": 86327538.0,
      "step": 49775
    },
    {
      "entropy": 5.43976936340332,
      "epoch": 3.8730597160085587,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035616899450112655,
      "loss": 4.8766,
      "mean_token_accuracy": 0.21599012911319732,
      "num_tokens": 86336351.0,
      "step": 49780
    },
    {
      "entropy": 5.435955476760864,
      "epoch": 3.8734487453802764,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035614314460631814,
      "loss": 4.8214,
      "mean_token_accuracy": 0.21309615671634674,
      "num_tokens": 86344591.0,
      "step": 49785
    },
    {
      "entropy": 5.337998294830323,
      "epoch": 3.8738377747519936,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003561172934802917,
      "loss": 4.8595,
      "mean_token_accuracy": 0.2178523898124695,
      "num_tokens": 86354155.0,
      "step": 49790
    },
    {
      "entropy": 5.335957479476929,
      "epoch": 3.8742268041237113,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003560914411234393,
      "loss": 4.838,
      "mean_token_accuracy": 0.21353577673435212,
      "num_tokens": 86363510.0,
      "step": 49795
    },
    {
      "entropy": 5.392968320846558,
      "epoch": 3.874615833495429,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003560655875361535,
      "loss": 4.774,
      "mean_token_accuracy": 0.22466313242912292,
      "num_tokens": 86372226.0,
      "step": 49800
    },
    {
      "entropy": 5.364480352401733,
      "epoch": 3.8750048628671463,
      "grad_norm": 1.25,
      "learning_rate": 0.0003560397327188263,
      "loss": 4.8801,
      "mean_token_accuracy": 0.20917447060346603,
      "num_tokens": 86380286.0,
      "step": 49805
    },
    {
      "entropy": 5.378635358810425,
      "epoch": 3.875393892238864,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003560138766718502,
      "loss": 4.8864,
      "mean_token_accuracy": 0.21298673003911972,
      "num_tokens": 86388566.0,
      "step": 49810
    },
    {
      "entropy": 5.425533771514893,
      "epoch": 3.8757829216105817,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035598801939561755,
      "loss": 4.7879,
      "mean_token_accuracy": 0.22374901920557022,
      "num_tokens": 86397151.0,
      "step": 49815
    },
    {
      "entropy": 5.374747610092163,
      "epoch": 3.8761719509822994,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035596216089052045,
      "loss": 4.901,
      "mean_token_accuracy": 0.20932171642780303,
      "num_tokens": 86405530.0,
      "step": 49820
    },
    {
      "entropy": 5.3583649635314945,
      "epoch": 3.8765609803540166,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035593630115695154,
      "loss": 4.7559,
      "mean_token_accuracy": 0.21755784004926682,
      "num_tokens": 86414292.0,
      "step": 49825
    },
    {
      "entropy": 5.431084680557251,
      "epoch": 3.8769500097257343,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003559104401953031,
      "loss": 4.832,
      "mean_token_accuracy": 0.22004285901784898,
      "num_tokens": 86422656.0,
      "step": 49830
    },
    {
      "entropy": 5.4137303829193115,
      "epoch": 3.8773390390974516,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035588457800596747,
      "loss": 4.9335,
      "mean_token_accuracy": 0.20738827288150788,
      "num_tokens": 86430783.0,
      "step": 49835
    },
    {
      "entropy": 5.35724983215332,
      "epoch": 3.8777280684691693,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003558587145893371,
      "loss": 4.7731,
      "mean_token_accuracy": 0.22251655012369156,
      "num_tokens": 86439137.0,
      "step": 49840
    },
    {
      "entropy": 5.359764957427979,
      "epoch": 3.878117097840887,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003558328499458044,
      "loss": 4.8205,
      "mean_token_accuracy": 0.22498827129602433,
      "num_tokens": 86447225.0,
      "step": 49845
    },
    {
      "entropy": 5.348977708816529,
      "epoch": 3.8785061272126047,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003558069840757619,
      "loss": 4.8664,
      "mean_token_accuracy": 0.2158392623066902,
      "num_tokens": 86456095.0,
      "step": 49850
    },
    {
      "entropy": 5.409688806533813,
      "epoch": 3.8788951565843224,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000355781116979602,
      "loss": 4.9126,
      "mean_token_accuracy": 0.2062249720096588,
      "num_tokens": 86464993.0,
      "step": 49855
    },
    {
      "entropy": 5.3507880687713625,
      "epoch": 3.8792841859560396,
      "grad_norm": 1.125,
      "learning_rate": 0.0003557552486577173,
      "loss": 4.7713,
      "mean_token_accuracy": 0.21847150921821595,
      "num_tokens": 86474269.0,
      "step": 49860
    },
    {
      "entropy": 5.514620399475097,
      "epoch": 3.8796732153277573,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003557293791105002,
      "loss": 4.9514,
      "mean_token_accuracy": 0.21215716302394866,
      "num_tokens": 86482523.0,
      "step": 49865
    },
    {
      "entropy": 5.338932514190674,
      "epoch": 3.8800622446994746,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035570350833834334,
      "loss": 4.8221,
      "mean_token_accuracy": 0.22817907929420472,
      "num_tokens": 86491091.0,
      "step": 49870
    },
    {
      "entropy": 5.4831572532653805,
      "epoch": 3.8804512740711923,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035567763634163904,
      "loss": 4.9798,
      "mean_token_accuracy": 0.20209675431251525,
      "num_tokens": 86500069.0,
      "step": 49875
    },
    {
      "entropy": 5.4740653991699215,
      "epoch": 3.88084030344291,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003556517631207802,
      "loss": 4.9376,
      "mean_token_accuracy": 0.21789976954460144,
      "num_tokens": 86508499.0,
      "step": 49880
    },
    {
      "entropy": 5.395103406906128,
      "epoch": 3.8812293328146277,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035562588867615907,
      "loss": 4.91,
      "mean_token_accuracy": 0.20759719163179396,
      "num_tokens": 86517036.0,
      "step": 49885
    },
    {
      "entropy": 5.43130087852478,
      "epoch": 3.881618362186345,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035560001300816846,
      "loss": 4.9017,
      "mean_token_accuracy": 0.21313188523054122,
      "num_tokens": 86525764.0,
      "step": 49890
    },
    {
      "entropy": 5.406771087646485,
      "epoch": 3.8820073915580626,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035557413611720095,
      "loss": 4.8244,
      "mean_token_accuracy": 0.22007715106010436,
      "num_tokens": 86534858.0,
      "step": 49895
    },
    {
      "entropy": 5.330812788009643,
      "epoch": 3.8823964209297803,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035554825800364917,
      "loss": 4.7761,
      "mean_token_accuracy": 0.22074450999498368,
      "num_tokens": 86543509.0,
      "step": 49900
    },
    {
      "entropy": 5.362614440917969,
      "epoch": 3.8827854503014976,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003555223786679058,
      "loss": 4.8269,
      "mean_token_accuracy": 0.22002876847982406,
      "num_tokens": 86552278.0,
      "step": 49905
    },
    {
      "entropy": 5.387450122833252,
      "epoch": 3.8831744796732153,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035549649811036346,
      "loss": 4.8301,
      "mean_token_accuracy": 0.21801259219646454,
      "num_tokens": 86560684.0,
      "step": 49910
    },
    {
      "entropy": 5.496552515029907,
      "epoch": 3.883563509044933,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035547061633141487,
      "loss": 4.9193,
      "mean_token_accuracy": 0.20778851360082626,
      "num_tokens": 86569015.0,
      "step": 49915
    },
    {
      "entropy": 5.353770112991333,
      "epoch": 3.8839525384166507,
      "grad_norm": 1.125,
      "learning_rate": 0.00035544473333145275,
      "loss": 4.8539,
      "mean_token_accuracy": 0.21856000125408173,
      "num_tokens": 86578544.0,
      "step": 49920
    },
    {
      "entropy": 5.311742973327637,
      "epoch": 3.884341567788368,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003554188491108699,
      "loss": 4.8162,
      "mean_token_accuracy": 0.20862941294908524,
      "num_tokens": 86587176.0,
      "step": 49925
    },
    {
      "entropy": 5.394986724853515,
      "epoch": 3.8847305971600856,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000355392963670059,
      "loss": 4.8289,
      "mean_token_accuracy": 0.21128001511096955,
      "num_tokens": 86595472.0,
      "step": 49930
    },
    {
      "entropy": 5.427108240127564,
      "epoch": 3.8851196265318033,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003553670770094129,
      "loss": 4.843,
      "mean_token_accuracy": 0.21036445796489717,
      "num_tokens": 86603958.0,
      "step": 49935
    },
    {
      "entropy": 5.466770029067993,
      "epoch": 3.8855086559035206,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035534118912932425,
      "loss": 4.9558,
      "mean_token_accuracy": 0.2056156128644943,
      "num_tokens": 86613128.0,
      "step": 49940
    },
    {
      "entropy": 5.4814760208129885,
      "epoch": 3.8858976852752383,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00035531530003018584,
      "loss": 4.8389,
      "mean_token_accuracy": 0.21914105266332626,
      "num_tokens": 86621790.0,
      "step": 49945
    },
    {
      "entropy": 5.364697074890136,
      "epoch": 3.886286714646956,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003552894097123907,
      "loss": 4.8606,
      "mean_token_accuracy": 0.21530188024044036,
      "num_tokens": 86630774.0,
      "step": 49950
    },
    {
      "entropy": 5.37090482711792,
      "epoch": 3.8866757440186737,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035526351817633157,
      "loss": 4.8602,
      "mean_token_accuracy": 0.21424614042043685,
      "num_tokens": 86639372.0,
      "step": 49955
    },
    {
      "entropy": 5.383608150482178,
      "epoch": 3.887064773390391,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003552376254224013,
      "loss": 4.8043,
      "mean_token_accuracy": 0.21705559343099595,
      "num_tokens": 86648114.0,
      "step": 49960
    },
    {
      "entropy": 5.378827476501465,
      "epoch": 3.8874538027621086,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003552117314509928,
      "loss": 4.8895,
      "mean_token_accuracy": 0.21185476183891297,
      "num_tokens": 86656823.0,
      "step": 49965
    },
    {
      "entropy": 5.310986471176148,
      "epoch": 3.887842832133826,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003551858362624989,
      "loss": 4.7835,
      "mean_token_accuracy": 0.22240580171346663,
      "num_tokens": 86664899.0,
      "step": 49970
    },
    {
      "entropy": 5.393229675292969,
      "epoch": 3.8882318615055436,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003551599398573125,
      "loss": 4.8125,
      "mean_token_accuracy": 0.2182297185063362,
      "num_tokens": 86673520.0,
      "step": 49975
    },
    {
      "entropy": 5.336479663848877,
      "epoch": 3.8886208908772613,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035513404223582666,
      "loss": 4.7969,
      "mean_token_accuracy": 0.21849294751882553,
      "num_tokens": 86682068.0,
      "step": 49980
    },
    {
      "entropy": 5.314722347259521,
      "epoch": 3.889009920248979,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035510814339843427,
      "loss": 4.8321,
      "mean_token_accuracy": 0.20771203190088272,
      "num_tokens": 86690569.0,
      "step": 49985
    },
    {
      "entropy": 5.413291501998901,
      "epoch": 3.889398949620696,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035508224334552835,
      "loss": 4.847,
      "mean_token_accuracy": 0.21426754742860793,
      "num_tokens": 86699243.0,
      "step": 49990
    },
    {
      "entropy": 5.329246997833252,
      "epoch": 3.889787978992414,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003550563420775018,
      "loss": 4.7639,
      "mean_token_accuracy": 0.21970055848360062,
      "num_tokens": 86707920.0,
      "step": 49995
    },
    {
      "entropy": 5.348993492126465,
      "epoch": 3.8901770083641316,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035503043959474775,
      "loss": 4.8013,
      "mean_token_accuracy": 0.220321723818779,
      "num_tokens": 86716793.0,
      "step": 50000
    },
    {
      "entropy": 5.385060739517212,
      "epoch": 3.890566037735849,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035500453589765906,
      "loss": 4.8675,
      "mean_token_accuracy": 0.21564192324876785,
      "num_tokens": 86725594.0,
      "step": 50005
    },
    {
      "entropy": 5.306797695159912,
      "epoch": 3.8909550671075666,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035497863098662893,
      "loss": 4.7659,
      "mean_token_accuracy": 0.22446653246879578,
      "num_tokens": 86734637.0,
      "step": 50010
    },
    {
      "entropy": 5.373083543777466,
      "epoch": 3.8913440964792843,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035495272486205033,
      "loss": 4.8023,
      "mean_token_accuracy": 0.2130637064576149,
      "num_tokens": 86743159.0,
      "step": 50015
    },
    {
      "entropy": 5.365118217468262,
      "epoch": 3.891733125851002,
      "grad_norm": 1.125,
      "learning_rate": 0.0003549268175243165,
      "loss": 4.7994,
      "mean_token_accuracy": 0.22018594890832902,
      "num_tokens": 86751345.0,
      "step": 50020
    },
    {
      "entropy": 5.384858179092407,
      "epoch": 3.892122155222719,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003549009089738204,
      "loss": 4.9259,
      "mean_token_accuracy": 0.21618716716766356,
      "num_tokens": 86760692.0,
      "step": 50025
    },
    {
      "entropy": 5.413939952850342,
      "epoch": 3.892511184594437,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003548749992109551,
      "loss": 4.8418,
      "mean_token_accuracy": 0.21532336473464966,
      "num_tokens": 86769560.0,
      "step": 50030
    },
    {
      "entropy": 5.478565311431884,
      "epoch": 3.8929002139661546,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003548490882361139,
      "loss": 4.9198,
      "mean_token_accuracy": 0.2093125730752945,
      "num_tokens": 86778223.0,
      "step": 50035
    },
    {
      "entropy": 5.33926362991333,
      "epoch": 3.893289243337872,
      "grad_norm": 1.125,
      "learning_rate": 0.0003548231760496898,
      "loss": 4.7762,
      "mean_token_accuracy": 0.22014688402414323,
      "num_tokens": 86786976.0,
      "step": 50040
    },
    {
      "entropy": 5.398734426498413,
      "epoch": 3.8936782727095895,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035479726265207604,
      "loss": 4.9146,
      "mean_token_accuracy": 0.21150938272476197,
      "num_tokens": 86795528.0,
      "step": 50045
    },
    {
      "entropy": 5.436475324630737,
      "epoch": 3.8940673020813072,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003547713480436659,
      "loss": 4.908,
      "mean_token_accuracy": 0.20612601488828658,
      "num_tokens": 86804160.0,
      "step": 50050
    },
    {
      "entropy": 5.452000761032105,
      "epoch": 3.894456331453025,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035474543222485254,
      "loss": 4.8701,
      "mean_token_accuracy": 0.21208083033561706,
      "num_tokens": 86812224.0,
      "step": 50055
    },
    {
      "entropy": 5.337144184112549,
      "epoch": 3.894845360824742,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035471951519602926,
      "loss": 4.7987,
      "mean_token_accuracy": 0.21609776467084885,
      "num_tokens": 86821649.0,
      "step": 50060
    },
    {
      "entropy": 5.410448455810547,
      "epoch": 3.89523439019646,
      "grad_norm": 1.125,
      "learning_rate": 0.00035469359695758905,
      "loss": 4.8561,
      "mean_token_accuracy": 0.2155676931142807,
      "num_tokens": 86830642.0,
      "step": 50065
    },
    {
      "entropy": 5.353119802474976,
      "epoch": 3.895623419568177,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003546676775099254,
      "loss": 4.7695,
      "mean_token_accuracy": 0.2249339923262596,
      "num_tokens": 86839306.0,
      "step": 50070
    },
    {
      "entropy": 5.316918659210205,
      "epoch": 3.896012448939895,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003546417568534315,
      "loss": 4.7637,
      "mean_token_accuracy": 0.21667125076055527,
      "num_tokens": 86847389.0,
      "step": 50075
    },
    {
      "entropy": 5.27915678024292,
      "epoch": 3.8964014783116125,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035461583498850083,
      "loss": 4.7949,
      "mean_token_accuracy": 0.21421377658843993,
      "num_tokens": 86857160.0,
      "step": 50080
    },
    {
      "entropy": 5.491142702102661,
      "epoch": 3.8967905076833302,
      "grad_norm": 1.25,
      "learning_rate": 0.0003545899119155265,
      "loss": 4.9521,
      "mean_token_accuracy": 0.2104384794831276,
      "num_tokens": 86865236.0,
      "step": 50085
    },
    {
      "entropy": 5.4923743724823,
      "epoch": 3.8971795370550475,
      "grad_norm": 1.125,
      "learning_rate": 0.00035456398763490193,
      "loss": 4.9075,
      "mean_token_accuracy": 0.20678262263536454,
      "num_tokens": 86873623.0,
      "step": 50090
    },
    {
      "entropy": 5.32187557220459,
      "epoch": 3.897568566426765,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003545380621470205,
      "loss": 4.7763,
      "mean_token_accuracy": 0.22341914623975753,
      "num_tokens": 86883075.0,
      "step": 50095
    },
    {
      "entropy": 5.388635873794556,
      "epoch": 3.897957595798483,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003545121354522756,
      "loss": 4.9122,
      "mean_token_accuracy": 0.20871062725782394,
      "num_tokens": 86892247.0,
      "step": 50100
    },
    {
      "entropy": 5.407228803634643,
      "epoch": 3.8983466251702,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035448620755106064,
      "loss": 4.8563,
      "mean_token_accuracy": 0.21403463035821915,
      "num_tokens": 86901795.0,
      "step": 50105
    },
    {
      "entropy": 5.368407440185547,
      "epoch": 3.898735654541918,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003544602784437689,
      "loss": 4.8201,
      "mean_token_accuracy": 0.21264290809631348,
      "num_tokens": 86910720.0,
      "step": 50110
    },
    {
      "entropy": 5.3248663425445555,
      "epoch": 3.8991246839136355,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035443434813079393,
      "loss": 4.8307,
      "mean_token_accuracy": 0.21907418966293335,
      "num_tokens": 86919350.0,
      "step": 50115
    },
    {
      "entropy": 5.405660581588745,
      "epoch": 3.8995137132853532,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003544084166125292,
      "loss": 4.8782,
      "mean_token_accuracy": 0.21362458765506745,
      "num_tokens": 86927612.0,
      "step": 50120
    },
    {
      "entropy": 5.410800552368164,
      "epoch": 3.8999027426570705,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003543824838893682,
      "loss": 4.8382,
      "mean_token_accuracy": 0.21727649122476578,
      "num_tokens": 86936416.0,
      "step": 50125
    },
    {
      "entropy": 5.385420131683349,
      "epoch": 3.900291772028788,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035435654996170417,
      "loss": 4.9067,
      "mean_token_accuracy": 0.20990137457847596,
      "num_tokens": 86945764.0,
      "step": 50130
    },
    {
      "entropy": 5.386343193054199,
      "epoch": 3.900680801400506,
      "grad_norm": 1.125,
      "learning_rate": 0.000354330614829931,
      "loss": 4.8428,
      "mean_token_accuracy": 0.21837088018655776,
      "num_tokens": 86953952.0,
      "step": 50135
    },
    {
      "entropy": 5.451874160766602,
      "epoch": 3.901069830772223,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003543046784944419,
      "loss": 4.8234,
      "mean_token_accuracy": 0.21493812501430512,
      "num_tokens": 86962295.0,
      "step": 50140
    },
    {
      "entropy": 5.499092388153076,
      "epoch": 3.901458860143941,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00035427874095563054,
      "loss": 4.9488,
      "mean_token_accuracy": 0.21205949783325195,
      "num_tokens": 86970361.0,
      "step": 50145
    },
    {
      "entropy": 5.358064937591553,
      "epoch": 3.9018478895156585,
      "grad_norm": 1.125,
      "learning_rate": 0.0003542528022138905,
      "loss": 4.7548,
      "mean_token_accuracy": 0.21842140108346939,
      "num_tokens": 86979131.0,
      "step": 50150
    },
    {
      "entropy": 5.293951845169067,
      "epoch": 3.9022369188873762,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003542268622696153,
      "loss": 4.8004,
      "mean_token_accuracy": 0.21640967279672624,
      "num_tokens": 86988088.0,
      "step": 50155
    },
    {
      "entropy": 5.490346097946167,
      "epoch": 3.9026259482590935,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003542009211231986,
      "loss": 4.9426,
      "mean_token_accuracy": 0.20485568046569824,
      "num_tokens": 86996792.0,
      "step": 50160
    },
    {
      "entropy": 5.436500406265258,
      "epoch": 3.903014977630811,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035417497877503396,
      "loss": 4.8969,
      "mean_token_accuracy": 0.21138523072004317,
      "num_tokens": 87005547.0,
      "step": 50165
    },
    {
      "entropy": 5.4955888271331785,
      "epoch": 3.9034040070025284,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035414903522551505,
      "loss": 5.0202,
      "mean_token_accuracy": 0.2018817499279976,
      "num_tokens": 87014923.0,
      "step": 50170
    },
    {
      "entropy": 5.402669048309326,
      "epoch": 3.903793036374246,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035412309047503553,
      "loss": 4.8752,
      "mean_token_accuracy": 0.21724965870380403,
      "num_tokens": 87023842.0,
      "step": 50175
    },
    {
      "entropy": 5.467564487457276,
      "epoch": 3.904182065745964,
      "grad_norm": 1.15625,
      "learning_rate": 0.000354097144523989,
      "loss": 4.8701,
      "mean_token_accuracy": 0.21566506326198578,
      "num_tokens": 87032251.0,
      "step": 50180
    },
    {
      "entropy": 5.4003565311431885,
      "epoch": 3.9045710951176815,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003540711973727693,
      "loss": 4.9059,
      "mean_token_accuracy": 0.20854803621768953,
      "num_tokens": 87041052.0,
      "step": 50185
    },
    {
      "entropy": 5.409228229522705,
      "epoch": 3.904960124489399,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003540452490217699,
      "loss": 4.8707,
      "mean_token_accuracy": 0.2156693994998932,
      "num_tokens": 87049908.0,
      "step": 50190
    },
    {
      "entropy": 5.407024812698364,
      "epoch": 3.9053491538611165,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003540192994713847,
      "loss": 4.8452,
      "mean_token_accuracy": 0.2093372568488121,
      "num_tokens": 87059458.0,
      "step": 50195
    },
    {
      "entropy": 5.462924337387085,
      "epoch": 3.905738183232834,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035399334872200747,
      "loss": 4.9253,
      "mean_token_accuracy": 0.20550031810998917,
      "num_tokens": 87067856.0,
      "step": 50200
    },
    {
      "entropy": 5.474410343170166,
      "epoch": 3.9061272126045514,
      "grad_norm": 1.25,
      "learning_rate": 0.0003539673967740318,
      "loss": 4.9476,
      "mean_token_accuracy": 0.21095436960458755,
      "num_tokens": 87076427.0,
      "step": 50205
    },
    {
      "entropy": 5.395767641067505,
      "epoch": 3.906516241976269,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003539414436278516,
      "loss": 4.8465,
      "mean_token_accuracy": 0.211288383603096,
      "num_tokens": 87084576.0,
      "step": 50210
    },
    {
      "entropy": 5.425954723358155,
      "epoch": 3.906905271347987,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035391548928386077,
      "loss": 4.9243,
      "mean_token_accuracy": 0.21403887569904329,
      "num_tokens": 87092788.0,
      "step": 50215
    },
    {
      "entropy": 5.3314220905303955,
      "epoch": 3.9072943007197045,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003538895337424529,
      "loss": 4.7132,
      "mean_token_accuracy": 0.2256551504135132,
      "num_tokens": 87101780.0,
      "step": 50220
    },
    {
      "entropy": 5.4562421321868895,
      "epoch": 3.9076833300914218,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035386357700402187,
      "loss": 4.9371,
      "mean_token_accuracy": 0.2142144039273262,
      "num_tokens": 87111234.0,
      "step": 50225
    },
    {
      "entropy": 5.338387966156006,
      "epoch": 3.9080723594631395,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003538376190689616,
      "loss": 4.7551,
      "mean_token_accuracy": 0.2238451972603798,
      "num_tokens": 87119488.0,
      "step": 50230
    },
    {
      "entropy": 5.433273983001709,
      "epoch": 3.908461388834857,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003538116599376661,
      "loss": 4.976,
      "mean_token_accuracy": 0.20233719497919084,
      "num_tokens": 87129479.0,
      "step": 50235
    },
    {
      "entropy": 5.407299757003784,
      "epoch": 3.9088504182065744,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035378569961052887,
      "loss": 4.7957,
      "mean_token_accuracy": 0.215228670835495,
      "num_tokens": 87137711.0,
      "step": 50240
    },
    {
      "entropy": 5.389112377166748,
      "epoch": 3.909239447578292,
      "grad_norm": 1.125,
      "learning_rate": 0.00035375973808794416,
      "loss": 4.8198,
      "mean_token_accuracy": 0.21694521903991698,
      "num_tokens": 87145858.0,
      "step": 50245
    },
    {
      "entropy": 5.316772031784057,
      "epoch": 3.90962847695001,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003537337753703058,
      "loss": 4.7949,
      "mean_token_accuracy": 0.22099514603614806,
      "num_tokens": 87154200.0,
      "step": 50250
    },
    {
      "entropy": 5.416736078262329,
      "epoch": 3.9100175063217275,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003537078114580077,
      "loss": 4.922,
      "mean_token_accuracy": 0.20738909542560577,
      "num_tokens": 87163329.0,
      "step": 50255
    },
    {
      "entropy": 5.468604612350464,
      "epoch": 3.9104065356934448,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003536818463514438,
      "loss": 4.8745,
      "mean_token_accuracy": 0.21667521446943283,
      "num_tokens": 87172459.0,
      "step": 50260
    },
    {
      "entropy": 5.432321214675904,
      "epoch": 3.9107955650651625,
      "grad_norm": 1.125,
      "learning_rate": 0.00035365588005100817,
      "loss": 4.8727,
      "mean_token_accuracy": 0.21274207830429076,
      "num_tokens": 87181155.0,
      "step": 50265
    },
    {
      "entropy": 5.360682439804077,
      "epoch": 3.9111845944368797,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003536299125570947,
      "loss": 4.8461,
      "mean_token_accuracy": 0.2174878776073456,
      "num_tokens": 87190191.0,
      "step": 50270
    },
    {
      "entropy": 5.444006156921387,
      "epoch": 3.9115736238085974,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003536039438700975,
      "loss": 4.889,
      "mean_token_accuracy": 0.214572736620903,
      "num_tokens": 87199023.0,
      "step": 50275
    },
    {
      "entropy": 5.4204004287719725,
      "epoch": 3.911962653180315,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035357797399041056,
      "loss": 4.9364,
      "mean_token_accuracy": 0.20699241161346435,
      "num_tokens": 87207694.0,
      "step": 50280
    },
    {
      "entropy": 5.394293880462646,
      "epoch": 3.912351682552033,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003535520029184279,
      "loss": 4.8517,
      "mean_token_accuracy": 0.21236144751310349,
      "num_tokens": 87216019.0,
      "step": 50285
    },
    {
      "entropy": 5.4463968753814695,
      "epoch": 3.9127407119237505,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003535260306545437,
      "loss": 4.8462,
      "mean_token_accuracy": 0.212478905916214,
      "num_tokens": 87224737.0,
      "step": 50290
    },
    {
      "entropy": 5.455323696136475,
      "epoch": 3.9131297412954678,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035350005719915196,
      "loss": 4.8895,
      "mean_token_accuracy": 0.2139474242925644,
      "num_tokens": 87233797.0,
      "step": 50295
    },
    {
      "entropy": 5.359736061096191,
      "epoch": 3.9135187706671855,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035347408255264674,
      "loss": 4.7508,
      "mean_token_accuracy": 0.2244486153125763,
      "num_tokens": 87242767.0,
      "step": 50300
    },
    {
      "entropy": 5.3581804752349855,
      "epoch": 3.9139078000389027,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003534481067154223,
      "loss": 4.8102,
      "mean_token_accuracy": 0.2247103199362755,
      "num_tokens": 87251684.0,
      "step": 50305
    },
    {
      "entropy": 5.437606859207153,
      "epoch": 3.9142968294106204,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003534221296878726,
      "loss": 4.9182,
      "mean_token_accuracy": 0.2135276824235916,
      "num_tokens": 87260730.0,
      "step": 50310
    },
    {
      "entropy": 5.41645245552063,
      "epoch": 3.914685858782338,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035339615147039196,
      "loss": 4.8792,
      "mean_token_accuracy": 0.2062883198261261,
      "num_tokens": 87270241.0,
      "step": 50315
    },
    {
      "entropy": 5.366668319702148,
      "epoch": 3.915074888154056,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035337017206337444,
      "loss": 4.8061,
      "mean_token_accuracy": 0.21917391270399095,
      "num_tokens": 87279463.0,
      "step": 50320
    },
    {
      "entropy": 5.418638610839844,
      "epoch": 3.915463917525773,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003533441914672144,
      "loss": 4.8483,
      "mean_token_accuracy": 0.20943626910448074,
      "num_tokens": 87287887.0,
      "step": 50325
    },
    {
      "entropy": 5.458248090744019,
      "epoch": 3.9158529468974907,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035331820968230595,
      "loss": 4.8679,
      "mean_token_accuracy": 0.2148660510778427,
      "num_tokens": 87296832.0,
      "step": 50330
    },
    {
      "entropy": 5.506937456130982,
      "epoch": 3.9162419762692084,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035329222670904325,
      "loss": 4.9774,
      "mean_token_accuracy": 0.20019400417804717,
      "num_tokens": 87304867.0,
      "step": 50335
    },
    {
      "entropy": 5.415652465820313,
      "epoch": 3.9166310056409257,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003532662425478206,
      "loss": 4.9068,
      "mean_token_accuracy": 0.21732494086027146,
      "num_tokens": 87313541.0,
      "step": 50340
    },
    {
      "entropy": 5.396382999420166,
      "epoch": 3.9170200350126434,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035324025719903236,
      "loss": 4.9029,
      "mean_token_accuracy": 0.21252136379480363,
      "num_tokens": 87322462.0,
      "step": 50345
    },
    {
      "entropy": 5.385149621963501,
      "epoch": 3.917409064384361,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035321427066307276,
      "loss": 4.8617,
      "mean_token_accuracy": 0.21457833051681519,
      "num_tokens": 87331519.0,
      "step": 50350
    },
    {
      "entropy": 5.468873500823975,
      "epoch": 3.917798093756079,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003531882829403361,
      "loss": 4.9295,
      "mean_token_accuracy": 0.20591126680374144,
      "num_tokens": 87340380.0,
      "step": 50355
    },
    {
      "entropy": 5.426613664627075,
      "epoch": 3.918187123127796,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035316229403121666,
      "loss": 4.8379,
      "mean_token_accuracy": 0.2114797681570053,
      "num_tokens": 87348022.0,
      "step": 50360
    },
    {
      "entropy": 5.329485321044922,
      "epoch": 3.9185761524995137,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003531363039361088,
      "loss": 4.7713,
      "mean_token_accuracy": 0.21548608988523482,
      "num_tokens": 87357080.0,
      "step": 50365
    },
    {
      "entropy": 5.359896516799926,
      "epoch": 3.9189651818712314,
      "grad_norm": 1.125,
      "learning_rate": 0.00035311031265540695,
      "loss": 4.8321,
      "mean_token_accuracy": 0.21243430823087692,
      "num_tokens": 87365576.0,
      "step": 50370
    },
    {
      "entropy": 5.424797439575196,
      "epoch": 3.9193542112429487,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035308432018950537,
      "loss": 4.9418,
      "mean_token_accuracy": 0.21028377562761308,
      "num_tokens": 87375149.0,
      "step": 50375
    },
    {
      "entropy": 5.405410671234131,
      "epoch": 3.9197432406146664,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035305832653879856,
      "loss": 4.8199,
      "mean_token_accuracy": 0.22140689492225646,
      "num_tokens": 87384299.0,
      "step": 50380
    },
    {
      "entropy": 5.367635917663574,
      "epoch": 3.920132269986384,
      "grad_norm": 1.125,
      "learning_rate": 0.00035303233170368083,
      "loss": 4.8212,
      "mean_token_accuracy": 0.21972196847200393,
      "num_tokens": 87392836.0,
      "step": 50385
    },
    {
      "entropy": 5.4342375755310055,
      "epoch": 3.920521299358102,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035300633568454673,
      "loss": 4.8996,
      "mean_token_accuracy": 0.2113592192530632,
      "num_tokens": 87401798.0,
      "step": 50390
    },
    {
      "entropy": 5.378832769393921,
      "epoch": 3.920910328729819,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035298033848179064,
      "loss": 4.7707,
      "mean_token_accuracy": 0.22155203223228453,
      "num_tokens": 87410487.0,
      "step": 50395
    },
    {
      "entropy": 5.472595691680908,
      "epoch": 3.9212993581015367,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000352954340095807,
      "loss": 5.0425,
      "mean_token_accuracy": 0.20302307456731797,
      "num_tokens": 87419664.0,
      "step": 50400
    },
    {
      "entropy": 5.471971940994263,
      "epoch": 3.921688387473254,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035292834052699036,
      "loss": 4.8879,
      "mean_token_accuracy": 0.21246059536933898,
      "num_tokens": 87428073.0,
      "step": 50405
    },
    {
      "entropy": 5.4853757381439205,
      "epoch": 3.9220774168449717,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003529023397757351,
      "loss": 4.9033,
      "mean_token_accuracy": 0.2101770117878914,
      "num_tokens": 87436250.0,
      "step": 50410
    },
    {
      "entropy": 5.358656597137451,
      "epoch": 3.9224664462166894,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035287633784243595,
      "loss": 4.769,
      "mean_token_accuracy": 0.21716036945581435,
      "num_tokens": 87445131.0,
      "step": 50415
    },
    {
      "entropy": 5.336965322494507,
      "epoch": 3.922855475588407,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035285033472748725,
      "loss": 4.8593,
      "mean_token_accuracy": 0.2167794182896614,
      "num_tokens": 87454428.0,
      "step": 50420
    },
    {
      "entropy": 5.4946959018707275,
      "epoch": 3.9232445049601243,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003528243304312837,
      "loss": 4.9406,
      "mean_token_accuracy": 0.20968398749828338,
      "num_tokens": 87462956.0,
      "step": 50425
    },
    {
      "entropy": 5.434648704528809,
      "epoch": 3.923633534331842,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003527983249542197,
      "loss": 4.8979,
      "mean_token_accuracy": 0.21172115355730056,
      "num_tokens": 87471437.0,
      "step": 50430
    },
    {
      "entropy": 5.3298088073730465,
      "epoch": 3.9240225637035597,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035277231829669006,
      "loss": 4.752,
      "mean_token_accuracy": 0.21640150249004364,
      "num_tokens": 87480327.0,
      "step": 50435
    },
    {
      "entropy": 5.359810543060303,
      "epoch": 3.924411593075277,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003527463104590892,
      "loss": 4.7786,
      "mean_token_accuracy": 0.22254247069358826,
      "num_tokens": 87489365.0,
      "step": 50440
    },
    {
      "entropy": 5.41688904762268,
      "epoch": 3.9248006224469947,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003527203014418119,
      "loss": 4.9171,
      "mean_token_accuracy": 0.21648739278316498,
      "num_tokens": 87499267.0,
      "step": 50445
    },
    {
      "entropy": 5.386222982406617,
      "epoch": 3.9251896518187124,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003526942912452526,
      "loss": 4.832,
      "mean_token_accuracy": 0.20916824489831926,
      "num_tokens": 87508713.0,
      "step": 50450
    },
    {
      "entropy": 5.417311000823974,
      "epoch": 3.92557868119043,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003526682798698062,
      "loss": 4.9265,
      "mean_token_accuracy": 0.20548066049814223,
      "num_tokens": 87517783.0,
      "step": 50455
    },
    {
      "entropy": 5.401719284057617,
      "epoch": 3.9259677105621473,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003526422673158673,
      "loss": 4.8462,
      "mean_token_accuracy": 0.21406918615102768,
      "num_tokens": 87525592.0,
      "step": 50460
    },
    {
      "entropy": 5.281982851028443,
      "epoch": 3.926356739933865,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003526162535838305,
      "loss": 4.8031,
      "mean_token_accuracy": 0.21481645703315735,
      "num_tokens": 87534540.0,
      "step": 50465
    },
    {
      "entropy": 5.431277942657471,
      "epoch": 3.9267457693055827,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035259023867409065,
      "loss": 4.9237,
      "mean_token_accuracy": 0.2117403730750084,
      "num_tokens": 87543540.0,
      "step": 50470
    },
    {
      "entropy": 5.495229721069336,
      "epoch": 3.9271347986773,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003525642225870423,
      "loss": 4.9142,
      "mean_token_accuracy": 0.21537659019231797,
      "num_tokens": 87552913.0,
      "step": 50475
    },
    {
      "entropy": 5.464085388183594,
      "epoch": 3.9275238280490177,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035253820532308043,
      "loss": 4.9236,
      "mean_token_accuracy": 0.21554374992847442,
      "num_tokens": 87561548.0,
      "step": 50480
    },
    {
      "entropy": 5.391648674011231,
      "epoch": 3.9279128574207354,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003525121868825997,
      "loss": 4.9049,
      "mean_token_accuracy": 0.211708103120327,
      "num_tokens": 87569877.0,
      "step": 50485
    },
    {
      "entropy": 5.412351989746094,
      "epoch": 3.928301886792453,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003524861672659949,
      "loss": 4.8396,
      "mean_token_accuracy": 0.2201072543859482,
      "num_tokens": 87578125.0,
      "step": 50490
    },
    {
      "entropy": 5.433688402175903,
      "epoch": 3.9286909161641703,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003524601464736609,
      "loss": 4.962,
      "mean_token_accuracy": 0.20101673007011414,
      "num_tokens": 87586637.0,
      "step": 50495
    },
    {
      "entropy": 5.480445623397827,
      "epoch": 3.929079945535888,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003524341245059924,
      "loss": 4.9545,
      "mean_token_accuracy": 0.20769410729408264,
      "num_tokens": 87595674.0,
      "step": 50500
    },
    {
      "entropy": 5.49881386756897,
      "epoch": 3.9294689749076053,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003524081013633843,
      "loss": 5.0103,
      "mean_token_accuracy": 0.20782867521047593,
      "num_tokens": 87605404.0,
      "step": 50505
    },
    {
      "entropy": 5.388889980316162,
      "epoch": 3.929858004279323,
      "grad_norm": 1.125,
      "learning_rate": 0.00035238207704623147,
      "loss": 4.7843,
      "mean_token_accuracy": 0.2169026866555214,
      "num_tokens": 87613511.0,
      "step": 50510
    },
    {
      "entropy": 5.363560676574707,
      "epoch": 3.9302470336510407,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035235605155492885,
      "loss": 4.8612,
      "mean_token_accuracy": 0.21667470484972,
      "num_tokens": 87622554.0,
      "step": 50515
    },
    {
      "entropy": 5.288969326019287,
      "epoch": 3.9306360630227584,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035233002488987123,
      "loss": 4.6937,
      "mean_token_accuracy": 0.22188831865787506,
      "num_tokens": 87630628.0,
      "step": 50520
    },
    {
      "entropy": 5.458449840545654,
      "epoch": 3.9310250923944756,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035230399705145365,
      "loss": 4.9094,
      "mean_token_accuracy": 0.20983887314796448,
      "num_tokens": 87639292.0,
      "step": 50525
    },
    {
      "entropy": 5.479224681854248,
      "epoch": 3.9314141217661933,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003522779680400709,
      "loss": 4.9229,
      "mean_token_accuracy": 0.20914674997329713,
      "num_tokens": 87646949.0,
      "step": 50530
    },
    {
      "entropy": 5.426835536956787,
      "epoch": 3.931803151137911,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035225193785611795,
      "loss": 4.8898,
      "mean_token_accuracy": 0.21294473856687546,
      "num_tokens": 87656247.0,
      "step": 50535
    },
    {
      "entropy": 5.392188215255738,
      "epoch": 3.9321921805096283,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003522259064999898,
      "loss": 4.9591,
      "mean_token_accuracy": 0.20332612097263336,
      "num_tokens": 87665297.0,
      "step": 50540
    },
    {
      "entropy": 5.387047481536865,
      "epoch": 3.932581209881346,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035219987397208137,
      "loss": 4.8664,
      "mean_token_accuracy": 0.21492185741662978,
      "num_tokens": 87674055.0,
      "step": 50545
    },
    {
      "entropy": 5.388765716552735,
      "epoch": 3.9329702392530637,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003521738402727878,
      "loss": 4.8441,
      "mean_token_accuracy": 0.21532511264085769,
      "num_tokens": 87682717.0,
      "step": 50550
    },
    {
      "entropy": 5.43828935623169,
      "epoch": 3.9333592686247814,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003521478054025041,
      "loss": 4.9688,
      "mean_token_accuracy": 0.20688313692808152,
      "num_tokens": 87691683.0,
      "step": 50555
    },
    {
      "entropy": 5.41693000793457,
      "epoch": 3.9337482979964986,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035212176936162524,
      "loss": 4.8088,
      "mean_token_accuracy": 0.2135939747095108,
      "num_tokens": 87699953.0,
      "step": 50560
    },
    {
      "entropy": 5.376069259643555,
      "epoch": 3.9341373273682163,
      "grad_norm": 1.125,
      "learning_rate": 0.0003520957321505462,
      "loss": 4.8584,
      "mean_token_accuracy": 0.21170159578323364,
      "num_tokens": 87708278.0,
      "step": 50565
    },
    {
      "entropy": 5.418861627578735,
      "epoch": 3.934526356739934,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035206969376966226,
      "loss": 4.8932,
      "mean_token_accuracy": 0.2074375182390213,
      "num_tokens": 87717737.0,
      "step": 50570
    },
    {
      "entropy": 5.4540667057037355,
      "epoch": 3.9349153861116513,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035204365421936826,
      "loss": 4.8771,
      "mean_token_accuracy": 0.21065657138824462,
      "num_tokens": 87726907.0,
      "step": 50575
    },
    {
      "entropy": 5.400401306152344,
      "epoch": 3.935304415483369,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003520176135000594,
      "loss": 4.852,
      "mean_token_accuracy": 0.21460242122411727,
      "num_tokens": 87735433.0,
      "step": 50580
    },
    {
      "entropy": 5.381544065475464,
      "epoch": 3.9356934448550867,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003519915716121309,
      "loss": 4.7991,
      "mean_token_accuracy": 0.21764929592609406,
      "num_tokens": 87744021.0,
      "step": 50585
    },
    {
      "entropy": 5.366984844207764,
      "epoch": 3.9360824742268044,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003519655285559779,
      "loss": 4.8807,
      "mean_token_accuracy": 0.20591518878936768,
      "num_tokens": 87752525.0,
      "step": 50590
    },
    {
      "entropy": 5.460814523696899,
      "epoch": 3.9364715035985216,
      "grad_norm": 1.125,
      "learning_rate": 0.00035193948433199547,
      "loss": 4.8732,
      "mean_token_accuracy": 0.2186979666352272,
      "num_tokens": 87760806.0,
      "step": 50595
    },
    {
      "entropy": 5.412738037109375,
      "epoch": 3.9368605329702393,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035191343894057885,
      "loss": 4.853,
      "mean_token_accuracy": 0.215633824467659,
      "num_tokens": 87769091.0,
      "step": 50600
    },
    {
      "entropy": 5.41365761756897,
      "epoch": 3.9372495623419566,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035188739238212316,
      "loss": 4.9637,
      "mean_token_accuracy": 0.2000601261854172,
      "num_tokens": 87778096.0,
      "step": 50605
    },
    {
      "entropy": 5.37999153137207,
      "epoch": 3.9376385917136743,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003518613446570237,
      "loss": 4.8672,
      "mean_token_accuracy": 0.21605005115270615,
      "num_tokens": 87787439.0,
      "step": 50610
    },
    {
      "entropy": 5.5152545928955075,
      "epoch": 3.938027621085392,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035183529576567567,
      "loss": 4.8825,
      "mean_token_accuracy": 0.21816454827785492,
      "num_tokens": 87796592.0,
      "step": 50615
    },
    {
      "entropy": 5.312342834472656,
      "epoch": 3.9384166504571096,
      "grad_norm": 1.125,
      "learning_rate": 0.0003518092457084743,
      "loss": 4.7643,
      "mean_token_accuracy": 0.2259387880563736,
      "num_tokens": 87805283.0,
      "step": 50620
    },
    {
      "entropy": 5.359522724151612,
      "epoch": 3.9388056798288273,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003517831944858149,
      "loss": 4.9049,
      "mean_token_accuracy": 0.21658210456371307,
      "num_tokens": 87813474.0,
      "step": 50625
    },
    {
      "entropy": 5.4475172996521,
      "epoch": 3.9391947092005446,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035175714209809275,
      "loss": 4.9483,
      "mean_token_accuracy": 0.21095891743898393,
      "num_tokens": 87822826.0,
      "step": 50630
    },
    {
      "entropy": 5.508661222457886,
      "epoch": 3.9395837385722623,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003517310885457032,
      "loss": 4.8517,
      "mean_token_accuracy": 0.21135308891534804,
      "num_tokens": 87831582.0,
      "step": 50635
    },
    {
      "entropy": 5.364058017730713,
      "epoch": 3.9399727679439795,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003517050338290414,
      "loss": 4.7872,
      "mean_token_accuracy": 0.21874913275241853,
      "num_tokens": 87839527.0,
      "step": 50640
    },
    {
      "entropy": 5.37795729637146,
      "epoch": 3.9403617973156972,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035167897794850285,
      "loss": 4.7791,
      "mean_token_accuracy": 0.22116046845912934,
      "num_tokens": 87847973.0,
      "step": 50645
    },
    {
      "entropy": 5.395332527160645,
      "epoch": 3.940750826687415,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035165292090448287,
      "loss": 4.9138,
      "mean_token_accuracy": 0.20966066420078278,
      "num_tokens": 87857802.0,
      "step": 50650
    },
    {
      "entropy": 5.307913208007813,
      "epoch": 3.9411398560591326,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035162686269737676,
      "loss": 4.7448,
      "mean_token_accuracy": 0.22276515662670135,
      "num_tokens": 87865771.0,
      "step": 50655
    },
    {
      "entropy": 5.328939485549927,
      "epoch": 3.94152888543085,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035160080332758006,
      "loss": 4.7614,
      "mean_token_accuracy": 0.22130703777074814,
      "num_tokens": 87874434.0,
      "step": 50660
    },
    {
      "entropy": 5.3313539028167725,
      "epoch": 3.9419179148025676,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000351574742795488,
      "loss": 4.7151,
      "mean_token_accuracy": 0.22628068178892136,
      "num_tokens": 87882509.0,
      "step": 50665
    },
    {
      "entropy": 5.398054361343384,
      "epoch": 3.9423069441742853,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003515486811014962,
      "loss": 4.9256,
      "mean_token_accuracy": 0.20837249457836152,
      "num_tokens": 87891440.0,
      "step": 50670
    },
    {
      "entropy": 5.322513437271118,
      "epoch": 3.9426959735460025,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003515226182459999,
      "loss": 4.7111,
      "mean_token_accuracy": 0.22443006932735443,
      "num_tokens": 87900275.0,
      "step": 50675
    },
    {
      "entropy": 5.410129642486572,
      "epoch": 3.9430850029177202,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035149655422939477,
      "loss": 4.8634,
      "mean_token_accuracy": 0.21086281090974807,
      "num_tokens": 87908473.0,
      "step": 50680
    },
    {
      "entropy": 5.292223358154297,
      "epoch": 3.943474032289438,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003514704890520761,
      "loss": 4.7199,
      "mean_token_accuracy": 0.23270262479782106,
      "num_tokens": 87917927.0,
      "step": 50685
    },
    {
      "entropy": 5.353907871246338,
      "epoch": 3.9438630616611556,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003514444227144395,
      "loss": 4.8536,
      "mean_token_accuracy": 0.21157536804676055,
      "num_tokens": 87926982.0,
      "step": 50690
    },
    {
      "entropy": 5.439753103256225,
      "epoch": 3.944252091032873,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003514183552168804,
      "loss": 4.8373,
      "mean_token_accuracy": 0.21096294820308686,
      "num_tokens": 87935539.0,
      "step": 50695
    },
    {
      "entropy": 5.466404628753662,
      "epoch": 3.9446411204045906,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035139228655979457,
      "loss": 4.8434,
      "mean_token_accuracy": 0.21368503719568252,
      "num_tokens": 87943285.0,
      "step": 50700
    },
    {
      "entropy": 5.406633901596069,
      "epoch": 3.945030149776308,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035136621674357717,
      "loss": 4.9069,
      "mean_token_accuracy": 0.2108253076672554,
      "num_tokens": 87952634.0,
      "step": 50705
    },
    {
      "entropy": 5.393271207809448,
      "epoch": 3.9454191791480255,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003513401457686242,
      "loss": 4.807,
      "mean_token_accuracy": 0.21426254361867905,
      "num_tokens": 87960453.0,
      "step": 50710
    },
    {
      "entropy": 5.444956350326538,
      "epoch": 3.9458082085197432,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035131407363533083,
      "loss": 4.8884,
      "mean_token_accuracy": 0.2068654254078865,
      "num_tokens": 87969696.0,
      "step": 50715
    },
    {
      "entropy": 5.389322710037232,
      "epoch": 3.946197237891461,
      "grad_norm": 1.125,
      "learning_rate": 0.0003512880003440929,
      "loss": 4.8475,
      "mean_token_accuracy": 0.21339569687843324,
      "num_tokens": 87978747.0,
      "step": 50720
    },
    {
      "entropy": 5.373222541809082,
      "epoch": 3.9465862672631786,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003512619258953061,
      "loss": 4.855,
      "mean_token_accuracy": 0.20999928712844848,
      "num_tokens": 87987043.0,
      "step": 50725
    },
    {
      "entropy": 5.343416309356689,
      "epoch": 3.946975296634896,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003512358502893658,
      "loss": 4.8547,
      "mean_token_accuracy": 0.21480091214179992,
      "num_tokens": 87996390.0,
      "step": 50730
    },
    {
      "entropy": 5.460316705703735,
      "epoch": 3.9473643260066136,
      "grad_norm": 1.125,
      "learning_rate": 0.000351209773526668,
      "loss": 4.947,
      "mean_token_accuracy": 0.20860592573881148,
      "num_tokens": 88004341.0,
      "step": 50735
    },
    {
      "entropy": 5.36925482749939,
      "epoch": 3.947753355378331,
      "grad_norm": 1.171875,
      "learning_rate": 0.000351183695607608,
      "loss": 4.8018,
      "mean_token_accuracy": 0.21551977694034577,
      "num_tokens": 88013087.0,
      "step": 50740
    },
    {
      "entropy": 5.393372678756714,
      "epoch": 3.9481423847500485,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035115761653258177,
      "loss": 4.8309,
      "mean_token_accuracy": 0.21662657409906388,
      "num_tokens": 88021631.0,
      "step": 50745
    },
    {
      "entropy": 5.393382167816162,
      "epoch": 3.9485314141217662,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035113153630198494,
      "loss": 4.8746,
      "mean_token_accuracy": 0.20753007233142853,
      "num_tokens": 88030201.0,
      "step": 50750
    },
    {
      "entropy": 5.28232159614563,
      "epoch": 3.948920443493484,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003511054549162132,
      "loss": 4.7049,
      "mean_token_accuracy": 0.22619131058454514,
      "num_tokens": 88038332.0,
      "step": 50755
    },
    {
      "entropy": 5.372884225845337,
      "epoch": 3.949309472865201,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035107937237566233,
      "loss": 4.8556,
      "mean_token_accuracy": 0.22293240278959275,
      "num_tokens": 88046606.0,
      "step": 50760
    },
    {
      "entropy": 5.351950788497925,
      "epoch": 3.949698502236919,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003510532886807281,
      "loss": 4.7752,
      "mean_token_accuracy": 0.21822787672281266,
      "num_tokens": 88054783.0,
      "step": 50765
    },
    {
      "entropy": 5.388687086105347,
      "epoch": 3.9500875316086366,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003510272038318062,
      "loss": 4.8176,
      "mean_token_accuracy": 0.2255735531449318,
      "num_tokens": 88062649.0,
      "step": 50770
    },
    {
      "entropy": 5.377196168899536,
      "epoch": 3.950476560980354,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003510011178292925,
      "loss": 4.8486,
      "mean_token_accuracy": 0.21770191788673401,
      "num_tokens": 88071597.0,
      "step": 50775
    },
    {
      "entropy": 5.422373342514038,
      "epoch": 3.9508655903520715,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003509750306735828,
      "loss": 4.8169,
      "mean_token_accuracy": 0.21476075649261475,
      "num_tokens": 88081229.0,
      "step": 50780
    },
    {
      "entropy": 5.401809549331665,
      "epoch": 3.951254619723789,
      "grad_norm": 1.140625,
      "learning_rate": 0.000350948942365073,
      "loss": 4.8413,
      "mean_token_accuracy": 0.2183124914765358,
      "num_tokens": 88089509.0,
      "step": 50785
    },
    {
      "entropy": 5.453243684768677,
      "epoch": 3.951643649095507,
      "grad_norm": 1.125,
      "learning_rate": 0.00035092285290415885,
      "loss": 4.9145,
      "mean_token_accuracy": 0.21121155619621276,
      "num_tokens": 88098420.0,
      "step": 50790
    },
    {
      "entropy": 5.4251936912536625,
      "epoch": 3.952032678467224,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003508967622912362,
      "loss": 4.8531,
      "mean_token_accuracy": 0.21462558060884476,
      "num_tokens": 88106956.0,
      "step": 50795
    },
    {
      "entropy": 5.395204401016235,
      "epoch": 3.952421707838942,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035087067052670115,
      "loss": 4.8112,
      "mean_token_accuracy": 0.22056625485420228,
      "num_tokens": 88116021.0,
      "step": 50800
    },
    {
      "entropy": 5.355083847045899,
      "epoch": 3.9528107372106596,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035084457761094925,
      "loss": 4.7053,
      "mean_token_accuracy": 0.22740679532289504,
      "num_tokens": 88124496.0,
      "step": 50805
    },
    {
      "entropy": 5.365011262893677,
      "epoch": 3.953199766582377,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003508184835443766,
      "loss": 4.8073,
      "mean_token_accuracy": 0.2163566529750824,
      "num_tokens": 88132468.0,
      "step": 50810
    },
    {
      "entropy": 5.40766978263855,
      "epoch": 3.9535887959540945,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035079238832737926,
      "loss": 4.9911,
      "mean_token_accuracy": 0.20699509531259536,
      "num_tokens": 88142028.0,
      "step": 50815
    },
    {
      "entropy": 5.375790119171143,
      "epoch": 3.953977825325812,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000350766291960353,
      "loss": 4.801,
      "mean_token_accuracy": 0.2165253773331642,
      "num_tokens": 88150309.0,
      "step": 50820
    },
    {
      "entropy": 5.3817614078521725,
      "epoch": 3.95436685469753,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003507401944436939,
      "loss": 4.8163,
      "mean_token_accuracy": 0.21508964151144028,
      "num_tokens": 88159271.0,
      "step": 50825
    },
    {
      "entropy": 5.473111867904663,
      "epoch": 3.954755884069247,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003507140957777979,
      "loss": 4.9356,
      "mean_token_accuracy": 0.21270115822553634,
      "num_tokens": 88168320.0,
      "step": 50830
    },
    {
      "entropy": 5.4673761367797855,
      "epoch": 3.955144913440965,
      "grad_norm": 1.125,
      "learning_rate": 0.00035068799596306096,
      "loss": 4.8992,
      "mean_token_accuracy": 0.2103220120072365,
      "num_tokens": 88177078.0,
      "step": 50835
    },
    {
      "entropy": 5.464989233016968,
      "epoch": 3.955533942812682,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003506618949998791,
      "loss": 4.8916,
      "mean_token_accuracy": 0.21005450338125228,
      "num_tokens": 88184843.0,
      "step": 50840
    },
    {
      "entropy": 5.464234972000122,
      "epoch": 3.9559229721844,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035063579288864853,
      "loss": 4.9172,
      "mean_token_accuracy": 0.2137076050043106,
      "num_tokens": 88193623.0,
      "step": 50845
    },
    {
      "entropy": 5.337448310852051,
      "epoch": 3.9563120015561175,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003506096896297651,
      "loss": 4.796,
      "mean_token_accuracy": 0.21619633138179778,
      "num_tokens": 88202097.0,
      "step": 50850
    },
    {
      "entropy": 5.382431602478027,
      "epoch": 3.956701030927835,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035058358522362505,
      "loss": 4.8758,
      "mean_token_accuracy": 0.2093554198741913,
      "num_tokens": 88210537.0,
      "step": 50855
    },
    {
      "entropy": 5.326663160324097,
      "epoch": 3.9570900602995525,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035055747967062433,
      "loss": 4.7984,
      "mean_token_accuracy": 0.2235502526164055,
      "num_tokens": 88219427.0,
      "step": 50860
    },
    {
      "entropy": 5.3365637302398685,
      "epoch": 3.95747908967127,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003505313729711592,
      "loss": 4.7544,
      "mean_token_accuracy": 0.22244355231523513,
      "num_tokens": 88227886.0,
      "step": 50865
    },
    {
      "entropy": 5.353641605377197,
      "epoch": 3.957868119042988,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035050526512562564,
      "loss": 4.7642,
      "mean_token_accuracy": 0.22035931199789047,
      "num_tokens": 88236783.0,
      "step": 50870
    },
    {
      "entropy": 5.392473363876343,
      "epoch": 3.958257148414705,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003504791561344199,
      "loss": 4.851,
      "mean_token_accuracy": 0.20735141038894653,
      "num_tokens": 88245576.0,
      "step": 50875
    },
    {
      "entropy": 5.371248054504394,
      "epoch": 3.958646177786423,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035045304599793807,
      "loss": 4.8579,
      "mean_token_accuracy": 0.21282961517572402,
      "num_tokens": 88253675.0,
      "step": 50880
    },
    {
      "entropy": 5.24962010383606,
      "epoch": 3.9590352071581405,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035042693471657637,
      "loss": 4.6674,
      "mean_token_accuracy": 0.2272094801068306,
      "num_tokens": 88261870.0,
      "step": 50885
    },
    {
      "entropy": 5.358071899414062,
      "epoch": 3.959424236529858,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035040082229073096,
      "loss": 4.8057,
      "mean_token_accuracy": 0.2221774011850357,
      "num_tokens": 88270204.0,
      "step": 50890
    },
    {
      "entropy": 5.465081310272216,
      "epoch": 3.9598132659015755,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003503747087207981,
      "loss": 4.9162,
      "mean_token_accuracy": 0.21159758418798447,
      "num_tokens": 88278722.0,
      "step": 50895
    },
    {
      "entropy": 5.424841451644897,
      "epoch": 3.960202295273293,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035034859400717407,
      "loss": 4.8761,
      "mean_token_accuracy": 0.21334708482027054,
      "num_tokens": 88288110.0,
      "step": 50900
    },
    {
      "entropy": 5.4268371105194095,
      "epoch": 3.960591324645011,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035032247815025504,
      "loss": 4.8613,
      "mean_token_accuracy": 0.21504829227924346,
      "num_tokens": 88297469.0,
      "step": 50905
    },
    {
      "entropy": 5.387601900100708,
      "epoch": 3.960980354016728,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035029636115043724,
      "loss": 4.8629,
      "mean_token_accuracy": 0.20823004692792893,
      "num_tokens": 88306588.0,
      "step": 50910
    },
    {
      "entropy": 5.386000347137451,
      "epoch": 3.961369383388446,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035027024300811707,
      "loss": 4.8725,
      "mean_token_accuracy": 0.21385190337896348,
      "num_tokens": 88316275.0,
      "step": 50915
    },
    {
      "entropy": 5.398382472991943,
      "epoch": 3.9617584127601635,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003502441237236907,
      "loss": 4.8805,
      "mean_token_accuracy": 0.20613760501146317,
      "num_tokens": 88325314.0,
      "step": 50920
    },
    {
      "entropy": 5.424760866165161,
      "epoch": 3.962147442131881,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003502180032975545,
      "loss": 4.8788,
      "mean_token_accuracy": 0.21617455333471297,
      "num_tokens": 88335252.0,
      "step": 50925
    },
    {
      "entropy": 5.349911165237427,
      "epoch": 3.9625364715035984,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003501918817301049,
      "loss": 4.7855,
      "mean_token_accuracy": 0.21747283041477203,
      "num_tokens": 88343467.0,
      "step": 50930
    },
    {
      "entropy": 5.402218675613403,
      "epoch": 3.962925500875316,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035016575902173814,
      "loss": 4.9044,
      "mean_token_accuracy": 0.21832928508520127,
      "num_tokens": 88351432.0,
      "step": 50935
    },
    {
      "entropy": 5.392431545257568,
      "epoch": 3.9633145302470334,
      "grad_norm": 1.125,
      "learning_rate": 0.00035013963517285064,
      "loss": 4.879,
      "mean_token_accuracy": 0.21080988496541977,
      "num_tokens": 88360665.0,
      "step": 50940
    },
    {
      "entropy": 5.3835530281066895,
      "epoch": 3.963703559618751,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035011351018383874,
      "loss": 4.7768,
      "mean_token_accuracy": 0.2142104059457779,
      "num_tokens": 88368980.0,
      "step": 50945
    },
    {
      "entropy": 5.3297926902771,
      "epoch": 3.964092588990469,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003500873840550989,
      "loss": 4.7768,
      "mean_token_accuracy": 0.22513724118471146,
      "num_tokens": 88377888.0,
      "step": 50950
    },
    {
      "entropy": 5.2133149147033695,
      "epoch": 3.9644816183621865,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035006125678702755,
      "loss": 4.7124,
      "mean_token_accuracy": 0.2217951625585556,
      "num_tokens": 88385845.0,
      "step": 50955
    },
    {
      "entropy": 5.368986511230469,
      "epoch": 3.964870647733904,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035003512838002103,
      "loss": 4.8254,
      "mean_token_accuracy": 0.2150809422135353,
      "num_tokens": 88394608.0,
      "step": 50960
    },
    {
      "entropy": 5.404612493515015,
      "epoch": 3.9652596771056214,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003500089988344759,
      "loss": 4.7876,
      "mean_token_accuracy": 0.22453247010707855,
      "num_tokens": 88402857.0,
      "step": 50965
    },
    {
      "entropy": 5.380050897598267,
      "epoch": 3.965648706477339,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034998286815078855,
      "loss": 4.8672,
      "mean_token_accuracy": 0.20557522773742676,
      "num_tokens": 88410977.0,
      "step": 50970
    },
    {
      "entropy": 5.505987882614136,
      "epoch": 3.9660377358490564,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034995673632935557,
      "loss": 5.0108,
      "mean_token_accuracy": 0.20224587470293046,
      "num_tokens": 88420409.0,
      "step": 50975
    },
    {
      "entropy": 5.378875350952148,
      "epoch": 3.966426765220774,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003499306033705735,
      "loss": 4.7935,
      "mean_token_accuracy": 0.21617303192615508,
      "num_tokens": 88429389.0,
      "step": 50980
    },
    {
      "entropy": 5.435645246505738,
      "epoch": 3.966815794592492,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034990446927483877,
      "loss": 4.8638,
      "mean_token_accuracy": 0.2078116089105606,
      "num_tokens": 88437758.0,
      "step": 50985
    },
    {
      "entropy": 5.380201148986816,
      "epoch": 3.9672048239642095,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003498783340425479,
      "loss": 4.7734,
      "mean_token_accuracy": 0.22040392756462096,
      "num_tokens": 88446205.0,
      "step": 50990
    },
    {
      "entropy": 5.3852544784545895,
      "epoch": 3.9675938533359267,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034985219767409747,
      "loss": 4.811,
      "mean_token_accuracy": 0.21162720322608947,
      "num_tokens": 88453808.0,
      "step": 50995
    },
    {
      "entropy": 5.288815355300903,
      "epoch": 3.9679828827076444,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034982606016988407,
      "loss": 4.8088,
      "mean_token_accuracy": 0.21983133107423783,
      "num_tokens": 88462559.0,
      "step": 51000
    },
    {
      "epoch": 3.9679828827076444,
      "eval_entropy": 5.098066503159465,
      "eval_loss": 4.967385292053223,
      "eval_mean_token_accuracy": 0.21750381170871513,
      "eval_num_tokens": 88462559.0,
      "eval_runtime": 28.7204,
      "eval_samples_per_second": 1446.984,
      "eval_steps_per_second": 180.882,
      "step": 51000
    },
    {
      "entropy": 5.343781900405884,
      "epoch": 3.968371912079362,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034979992153030424,
      "loss": 4.8418,
      "mean_token_accuracy": 0.21697256416082383,
      "num_tokens": 88471633.0,
      "step": 51005
    },
    {
      "entropy": 5.355787229537964,
      "epoch": 3.9687609414510794,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003497737817557548,
      "loss": 4.888,
      "mean_token_accuracy": 0.20885892510414122,
      "num_tokens": 88481362.0,
      "step": 51010
    },
    {
      "entropy": 5.341981029510498,
      "epoch": 3.969149970822797,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034974764084663213,
      "loss": 4.7324,
      "mean_token_accuracy": 0.22860104441642762,
      "num_tokens": 88489742.0,
      "step": 51015
    },
    {
      "entropy": 5.323961877822876,
      "epoch": 3.9695390001945148,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034972149880333316,
      "loss": 4.7854,
      "mean_token_accuracy": 0.2207481548190117,
      "num_tokens": 88499049.0,
      "step": 51020
    },
    {
      "entropy": 5.324609184265137,
      "epoch": 3.9699280295662325,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003496953556262542,
      "loss": 4.794,
      "mean_token_accuracy": 0.2267455130815506,
      "num_tokens": 88507810.0,
      "step": 51025
    },
    {
      "entropy": 5.3938860416412355,
      "epoch": 3.9703170589379497,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034966921131579216,
      "loss": 4.7855,
      "mean_token_accuracy": 0.21184342801570893,
      "num_tokens": 88515807.0,
      "step": 51030
    },
    {
      "entropy": 5.374936819076538,
      "epoch": 3.9707060883096674,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034964306587234367,
      "loss": 4.8588,
      "mean_token_accuracy": 0.22285531908273698,
      "num_tokens": 88524493.0,
      "step": 51035
    },
    {
      "entropy": 5.412810373306274,
      "epoch": 3.9710951176813847,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034961691929630544,
      "loss": 4.8959,
      "mean_token_accuracy": 0.20716042667627335,
      "num_tokens": 88533241.0,
      "step": 51040
    },
    {
      "entropy": 5.443279027938843,
      "epoch": 3.9714841470531024,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003495907715880743,
      "loss": 4.96,
      "mean_token_accuracy": 0.21013213843107223,
      "num_tokens": 88542889.0,
      "step": 51045
    },
    {
      "entropy": 5.385065412521362,
      "epoch": 3.97187317642482,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003495646227480469,
      "loss": 4.8166,
      "mean_token_accuracy": 0.21199689507484437,
      "num_tokens": 88551894.0,
      "step": 51050
    },
    {
      "entropy": 5.3824756145477295,
      "epoch": 3.9722622057965378,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034953847277662,
      "loss": 4.8854,
      "mean_token_accuracy": 0.22265829443931578,
      "num_tokens": 88560518.0,
      "step": 51055
    },
    {
      "entropy": 5.382286262512207,
      "epoch": 3.9726512351682555,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034951232167419044,
      "loss": 4.885,
      "mean_token_accuracy": 0.2071485236287117,
      "num_tokens": 88569270.0,
      "step": 51060
    },
    {
      "entropy": 5.3590740203857425,
      "epoch": 3.9730402645399727,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034948616944115504,
      "loss": 4.7976,
      "mean_token_accuracy": 0.21538321524858475,
      "num_tokens": 88579208.0,
      "step": 51065
    },
    {
      "entropy": 5.3422284603118895,
      "epoch": 3.9734292939116904,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003494600160779105,
      "loss": 4.7529,
      "mean_token_accuracy": 0.22014327645301818,
      "num_tokens": 88587654.0,
      "step": 51070
    },
    {
      "entropy": 5.39596996307373,
      "epoch": 3.9738183232834077,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003494338615848537,
      "loss": 4.869,
      "mean_token_accuracy": 0.2166467010974884,
      "num_tokens": 88596049.0,
      "step": 51075
    },
    {
      "entropy": 5.306340265274048,
      "epoch": 3.9742073526551254,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003494077059623816,
      "loss": 4.8067,
      "mean_token_accuracy": 0.21405717134475707,
      "num_tokens": 88604939.0,
      "step": 51080
    },
    {
      "entropy": 5.349564552307129,
      "epoch": 3.974596382026843,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034938154921089097,
      "loss": 4.8202,
      "mean_token_accuracy": 0.21656274646520615,
      "num_tokens": 88613191.0,
      "step": 51085
    },
    {
      "entropy": 5.442694282531738,
      "epoch": 3.9749854113985608,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003493553913307788,
      "loss": 4.8909,
      "mean_token_accuracy": 0.21238974928855897,
      "num_tokens": 88621520.0,
      "step": 51090
    },
    {
      "entropy": 5.3690577983856205,
      "epoch": 3.975374440770278,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034932923232244183,
      "loss": 4.8444,
      "mean_token_accuracy": 0.21954160183668137,
      "num_tokens": 88630216.0,
      "step": 51095
    },
    {
      "entropy": 5.463378095626831,
      "epoch": 3.9757634701419957,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00034930307218627706,
      "loss": 4.8806,
      "mean_token_accuracy": 0.2092505007982254,
      "num_tokens": 88638677.0,
      "step": 51100
    },
    {
      "entropy": 5.429535388946533,
      "epoch": 3.9761524995137134,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003492769109226815,
      "loss": 4.8571,
      "mean_token_accuracy": 0.20684039145708083,
      "num_tokens": 88646955.0,
      "step": 51105
    },
    {
      "entropy": 5.297150468826294,
      "epoch": 3.9765415288854307,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034925074853205194,
      "loss": 4.6729,
      "mean_token_accuracy": 0.23237662613391877,
      "num_tokens": 88655120.0,
      "step": 51110
    },
    {
      "entropy": 5.32683916091919,
      "epoch": 3.9769305582571484,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034922458501478556,
      "loss": 4.9106,
      "mean_token_accuracy": 0.2098128542304039,
      "num_tokens": 88663942.0,
      "step": 51115
    },
    {
      "entropy": 5.4396649360656735,
      "epoch": 3.977319587628866,
      "grad_norm": 1.125,
      "learning_rate": 0.0003491984203712792,
      "loss": 4.8684,
      "mean_token_accuracy": 0.21191660761833192,
      "num_tokens": 88672490.0,
      "step": 51120
    },
    {
      "entropy": 5.507908964157105,
      "epoch": 3.9777086170005838,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003491722546019299,
      "loss": 4.9225,
      "mean_token_accuracy": 0.21909769028425216,
      "num_tokens": 88681091.0,
      "step": 51125
    },
    {
      "entropy": 5.321393346786499,
      "epoch": 3.978097646372301,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003491460877071347,
      "loss": 4.8268,
      "mean_token_accuracy": 0.21862240731716157,
      "num_tokens": 88690473.0,
      "step": 51130
    },
    {
      "entropy": 5.527681159973144,
      "epoch": 3.9784866757440187,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003491199196872906,
      "loss": 4.9713,
      "mean_token_accuracy": 0.20171018540859223,
      "num_tokens": 88699036.0,
      "step": 51135
    },
    {
      "entropy": 5.476132535934449,
      "epoch": 3.978875705115736,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003490937505427947,
      "loss": 4.8899,
      "mean_token_accuracy": 0.21066357046365738,
      "num_tokens": 88707893.0,
      "step": 51140
    },
    {
      "entropy": 5.432781934738159,
      "epoch": 3.9792647344874537,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003490675802740441,
      "loss": 4.809,
      "mean_token_accuracy": 0.21702309548854828,
      "num_tokens": 88716743.0,
      "step": 51145
    },
    {
      "entropy": 5.354176187515259,
      "epoch": 3.9796537638591714,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003490414088814359,
      "loss": 4.9084,
      "mean_token_accuracy": 0.21376069039106368,
      "num_tokens": 88724970.0,
      "step": 51150
    },
    {
      "entropy": 5.312931823730469,
      "epoch": 3.980042793230889,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003490152363653671,
      "loss": 4.7129,
      "mean_token_accuracy": 0.22563090771436692,
      "num_tokens": 88733207.0,
      "step": 51155
    },
    {
      "entropy": 5.443997240066528,
      "epoch": 3.9804318226026068,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003489890627262349,
      "loss": 4.9283,
      "mean_token_accuracy": 0.20759814381599426,
      "num_tokens": 88741907.0,
      "step": 51160
    },
    {
      "entropy": 5.45283145904541,
      "epoch": 3.980820851974324,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034896288796443654,
      "loss": 4.8055,
      "mean_token_accuracy": 0.20930091738700868,
      "num_tokens": 88751080.0,
      "step": 51165
    },
    {
      "entropy": 5.3644510269165036,
      "epoch": 3.9812098813460417,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00034893671208036895,
      "loss": 4.8314,
      "mean_token_accuracy": 0.21665865033864976,
      "num_tokens": 88759387.0,
      "step": 51170
    },
    {
      "entropy": 5.416509389877319,
      "epoch": 3.981598910717759,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034891053507442944,
      "loss": 4.8495,
      "mean_token_accuracy": 0.21289546191692352,
      "num_tokens": 88768785.0,
      "step": 51175
    },
    {
      "entropy": 5.474390172958374,
      "epoch": 3.9819879400894767,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003488843569470153,
      "loss": 4.8722,
      "mean_token_accuracy": 0.2128552094101906,
      "num_tokens": 88777835.0,
      "step": 51180
    },
    {
      "entropy": 5.399246501922607,
      "epoch": 3.9823769694611943,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034885817769852355,
      "loss": 4.8402,
      "mean_token_accuracy": 0.20878460109233857,
      "num_tokens": 88786447.0,
      "step": 51185
    },
    {
      "entropy": 5.392108345031739,
      "epoch": 3.982765998832912,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034883199732935156,
      "loss": 4.8271,
      "mean_token_accuracy": 0.2162730023264885,
      "num_tokens": 88795235.0,
      "step": 51190
    },
    {
      "entropy": 5.397614192962647,
      "epoch": 3.9831550282046293,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034880581583989655,
      "loss": 4.8213,
      "mean_token_accuracy": 0.2091386005282402,
      "num_tokens": 88803975.0,
      "step": 51195
    },
    {
      "entropy": 5.4236688137054445,
      "epoch": 3.983544057576347,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003487796332305557,
      "loss": 4.9108,
      "mean_token_accuracy": 0.21045883893966674,
      "num_tokens": 88813332.0,
      "step": 51200
    },
    {
      "entropy": 5.440752410888672,
      "epoch": 3.9839330869480647,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034875344950172635,
      "loss": 4.9397,
      "mean_token_accuracy": 0.20393143594264984,
      "num_tokens": 88822230.0,
      "step": 51205
    },
    {
      "entropy": 5.399003362655639,
      "epoch": 3.984322116319782,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003487272646538058,
      "loss": 4.7896,
      "mean_token_accuracy": 0.2177165001630783,
      "num_tokens": 88830790.0,
      "step": 51210
    },
    {
      "entropy": 5.486355257034302,
      "epoch": 3.9847111456914996,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00034870107868719135,
      "loss": 4.9464,
      "mean_token_accuracy": 0.2047772690653801,
      "num_tokens": 88839191.0,
      "step": 51215
    },
    {
      "entropy": 5.357129955291748,
      "epoch": 3.9851001750632173,
      "grad_norm": 1.125,
      "learning_rate": 0.0003486748916022804,
      "loss": 4.7712,
      "mean_token_accuracy": 0.22061035931110382,
      "num_tokens": 88848066.0,
      "step": 51220
    },
    {
      "entropy": 5.340689468383789,
      "epoch": 3.985489204434935,
      "grad_norm": 1.125,
      "learning_rate": 0.0003486487033994702,
      "loss": 4.8437,
      "mean_token_accuracy": 0.21492844223976135,
      "num_tokens": 88856323.0,
      "step": 51225
    },
    {
      "entropy": 5.3923829078674315,
      "epoch": 3.9858782338066523,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034862251407915813,
      "loss": 4.8309,
      "mean_token_accuracy": 0.21588771641254426,
      "num_tokens": 88865294.0,
      "step": 51230
    },
    {
      "entropy": 5.446277713775634,
      "epoch": 3.98626726317837,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034859632364174167,
      "loss": 4.8728,
      "mean_token_accuracy": 0.21527935415506363,
      "num_tokens": 88873572.0,
      "step": 51235
    },
    {
      "entropy": 5.432677698135376,
      "epoch": 3.9866562925500877,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000348570132087618,
      "loss": 4.8293,
      "mean_token_accuracy": 0.21862333565950393,
      "num_tokens": 88881861.0,
      "step": 51240
    },
    {
      "entropy": 5.368596982955933,
      "epoch": 3.987045321921805,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003485439394171847,
      "loss": 4.9306,
      "mean_token_accuracy": 0.2149570792913437,
      "num_tokens": 88891805.0,
      "step": 51245
    },
    {
      "entropy": 5.431462240219116,
      "epoch": 3.9874343512935226,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034851774563083925,
      "loss": 4.9155,
      "mean_token_accuracy": 0.20740412324666976,
      "num_tokens": 88899913.0,
      "step": 51250
    },
    {
      "entropy": 5.436556482315064,
      "epoch": 3.9878233806652403,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003484915507289791,
      "loss": 4.841,
      "mean_token_accuracy": 0.2212355688214302,
      "num_tokens": 88908684.0,
      "step": 51255
    },
    {
      "entropy": 5.433439064025879,
      "epoch": 3.988212410036958,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034846535471200153,
      "loss": 4.9055,
      "mean_token_accuracy": 0.21381947845220567,
      "num_tokens": 88917514.0,
      "step": 51260
    },
    {
      "entropy": 5.319541311264038,
      "epoch": 3.9886014394086753,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003484391575803041,
      "loss": 4.8479,
      "mean_token_accuracy": 0.22130178660154343,
      "num_tokens": 88926065.0,
      "step": 51265
    },
    {
      "entropy": 5.392234754562378,
      "epoch": 3.988990468780393,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003484129593342844,
      "loss": 4.8001,
      "mean_token_accuracy": 0.21393730044364928,
      "num_tokens": 88934389.0,
      "step": 51270
    },
    {
      "entropy": 5.457669925689697,
      "epoch": 3.9893794981521102,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034838675997433987,
      "loss": 4.8342,
      "mean_token_accuracy": 0.2130183070898056,
      "num_tokens": 88943691.0,
      "step": 51275
    },
    {
      "entropy": 5.409145450592041,
      "epoch": 3.989768527523828,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003483605595008681,
      "loss": 4.9274,
      "mean_token_accuracy": 0.20783212184906005,
      "num_tokens": 88952318.0,
      "step": 51280
    },
    {
      "entropy": 5.361263656616211,
      "epoch": 3.9901575568955456,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034833435791426653,
      "loss": 4.7326,
      "mean_token_accuracy": 0.2281709060072899,
      "num_tokens": 88960689.0,
      "step": 51285
    },
    {
      "entropy": 5.306232213973999,
      "epoch": 3.9905465862672633,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034830815521493295,
      "loss": 4.7723,
      "mean_token_accuracy": 0.22109062671661378,
      "num_tokens": 88969533.0,
      "step": 51290
    },
    {
      "entropy": 5.422588348388672,
      "epoch": 3.9909356156389806,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003482819514032647,
      "loss": 4.8298,
      "mean_token_accuracy": 0.21314404010772706,
      "num_tokens": 88978512.0,
      "step": 51295
    },
    {
      "entropy": 5.420740222930908,
      "epoch": 3.9913246450106983,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034825574647965945,
      "loss": 4.8398,
      "mean_token_accuracy": 0.21670727878808976,
      "num_tokens": 88987120.0,
      "step": 51300
    },
    {
      "entropy": 5.417478132247925,
      "epoch": 3.991713674382416,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034822954044451485,
      "loss": 4.8883,
      "mean_token_accuracy": 0.210420124232769,
      "num_tokens": 88996271.0,
      "step": 51305
    },
    {
      "entropy": 5.412640762329102,
      "epoch": 3.9921027037541332,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003482033332982286,
      "loss": 4.8467,
      "mean_token_accuracy": 0.20758840143680574,
      "num_tokens": 89004538.0,
      "step": 51310
    },
    {
      "entropy": 5.430873775482178,
      "epoch": 3.992491733125851,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003481771250411982,
      "loss": 4.8945,
      "mean_token_accuracy": 0.21125626415014268,
      "num_tokens": 89013662.0,
      "step": 51315
    },
    {
      "entropy": 5.3577282428741455,
      "epoch": 3.9928807624975686,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003481509156738214,
      "loss": 4.8276,
      "mean_token_accuracy": 0.2193271279335022,
      "num_tokens": 89022451.0,
      "step": 51320
    },
    {
      "entropy": 5.345005798339844,
      "epoch": 3.9932697918692863,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034812470519649593,
      "loss": 4.8155,
      "mean_token_accuracy": 0.20838116258382797,
      "num_tokens": 89030968.0,
      "step": 51325
    },
    {
      "entropy": 5.368816232681274,
      "epoch": 3.9936588212410036,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003480984936096194,
      "loss": 4.8481,
      "mean_token_accuracy": 0.21497409045696259,
      "num_tokens": 89039614.0,
      "step": 51330
    },
    {
      "entropy": 5.368923711776733,
      "epoch": 3.9940478506127213,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034807228091358964,
      "loss": 4.7879,
      "mean_token_accuracy": 0.2193361535668373,
      "num_tokens": 89048245.0,
      "step": 51335
    },
    {
      "entropy": 5.336399888992309,
      "epoch": 3.994436879984439,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003480460671088043,
      "loss": 4.7835,
      "mean_token_accuracy": 0.21289712935686111,
      "num_tokens": 89057486.0,
      "step": 51340
    },
    {
      "entropy": 5.4264050960540775,
      "epoch": 3.9948259093561562,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034801985219566124,
      "loss": 4.8657,
      "mean_token_accuracy": 0.2153733879327774,
      "num_tokens": 89065833.0,
      "step": 51345
    },
    {
      "entropy": 5.333083200454712,
      "epoch": 3.995214938727874,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000347993636174558,
      "loss": 4.8685,
      "mean_token_accuracy": 0.21378479450941085,
      "num_tokens": 89074562.0,
      "step": 51350
    },
    {
      "entropy": 5.3985615253448485,
      "epoch": 3.9956039680995916,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034796741904589263,
      "loss": 4.9388,
      "mean_token_accuracy": 0.21665503531694413,
      "num_tokens": 89083013.0,
      "step": 51355
    },
    {
      "entropy": 5.411608934402466,
      "epoch": 3.9959929974713093,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034794120081006275,
      "loss": 4.8365,
      "mean_token_accuracy": 0.21547756642103194,
      "num_tokens": 89090926.0,
      "step": 51360
    },
    {
      "entropy": 5.397363710403442,
      "epoch": 3.9963820268430266,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003479149814674663,
      "loss": 4.8134,
      "mean_token_accuracy": 0.22248280942440032,
      "num_tokens": 89099251.0,
      "step": 51365
    },
    {
      "entropy": 5.301083898544311,
      "epoch": 3.9967710562147443,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000347888761018501,
      "loss": 4.8713,
      "mean_token_accuracy": 0.20601505041122437,
      "num_tokens": 89108227.0,
      "step": 51370
    },
    {
      "entropy": 5.3769653797149655,
      "epoch": 3.9971600855864615,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003478625394635648,
      "loss": 4.8394,
      "mean_token_accuracy": 0.21165846586227416,
      "num_tokens": 89116434.0,
      "step": 51375
    },
    {
      "entropy": 5.3993518352508545,
      "epoch": 3.997549114958179,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034783631680305554,
      "loss": 4.7999,
      "mean_token_accuracy": 0.21658010482788087,
      "num_tokens": 89124855.0,
      "step": 51380
    },
    {
      "entropy": 5.321805191040039,
      "epoch": 3.997938144329897,
      "grad_norm": 1.125,
      "learning_rate": 0.0003478100930373711,
      "loss": 4.7863,
      "mean_token_accuracy": 0.2177918165922165,
      "num_tokens": 89133476.0,
      "step": 51385
    },
    {
      "entropy": 5.29525318145752,
      "epoch": 3.9983271737016146,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003477838681669094,
      "loss": 4.8008,
      "mean_token_accuracy": 0.22052622586488724,
      "num_tokens": 89142306.0,
      "step": 51390
    },
    {
      "entropy": 5.36322546005249,
      "epoch": 3.9987162030733323,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003477576421920682,
      "loss": 4.8024,
      "mean_token_accuracy": 0.22153883278369904,
      "num_tokens": 89151101.0,
      "step": 51395
    },
    {
      "entropy": 5.404524421691894,
      "epoch": 3.9991052324450496,
      "grad_norm": 1.125,
      "learning_rate": 0.0003477314151132458,
      "loss": 4.835,
      "mean_token_accuracy": 0.2151877373456955,
      "num_tokens": 89159311.0,
      "step": 51400
    },
    {
      "entropy": 5.4113623142242435,
      "epoch": 3.9994942618167673,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034770518693083986,
      "loss": 4.8516,
      "mean_token_accuracy": 0.2121650144457817,
      "num_tokens": 89167130.0,
      "step": 51405
    },
    {
      "entropy": 5.406304502487183,
      "epoch": 3.9998832911884845,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034767895764524845,
      "loss": 4.9047,
      "mean_token_accuracy": 0.20899797976016998,
      "num_tokens": 89175732.0,
      "step": 51410
    },
    {
      "entropy": 5.426645225948757,
      "epoch": 4.000233417623031,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034765272725686947,
      "loss": 4.8293,
      "mean_token_accuracy": 0.2202550686068005,
      "num_tokens": 89182536.0,
      "step": 51415
    },
    {
      "entropy": 5.370031404495239,
      "epoch": 4.000622446994748,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034762649576610105,
      "loss": 4.7549,
      "mean_token_accuracy": 0.21045888513326644,
      "num_tokens": 89191431.0,
      "step": 51420
    },
    {
      "entropy": 5.351933288574219,
      "epoch": 4.0010114763664655,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034760026317334113,
      "loss": 4.7834,
      "mean_token_accuracy": 0.21218374818563462,
      "num_tokens": 89200365.0,
      "step": 51425
    },
    {
      "entropy": 5.404979085922241,
      "epoch": 4.001400505738183,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003475740294789878,
      "loss": 4.8271,
      "mean_token_accuracy": 0.2177969053387642,
      "num_tokens": 89208968.0,
      "step": 51430
    },
    {
      "entropy": 5.470998620986938,
      "epoch": 4.001789535109901,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000347547794683439,
      "loss": 4.9017,
      "mean_token_accuracy": 0.2125532314181328,
      "num_tokens": 89217543.0,
      "step": 51435
    },
    {
      "entropy": 5.358471393585205,
      "epoch": 4.002178564481619,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003475215587870929,
      "loss": 4.6941,
      "mean_token_accuracy": 0.2282708540558815,
      "num_tokens": 89225847.0,
      "step": 51440
    },
    {
      "entropy": 5.477706241607666,
      "epoch": 4.002567593853336,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003474953217903477,
      "loss": 4.9038,
      "mean_token_accuracy": 0.20762912780046464,
      "num_tokens": 89235010.0,
      "step": 51445
    },
    {
      "entropy": 5.473307991027832,
      "epoch": 4.002956623225053,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003474690836936013,
      "loss": 4.8143,
      "mean_token_accuracy": 0.21996689587831497,
      "num_tokens": 89243827.0,
      "step": 51450
    },
    {
      "entropy": 5.425415277481079,
      "epoch": 4.003345652596771,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003474428444972519,
      "loss": 4.8921,
      "mean_token_accuracy": 0.21453977525234222,
      "num_tokens": 89252574.0,
      "step": 51455
    },
    {
      "entropy": 5.350248718261719,
      "epoch": 4.0037346819684885,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003474166042016977,
      "loss": 4.7973,
      "mean_token_accuracy": 0.2211287260055542,
      "num_tokens": 89260827.0,
      "step": 51460
    },
    {
      "entropy": 5.3988597869873045,
      "epoch": 4.004123711340206,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034739036280733674,
      "loss": 4.8422,
      "mean_token_accuracy": 0.2146250605583191,
      "num_tokens": 89269635.0,
      "step": 51465
    },
    {
      "entropy": 5.485918378829956,
      "epoch": 4.004512740711924,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034736412031456714,
      "loss": 4.8878,
      "mean_token_accuracy": 0.21481534987688064,
      "num_tokens": 89278490.0,
      "step": 51470
    },
    {
      "entropy": 5.497883892059326,
      "epoch": 4.004901770083642,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034733787672378734,
      "loss": 4.8299,
      "mean_token_accuracy": 0.2133852571249008,
      "num_tokens": 89287337.0,
      "step": 51475
    },
    {
      "entropy": 5.386208534240723,
      "epoch": 4.005290799455359,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003473116320353953,
      "loss": 4.7574,
      "mean_token_accuracy": 0.22357117533683776,
      "num_tokens": 89295703.0,
      "step": 51480
    },
    {
      "entropy": 5.420983934402466,
      "epoch": 4.005679828827076,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003472853862497895,
      "loss": 4.8394,
      "mean_token_accuracy": 0.22127819806337357,
      "num_tokens": 89304470.0,
      "step": 51485
    },
    {
      "entropy": 5.225989484786988,
      "epoch": 4.006068858198794,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003472591393673679,
      "loss": 4.6128,
      "mean_token_accuracy": 0.2307997763156891,
      "num_tokens": 89313344.0,
      "step": 51490
    },
    {
      "entropy": 5.289905214309693,
      "epoch": 4.0064578875705115,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034723289138852885,
      "loss": 4.7212,
      "mean_token_accuracy": 0.22366084307432174,
      "num_tokens": 89322112.0,
      "step": 51495
    },
    {
      "entropy": 5.383905601501465,
      "epoch": 4.006846916942229,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003472066423136707,
      "loss": 4.7316,
      "mean_token_accuracy": 0.22573547810316086,
      "num_tokens": 89331972.0,
      "step": 51500
    },
    {
      "entropy": 5.3920598983764645,
      "epoch": 4.007235946313947,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003471803921431917,
      "loss": 4.7119,
      "mean_token_accuracy": 0.233779576420784,
      "num_tokens": 89340343.0,
      "step": 51505
    },
    {
      "entropy": 5.328002643585205,
      "epoch": 4.007624975685665,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034715414087749003,
      "loss": 4.8527,
      "mean_token_accuracy": 0.2161891222000122,
      "num_tokens": 89349063.0,
      "step": 51510
    },
    {
      "entropy": 5.305205011367798,
      "epoch": 4.008014005057382,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003471278885169642,
      "loss": 4.7423,
      "mean_token_accuracy": 0.2289646103978157,
      "num_tokens": 89357648.0,
      "step": 51515
    },
    {
      "entropy": 5.420944738388061,
      "epoch": 4.008403034429099,
      "grad_norm": 1.125,
      "learning_rate": 0.00034710163506201247,
      "loss": 4.8851,
      "mean_token_accuracy": 0.21347897052764891,
      "num_tokens": 89366168.0,
      "step": 51520
    },
    {
      "entropy": 5.285510778427124,
      "epoch": 4.008792063800817,
      "grad_norm": 1.125,
      "learning_rate": 0.00034707538051303316,
      "loss": 4.6728,
      "mean_token_accuracy": 0.2204428121447563,
      "num_tokens": 89374314.0,
      "step": 51525
    },
    {
      "entropy": 5.372182559967041,
      "epoch": 4.0091810931725345,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003470491248704247,
      "loss": 4.8202,
      "mean_token_accuracy": 0.21968117654323577,
      "num_tokens": 89383377.0,
      "step": 51530
    },
    {
      "entropy": 5.414815950393677,
      "epoch": 4.009570122544252,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034702286813458543,
      "loss": 4.7916,
      "mean_token_accuracy": 0.21997666656970977,
      "num_tokens": 89392456.0,
      "step": 51535
    },
    {
      "entropy": 5.4210433006286625,
      "epoch": 4.00995915191597,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034699661030591374,
      "loss": 4.9036,
      "mean_token_accuracy": 0.21345265507698058,
      "num_tokens": 89401398.0,
      "step": 51540
    },
    {
      "entropy": 5.460687255859375,
      "epoch": 4.0103481812876876,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003469703513848081,
      "loss": 4.8225,
      "mean_token_accuracy": 0.21796278953552245,
      "num_tokens": 89411138.0,
      "step": 51545
    },
    {
      "entropy": 5.348271608352661,
      "epoch": 4.010737210659404,
      "grad_norm": 1.1875,
      "learning_rate": 0.000346944091371667,
      "loss": 4.7235,
      "mean_token_accuracy": 0.22763839215040207,
      "num_tokens": 89420121.0,
      "step": 51550
    },
    {
      "entropy": 5.410749769210815,
      "epoch": 4.011126240031122,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003469178302668888,
      "loss": 4.8223,
      "mean_token_accuracy": 0.20881318002939225,
      "num_tokens": 89429381.0,
      "step": 51555
    },
    {
      "entropy": 5.349842977523804,
      "epoch": 4.01151526940284,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003468915680708719,
      "loss": 4.806,
      "mean_token_accuracy": 0.22515250444412233,
      "num_tokens": 89438658.0,
      "step": 51560
    },
    {
      "entropy": 5.353598499298096,
      "epoch": 4.0119042987745575,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000346865304784015,
      "loss": 4.7728,
      "mean_token_accuracy": 0.21572664976119996,
      "num_tokens": 89447336.0,
      "step": 51565
    },
    {
      "entropy": 5.344105005264282,
      "epoch": 4.012293328146275,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003468390404067164,
      "loss": 4.6714,
      "mean_token_accuracy": 0.2243387967348099,
      "num_tokens": 89456006.0,
      "step": 51570
    },
    {
      "entropy": 5.423496532440185,
      "epoch": 4.012682357517993,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003468127749393747,
      "loss": 4.8228,
      "mean_token_accuracy": 0.21916327029466628,
      "num_tokens": 89465238.0,
      "step": 51575
    },
    {
      "entropy": 5.405986309051514,
      "epoch": 4.0130713868897105,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003467865083823885,
      "loss": 4.8664,
      "mean_token_accuracy": 0.2188045084476471,
      "num_tokens": 89474969.0,
      "step": 51580
    },
    {
      "entropy": 5.360036849975586,
      "epoch": 4.013460416261427,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003467602407361563,
      "loss": 4.7646,
      "mean_token_accuracy": 0.21831746846437455,
      "num_tokens": 89483678.0,
      "step": 51585
    },
    {
      "entropy": 5.331708669662476,
      "epoch": 4.013849445633145,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034673397200107663,
      "loss": 4.7415,
      "mean_token_accuracy": 0.22091909348964692,
      "num_tokens": 89492640.0,
      "step": 51590
    },
    {
      "entropy": 5.357826995849609,
      "epoch": 4.014238475004863,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003467077021775481,
      "loss": 4.7211,
      "mean_token_accuracy": 0.22746371626853942,
      "num_tokens": 89500771.0,
      "step": 51595
    },
    {
      "entropy": 5.39482011795044,
      "epoch": 4.0146275043765804,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034668143126596933,
      "loss": 4.8235,
      "mean_token_accuracy": 0.21425239443778993,
      "num_tokens": 89509182.0,
      "step": 51600
    },
    {
      "entropy": 5.3698913097381595,
      "epoch": 4.015016533748298,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034665515926673903,
      "loss": 4.7326,
      "mean_token_accuracy": 0.22454380840063096,
      "num_tokens": 89517747.0,
      "step": 51605
    },
    {
      "entropy": 5.458690452575683,
      "epoch": 4.015405563120016,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003466288861802556,
      "loss": 4.8742,
      "mean_token_accuracy": 0.20842254310846328,
      "num_tokens": 89527386.0,
      "step": 51610
    },
    {
      "entropy": 5.416431140899658,
      "epoch": 4.0157945924917335,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003466026120069179,
      "loss": 4.7325,
      "mean_token_accuracy": 0.22741924226284027,
      "num_tokens": 89537116.0,
      "step": 51615
    },
    {
      "entropy": 5.391231489181519,
      "epoch": 4.01618362186345,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003465763367471245,
      "loss": 4.8135,
      "mean_token_accuracy": 0.22424491196870805,
      "num_tokens": 89545725.0,
      "step": 51620
    },
    {
      "entropy": 5.360003852844239,
      "epoch": 4.016572651235168,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034655006040127415,
      "loss": 4.7459,
      "mean_token_accuracy": 0.22644126117229463,
      "num_tokens": 89554055.0,
      "step": 51625
    },
    {
      "entropy": 5.381274461746216,
      "epoch": 4.016961680606886,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003465237829697655,
      "loss": 4.7756,
      "mean_token_accuracy": 0.22149696201086044,
      "num_tokens": 89563438.0,
      "step": 51630
    },
    {
      "entropy": 5.368740367889404,
      "epoch": 4.017350709978603,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034649750445299725,
      "loss": 4.7963,
      "mean_token_accuracy": 0.22239546626806259,
      "num_tokens": 89571847.0,
      "step": 51635
    },
    {
      "entropy": 5.42005934715271,
      "epoch": 4.017739739350321,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003464712248513682,
      "loss": 4.8972,
      "mean_token_accuracy": 0.21729420721530915,
      "num_tokens": 89581295.0,
      "step": 51640
    },
    {
      "entropy": 5.393592882156372,
      "epoch": 4.018128768722039,
      "grad_norm": 1.125,
      "learning_rate": 0.0003464449441652771,
      "loss": 4.7141,
      "mean_token_accuracy": 0.22602360397577287,
      "num_tokens": 89590029.0,
      "step": 51645
    },
    {
      "entropy": 5.368975257873535,
      "epoch": 4.018517798093756,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003464186623951227,
      "loss": 4.7423,
      "mean_token_accuracy": 0.21539913415908812,
      "num_tokens": 89598508.0,
      "step": 51650
    },
    {
      "entropy": 5.4092296123504635,
      "epoch": 4.018906827465473,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003463923795413037,
      "loss": 4.8519,
      "mean_token_accuracy": 0.21575629562139512,
      "num_tokens": 89607199.0,
      "step": 51655
    },
    {
      "entropy": 5.349124622344971,
      "epoch": 4.019295856837191,
      "grad_norm": 1.125,
      "learning_rate": 0.00034636609560421906,
      "loss": 4.7666,
      "mean_token_accuracy": 0.22525476664304733,
      "num_tokens": 89616048.0,
      "step": 51660
    },
    {
      "entropy": 5.341390657424927,
      "epoch": 4.019684886208909,
      "grad_norm": 1.25,
      "learning_rate": 0.00034633981058426744,
      "loss": 4.7437,
      "mean_token_accuracy": 0.22339099794626235,
      "num_tokens": 89624652.0,
      "step": 51665
    },
    {
      "entropy": 5.380967617034912,
      "epoch": 4.020073915580626,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003463135244818477,
      "loss": 4.7328,
      "mean_token_accuracy": 0.222388331592083,
      "num_tokens": 89633132.0,
      "step": 51670
    },
    {
      "entropy": 5.266452074050903,
      "epoch": 4.020462944952344,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003462872372973588,
      "loss": 4.7421,
      "mean_token_accuracy": 0.2155402272939682,
      "num_tokens": 89641590.0,
      "step": 51675
    },
    {
      "entropy": 5.4299884796142575,
      "epoch": 4.020851974324062,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003462609490311996,
      "loss": 4.8594,
      "mean_token_accuracy": 0.21134717166423797,
      "num_tokens": 89650269.0,
      "step": 51680
    },
    {
      "entropy": 5.41226978302002,
      "epoch": 4.021241003695779,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003462346596837689,
      "loss": 4.7661,
      "mean_token_accuracy": 0.22699035406112672,
      "num_tokens": 89658493.0,
      "step": 51685
    },
    {
      "entropy": 5.371568489074707,
      "epoch": 4.021630033067496,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003462083692554655,
      "loss": 4.6814,
      "mean_token_accuracy": 0.2242964118719101,
      "num_tokens": 89667128.0,
      "step": 51690
    },
    {
      "entropy": 5.345443916320801,
      "epoch": 4.022019062439214,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034618207774668854,
      "loss": 4.8148,
      "mean_token_accuracy": 0.21781099289655687,
      "num_tokens": 89675728.0,
      "step": 51695
    },
    {
      "entropy": 5.363828516006469,
      "epoch": 4.022408091810932,
      "grad_norm": 1.125,
      "learning_rate": 0.00034615578515783687,
      "loss": 4.7548,
      "mean_token_accuracy": 0.2261461451649666,
      "num_tokens": 89683392.0,
      "step": 51700
    },
    {
      "entropy": 5.3925196647644045,
      "epoch": 4.022797121182649,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034612949148930935,
      "loss": 4.7678,
      "mean_token_accuracy": 0.2247040182352066,
      "num_tokens": 89691541.0,
      "step": 51705
    },
    {
      "entropy": 5.3680929183959964,
      "epoch": 4.023186150554367,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000346103196741505,
      "loss": 4.7251,
      "mean_token_accuracy": 0.2261888012290001,
      "num_tokens": 89700445.0,
      "step": 51710
    },
    {
      "entropy": 5.372253704071045,
      "epoch": 4.023575179926085,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00034607690091482284,
      "loss": 4.7574,
      "mean_token_accuracy": 0.21877697110176086,
      "num_tokens": 89708902.0,
      "step": 51715
    },
    {
      "entropy": 5.305436992645264,
      "epoch": 4.023964209297802,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003460506040096619,
      "loss": 4.6567,
      "mean_token_accuracy": 0.2233501300215721,
      "num_tokens": 89716804.0,
      "step": 51720
    },
    {
      "entropy": 5.340112638473511,
      "epoch": 4.024353238669519,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003460243060264211,
      "loss": 4.763,
      "mean_token_accuracy": 0.22873058915138245,
      "num_tokens": 89726046.0,
      "step": 51725
    },
    {
      "entropy": 5.324047994613648,
      "epoch": 4.024742268041237,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034599800696549944,
      "loss": 4.7215,
      "mean_token_accuracy": 0.22683578580617905,
      "num_tokens": 89735106.0,
      "step": 51730
    },
    {
      "entropy": 5.37718710899353,
      "epoch": 4.025131297412955,
      "grad_norm": 1.203125,
      "learning_rate": 0.000345971706827296,
      "loss": 4.7737,
      "mean_token_accuracy": 0.21645358353853225,
      "num_tokens": 89744456.0,
      "step": 51735
    },
    {
      "entropy": 5.377193117141724,
      "epoch": 4.025520326784672,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034594540561221,
      "loss": 4.8117,
      "mean_token_accuracy": 0.220954991877079,
      "num_tokens": 89753179.0,
      "step": 51740
    },
    {
      "entropy": 5.339804601669312,
      "epoch": 4.02590935615639,
      "grad_norm": 1.125,
      "learning_rate": 0.00034591910332064027,
      "loss": 4.7814,
      "mean_token_accuracy": 0.22279227524995804,
      "num_tokens": 89762063.0,
      "step": 51745
    },
    {
      "entropy": 5.408343887329101,
      "epoch": 4.026298385528108,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003458927999529861,
      "loss": 4.831,
      "mean_token_accuracy": 0.21686457395553588,
      "num_tokens": 89770501.0,
      "step": 51750
    },
    {
      "entropy": 5.429909753799438,
      "epoch": 4.026687414899825,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034586649550964656,
      "loss": 4.7762,
      "mean_token_accuracy": 0.21429161876440048,
      "num_tokens": 89778921.0,
      "step": 51755
    },
    {
      "entropy": 5.431669473648071,
      "epoch": 4.027076444271542,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034584018999102073,
      "loss": 4.7999,
      "mean_token_accuracy": 0.21284226030111314,
      "num_tokens": 89786663.0,
      "step": 51760
    },
    {
      "entropy": 5.41167106628418,
      "epoch": 4.02746547364326,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034581388339750775,
      "loss": 4.8063,
      "mean_token_accuracy": 0.21779376417398452,
      "num_tokens": 89795614.0,
      "step": 51765
    },
    {
      "entropy": 5.333299350738526,
      "epoch": 4.027854503014978,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003457875757295067,
      "loss": 4.6423,
      "mean_token_accuracy": 0.22810730040073396,
      "num_tokens": 89804145.0,
      "step": 51770
    },
    {
      "entropy": 5.377498388290405,
      "epoch": 4.028243532386695,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034576126698741697,
      "loss": 4.739,
      "mean_token_accuracy": 0.21500207334756852,
      "num_tokens": 89812229.0,
      "step": 51775
    },
    {
      "entropy": 5.352486848831177,
      "epoch": 4.028632561758413,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003457349571716376,
      "loss": 4.7681,
      "mean_token_accuracy": 0.22426032721996308,
      "num_tokens": 89820665.0,
      "step": 51780
    },
    {
      "entropy": 5.344851875305176,
      "epoch": 4.02902159113013,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034570864628256795,
      "loss": 4.693,
      "mean_token_accuracy": 0.22825457155704498,
      "num_tokens": 89829219.0,
      "step": 51785
    },
    {
      "entropy": 5.333939695358277,
      "epoch": 4.029410620501848,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034568233432060705,
      "loss": 4.6858,
      "mean_token_accuracy": 0.22687558233737945,
      "num_tokens": 89837059.0,
      "step": 51790
    },
    {
      "entropy": 5.3606898307800295,
      "epoch": 4.029799649873565,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003456560212861543,
      "loss": 4.7914,
      "mean_token_accuracy": 0.2176296055316925,
      "num_tokens": 89845431.0,
      "step": 51795
    },
    {
      "entropy": 5.4825756549835205,
      "epoch": 4.030188679245283,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003456297071796087,
      "loss": 4.9273,
      "mean_token_accuracy": 0.21746211498975754,
      "num_tokens": 89854602.0,
      "step": 51800
    },
    {
      "entropy": 5.295718479156494,
      "epoch": 4.030577708617001,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034560339200136983,
      "loss": 4.648,
      "mean_token_accuracy": 0.2252985119819641,
      "num_tokens": 89863156.0,
      "step": 51805
    },
    {
      "entropy": 5.360506677627564,
      "epoch": 4.030966737988718,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034557707575183684,
      "loss": 4.758,
      "mean_token_accuracy": 0.22467788308858871,
      "num_tokens": 89872484.0,
      "step": 51810
    },
    {
      "entropy": 5.456721258163452,
      "epoch": 4.031355767360436,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034555075843140917,
      "loss": 4.7935,
      "mean_token_accuracy": 0.22436410039663315,
      "num_tokens": 89880955.0,
      "step": 51815
    },
    {
      "entropy": 5.400187301635742,
      "epoch": 4.031744796732153,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034552444004048587,
      "loss": 4.7822,
      "mean_token_accuracy": 0.21881779581308364,
      "num_tokens": 89889025.0,
      "step": 51820
    },
    {
      "entropy": 5.370055198669434,
      "epoch": 4.032133826103871,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034549812057946657,
      "loss": 4.7721,
      "mean_token_accuracy": 0.2208339214324951,
      "num_tokens": 89897319.0,
      "step": 51825
    },
    {
      "entropy": 5.323062372207642,
      "epoch": 4.032522855475588,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003454718000487505,
      "loss": 4.718,
      "mean_token_accuracy": 0.23115455806255342,
      "num_tokens": 89905821.0,
      "step": 51830
    },
    {
      "entropy": 5.426640844345092,
      "epoch": 4.032911884847306,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003454454784487369,
      "loss": 4.7817,
      "mean_token_accuracy": 0.22063681930303575,
      "num_tokens": 89913982.0,
      "step": 51835
    },
    {
      "entropy": 5.419381141662598,
      "epoch": 4.033300914219024,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003454191557798254,
      "loss": 4.7784,
      "mean_token_accuracy": 0.22214216738939285,
      "num_tokens": 89923237.0,
      "step": 51840
    },
    {
      "entropy": 5.390461397171021,
      "epoch": 4.033689943590741,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034539283204241525,
      "loss": 4.8247,
      "mean_token_accuracy": 0.218178454041481,
      "num_tokens": 89932836.0,
      "step": 51845
    },
    {
      "entropy": 5.399798345565796,
      "epoch": 4.034078972962459,
      "grad_norm": 1.125,
      "learning_rate": 0.00034536650723690596,
      "loss": 4.7204,
      "mean_token_accuracy": 0.22843819707632065,
      "num_tokens": 89941585.0,
      "step": 51850
    },
    {
      "entropy": 5.276509380340576,
      "epoch": 4.034468002334176,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034534018136369687,
      "loss": 4.6914,
      "mean_token_accuracy": 0.22913856208324432,
      "num_tokens": 89950561.0,
      "step": 51855
    },
    {
      "entropy": 5.333795595169067,
      "epoch": 4.034857031705894,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003453138544231875,
      "loss": 4.7551,
      "mean_token_accuracy": 0.22338641285896302,
      "num_tokens": 89959471.0,
      "step": 51860
    },
    {
      "entropy": 5.373469829559326,
      "epoch": 4.035246061077611,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003452875264157774,
      "loss": 4.7661,
      "mean_token_accuracy": 0.21219884157180785,
      "num_tokens": 89967394.0,
      "step": 51865
    },
    {
      "entropy": 5.357634162902832,
      "epoch": 4.035635090449329,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034526119734186586,
      "loss": 4.7573,
      "mean_token_accuracy": 0.2217332974076271,
      "num_tokens": 89975074.0,
      "step": 51870
    },
    {
      "entropy": 5.34039716720581,
      "epoch": 4.036024119821047,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034523486720185246,
      "loss": 4.7464,
      "mean_token_accuracy": 0.2247933715581894,
      "num_tokens": 89983353.0,
      "step": 51875
    },
    {
      "entropy": 5.352906131744385,
      "epoch": 4.036413149192764,
      "grad_norm": 1.25,
      "learning_rate": 0.00034520853599613675,
      "loss": 4.8253,
      "mean_token_accuracy": 0.2196851924061775,
      "num_tokens": 89991982.0,
      "step": 51880
    },
    {
      "entropy": 5.2764379501342775,
      "epoch": 4.036802178564481,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034518220372511835,
      "loss": 4.6973,
      "mean_token_accuracy": 0.2330554708838463,
      "num_tokens": 90000804.0,
      "step": 51885
    },
    {
      "entropy": 5.371105432510376,
      "epoch": 4.037191207936199,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003451558703891967,
      "loss": 4.7999,
      "mean_token_accuracy": 0.2268224135041237,
      "num_tokens": 90009500.0,
      "step": 51890
    },
    {
      "entropy": 5.4443882465362545,
      "epoch": 4.037580237307917,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003451295359887713,
      "loss": 4.7912,
      "mean_token_accuracy": 0.21527750343084334,
      "num_tokens": 90017753.0,
      "step": 51895
    },
    {
      "entropy": 5.282799100875854,
      "epoch": 4.037969266679634,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003451032005242418,
      "loss": 4.6201,
      "mean_token_accuracy": 0.2321004018187523,
      "num_tokens": 90027475.0,
      "step": 51900
    },
    {
      "entropy": 5.378875684738159,
      "epoch": 4.038358296051352,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034507686399600786,
      "loss": 4.8012,
      "mean_token_accuracy": 0.2131760910153389,
      "num_tokens": 90035510.0,
      "step": 51905
    },
    {
      "entropy": 5.3661887645721436,
      "epoch": 4.03874732542307,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034505052640446907,
      "loss": 4.745,
      "mean_token_accuracy": 0.22031114548444747,
      "num_tokens": 90043858.0,
      "step": 51910
    },
    {
      "entropy": 5.422360420227051,
      "epoch": 4.039136354794787,
      "grad_norm": 1.21875,
      "learning_rate": 0.000345024187750025,
      "loss": 4.9155,
      "mean_token_accuracy": 0.21772330701351167,
      "num_tokens": 90052361.0,
      "step": 51915
    },
    {
      "entropy": 5.372708034515381,
      "epoch": 4.039525384166504,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034499784803307534,
      "loss": 4.7604,
      "mean_token_accuracy": 0.22407192289829253,
      "num_tokens": 90060856.0,
      "step": 51920
    },
    {
      "entropy": 5.409184503555298,
      "epoch": 4.039914413538222,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003449715072540198,
      "loss": 4.7884,
      "mean_token_accuracy": 0.2158753827214241,
      "num_tokens": 90069452.0,
      "step": 51925
    },
    {
      "entropy": 5.34556212425232,
      "epoch": 4.04030344290994,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034494516541325795,
      "loss": 4.7443,
      "mean_token_accuracy": 0.2192401483654976,
      "num_tokens": 90078077.0,
      "step": 51930
    },
    {
      "entropy": 5.3824131965637205,
      "epoch": 4.040692472281657,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003449188225111895,
      "loss": 4.8016,
      "mean_token_accuracy": 0.21611636132001877,
      "num_tokens": 90087492.0,
      "step": 51935
    },
    {
      "entropy": 5.353448343276978,
      "epoch": 4.041081501653375,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034489247854821426,
      "loss": 4.7526,
      "mean_token_accuracy": 0.225431627035141,
      "num_tokens": 90095988.0,
      "step": 51940
    },
    {
      "entropy": 5.342483568191528,
      "epoch": 4.041470531025093,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034486613352473195,
      "loss": 4.8083,
      "mean_token_accuracy": 0.23163609057664872,
      "num_tokens": 90105135.0,
      "step": 51945
    },
    {
      "entropy": 5.344291925430298,
      "epoch": 4.04185956039681,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034483978744114224,
      "loss": 4.7672,
      "mean_token_accuracy": 0.21563464254140854,
      "num_tokens": 90113800.0,
      "step": 51950
    },
    {
      "entropy": 5.428956460952759,
      "epoch": 4.042248589768527,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003448134402978449,
      "loss": 4.8313,
      "mean_token_accuracy": 0.22251420170068742,
      "num_tokens": 90121828.0,
      "step": 51955
    },
    {
      "entropy": 5.4242266654968265,
      "epoch": 4.042637619140245,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034478709209523974,
      "loss": 4.7672,
      "mean_token_accuracy": 0.22554436773061753,
      "num_tokens": 90129688.0,
      "step": 51960
    },
    {
      "entropy": 5.393194580078125,
      "epoch": 4.043026648511963,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003447607428337265,
      "loss": 4.7746,
      "mean_token_accuracy": 0.21797650009393693,
      "num_tokens": 90138282.0,
      "step": 51965
    },
    {
      "entropy": 5.401811408996582,
      "epoch": 4.04341567788368,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034473439251370505,
      "loss": 4.8303,
      "mean_token_accuracy": 0.22346438318490983,
      "num_tokens": 90147356.0,
      "step": 51970
    },
    {
      "entropy": 5.399550580978394,
      "epoch": 4.043804707255398,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003447080411355752,
      "loss": 4.8585,
      "mean_token_accuracy": 0.21167974472045897,
      "num_tokens": 90156135.0,
      "step": 51975
    },
    {
      "entropy": 5.394518232345581,
      "epoch": 4.044193736627116,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034468168869973673,
      "loss": 4.8047,
      "mean_token_accuracy": 0.21095367819070815,
      "num_tokens": 90164691.0,
      "step": 51980
    },
    {
      "entropy": 5.352939939498901,
      "epoch": 4.0445827659988325,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034465533520658967,
      "loss": 4.7561,
      "mean_token_accuracy": 0.2213965818285942,
      "num_tokens": 90173258.0,
      "step": 51985
    },
    {
      "entropy": 5.370508098602295,
      "epoch": 4.04497179537055,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034462898065653366,
      "loss": 4.7769,
      "mean_token_accuracy": 0.22174972742795945,
      "num_tokens": 90182228.0,
      "step": 51990
    },
    {
      "entropy": 5.365706586837769,
      "epoch": 4.045360824742268,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003446026250499687,
      "loss": 4.7344,
      "mean_token_accuracy": 0.22278038859367372,
      "num_tokens": 90191046.0,
      "step": 51995
    },
    {
      "entropy": 5.374742317199707,
      "epoch": 4.045749854113986,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003445762683872946,
      "loss": 4.7289,
      "mean_token_accuracy": 0.2245160937309265,
      "num_tokens": 90200201.0,
      "step": 52000
    },
    {
      "entropy": 5.435004615783692,
      "epoch": 4.046138883485703,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003445499106689115,
      "loss": 4.8033,
      "mean_token_accuracy": 0.21891851276159285,
      "num_tokens": 90209817.0,
      "step": 52005
    },
    {
      "entropy": 5.3787538528442385,
      "epoch": 4.046527912857421,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00034452355189521915,
      "loss": 4.8306,
      "mean_token_accuracy": 0.21442177295684814,
      "num_tokens": 90219560.0,
      "step": 52010
    },
    {
      "entropy": 5.410332155227661,
      "epoch": 4.046916942229139,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003444971920666176,
      "loss": 4.8281,
      "mean_token_accuracy": 0.21440242379903793,
      "num_tokens": 90228549.0,
      "step": 52015
    },
    {
      "entropy": 5.389546060562134,
      "epoch": 4.0473059716008555,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003444708311835068,
      "loss": 4.766,
      "mean_token_accuracy": 0.21882151961326599,
      "num_tokens": 90237415.0,
      "step": 52020
    },
    {
      "entropy": 5.278520250320435,
      "epoch": 4.047695000972573,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034444446924628664,
      "loss": 4.6638,
      "mean_token_accuracy": 0.23016985058784484,
      "num_tokens": 90246546.0,
      "step": 52025
    },
    {
      "entropy": 5.330224514007568,
      "epoch": 4.048084030344291,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034441810625535725,
      "loss": 4.8432,
      "mean_token_accuracy": 0.211902217566967,
      "num_tokens": 90254657.0,
      "step": 52030
    },
    {
      "entropy": 5.352524185180664,
      "epoch": 4.048473059716009,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003443917422111185,
      "loss": 4.7261,
      "mean_token_accuracy": 0.22042022943496703,
      "num_tokens": 90263760.0,
      "step": 52035
    },
    {
      "entropy": 5.380495691299439,
      "epoch": 4.048862089087726,
      "grad_norm": 1.125,
      "learning_rate": 0.00034436537711397054,
      "loss": 4.7062,
      "mean_token_accuracy": 0.23094047158956527,
      "num_tokens": 90272483.0,
      "step": 52040
    },
    {
      "entropy": 5.2680768966674805,
      "epoch": 4.049251118459444,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003443390109643134,
      "loss": 4.6678,
      "mean_token_accuracy": 0.22946900874376297,
      "num_tokens": 90280789.0,
      "step": 52045
    },
    {
      "entropy": 5.279154491424561,
      "epoch": 4.049640147831162,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003443126437625472,
      "loss": 4.7905,
      "mean_token_accuracy": 0.2201910838484764,
      "num_tokens": 90289511.0,
      "step": 52050
    },
    {
      "entropy": 5.324139785766602,
      "epoch": 4.0500291772028785,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003442862755090719,
      "loss": 4.7942,
      "mean_token_accuracy": 0.21852817982435227,
      "num_tokens": 90298716.0,
      "step": 52055
    },
    {
      "entropy": 5.397677564620972,
      "epoch": 4.050418206574596,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034425990620428763,
      "loss": 4.7247,
      "mean_token_accuracy": 0.22933153808116913,
      "num_tokens": 90307072.0,
      "step": 52060
    },
    {
      "entropy": 5.374121379852295,
      "epoch": 4.050807235946314,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003442335358485946,
      "loss": 4.7779,
      "mean_token_accuracy": 0.22530280202627181,
      "num_tokens": 90315388.0,
      "step": 52065
    },
    {
      "entropy": 5.418425369262695,
      "epoch": 4.051196265318032,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003442071644423928,
      "loss": 4.8146,
      "mean_token_accuracy": 0.21885016113519667,
      "num_tokens": 90323858.0,
      "step": 52070
    },
    {
      "entropy": 5.371873235702514,
      "epoch": 4.051585294689749,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003441807919860824,
      "loss": 4.8161,
      "mean_token_accuracy": 0.2153049737215042,
      "num_tokens": 90332425.0,
      "step": 52075
    },
    {
      "entropy": 5.423493957519531,
      "epoch": 4.051974324061467,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034415441848006364,
      "loss": 4.816,
      "mean_token_accuracy": 0.2210131138563156,
      "num_tokens": 90340982.0,
      "step": 52080
    },
    {
      "entropy": 5.348751974105835,
      "epoch": 4.052363353433184,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034412804392473665,
      "loss": 4.72,
      "mean_token_accuracy": 0.224295973777771,
      "num_tokens": 90348876.0,
      "step": 52085
    },
    {
      "entropy": 5.307561922073364,
      "epoch": 4.0527523828049015,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003441016683205016,
      "loss": 4.6865,
      "mean_token_accuracy": 0.22735846489667894,
      "num_tokens": 90357531.0,
      "step": 52090
    },
    {
      "entropy": 5.412411785125732,
      "epoch": 4.053141412176619,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034407529166775874,
      "loss": 4.8784,
      "mean_token_accuracy": 0.2177929922938347,
      "num_tokens": 90365542.0,
      "step": 52095
    },
    {
      "entropy": 5.419761800765992,
      "epoch": 4.053530441548337,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003440489139669083,
      "loss": 4.7881,
      "mean_token_accuracy": 0.22033295035362244,
      "num_tokens": 90374295.0,
      "step": 52100
    },
    {
      "entropy": 5.401559162139892,
      "epoch": 4.053919470920055,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034402253521835036,
      "loss": 4.7802,
      "mean_token_accuracy": 0.21651293188333512,
      "num_tokens": 90382605.0,
      "step": 52105
    },
    {
      "entropy": 5.3376569747924805,
      "epoch": 4.054308500291772,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003439961554224855,
      "loss": 4.7504,
      "mean_token_accuracy": 0.21553211361169816,
      "num_tokens": 90391220.0,
      "step": 52110
    },
    {
      "entropy": 5.328606653213501,
      "epoch": 4.05469752966349,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003439697745797137,
      "loss": 4.7286,
      "mean_token_accuracy": 0.2215635061264038,
      "num_tokens": 90398986.0,
      "step": 52115
    },
    {
      "entropy": 5.3276190757751465,
      "epoch": 4.055086559035207,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034394339269043533,
      "loss": 4.7711,
      "mean_token_accuracy": 0.21543273627758025,
      "num_tokens": 90407671.0,
      "step": 52120
    },
    {
      "entropy": 5.425623989105224,
      "epoch": 4.0554755884069245,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003439170097550507,
      "loss": 4.8233,
      "mean_token_accuracy": 0.21875323951244355,
      "num_tokens": 90416184.0,
      "step": 52125
    },
    {
      "entropy": 5.393656826019287,
      "epoch": 4.055864617778642,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003438906257739602,
      "loss": 4.7627,
      "mean_token_accuracy": 0.2208339527249336,
      "num_tokens": 90424992.0,
      "step": 52130
    },
    {
      "entropy": 5.330661869049072,
      "epoch": 4.05625364715036,
      "grad_norm": 1.125,
      "learning_rate": 0.0003438642407475641,
      "loss": 4.751,
      "mean_token_accuracy": 0.22340321242809297,
      "num_tokens": 90433802.0,
      "step": 52135
    },
    {
      "entropy": 5.35253119468689,
      "epoch": 4.0566426765220776,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003438378546762627,
      "loss": 4.8403,
      "mean_token_accuracy": 0.22153252959251404,
      "num_tokens": 90441831.0,
      "step": 52140
    },
    {
      "entropy": 5.432820081710815,
      "epoch": 4.057031705893795,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003438114675604565,
      "loss": 4.7716,
      "mean_token_accuracy": 0.21511563062667846,
      "num_tokens": 90450327.0,
      "step": 52145
    },
    {
      "entropy": 5.429895305633545,
      "epoch": 4.057420735265513,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003437850794005457,
      "loss": 4.7517,
      "mean_token_accuracy": 0.21933069527149202,
      "num_tokens": 90459012.0,
      "step": 52150
    },
    {
      "entropy": 5.388191986083984,
      "epoch": 4.05780976463723,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003437586901969309,
      "loss": 4.7638,
      "mean_token_accuracy": 0.22187469899654388,
      "num_tokens": 90467296.0,
      "step": 52155
    },
    {
      "entropy": 5.310207080841065,
      "epoch": 4.0581987940089475,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003437322999500124,
      "loss": 4.7904,
      "mean_token_accuracy": 0.22073110342025756,
      "num_tokens": 90475953.0,
      "step": 52160
    },
    {
      "entropy": 5.376143503189087,
      "epoch": 4.058587823380665,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034370590866019063,
      "loss": 4.7787,
      "mean_token_accuracy": 0.2197295218706131,
      "num_tokens": 90483562.0,
      "step": 52165
    },
    {
      "entropy": 5.386682224273682,
      "epoch": 4.058976852752383,
      "grad_norm": 1.1875,
      "learning_rate": 0.000343679516327866,
      "loss": 4.8061,
      "mean_token_accuracy": 0.21905774623155594,
      "num_tokens": 90491479.0,
      "step": 52170
    },
    {
      "entropy": 5.352588415145874,
      "epoch": 4.0593658821241005,
      "grad_norm": 1.125,
      "learning_rate": 0.0003436531229534391,
      "loss": 4.7613,
      "mean_token_accuracy": 0.2201991781592369,
      "num_tokens": 90499674.0,
      "step": 52175
    },
    {
      "entropy": 5.270554113388061,
      "epoch": 4.059754911495818,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003436267285373103,
      "loss": 4.6855,
      "mean_token_accuracy": 0.22640990167856218,
      "num_tokens": 90508453.0,
      "step": 52180
    },
    {
      "entropy": 5.3475282192230225,
      "epoch": 4.060143940867536,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034360033307988014,
      "loss": 4.7079,
      "mean_token_accuracy": 0.2266128733754158,
      "num_tokens": 90516698.0,
      "step": 52185
    },
    {
      "entropy": 5.358474397659302,
      "epoch": 4.060532970239253,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034357393658154914,
      "loss": 4.7585,
      "mean_token_accuracy": 0.21921961307525634,
      "num_tokens": 90525260.0,
      "step": 52190
    },
    {
      "entropy": 5.394729471206665,
      "epoch": 4.0609219996109704,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003435475390427178,
      "loss": 4.8099,
      "mean_token_accuracy": 0.22491785287857055,
      "num_tokens": 90534468.0,
      "step": 52195
    },
    {
      "entropy": 5.300397253036499,
      "epoch": 4.061311028982688,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034352114046378655,
      "loss": 4.698,
      "mean_token_accuracy": 0.22872281819581985,
      "num_tokens": 90543559.0,
      "step": 52200
    },
    {
      "entropy": 5.439226150512695,
      "epoch": 4.061700058354406,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034349474084515607,
      "loss": 4.8764,
      "mean_token_accuracy": 0.21539284586906432,
      "num_tokens": 90552161.0,
      "step": 52205
    },
    {
      "entropy": 5.408782911300659,
      "epoch": 4.0620890877261235,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034346834018722706,
      "loss": 4.8167,
      "mean_token_accuracy": 0.21707588732242583,
      "num_tokens": 90560175.0,
      "step": 52210
    },
    {
      "entropy": 5.357005310058594,
      "epoch": 4.062478117097841,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034344193849039997,
      "loss": 4.7726,
      "mean_token_accuracy": 0.2200980708003044,
      "num_tokens": 90569791.0,
      "step": 52215
    },
    {
      "entropy": 5.315792036056519,
      "epoch": 4.062867146469558,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003434155357550753,
      "loss": 4.7357,
      "mean_token_accuracy": 0.21849081963300704,
      "num_tokens": 90578407.0,
      "step": 52220
    },
    {
      "entropy": 5.37729549407959,
      "epoch": 4.063256175841276,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034338913198165373,
      "loss": 4.7488,
      "mean_token_accuracy": 0.21839040368795395,
      "num_tokens": 90586705.0,
      "step": 52225
    },
    {
      "entropy": 5.375359725952149,
      "epoch": 4.063645205212993,
      "grad_norm": 1.171875,
      "learning_rate": 0.000343362727170536,
      "loss": 4.7409,
      "mean_token_accuracy": 0.22002536803483963,
      "num_tokens": 90595766.0,
      "step": 52230
    },
    {
      "entropy": 5.411321210861206,
      "epoch": 4.064034234584711,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003433363213221227,
      "loss": 4.791,
      "mean_token_accuracy": 0.21863074898719786,
      "num_tokens": 90604550.0,
      "step": 52235
    },
    {
      "entropy": 5.475142478942871,
      "epoch": 4.064423263956429,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034330991443681444,
      "loss": 4.8815,
      "mean_token_accuracy": 0.21274033635854722,
      "num_tokens": 90613070.0,
      "step": 52240
    },
    {
      "entropy": 5.409522628784179,
      "epoch": 4.0648122933281465,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000343283506515012,
      "loss": 4.8252,
      "mean_token_accuracy": 0.21623256504535676,
      "num_tokens": 90623073.0,
      "step": 52245
    },
    {
      "entropy": 5.384593105316162,
      "epoch": 4.065201322699864,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034325709755711606,
      "loss": 4.7948,
      "mean_token_accuracy": 0.22112827003002167,
      "num_tokens": 90631944.0,
      "step": 52250
    },
    {
      "entropy": 5.312540864944458,
      "epoch": 4.065590352071581,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034323068756352725,
      "loss": 4.6653,
      "mean_token_accuracy": 0.22913322001695632,
      "num_tokens": 90641007.0,
      "step": 52255
    },
    {
      "entropy": 5.4324099063873295,
      "epoch": 4.065979381443299,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003432042765346464,
      "loss": 4.8906,
      "mean_token_accuracy": 0.20163841545581818,
      "num_tokens": 90649628.0,
      "step": 52260
    },
    {
      "entropy": 5.344708204269409,
      "epoch": 4.066368410815016,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003431778644708742,
      "loss": 4.7584,
      "mean_token_accuracy": 0.2204756811261177,
      "num_tokens": 90658301.0,
      "step": 52265
    },
    {
      "entropy": 5.344548940658569,
      "epoch": 4.066757440186734,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003431514513726114,
      "loss": 4.7804,
      "mean_token_accuracy": 0.22245070934295655,
      "num_tokens": 90667080.0,
      "step": 52270
    },
    {
      "entropy": 5.3291051387786865,
      "epoch": 4.067146469558452,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003431250372402588,
      "loss": 4.7805,
      "mean_token_accuracy": 0.2230556935071945,
      "num_tokens": 90675658.0,
      "step": 52275
    },
    {
      "entropy": 5.3004528999328615,
      "epoch": 4.0675354989301695,
      "grad_norm": 1.125,
      "learning_rate": 0.00034309862207421724,
      "loss": 4.7251,
      "mean_token_accuracy": 0.22948587089776992,
      "num_tokens": 90684345.0,
      "step": 52280
    },
    {
      "entropy": 5.330675983428955,
      "epoch": 4.067924528301887,
      "grad_norm": 1.328125,
      "learning_rate": 0.00034307220587488743,
      "loss": 4.7687,
      "mean_token_accuracy": 0.21936175376176834,
      "num_tokens": 90693342.0,
      "step": 52285
    },
    {
      "entropy": 5.348571157455444,
      "epoch": 4.068313557673604,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034304578864267026,
      "loss": 4.6922,
      "mean_token_accuracy": 0.22144949436187744,
      "num_tokens": 90701948.0,
      "step": 52290
    },
    {
      "entropy": 5.356941604614258,
      "epoch": 4.068702587045322,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034301937037796654,
      "loss": 4.7113,
      "mean_token_accuracy": 0.22795185148715974,
      "num_tokens": 90710273.0,
      "step": 52295
    },
    {
      "entropy": 5.2557933807373045,
      "epoch": 4.069091616417039,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034299295108117716,
      "loss": 4.7266,
      "mean_token_accuracy": 0.22502560913562775,
      "num_tokens": 90718993.0,
      "step": 52300
    },
    {
      "entropy": 5.340765380859375,
      "epoch": 4.069480645788757,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034296653075270296,
      "loss": 4.7707,
      "mean_token_accuracy": 0.22065045386552812,
      "num_tokens": 90727655.0,
      "step": 52305
    },
    {
      "entropy": 5.355353498458863,
      "epoch": 4.069869675160475,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034294010939294486,
      "loss": 4.7511,
      "mean_token_accuracy": 0.22449631243944168,
      "num_tokens": 90735512.0,
      "step": 52310
    },
    {
      "entropy": 5.328531408309937,
      "epoch": 4.0702587045321925,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034291368700230376,
      "loss": 4.6929,
      "mean_token_accuracy": 0.23198579698801042,
      "num_tokens": 90743858.0,
      "step": 52315
    },
    {
      "entropy": 5.435466384887695,
      "epoch": 4.070647733903909,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003428872635811804,
      "loss": 4.837,
      "mean_token_accuracy": 0.20896072685718536,
      "num_tokens": 90752194.0,
      "step": 52320
    },
    {
      "entropy": 5.345551061630249,
      "epoch": 4.071036763275627,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000342860839129976,
      "loss": 4.8004,
      "mean_token_accuracy": 0.21859817653894426,
      "num_tokens": 90761213.0,
      "step": 52325
    },
    {
      "entropy": 5.322601318359375,
      "epoch": 4.071425792647345,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003428344136490914,
      "loss": 4.7177,
      "mean_token_accuracy": 0.2264167070388794,
      "num_tokens": 90770634.0,
      "step": 52330
    },
    {
      "entropy": 5.380029678344727,
      "epoch": 4.071814822019062,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034280798713892744,
      "loss": 4.777,
      "mean_token_accuracy": 0.2196333095431328,
      "num_tokens": 90779861.0,
      "step": 52335
    },
    {
      "entropy": 5.300431442260742,
      "epoch": 4.07220385139078,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003427815595998853,
      "loss": 4.667,
      "mean_token_accuracy": 0.2213229089975357,
      "num_tokens": 90788773.0,
      "step": 52340
    },
    {
      "entropy": 5.371387481689453,
      "epoch": 4.072592880762498,
      "grad_norm": 1.125,
      "learning_rate": 0.0003427551310323658,
      "loss": 4.8542,
      "mean_token_accuracy": 0.21457189321517944,
      "num_tokens": 90797372.0,
      "step": 52345
    },
    {
      "entropy": 5.396743726730347,
      "epoch": 4.0729819101342155,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034272870143677015,
      "loss": 4.787,
      "mean_token_accuracy": 0.21719556748867036,
      "num_tokens": 90806065.0,
      "step": 52350
    },
    {
      "entropy": 5.430783033370972,
      "epoch": 4.073370939505932,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034270227081349913,
      "loss": 4.8275,
      "mean_token_accuracy": 0.22698843777179717,
      "num_tokens": 90815531.0,
      "step": 52355
    },
    {
      "entropy": 5.491317224502564,
      "epoch": 4.07375996887765,
      "grad_norm": 1.171875,
      "learning_rate": 0.000342675839162954,
      "loss": 4.9478,
      "mean_token_accuracy": 0.2141076698899269,
      "num_tokens": 90823416.0,
      "step": 52360
    },
    {
      "entropy": 5.445312213897705,
      "epoch": 4.074148998249368,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034264940648553565,
      "loss": 4.8353,
      "mean_token_accuracy": 0.21536418199539184,
      "num_tokens": 90832441.0,
      "step": 52365
    },
    {
      "entropy": 5.267932319641114,
      "epoch": 4.074538027621085,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003426229727816453,
      "loss": 4.6378,
      "mean_token_accuracy": 0.2348220467567444,
      "num_tokens": 90840453.0,
      "step": 52370
    },
    {
      "entropy": 5.374851131439209,
      "epoch": 4.074927056992803,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003425965380516839,
      "loss": 4.7752,
      "mean_token_accuracy": 0.22357335835695266,
      "num_tokens": 90848802.0,
      "step": 52375
    },
    {
      "entropy": 5.33278226852417,
      "epoch": 4.075316086364521,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003425701022960527,
      "loss": 4.9172,
      "mean_token_accuracy": 0.21150824427604675,
      "num_tokens": 90856950.0,
      "step": 52380
    },
    {
      "entropy": 5.408066129684448,
      "epoch": 4.0757051157362385,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034254366551515276,
      "loss": 4.778,
      "mean_token_accuracy": 0.2194082260131836,
      "num_tokens": 90866100.0,
      "step": 52385
    },
    {
      "entropy": 5.438610649108886,
      "epoch": 4.076094145107955,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003425172277093852,
      "loss": 4.8348,
      "mean_token_accuracy": 0.21268681287765503,
      "num_tokens": 90875323.0,
      "step": 52390
    },
    {
      "entropy": 5.371421813964844,
      "epoch": 4.076483174479673,
      "grad_norm": 1.046875,
      "learning_rate": 0.000342490788879151,
      "loss": 4.7799,
      "mean_token_accuracy": 0.22356570959091188,
      "num_tokens": 90884287.0,
      "step": 52395
    },
    {
      "entropy": 5.38052773475647,
      "epoch": 4.076872203851391,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034246434902485156,
      "loss": 4.7975,
      "mean_token_accuracy": 0.22307640165090561,
      "num_tokens": 90893311.0,
      "step": 52400
    },
    {
      "entropy": 5.344840240478516,
      "epoch": 4.077261233223108,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034243790814688816,
      "loss": 4.746,
      "mean_token_accuracy": 0.22306982427835464,
      "num_tokens": 90901977.0,
      "step": 52405
    },
    {
      "entropy": 5.308395004272461,
      "epoch": 4.077650262594826,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034241146624566175,
      "loss": 4.7537,
      "mean_token_accuracy": 0.2193320170044899,
      "num_tokens": 90910739.0,
      "step": 52410
    },
    {
      "entropy": 5.483261489868164,
      "epoch": 4.078039291966544,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003423850233215737,
      "loss": 4.946,
      "mean_token_accuracy": 0.20557509809732438,
      "num_tokens": 90920329.0,
      "step": 52415
    },
    {
      "entropy": 5.272234201431274,
      "epoch": 4.078428321338261,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003423585793750251,
      "loss": 4.6727,
      "mean_token_accuracy": 0.2269975021481514,
      "num_tokens": 90928697.0,
      "step": 52420
    },
    {
      "entropy": 5.417020320892334,
      "epoch": 4.078817350709978,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034233213440641724,
      "loss": 4.8119,
      "mean_token_accuracy": 0.22041169852018355,
      "num_tokens": 90937483.0,
      "step": 52425
    },
    {
      "entropy": 5.394551563262939,
      "epoch": 4.079206380081696,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034230568841615145,
      "loss": 4.8406,
      "mean_token_accuracy": 0.21279986053705216,
      "num_tokens": 90946467.0,
      "step": 52430
    },
    {
      "entropy": 5.407749223709106,
      "epoch": 4.079595409453414,
      "grad_norm": 1.234375,
      "learning_rate": 0.000342279241404629,
      "loss": 4.7601,
      "mean_token_accuracy": 0.21768074631690978,
      "num_tokens": 90954498.0,
      "step": 52435
    },
    {
      "entropy": 5.437407302856445,
      "epoch": 4.079984438825131,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034225279337225106,
      "loss": 4.865,
      "mean_token_accuracy": 0.211087167263031,
      "num_tokens": 90962683.0,
      "step": 52440
    },
    {
      "entropy": 5.305998373031616,
      "epoch": 4.080373468196849,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003422263443194191,
      "loss": 4.7013,
      "mean_token_accuracy": 0.23345850706100463,
      "num_tokens": 90971083.0,
      "step": 52445
    },
    {
      "entropy": 5.330578994750977,
      "epoch": 4.080762497568567,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003421998942465344,
      "loss": 4.8039,
      "mean_token_accuracy": 0.22398509979248046,
      "num_tokens": 90980369.0,
      "step": 52450
    },
    {
      "entropy": 5.2651355266571045,
      "epoch": 4.081151526940284,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034217344315399823,
      "loss": 4.6888,
      "mean_token_accuracy": 0.2270424857735634,
      "num_tokens": 90988833.0,
      "step": 52455
    },
    {
      "entropy": 5.4009312152862545,
      "epoch": 4.081540556312001,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000342146991042212,
      "loss": 4.8323,
      "mean_token_accuracy": 0.21593854576349258,
      "num_tokens": 90997281.0,
      "step": 52460
    },
    {
      "entropy": 5.453297853469849,
      "epoch": 4.081929585683719,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034212053791157695,
      "loss": 4.8856,
      "mean_token_accuracy": 0.21697194129228592,
      "num_tokens": 91006471.0,
      "step": 52465
    },
    {
      "entropy": 5.472080135345459,
      "epoch": 4.082318615055437,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034209408376249464,
      "loss": 4.7784,
      "mean_token_accuracy": 0.2218833714723587,
      "num_tokens": 91016090.0,
      "step": 52470
    },
    {
      "entropy": 5.295069837570191,
      "epoch": 4.082707644427154,
      "grad_norm": 1.125,
      "learning_rate": 0.0003420676285953664,
      "loss": 4.7043,
      "mean_token_accuracy": 0.22152167111635207,
      "num_tokens": 91025778.0,
      "step": 52475
    },
    {
      "entropy": 5.37457857131958,
      "epoch": 4.083096673798872,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003420411724105937,
      "loss": 4.8179,
      "mean_token_accuracy": 0.21502785980701447,
      "num_tokens": 91033401.0,
      "step": 52480
    },
    {
      "entropy": 5.330930137634278,
      "epoch": 4.08348570317059,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034201471520857793,
      "loss": 4.711,
      "mean_token_accuracy": 0.2322893038392067,
      "num_tokens": 91042792.0,
      "step": 52485
    },
    {
      "entropy": 5.460653066635132,
      "epoch": 4.083874732542307,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003419882569897204,
      "loss": 4.8454,
      "mean_token_accuracy": 0.21239145547151567,
      "num_tokens": 91052060.0,
      "step": 52490
    },
    {
      "entropy": 5.326344108581543,
      "epoch": 4.084263761914024,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034196179775442274,
      "loss": 4.7696,
      "mean_token_accuracy": 0.2205837771296501,
      "num_tokens": 91060676.0,
      "step": 52495
    },
    {
      "entropy": 5.362199783325195,
      "epoch": 4.084652791285742,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003419353375030864,
      "loss": 4.7892,
      "mean_token_accuracy": 0.21621714234352113,
      "num_tokens": 91068986.0,
      "step": 52500
    },
    {
      "entropy": 5.402423095703125,
      "epoch": 4.08504182065746,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003419088762361128,
      "loss": 4.7619,
      "mean_token_accuracy": 0.2265496239066124,
      "num_tokens": 91077045.0,
      "step": 52505
    },
    {
      "entropy": 5.382566118240357,
      "epoch": 4.085430850029177,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003418824139539035,
      "loss": 4.8393,
      "mean_token_accuracy": 0.2142954096198082,
      "num_tokens": 91085352.0,
      "step": 52510
    },
    {
      "entropy": 5.373075199127197,
      "epoch": 4.085819879400895,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003418559506568601,
      "loss": 4.7863,
      "mean_token_accuracy": 0.22040842175483705,
      "num_tokens": 91094369.0,
      "step": 52515
    },
    {
      "entropy": 5.443074703216553,
      "epoch": 4.086208908772612,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034182948634538403,
      "loss": 4.8128,
      "mean_token_accuracy": 0.227288319170475,
      "num_tokens": 91102452.0,
      "step": 52520
    },
    {
      "entropy": 5.430651569366455,
      "epoch": 4.08659793814433,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003418030210198769,
      "loss": 4.8689,
      "mean_token_accuracy": 0.21718371361494065,
      "num_tokens": 91111202.0,
      "step": 52525
    },
    {
      "entropy": 5.419070148468018,
      "epoch": 4.086986967516047,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034177655468074027,
      "loss": 4.7991,
      "mean_token_accuracy": 0.23089154958724975,
      "num_tokens": 91119245.0,
      "step": 52530
    },
    {
      "entropy": 5.372449445724487,
      "epoch": 4.087375996887765,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003417500873283756,
      "loss": 4.8311,
      "mean_token_accuracy": 0.2131991684436798,
      "num_tokens": 91128504.0,
      "step": 52535
    },
    {
      "entropy": 5.318690061569214,
      "epoch": 4.087765026259483,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003417236189631846,
      "loss": 4.7833,
      "mean_token_accuracy": 0.22270189523696898,
      "num_tokens": 91137484.0,
      "step": 52540
    },
    {
      "entropy": 5.385479879379273,
      "epoch": 4.0881540556312,
      "grad_norm": 1.140625,
      "learning_rate": 0.000341697149585569,
      "loss": 4.7931,
      "mean_token_accuracy": 0.21593436151742934,
      "num_tokens": 91146318.0,
      "step": 52545
    },
    {
      "entropy": 5.429203224182129,
      "epoch": 4.088543085002918,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003416706791959302,
      "loss": 4.8558,
      "mean_token_accuracy": 0.2158903583884239,
      "num_tokens": 91153813.0,
      "step": 52550
    },
    {
      "entropy": 5.398557519912719,
      "epoch": 4.088932114374635,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034164420779467003,
      "loss": 4.8257,
      "mean_token_accuracy": 0.22059576362371444,
      "num_tokens": 91162975.0,
      "step": 52555
    },
    {
      "entropy": 5.262619590759277,
      "epoch": 4.089321143746353,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003416177353821901,
      "loss": 4.622,
      "mean_token_accuracy": 0.23538741171360017,
      "num_tokens": 91171489.0,
      "step": 52560
    },
    {
      "entropy": 5.2323156833648685,
      "epoch": 4.08971017311807,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000341591261958892,
      "loss": 4.7674,
      "mean_token_accuracy": 0.21679968386888504,
      "num_tokens": 91180199.0,
      "step": 52565
    },
    {
      "entropy": 5.481382989883423,
      "epoch": 4.090099202489788,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034156478752517754,
      "loss": 4.8681,
      "mean_token_accuracy": 0.2065044417977333,
      "num_tokens": 91189525.0,
      "step": 52570
    },
    {
      "entropy": 5.477007675170898,
      "epoch": 4.090488231861506,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034153831208144837,
      "loss": 4.8746,
      "mean_token_accuracy": 0.21310272812843323,
      "num_tokens": 91198324.0,
      "step": 52575
    },
    {
      "entropy": 5.33432903289795,
      "epoch": 4.090877261233223,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003415118356281062,
      "loss": 4.699,
      "mean_token_accuracy": 0.22883144170045852,
      "num_tokens": 91206988.0,
      "step": 52580
    },
    {
      "entropy": 5.275058698654175,
      "epoch": 4.091266290604941,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003414853581655528,
      "loss": 4.7462,
      "mean_token_accuracy": 0.22622264921665192,
      "num_tokens": 91215646.0,
      "step": 52585
    },
    {
      "entropy": 5.4465028762817385,
      "epoch": 4.091655319976658,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034145887969419,
      "loss": 4.8883,
      "mean_token_accuracy": 0.217190720140934,
      "num_tokens": 91224816.0,
      "step": 52590
    },
    {
      "entropy": 5.471168279647827,
      "epoch": 4.092044349348376,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034143240021441935,
      "loss": 4.8469,
      "mean_token_accuracy": 0.22079011797904968,
      "num_tokens": 91232962.0,
      "step": 52595
    },
    {
      "entropy": 5.428801393508911,
      "epoch": 4.092433378720093,
      "grad_norm": 1.125,
      "learning_rate": 0.0003414059197266428,
      "loss": 4.747,
      "mean_token_accuracy": 0.22090332955121994,
      "num_tokens": 91242508.0,
      "step": 52600
    },
    {
      "entropy": 5.371498870849609,
      "epoch": 4.092822408091811,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034137943823126215,
      "loss": 4.7702,
      "mean_token_accuracy": 0.21542719304561614,
      "num_tokens": 91250632.0,
      "step": 52605
    },
    {
      "entropy": 5.372362756729126,
      "epoch": 4.093211437463529,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003413529557286792,
      "loss": 4.8106,
      "mean_token_accuracy": 0.21525839865207672,
      "num_tokens": 91259835.0,
      "step": 52610
    },
    {
      "entropy": 5.336719369888305,
      "epoch": 4.093600466835246,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003413264722192957,
      "loss": 4.7208,
      "mean_token_accuracy": 0.224212945997715,
      "num_tokens": 91267999.0,
      "step": 52615
    },
    {
      "entropy": 5.338137769699097,
      "epoch": 4.093989496206964,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003412999877035136,
      "loss": 4.7428,
      "mean_token_accuracy": 0.22048081755638121,
      "num_tokens": 91276695.0,
      "step": 52620
    },
    {
      "entropy": 5.3595024108886715,
      "epoch": 4.094378525578681,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034127350218173463,
      "loss": 4.7554,
      "mean_token_accuracy": 0.21681054085493087,
      "num_tokens": 91285474.0,
      "step": 52625
    },
    {
      "entropy": 5.410590028762817,
      "epoch": 4.094767554950399,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034124701565436076,
      "loss": 4.8217,
      "mean_token_accuracy": 0.21871497482061386,
      "num_tokens": 91293333.0,
      "step": 52630
    },
    {
      "entropy": 5.374338865280151,
      "epoch": 4.095156584322116,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000341220528121794,
      "loss": 4.7879,
      "mean_token_accuracy": 0.22420702427625655,
      "num_tokens": 91301881.0,
      "step": 52635
    },
    {
      "entropy": 5.310370254516601,
      "epoch": 4.095545613693834,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000341194039584436,
      "loss": 4.7387,
      "mean_token_accuracy": 0.22733586579561232,
      "num_tokens": 91309922.0,
      "step": 52640
    },
    {
      "entropy": 5.357850551605225,
      "epoch": 4.095934643065552,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003411675500426888,
      "loss": 4.8089,
      "mean_token_accuracy": 0.21542946994304657,
      "num_tokens": 91317844.0,
      "step": 52645
    },
    {
      "entropy": 5.439244270324707,
      "epoch": 4.096323672437269,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034114105949695445,
      "loss": 4.8716,
      "mean_token_accuracy": 0.21526251137256622,
      "num_tokens": 91326754.0,
      "step": 52650
    },
    {
      "entropy": 5.485074043273926,
      "epoch": 4.096712701808986,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003411145679476347,
      "loss": 4.8595,
      "mean_token_accuracy": 0.22414433509111403,
      "num_tokens": 91335885.0,
      "step": 52655
    },
    {
      "entropy": 5.3343198776245115,
      "epoch": 4.097101731180704,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003410880753951317,
      "loss": 4.7976,
      "mean_token_accuracy": 0.21626466810703276,
      "num_tokens": 91345120.0,
      "step": 52660
    },
    {
      "entropy": 5.3816393375396725,
      "epoch": 4.097490760552422,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003410615818398473,
      "loss": 4.8225,
      "mean_token_accuracy": 0.21757972687482835,
      "num_tokens": 91353854.0,
      "step": 52665
    },
    {
      "entropy": 5.377091121673584,
      "epoch": 4.097879789924139,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003410350872821835,
      "loss": 4.7462,
      "mean_token_accuracy": 0.22134456932544708,
      "num_tokens": 91362787.0,
      "step": 52670
    },
    {
      "entropy": 5.319772815704345,
      "epoch": 4.098268819295857,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034100859172254245,
      "loss": 4.7457,
      "mean_token_accuracy": 0.22878327667713166,
      "num_tokens": 91372332.0,
      "step": 52675
    },
    {
      "entropy": 5.300218725204468,
      "epoch": 4.098657848667575,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034098209516132607,
      "loss": 4.6808,
      "mean_token_accuracy": 0.22906354069709778,
      "num_tokens": 91380760.0,
      "step": 52680
    },
    {
      "entropy": 5.4391580581665036,
      "epoch": 4.099046878039292,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003409555975989363,
      "loss": 4.8978,
      "mean_token_accuracy": 0.2069607600569725,
      "num_tokens": 91390139.0,
      "step": 52685
    },
    {
      "entropy": 5.328381061553955,
      "epoch": 4.099435907411009,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034092909903577547,
      "loss": 4.7922,
      "mean_token_accuracy": 0.2183182030916214,
      "num_tokens": 91398460.0,
      "step": 52690
    },
    {
      "entropy": 5.405427169799805,
      "epoch": 4.099824936782727,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003409025994722454,
      "loss": 4.8356,
      "mean_token_accuracy": 0.22259434014558793,
      "num_tokens": 91407689.0,
      "step": 52695
    },
    {
      "entropy": 5.455458068847657,
      "epoch": 4.100213966154445,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003408760989087482,
      "loss": 4.8761,
      "mean_token_accuracy": 0.21131756901741028,
      "num_tokens": 91416559.0,
      "step": 52700
    },
    {
      "entropy": 5.312933683395386,
      "epoch": 4.100602995526162,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034084959734568616,
      "loss": 4.685,
      "mean_token_accuracy": 0.2254567936062813,
      "num_tokens": 91425785.0,
      "step": 52705
    },
    {
      "entropy": 5.355159425735474,
      "epoch": 4.10099202489788,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034082309478346127,
      "loss": 4.8119,
      "mean_token_accuracy": 0.2128932312130928,
      "num_tokens": 91433710.0,
      "step": 52710
    },
    {
      "entropy": 5.323598766326905,
      "epoch": 4.101381054269598,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034079659122247574,
      "loss": 4.7181,
      "mean_token_accuracy": 0.23461543768644333,
      "num_tokens": 91441943.0,
      "step": 52715
    },
    {
      "entropy": 5.473953580856323,
      "epoch": 4.101770083641315,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003407700866631317,
      "loss": 4.8428,
      "mean_token_accuracy": 0.21693453788757325,
      "num_tokens": 91450928.0,
      "step": 52720
    },
    {
      "entropy": 5.404916524887085,
      "epoch": 4.102159113013032,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003407435811058312,
      "loss": 4.8044,
      "mean_token_accuracy": 0.21938394010066986,
      "num_tokens": 91458893.0,
      "step": 52725
    },
    {
      "entropy": 5.381249284744262,
      "epoch": 4.10254814238475,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003407170745509764,
      "loss": 4.8272,
      "mean_token_accuracy": 0.22328583896160126,
      "num_tokens": 91467545.0,
      "step": 52730
    },
    {
      "entropy": 5.439403104782104,
      "epoch": 4.1029371717564675,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034069056699896974,
      "loss": 4.8844,
      "mean_token_accuracy": 0.20948629081249237,
      "num_tokens": 91476288.0,
      "step": 52735
    },
    {
      "entropy": 5.462603044509888,
      "epoch": 4.103326201128185,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003406640584502132,
      "loss": 4.8492,
      "mean_token_accuracy": 0.2114051342010498,
      "num_tokens": 91484302.0,
      "step": 52740
    },
    {
      "entropy": 5.411507272720337,
      "epoch": 4.103715230499903,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003406375489051091,
      "loss": 4.8208,
      "mean_token_accuracy": 0.21463291794061662,
      "num_tokens": 91492084.0,
      "step": 52745
    },
    {
      "entropy": 5.307044124603271,
      "epoch": 4.104104259871621,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003406110383640597,
      "loss": 4.7261,
      "mean_token_accuracy": 0.22005556523799896,
      "num_tokens": 91500512.0,
      "step": 52750
    },
    {
      "entropy": 5.304315710067749,
      "epoch": 4.1044932892433375,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034058452682746734,
      "loss": 4.688,
      "mean_token_accuracy": 0.23027225434780121,
      "num_tokens": 91508720.0,
      "step": 52755
    },
    {
      "entropy": 5.358227634429932,
      "epoch": 4.104882318615055,
      "grad_norm": 1.21875,
      "learning_rate": 0.000340558014295734,
      "loss": 4.8009,
      "mean_token_accuracy": 0.2190185621380806,
      "num_tokens": 91517372.0,
      "step": 52760
    },
    {
      "entropy": 5.293401193618775,
      "epoch": 4.105271347986773,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034053150076926214,
      "loss": 4.6451,
      "mean_token_accuracy": 0.22750626057386397,
      "num_tokens": 91525047.0,
      "step": 52765
    },
    {
      "entropy": 5.397005796432495,
      "epoch": 4.1056603773584905,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003405049862484541,
      "loss": 4.8239,
      "mean_token_accuracy": 0.20831654369831085,
      "num_tokens": 91533819.0,
      "step": 52770
    },
    {
      "entropy": 5.3624519348144535,
      "epoch": 4.106049406730208,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003404784707337122,
      "loss": 4.8716,
      "mean_token_accuracy": 0.21115259528160096,
      "num_tokens": 91542846.0,
      "step": 52775
    },
    {
      "entropy": 5.445082950592041,
      "epoch": 4.106438436101926,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034045195422543865,
      "loss": 4.8544,
      "mean_token_accuracy": 0.22222942858934402,
      "num_tokens": 91552402.0,
      "step": 52780
    },
    {
      "entropy": 5.344077110290527,
      "epoch": 4.106827465473644,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034042543672403594,
      "loss": 4.7557,
      "mean_token_accuracy": 0.2256408542394638,
      "num_tokens": 91560490.0,
      "step": 52785
    },
    {
      "entropy": 5.374427127838135,
      "epoch": 4.10721649484536,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034039891822990634,
      "loss": 4.7493,
      "mean_token_accuracy": 0.21630677133798598,
      "num_tokens": 91569221.0,
      "step": 52790
    },
    {
      "entropy": 5.303769636154175,
      "epoch": 4.107605524217078,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003403723987434523,
      "loss": 4.6976,
      "mean_token_accuracy": 0.22296379506587982,
      "num_tokens": 91577546.0,
      "step": 52795
    },
    {
      "entropy": 5.393552684783936,
      "epoch": 4.107994553588796,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034034587826507604,
      "loss": 4.8156,
      "mean_token_accuracy": 0.21757666021585464,
      "num_tokens": 91586273.0,
      "step": 52800
    },
    {
      "entropy": 5.465078973770142,
      "epoch": 4.1083835829605135,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034031935679518016,
      "loss": 4.846,
      "mean_token_accuracy": 0.211498661339283,
      "num_tokens": 91595478.0,
      "step": 52805
    },
    {
      "entropy": 5.406665325164795,
      "epoch": 4.108772612332231,
      "grad_norm": 1.15625,
      "learning_rate": 0.000340292834334167,
      "loss": 4.842,
      "mean_token_accuracy": 0.21079651564359664,
      "num_tokens": 91604983.0,
      "step": 52810
    },
    {
      "entropy": 5.395997714996338,
      "epoch": 4.109161641703949,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000340266310882439,
      "loss": 4.7654,
      "mean_token_accuracy": 0.22088808417320252,
      "num_tokens": 91613404.0,
      "step": 52815
    },
    {
      "entropy": 5.385249042510987,
      "epoch": 4.109550671075667,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034023978644039864,
      "loss": 4.8536,
      "mean_token_accuracy": 0.2169533208012581,
      "num_tokens": 91622511.0,
      "step": 52820
    },
    {
      "entropy": 5.405751466751099,
      "epoch": 4.109939700447383,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003402132610084483,
      "loss": 4.8226,
      "mean_token_accuracy": 0.21513909697532654,
      "num_tokens": 91630722.0,
      "step": 52825
    },
    {
      "entropy": 5.419298791885376,
      "epoch": 4.110328729819101,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034018673458699055,
      "loss": 4.8085,
      "mean_token_accuracy": 0.2190310224890709,
      "num_tokens": 91638747.0,
      "step": 52830
    },
    {
      "entropy": 5.3629697322845455,
      "epoch": 4.110717759190819,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003401602071764279,
      "loss": 4.7599,
      "mean_token_accuracy": 0.22039033621549606,
      "num_tokens": 91646990.0,
      "step": 52835
    },
    {
      "entropy": 5.344353342056275,
      "epoch": 4.1111067885625365,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034013367877716284,
      "loss": 4.7984,
      "mean_token_accuracy": 0.2251903533935547,
      "num_tokens": 91655205.0,
      "step": 52840
    },
    {
      "entropy": 5.306543159484863,
      "epoch": 4.111495817934254,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003401071493895977,
      "loss": 4.6656,
      "mean_token_accuracy": 0.22465095669031143,
      "num_tokens": 91663869.0,
      "step": 52845
    },
    {
      "entropy": 5.3391838550567625,
      "epoch": 4.111884847305972,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003400806190141354,
      "loss": 4.7357,
      "mean_token_accuracy": 0.22322702407836914,
      "num_tokens": 91672048.0,
      "step": 52850
    },
    {
      "entropy": 5.37914810180664,
      "epoch": 4.112273876677689,
      "grad_norm": 1.125,
      "learning_rate": 0.00034005408765117815,
      "loss": 4.7619,
      "mean_token_accuracy": 0.2206842929124832,
      "num_tokens": 91680669.0,
      "step": 52855
    },
    {
      "entropy": 5.421321058273316,
      "epoch": 4.112662906049406,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034002755530112877,
      "loss": 4.7918,
      "mean_token_accuracy": 0.22741796523332597,
      "num_tokens": 91690060.0,
      "step": 52860
    },
    {
      "entropy": 5.373548412322998,
      "epoch": 4.113051935421124,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003400010219643897,
      "loss": 4.6964,
      "mean_token_accuracy": 0.23401701003313063,
      "num_tokens": 91698108.0,
      "step": 52865
    },
    {
      "entropy": 5.3752960681915285,
      "epoch": 4.113440964792842,
      "grad_norm": 1.25,
      "learning_rate": 0.0003399744876413636,
      "loss": 4.7782,
      "mean_token_accuracy": 0.21769324839115142,
      "num_tokens": 91707003.0,
      "step": 52870
    },
    {
      "entropy": 5.371784496307373,
      "epoch": 4.1138299941645595,
      "grad_norm": 1.125,
      "learning_rate": 0.000339947952332453,
      "loss": 4.7932,
      "mean_token_accuracy": 0.21545062512159346,
      "num_tokens": 91715539.0,
      "step": 52875
    },
    {
      "entropy": 5.375323295593262,
      "epoch": 4.114219023536277,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033992141603806064,
      "loss": 4.7994,
      "mean_token_accuracy": 0.21191147565841675,
      "num_tokens": 91724410.0,
      "step": 52880
    },
    {
      "entropy": 5.394019031524659,
      "epoch": 4.114608052907995,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033989487875858915,
      "loss": 4.7888,
      "mean_token_accuracy": 0.22593642622232438,
      "num_tokens": 91732744.0,
      "step": 52885
    },
    {
      "entropy": 5.3799807071685795,
      "epoch": 4.114997082279712,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033986834049444113,
      "loss": 4.7167,
      "mean_token_accuracy": 0.2253296598792076,
      "num_tokens": 91741406.0,
      "step": 52890
    },
    {
      "entropy": 5.356348991394043,
      "epoch": 4.115386111651429,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033984180124601936,
      "loss": 4.7753,
      "mean_token_accuracy": 0.2189260959625244,
      "num_tokens": 91749998.0,
      "step": 52895
    },
    {
      "entropy": 5.352110862731934,
      "epoch": 4.115775141023147,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033981526101372635,
      "loss": 4.7781,
      "mean_token_accuracy": 0.21984853893518447,
      "num_tokens": 91758175.0,
      "step": 52900
    },
    {
      "entropy": 5.332473945617676,
      "epoch": 4.116164170394865,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000339788719797965,
      "loss": 4.7111,
      "mean_token_accuracy": 0.22017756700515748,
      "num_tokens": 91766606.0,
      "step": 52905
    },
    {
      "entropy": 5.349628067016601,
      "epoch": 4.1165531997665825,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003397621775991379,
      "loss": 4.7126,
      "mean_token_accuracy": 0.2264467492699623,
      "num_tokens": 91775446.0,
      "step": 52910
    },
    {
      "entropy": 5.349782419204712,
      "epoch": 4.1169422291383,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003397356344176479,
      "loss": 4.7166,
      "mean_token_accuracy": 0.22759489119052886,
      "num_tokens": 91783954.0,
      "step": 52915
    },
    {
      "entropy": 5.2686354637146,
      "epoch": 4.117331258510018,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003397090902538976,
      "loss": 4.6661,
      "mean_token_accuracy": 0.23432971835136412,
      "num_tokens": 91793169.0,
      "step": 52920
    },
    {
      "entropy": 5.3940552234649655,
      "epoch": 4.117720287881735,
      "grad_norm": 1.25,
      "learning_rate": 0.00033968254510828995,
      "loss": 4.849,
      "mean_token_accuracy": 0.21371231228113174,
      "num_tokens": 91802011.0,
      "step": 52925
    },
    {
      "entropy": 5.394496202468872,
      "epoch": 4.118109317253452,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003396559989812275,
      "loss": 4.8102,
      "mean_token_accuracy": 0.21799286901950837,
      "num_tokens": 91811120.0,
      "step": 52930
    },
    {
      "entropy": 5.471392631530762,
      "epoch": 4.11849834662517,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003396294518731133,
      "loss": 4.8759,
      "mean_token_accuracy": 0.214100281894207,
      "num_tokens": 91819837.0,
      "step": 52935
    },
    {
      "entropy": 5.3466535091400145,
      "epoch": 4.118887375996888,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033960290378435,
      "loss": 4.7262,
      "mean_token_accuracy": 0.22798630297183992,
      "num_tokens": 91828955.0,
      "step": 52940
    },
    {
      "entropy": 5.306018447875976,
      "epoch": 4.1192764053686055,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033957635471534045,
      "loss": 4.7052,
      "mean_token_accuracy": 0.21883408576250077,
      "num_tokens": 91837508.0,
      "step": 52945
    },
    {
      "entropy": 5.327532482147217,
      "epoch": 4.119665434740323,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003395498046664875,
      "loss": 4.7455,
      "mean_token_accuracy": 0.21889119148254393,
      "num_tokens": 91847030.0,
      "step": 52950
    },
    {
      "entropy": 5.389635610580444,
      "epoch": 4.120054464112041,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000339523253638194,
      "loss": 4.7255,
      "mean_token_accuracy": 0.22487041354179382,
      "num_tokens": 91855823.0,
      "step": 52955
    },
    {
      "entropy": 5.3975457668304445,
      "epoch": 4.120443493483758,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003394967016308629,
      "loss": 4.8093,
      "mean_token_accuracy": 0.2149109274148941,
      "num_tokens": 91864741.0,
      "step": 52960
    },
    {
      "entropy": 5.321229791641235,
      "epoch": 4.120832522855475,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003394701486448968,
      "loss": 4.7394,
      "mean_token_accuracy": 0.22114581763744354,
      "num_tokens": 91872780.0,
      "step": 52965
    },
    {
      "entropy": 5.396734046936035,
      "epoch": 4.121221552227193,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033944359468069903,
      "loss": 4.7836,
      "mean_token_accuracy": 0.22365962117910385,
      "num_tokens": 91881394.0,
      "step": 52970
    },
    {
      "entropy": 5.315213775634765,
      "epoch": 4.121610581598911,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033941703973867216,
      "loss": 4.7488,
      "mean_token_accuracy": 0.22400761395692825,
      "num_tokens": 91889758.0,
      "step": 52975
    },
    {
      "entropy": 5.367676734924316,
      "epoch": 4.1219996109706285,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033939048381921935,
      "loss": 4.7108,
      "mean_token_accuracy": 0.22394270598888397,
      "num_tokens": 91897799.0,
      "step": 52980
    },
    {
      "entropy": 5.345440673828125,
      "epoch": 4.122388640342346,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003393639269227434,
      "loss": 4.7674,
      "mean_token_accuracy": 0.22346251159906388,
      "num_tokens": 91907016.0,
      "step": 52985
    },
    {
      "entropy": 5.255563116073608,
      "epoch": 4.122777669714063,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003393373690496473,
      "loss": 4.6656,
      "mean_token_accuracy": 0.23237578868865966,
      "num_tokens": 91914934.0,
      "step": 52990
    },
    {
      "entropy": 5.385871887207031,
      "epoch": 4.123166699085781,
      "grad_norm": 1.171875,
      "learning_rate": 0.000339310810200334,
      "loss": 4.8255,
      "mean_token_accuracy": 0.21754276603460312,
      "num_tokens": 91923640.0,
      "step": 52995
    },
    {
      "entropy": 5.407751607894897,
      "epoch": 4.123555728457498,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033928425037520643,
      "loss": 4.8318,
      "mean_token_accuracy": 0.2173330694437027,
      "num_tokens": 91931768.0,
      "step": 53000
    },
    {
      "entropy": 5.405064725875855,
      "epoch": 4.123944757829216,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003392576895746678,
      "loss": 4.7851,
      "mean_token_accuracy": 0.21746721714735032,
      "num_tokens": 91940681.0,
      "step": 53005
    },
    {
      "entropy": 5.5169349193573,
      "epoch": 4.124333787200934,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003392311277991209,
      "loss": 4.9074,
      "mean_token_accuracy": 0.20890201181173323,
      "num_tokens": 91948578.0,
      "step": 53010
    },
    {
      "entropy": 5.358961153030395,
      "epoch": 4.1247228165726515,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033920456504896895,
      "loss": 4.7208,
      "mean_token_accuracy": 0.2272202715277672,
      "num_tokens": 91956706.0,
      "step": 53015
    },
    {
      "entropy": 5.333904457092285,
      "epoch": 4.125111845944369,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033917800132461493,
      "loss": 4.739,
      "mean_token_accuracy": 0.2236358180642128,
      "num_tokens": 91966071.0,
      "step": 53020
    },
    {
      "entropy": 5.398079252243042,
      "epoch": 4.125500875316086,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003391514366264619,
      "loss": 4.7747,
      "mean_token_accuracy": 0.22605164498090743,
      "num_tokens": 91973713.0,
      "step": 53025
    },
    {
      "entropy": 5.3161111831665036,
      "epoch": 4.125889904687804,
      "grad_norm": 1.25,
      "learning_rate": 0.0003391248709549129,
      "loss": 4.7613,
      "mean_token_accuracy": 0.21622918248176576,
      "num_tokens": 91982945.0,
      "step": 53030
    },
    {
      "entropy": 5.366640424728393,
      "epoch": 4.126278934059521,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003390983043103711,
      "loss": 4.8909,
      "mean_token_accuracy": 0.21523632258176803,
      "num_tokens": 91991943.0,
      "step": 53035
    },
    {
      "entropy": 5.397830486297607,
      "epoch": 4.126667963431239,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003390717366932395,
      "loss": 4.8211,
      "mean_token_accuracy": 0.21124104112386705,
      "num_tokens": 92000480.0,
      "step": 53040
    },
    {
      "entropy": 5.4509178638458256,
      "epoch": 4.127056992802957,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003390451681039212,
      "loss": 4.8706,
      "mean_token_accuracy": 0.2118455797433853,
      "num_tokens": 92009922.0,
      "step": 53045
    },
    {
      "entropy": 5.356840991973877,
      "epoch": 4.1274460221746745,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033901859854281954,
      "loss": 4.67,
      "mean_token_accuracy": 0.22865425199270248,
      "num_tokens": 92017978.0,
      "step": 53050
    },
    {
      "entropy": 5.342997121810913,
      "epoch": 4.127835051546391,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003389920280103375,
      "loss": 4.7703,
      "mean_token_accuracy": 0.2188594475388527,
      "num_tokens": 92026649.0,
      "step": 53055
    },
    {
      "entropy": 5.30946102142334,
      "epoch": 4.128224080918109,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033896545650687824,
      "loss": 4.726,
      "mean_token_accuracy": 0.22779389917850495,
      "num_tokens": 92035393.0,
      "step": 53060
    },
    {
      "entropy": 5.43549165725708,
      "epoch": 4.128613110289827,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003389388840328451,
      "loss": 4.7823,
      "mean_token_accuracy": 0.2170545592904091,
      "num_tokens": 92043644.0,
      "step": 53065
    },
    {
      "entropy": 5.427477407455444,
      "epoch": 4.129002139661544,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003389123105886411,
      "loss": 4.8226,
      "mean_token_accuracy": 0.21947708874940872,
      "num_tokens": 92052469.0,
      "step": 53070
    },
    {
      "entropy": 5.337659311294556,
      "epoch": 4.129391169033262,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003388857361746695,
      "loss": 4.7218,
      "mean_token_accuracy": 0.22316092848777772,
      "num_tokens": 92060887.0,
      "step": 53075
    },
    {
      "entropy": 5.324646091461181,
      "epoch": 4.12978019840498,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033885916079133354,
      "loss": 4.7082,
      "mean_token_accuracy": 0.22857169359922408,
      "num_tokens": 92069344.0,
      "step": 53080
    },
    {
      "entropy": 5.3360028743743895,
      "epoch": 4.1301692277766975,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033883258443903646,
      "loss": 4.7039,
      "mean_token_accuracy": 0.22247749716043472,
      "num_tokens": 92078634.0,
      "step": 53085
    },
    {
      "entropy": 5.293679666519165,
      "epoch": 4.130558257148414,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033880600711818157,
      "loss": 4.6323,
      "mean_token_accuracy": 0.22789670079946517,
      "num_tokens": 92086790.0,
      "step": 53090
    },
    {
      "entropy": 5.409792900085449,
      "epoch": 4.130947286520132,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033877942882917203,
      "loss": 4.7792,
      "mean_token_accuracy": 0.22417913675308226,
      "num_tokens": 92094944.0,
      "step": 53095
    },
    {
      "entropy": 5.346992635726929,
      "epoch": 4.13133631589185,
      "grad_norm": 1.359375,
      "learning_rate": 0.00033875284957241117,
      "loss": 4.7929,
      "mean_token_accuracy": 0.2221509724855423,
      "num_tokens": 92102845.0,
      "step": 53100
    },
    {
      "entropy": 5.368960332870484,
      "epoch": 4.131725345263567,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003387262693483023,
      "loss": 4.7728,
      "mean_token_accuracy": 0.22582782655954362,
      "num_tokens": 92110895.0,
      "step": 53105
    },
    {
      "entropy": 5.364451217651367,
      "epoch": 4.132114374635285,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033869968815724866,
      "loss": 4.7691,
      "mean_token_accuracy": 0.22725868821144105,
      "num_tokens": 92119400.0,
      "step": 53110
    },
    {
      "entropy": 5.459622573852539,
      "epoch": 4.132503404007003,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003386731059996537,
      "loss": 4.899,
      "mean_token_accuracy": 0.21772991269826888,
      "num_tokens": 92129014.0,
      "step": 53115
    },
    {
      "entropy": 5.451181888580322,
      "epoch": 4.1328924333787205,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033864652287592074,
      "loss": 4.9078,
      "mean_token_accuracy": 0.216256944835186,
      "num_tokens": 92137659.0,
      "step": 53120
    },
    {
      "entropy": 5.3701056957244875,
      "epoch": 4.133281462750437,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033861993878645305,
      "loss": 4.7603,
      "mean_token_accuracy": 0.22163394391536712,
      "num_tokens": 92146467.0,
      "step": 53125
    },
    {
      "entropy": 5.431371784210205,
      "epoch": 4.133670492122155,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000338593353731654,
      "loss": 4.7489,
      "mean_token_accuracy": 0.22166501432657243,
      "num_tokens": 92155180.0,
      "step": 53130
    },
    {
      "entropy": 5.364253807067871,
      "epoch": 4.134059521493873,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003385667677119271,
      "loss": 4.8099,
      "mean_token_accuracy": 0.22661019563674928,
      "num_tokens": 92163531.0,
      "step": 53135
    },
    {
      "entropy": 5.343440008163452,
      "epoch": 4.13444855086559,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033854018072767574,
      "loss": 4.8356,
      "mean_token_accuracy": 0.21555633693933487,
      "num_tokens": 92172148.0,
      "step": 53140
    },
    {
      "entropy": 5.381349515914917,
      "epoch": 4.134837580237308,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00033851359277930316,
      "loss": 4.7291,
      "mean_token_accuracy": 0.22282301485538483,
      "num_tokens": 92180666.0,
      "step": 53145
    },
    {
      "entropy": 5.274835538864136,
      "epoch": 4.135226609609026,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003384870038672129,
      "loss": 4.6996,
      "mean_token_accuracy": 0.22787546813488008,
      "num_tokens": 92189894.0,
      "step": 53150
    },
    {
      "entropy": 5.417509412765503,
      "epoch": 4.1356156389807435,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003384604139918085,
      "loss": 4.7822,
      "mean_token_accuracy": 0.22020720243453978,
      "num_tokens": 92199029.0,
      "step": 53155
    },
    {
      "entropy": 5.339353847503662,
      "epoch": 4.13600466835246,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033843382315349325,
      "loss": 4.7265,
      "mean_token_accuracy": 0.22383989542722701,
      "num_tokens": 92207388.0,
      "step": 53160
    },
    {
      "entropy": 5.344620895385742,
      "epoch": 4.136393697724178,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033840723135267073,
      "loss": 4.7428,
      "mean_token_accuracy": 0.22348374873399734,
      "num_tokens": 92216194.0,
      "step": 53165
    },
    {
      "entropy": 5.334713172912598,
      "epoch": 4.136782727095896,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033838063858974434,
      "loss": 4.7425,
      "mean_token_accuracy": 0.21988310515880585,
      "num_tokens": 92224322.0,
      "step": 53170
    },
    {
      "entropy": 5.342411756515503,
      "epoch": 4.137171756467613,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003383540448651177,
      "loss": 4.7318,
      "mean_token_accuracy": 0.22171326726675034,
      "num_tokens": 92232635.0,
      "step": 53175
    },
    {
      "entropy": 5.395830678939819,
      "epoch": 4.137560785839331,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033832745017919424,
      "loss": 4.772,
      "mean_token_accuracy": 0.224443481862545,
      "num_tokens": 92240609.0,
      "step": 53180
    },
    {
      "entropy": 5.299058485031128,
      "epoch": 4.137949815211049,
      "grad_norm": 1.125,
      "learning_rate": 0.00033830085453237757,
      "loss": 4.6726,
      "mean_token_accuracy": 0.22355888783931732,
      "num_tokens": 92248699.0,
      "step": 53185
    },
    {
      "entropy": 5.369287300109863,
      "epoch": 4.138338844582766,
      "grad_norm": 1.125,
      "learning_rate": 0.0003382742579250712,
      "loss": 4.8298,
      "mean_token_accuracy": 0.2149367719888687,
      "num_tokens": 92256882.0,
      "step": 53190
    },
    {
      "entropy": 5.305195569992065,
      "epoch": 4.138727873954483,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003382476603576785,
      "loss": 4.7099,
      "mean_token_accuracy": 0.2203458368778229,
      "num_tokens": 92264729.0,
      "step": 53195
    },
    {
      "entropy": 5.404139661788941,
      "epoch": 4.139116903326201,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003382210618306034,
      "loss": 4.9153,
      "mean_token_accuracy": 0.21020956635475158,
      "num_tokens": 92273924.0,
      "step": 53200
    },
    {
      "entropy": 5.395939731597901,
      "epoch": 4.139505932697919,
      "grad_norm": 1.25,
      "learning_rate": 0.0003381944623442492,
      "loss": 4.7289,
      "mean_token_accuracy": 0.22420473247766495,
      "num_tokens": 92282116.0,
      "step": 53205
    },
    {
      "entropy": 5.375788593292237,
      "epoch": 4.139894962069636,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003381678618990197,
      "loss": 4.7095,
      "mean_token_accuracy": 0.22707545310258864,
      "num_tokens": 92290478.0,
      "step": 53210
    },
    {
      "entropy": 5.279119253158569,
      "epoch": 4.140283991441354,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003381412604953185,
      "loss": 4.7353,
      "mean_token_accuracy": 0.22536227703094483,
      "num_tokens": 92298524.0,
      "step": 53215
    },
    {
      "entropy": 5.2842484474182125,
      "epoch": 4.140673020813072,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003381146581335491,
      "loss": 4.687,
      "mean_token_accuracy": 0.22839209586381912,
      "num_tokens": 92306532.0,
      "step": 53220
    },
    {
      "entropy": 5.476403284072876,
      "epoch": 4.141062050184789,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003380880548141152,
      "loss": 4.9156,
      "mean_token_accuracy": 0.20746221393346786,
      "num_tokens": 92316194.0,
      "step": 53225
    },
    {
      "entropy": 5.3446704864501955,
      "epoch": 4.141451079556506,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033806145053742043,
      "loss": 4.7167,
      "mean_token_accuracy": 0.2221660539507866,
      "num_tokens": 92324233.0,
      "step": 53230
    },
    {
      "entropy": 5.378196477890015,
      "epoch": 4.141840108928224,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003380348453038686,
      "loss": 4.8449,
      "mean_token_accuracy": 0.22004768550395964,
      "num_tokens": 92333823.0,
      "step": 53235
    },
    {
      "entropy": 5.465040302276611,
      "epoch": 4.142229138299942,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003380082391138633,
      "loss": 4.8263,
      "mean_token_accuracy": 0.22028243839740752,
      "num_tokens": 92342584.0,
      "step": 53240
    },
    {
      "entropy": 5.4392725944519045,
      "epoch": 4.142618167671659,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003379816319678083,
      "loss": 4.7706,
      "mean_token_accuracy": 0.2245413839817047,
      "num_tokens": 92350111.0,
      "step": 53245
    },
    {
      "entropy": 5.350080347061157,
      "epoch": 4.143007197043377,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003379550238661072,
      "loss": 4.8457,
      "mean_token_accuracy": 0.22499995827674865,
      "num_tokens": 92358731.0,
      "step": 53250
    },
    {
      "entropy": 5.304732084274292,
      "epoch": 4.143396226415095,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033792841480916394,
      "loss": 4.7235,
      "mean_token_accuracy": 0.22588523626327514,
      "num_tokens": 92367426.0,
      "step": 53255
    },
    {
      "entropy": 5.382920169830323,
      "epoch": 4.143785255786812,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033790180479738205,
      "loss": 4.759,
      "mean_token_accuracy": 0.21844741553068162,
      "num_tokens": 92376937.0,
      "step": 53260
    },
    {
      "entropy": 5.387865829467773,
      "epoch": 4.144174285158529,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003378751938311654,
      "loss": 4.778,
      "mean_token_accuracy": 0.22013824582099914,
      "num_tokens": 92385562.0,
      "step": 53265
    },
    {
      "entropy": 5.340405654907227,
      "epoch": 4.144563314530247,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003378485819109177,
      "loss": 4.7442,
      "mean_token_accuracy": 0.22913311421871185,
      "num_tokens": 92393774.0,
      "step": 53270
    },
    {
      "entropy": 5.332422590255737,
      "epoch": 4.144952343901965,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033782196903704286,
      "loss": 4.7593,
      "mean_token_accuracy": 0.22248225659132004,
      "num_tokens": 92402711.0,
      "step": 53275
    },
    {
      "entropy": 5.395894479751587,
      "epoch": 4.145341373273682,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003377953552099447,
      "loss": 4.8072,
      "mean_token_accuracy": 0.21800800710916518,
      "num_tokens": 92411071.0,
      "step": 53280
    },
    {
      "entropy": 5.394516468048096,
      "epoch": 4.1457304026454,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003377687404300269,
      "loss": 4.8227,
      "mean_token_accuracy": 0.21692912876605988,
      "num_tokens": 92420842.0,
      "step": 53285
    },
    {
      "entropy": 5.472236299514771,
      "epoch": 4.146119432017118,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003377421246976934,
      "loss": 4.9298,
      "mean_token_accuracy": 0.2098902463912964,
      "num_tokens": 92430157.0,
      "step": 53290
    },
    {
      "entropy": 5.366793060302735,
      "epoch": 4.1465084613888346,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033771550801334806,
      "loss": 4.7286,
      "mean_token_accuracy": 0.22010661214590072,
      "num_tokens": 92439171.0,
      "step": 53295
    },
    {
      "entropy": 5.319917392730713,
      "epoch": 4.146897490760552,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033768889037739464,
      "loss": 4.721,
      "mean_token_accuracy": 0.2218250811100006,
      "num_tokens": 92447906.0,
      "step": 53300
    },
    {
      "entropy": 5.304881763458252,
      "epoch": 4.14728652013227,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033766227179023704,
      "loss": 4.8003,
      "mean_token_accuracy": 0.20814212113618852,
      "num_tokens": 92456676.0,
      "step": 53305
    },
    {
      "entropy": 5.349608850479126,
      "epoch": 4.147675549503988,
      "grad_norm": 1.125,
      "learning_rate": 0.00033763565225227934,
      "loss": 4.727,
      "mean_token_accuracy": 0.2260004013776779,
      "num_tokens": 92465388.0,
      "step": 53310
    },
    {
      "entropy": 5.425984525680542,
      "epoch": 4.148064578875705,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003376090317639252,
      "loss": 4.8329,
      "mean_token_accuracy": 0.21115414053201675,
      "num_tokens": 92473347.0,
      "step": 53315
    },
    {
      "entropy": 5.299237680435181,
      "epoch": 4.148453608247423,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033758241032557875,
      "loss": 4.7505,
      "mean_token_accuracy": 0.2238055869936943,
      "num_tokens": 92482375.0,
      "step": 53320
    },
    {
      "entropy": 5.30037522315979,
      "epoch": 4.14884263761914,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033755578793764375,
      "loss": 4.7579,
      "mean_token_accuracy": 0.21774219572544098,
      "num_tokens": 92490680.0,
      "step": 53325
    },
    {
      "entropy": 5.346386289596557,
      "epoch": 4.1492316669908575,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003375291646005243,
      "loss": 4.7717,
      "mean_token_accuracy": 0.2234201028943062,
      "num_tokens": 92499627.0,
      "step": 53330
    },
    {
      "entropy": 5.392048120498657,
      "epoch": 4.149620696362575,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033750254031462425,
      "loss": 4.7491,
      "mean_token_accuracy": 0.2158157929778099,
      "num_tokens": 92507895.0,
      "step": 53335
    },
    {
      "entropy": 5.333922672271728,
      "epoch": 4.150009725734293,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033747591508034765,
      "loss": 4.7737,
      "mean_token_accuracy": 0.22023155838251113,
      "num_tokens": 92516569.0,
      "step": 53340
    },
    {
      "entropy": 5.290148735046387,
      "epoch": 4.150398755106011,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003374492888980985,
      "loss": 4.7721,
      "mean_token_accuracy": 0.2167720004916191,
      "num_tokens": 92525389.0,
      "step": 53345
    },
    {
      "entropy": 5.36008038520813,
      "epoch": 4.150787784477728,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033742266176828073,
      "loss": 4.7504,
      "mean_token_accuracy": 0.22173051536083221,
      "num_tokens": 92533734.0,
      "step": 53350
    },
    {
      "entropy": 5.360470819473266,
      "epoch": 4.151176813849446,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033739603369129846,
      "loss": 4.7059,
      "mean_token_accuracy": 0.22798762023448943,
      "num_tokens": 92541525.0,
      "step": 53355
    },
    {
      "entropy": 5.404436445236206,
      "epoch": 4.151565843221163,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033736940466755566,
      "loss": 4.8137,
      "mean_token_accuracy": 0.22372468262910844,
      "num_tokens": 92550344.0,
      "step": 53360
    },
    {
      "entropy": 5.287572431564331,
      "epoch": 4.1519548725928805,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033734277469745644,
      "loss": 4.7336,
      "mean_token_accuracy": 0.2302241951227188,
      "num_tokens": 92559661.0,
      "step": 53365
    },
    {
      "entropy": 5.372905969619751,
      "epoch": 4.152343901964598,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003373161437814049,
      "loss": 4.8226,
      "mean_token_accuracy": 0.21798766106367112,
      "num_tokens": 92567810.0,
      "step": 53370
    },
    {
      "entropy": 5.435521125793457,
      "epoch": 4.152732931336316,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033728951191980505,
      "loss": 4.8516,
      "mean_token_accuracy": 0.21409981399774553,
      "num_tokens": 92576051.0,
      "step": 53375
    },
    {
      "entropy": 5.4103669166564945,
      "epoch": 4.153121960708034,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033726287911306095,
      "loss": 4.9404,
      "mean_token_accuracy": 0.21267235428094863,
      "num_tokens": 92584990.0,
      "step": 53380
    },
    {
      "entropy": 5.422621154785157,
      "epoch": 4.153510990079751,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003372362453615768,
      "loss": 4.8711,
      "mean_token_accuracy": 0.21617391109466552,
      "num_tokens": 92593598.0,
      "step": 53385
    },
    {
      "entropy": 5.450185871124267,
      "epoch": 4.153900019451468,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003372096106657566,
      "loss": 4.8223,
      "mean_token_accuracy": 0.21277300268411636,
      "num_tokens": 92602757.0,
      "step": 53390
    },
    {
      "entropy": 5.373446416854859,
      "epoch": 4.154289048823186,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033718297502600465,
      "loss": 4.747,
      "mean_token_accuracy": 0.21843497455120087,
      "num_tokens": 92611316.0,
      "step": 53395
    },
    {
      "entropy": 5.378171730041504,
      "epoch": 4.1546780781949035,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033715633844272504,
      "loss": 4.731,
      "mean_token_accuracy": 0.22910406589508056,
      "num_tokens": 92619467.0,
      "step": 53400
    },
    {
      "entropy": 5.344860935211182,
      "epoch": 4.155067107566621,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033712970091632187,
      "loss": 4.7521,
      "mean_token_accuracy": 0.22574177980422974,
      "num_tokens": 92627858.0,
      "step": 53405
    },
    {
      "entropy": 5.317840480804444,
      "epoch": 4.155456136938339,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033710306244719954,
      "loss": 4.7173,
      "mean_token_accuracy": 0.22209156453609466,
      "num_tokens": 92636553.0,
      "step": 53410
    },
    {
      "entropy": 5.3687304019927975,
      "epoch": 4.155845166310057,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003370764230357619,
      "loss": 4.7716,
      "mean_token_accuracy": 0.22491025775671006,
      "num_tokens": 92645100.0,
      "step": 53415
    },
    {
      "entropy": 5.299055814743042,
      "epoch": 4.156234195681774,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003370497826824134,
      "loss": 4.6551,
      "mean_token_accuracy": 0.22994706332683562,
      "num_tokens": 92653885.0,
      "step": 53420
    },
    {
      "entropy": 5.319966888427734,
      "epoch": 4.156623225053491,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003370231413875582,
      "loss": 4.7397,
      "mean_token_accuracy": 0.22184529155492783,
      "num_tokens": 92662534.0,
      "step": 53425
    },
    {
      "entropy": 5.324884462356567,
      "epoch": 4.157012254425209,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033699649915160053,
      "loss": 4.6748,
      "mean_token_accuracy": 0.23044501394033431,
      "num_tokens": 92670616.0,
      "step": 53430
    },
    {
      "entropy": 5.304740905761719,
      "epoch": 4.1574012837969265,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033696985597494474,
      "loss": 4.7177,
      "mean_token_accuracy": 0.2221129685640335,
      "num_tokens": 92678403.0,
      "step": 53435
    },
    {
      "entropy": 5.293106508255005,
      "epoch": 4.157790313168644,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000336943211857995,
      "loss": 4.7587,
      "mean_token_accuracy": 0.2160153269767761,
      "num_tokens": 92688122.0,
      "step": 53440
    },
    {
      "entropy": 5.4493824481964115,
      "epoch": 4.158179342540362,
      "grad_norm": 1.25,
      "learning_rate": 0.0003369165668011556,
      "loss": 4.787,
      "mean_token_accuracy": 0.22683054953813553,
      "num_tokens": 92696909.0,
      "step": 53445
    },
    {
      "entropy": 5.375456142425537,
      "epoch": 4.15856837191208,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003368899208048308,
      "loss": 4.7411,
      "mean_token_accuracy": 0.2240825116634369,
      "num_tokens": 92705773.0,
      "step": 53450
    },
    {
      "entropy": 5.378240442276001,
      "epoch": 4.158957401283797,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033686327386942504,
      "loss": 4.8966,
      "mean_token_accuracy": 0.220670647919178,
      "num_tokens": 92713746.0,
      "step": 53455
    },
    {
      "entropy": 5.371901369094848,
      "epoch": 4.159346430655514,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003368366259953425,
      "loss": 4.7896,
      "mean_token_accuracy": 0.22070140689611434,
      "num_tokens": 92722144.0,
      "step": 53460
    },
    {
      "entropy": 5.340295076370239,
      "epoch": 4.159735460027232,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033680997718298756,
      "loss": 4.6821,
      "mean_token_accuracy": 0.23180922269821166,
      "num_tokens": 92730472.0,
      "step": 53465
    },
    {
      "entropy": 5.356926345825196,
      "epoch": 4.1601244893989495,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033678332743276466,
      "loss": 4.8201,
      "mean_token_accuracy": 0.22179990112781525,
      "num_tokens": 92738722.0,
      "step": 53470
    },
    {
      "entropy": 5.275622797012329,
      "epoch": 4.160513518770667,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003367566767450781,
      "loss": 4.765,
      "mean_token_accuracy": 0.22227921485900878,
      "num_tokens": 92748115.0,
      "step": 53475
    },
    {
      "entropy": 5.31640043258667,
      "epoch": 4.160902548142385,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033673002512033223,
      "loss": 4.6787,
      "mean_token_accuracy": 0.22584378272294997,
      "num_tokens": 92756346.0,
      "step": 53480
    },
    {
      "entropy": 5.401664400100708,
      "epoch": 4.161291577514103,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003367033725589315,
      "loss": 4.8542,
      "mean_token_accuracy": 0.21716010868549346,
      "num_tokens": 92764779.0,
      "step": 53485
    },
    {
      "entropy": 5.38063645362854,
      "epoch": 4.16168060688582,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003366767190612804,
      "loss": 4.7548,
      "mean_token_accuracy": 0.2241999924182892,
      "num_tokens": 92773011.0,
      "step": 53490
    },
    {
      "entropy": 5.300395250320435,
      "epoch": 4.162069636257537,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003366500646277831,
      "loss": 4.779,
      "mean_token_accuracy": 0.21551187187433243,
      "num_tokens": 92781928.0,
      "step": 53495
    },
    {
      "entropy": 5.281224012374878,
      "epoch": 4.162458665629255,
      "grad_norm": 1.125,
      "learning_rate": 0.0003366234092588443,
      "loss": 4.6711,
      "mean_token_accuracy": 0.23258210122585296,
      "num_tokens": 92790481.0,
      "step": 53500
    },
    {
      "entropy": 5.4500421524047855,
      "epoch": 4.1628476950009725,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033659675295486826,
      "loss": 4.8466,
      "mean_token_accuracy": 0.21680304557085037,
      "num_tokens": 92799528.0,
      "step": 53505
    },
    {
      "entropy": 5.437161588668824,
      "epoch": 4.16323672437269,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033657009571625964,
      "loss": 4.8368,
      "mean_token_accuracy": 0.2186627596616745,
      "num_tokens": 92807964.0,
      "step": 53510
    },
    {
      "entropy": 5.33622899055481,
      "epoch": 4.163625753744408,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003365434375434227,
      "loss": 4.7753,
      "mean_token_accuracy": 0.2174113780260086,
      "num_tokens": 92817010.0,
      "step": 53515
    },
    {
      "entropy": 5.35538878440857,
      "epoch": 4.164014783116126,
      "grad_norm": 1.125,
      "learning_rate": 0.00033651677843676213,
      "loss": 4.6977,
      "mean_token_accuracy": 0.2303978681564331,
      "num_tokens": 92826080.0,
      "step": 53520
    },
    {
      "entropy": 5.357707834243774,
      "epoch": 4.164403812487842,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003364901183966824,
      "loss": 4.7457,
      "mean_token_accuracy": 0.22311279326677322,
      "num_tokens": 92834837.0,
      "step": 53525
    },
    {
      "entropy": 5.428171539306641,
      "epoch": 4.16479284185956,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003364634574235879,
      "loss": 4.8468,
      "mean_token_accuracy": 0.20580584406852723,
      "num_tokens": 92843618.0,
      "step": 53530
    },
    {
      "entropy": 5.441087245941162,
      "epoch": 4.165181871231278,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033643679551788325,
      "loss": 4.8529,
      "mean_token_accuracy": 0.21888138800859452,
      "num_tokens": 92851762.0,
      "step": 53535
    },
    {
      "entropy": 5.421589756011963,
      "epoch": 4.1655709006029955,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033641013267997317,
      "loss": 4.8889,
      "mean_token_accuracy": 0.21543263494968415,
      "num_tokens": 92861266.0,
      "step": 53540
    },
    {
      "entropy": 5.364040565490723,
      "epoch": 4.165959929974713,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033638346891026196,
      "loss": 4.6628,
      "mean_token_accuracy": 0.23844515085220336,
      "num_tokens": 92869519.0,
      "step": 53545
    },
    {
      "entropy": 5.4244238376617435,
      "epoch": 4.166348959346431,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033635680420915424,
      "loss": 4.8192,
      "mean_token_accuracy": 0.21743252426385878,
      "num_tokens": 92878363.0,
      "step": 53550
    },
    {
      "entropy": 5.3774501323699955,
      "epoch": 4.166737988718149,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003363301385770548,
      "loss": 4.8375,
      "mean_token_accuracy": 0.22313880771398545,
      "num_tokens": 92887319.0,
      "step": 53555
    },
    {
      "entropy": 5.414543676376343,
      "epoch": 4.167127018089865,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033630347201436806,
      "loss": 4.738,
      "mean_token_accuracy": 0.22755542248487473,
      "num_tokens": 92896066.0,
      "step": 53560
    },
    {
      "entropy": 5.349680709838867,
      "epoch": 4.167516047461583,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033627680452149873,
      "loss": 4.7576,
      "mean_token_accuracy": 0.22861727476119995,
      "num_tokens": 92904203.0,
      "step": 53565
    },
    {
      "entropy": 5.45750937461853,
      "epoch": 4.167905076833301,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003362501360988514,
      "loss": 4.8241,
      "mean_token_accuracy": 0.2139332264661789,
      "num_tokens": 92913349.0,
      "step": 53570
    },
    {
      "entropy": 5.360860681533813,
      "epoch": 4.1682941062050185,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003362234667468308,
      "loss": 4.7549,
      "mean_token_accuracy": 0.21687950789928437,
      "num_tokens": 92922250.0,
      "step": 53575
    },
    {
      "entropy": 5.3319714069366455,
      "epoch": 4.168683135576736,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003361967964658415,
      "loss": 4.7219,
      "mean_token_accuracy": 0.22402497082948686,
      "num_tokens": 92930310.0,
      "step": 53580
    },
    {
      "entropy": 5.370862197875977,
      "epoch": 4.169072164948454,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003361701252562882,
      "loss": 4.8233,
      "mean_token_accuracy": 0.2208467900753021,
      "num_tokens": 92938483.0,
      "step": 53585
    },
    {
      "entropy": 5.399597024917602,
      "epoch": 4.169461194320172,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033614345311857566,
      "loss": 4.8123,
      "mean_token_accuracy": 0.21261609196662903,
      "num_tokens": 92948121.0,
      "step": 53590
    },
    {
      "entropy": 5.345770692825317,
      "epoch": 4.169850223691888,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003361167800531085,
      "loss": 4.73,
      "mean_token_accuracy": 0.22886659651994706,
      "num_tokens": 92956714.0,
      "step": 53595
    },
    {
      "entropy": 5.400656270980835,
      "epoch": 4.170239253063606,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003360901060602915,
      "loss": 4.8349,
      "mean_token_accuracy": 0.21506787836551666,
      "num_tokens": 92965494.0,
      "step": 53600
    },
    {
      "entropy": 5.376948642730713,
      "epoch": 4.170628282435324,
      "grad_norm": 1.125,
      "learning_rate": 0.00033606343114052936,
      "loss": 4.7835,
      "mean_token_accuracy": 0.222150619328022,
      "num_tokens": 92974909.0,
      "step": 53605
    },
    {
      "entropy": 5.394308757781983,
      "epoch": 4.1710173118070415,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033603675529422685,
      "loss": 4.8198,
      "mean_token_accuracy": 0.2192682832479477,
      "num_tokens": 92983969.0,
      "step": 53610
    },
    {
      "entropy": 5.358304738998413,
      "epoch": 4.171406341178759,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003360100785217887,
      "loss": 4.7665,
      "mean_token_accuracy": 0.22236955910921097,
      "num_tokens": 92992461.0,
      "step": 53615
    },
    {
      "entropy": 5.409924650192261,
      "epoch": 4.171795370550477,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003359834008236198,
      "loss": 4.8436,
      "mean_token_accuracy": 0.20980946868658065,
      "num_tokens": 93000821.0,
      "step": 53620
    },
    {
      "entropy": 5.425667238235474,
      "epoch": 4.172184399922194,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003359567222001248,
      "loss": 4.794,
      "mean_token_accuracy": 0.22910916209220886,
      "num_tokens": 93009092.0,
      "step": 53625
    },
    {
      "entropy": 5.350609874725341,
      "epoch": 4.172573429293911,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003359300426517085,
      "loss": 4.7061,
      "mean_token_accuracy": 0.2254367798566818,
      "num_tokens": 93017435.0,
      "step": 53630
    },
    {
      "entropy": 5.2702055931091305,
      "epoch": 4.172962458665629,
      "grad_norm": 1.125,
      "learning_rate": 0.0003359033621787759,
      "loss": 4.6885,
      "mean_token_accuracy": 0.2324233964085579,
      "num_tokens": 93026209.0,
      "step": 53635
    },
    {
      "entropy": 5.426909589767456,
      "epoch": 4.173351488037347,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003358766807817317,
      "loss": 4.9137,
      "mean_token_accuracy": 0.21116809844970702,
      "num_tokens": 93035512.0,
      "step": 53640
    },
    {
      "entropy": 5.40611629486084,
      "epoch": 4.1737405174090645,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003358499984609807,
      "loss": 4.8119,
      "mean_token_accuracy": 0.21704673916101455,
      "num_tokens": 93043837.0,
      "step": 53645
    },
    {
      "entropy": 5.3912139415740965,
      "epoch": 4.174129546780782,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003358233152169278,
      "loss": 4.7929,
      "mean_token_accuracy": 0.21657272428274155,
      "num_tokens": 93053253.0,
      "step": 53650
    },
    {
      "entropy": 5.395614862442017,
      "epoch": 4.1745185761525,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033579663104997803,
      "loss": 4.795,
      "mean_token_accuracy": 0.22104662358760835,
      "num_tokens": 93061400.0,
      "step": 53655
    },
    {
      "entropy": 5.313544464111328,
      "epoch": 4.174907605524217,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033576994596053606,
      "loss": 4.7035,
      "mean_token_accuracy": 0.2292048841714859,
      "num_tokens": 93070421.0,
      "step": 53660
    },
    {
      "entropy": 5.337964057922363,
      "epoch": 4.175296634895934,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003357432599490069,
      "loss": 4.8215,
      "mean_token_accuracy": 0.2203061267733574,
      "num_tokens": 93079629.0,
      "step": 53665
    },
    {
      "entropy": 5.343481349945068,
      "epoch": 4.175685664267652,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033571657301579547,
      "loss": 4.7341,
      "mean_token_accuracy": 0.22378868013620376,
      "num_tokens": 93087778.0,
      "step": 53670
    },
    {
      "entropy": 5.284874677658081,
      "epoch": 4.17607469363937,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003356898851613067,
      "loss": 4.683,
      "mean_token_accuracy": 0.21865817308425903,
      "num_tokens": 93096450.0,
      "step": 53675
    },
    {
      "entropy": 5.316546297073364,
      "epoch": 4.1764637230110875,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003356631963859455,
      "loss": 4.783,
      "mean_token_accuracy": 0.22254587858915328,
      "num_tokens": 93105225.0,
      "step": 53680
    },
    {
      "entropy": 5.470517635345459,
      "epoch": 4.176852752382805,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003356365066901168,
      "loss": 4.8655,
      "mean_token_accuracy": 0.2184503570199013,
      "num_tokens": 93114386.0,
      "step": 53685
    },
    {
      "entropy": 5.485641002655029,
      "epoch": 4.177241781754523,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033560981607422567,
      "loss": 4.8479,
      "mean_token_accuracy": 0.22222215682268143,
      "num_tokens": 93122900.0,
      "step": 53690
    },
    {
      "entropy": 5.337961912155151,
      "epoch": 4.17763081112624,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033558312453867703,
      "loss": 4.7526,
      "mean_token_accuracy": 0.21666055917739868,
      "num_tokens": 93131542.0,
      "step": 53695
    },
    {
      "entropy": 5.365678167343139,
      "epoch": 4.178019840497957,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033555643208387593,
      "loss": 4.7946,
      "mean_token_accuracy": 0.2182489812374115,
      "num_tokens": 93140128.0,
      "step": 53700
    },
    {
      "entropy": 5.33752555847168,
      "epoch": 4.178408869869675,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033552973871022737,
      "loss": 4.7898,
      "mean_token_accuracy": 0.2201438680291176,
      "num_tokens": 93147966.0,
      "step": 53705
    },
    {
      "entropy": 5.431756639480591,
      "epoch": 4.178797899241393,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033550304441813634,
      "loss": 4.9159,
      "mean_token_accuracy": 0.20920156687498093,
      "num_tokens": 93156915.0,
      "step": 53710
    },
    {
      "entropy": 5.383739852905274,
      "epoch": 4.1791869286131105,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003354763492080079,
      "loss": 4.7606,
      "mean_token_accuracy": 0.22537904977798462,
      "num_tokens": 93165844.0,
      "step": 53715
    },
    {
      "entropy": 5.364845275878906,
      "epoch": 4.179575957984828,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003354496530802472,
      "loss": 4.7742,
      "mean_token_accuracy": 0.2161043554544449,
      "num_tokens": 93174653.0,
      "step": 53720
    },
    {
      "entropy": 5.390312671661377,
      "epoch": 4.179964987356545,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003354229560352592,
      "loss": 4.7963,
      "mean_token_accuracy": 0.22337675392627715,
      "num_tokens": 93183407.0,
      "step": 53725
    },
    {
      "entropy": 5.46422643661499,
      "epoch": 4.180354016728263,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003353962580734489,
      "loss": 4.8527,
      "mean_token_accuracy": 0.22105470001697541,
      "num_tokens": 93192103.0,
      "step": 53730
    },
    {
      "entropy": 5.30750994682312,
      "epoch": 4.18074304609998,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033536955919522167,
      "loss": 4.6999,
      "mean_token_accuracy": 0.2252844676375389,
      "num_tokens": 93200018.0,
      "step": 53735
    },
    {
      "entropy": 5.419745445251465,
      "epoch": 4.181132075471698,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003353428594009824,
      "loss": 4.7868,
      "mean_token_accuracy": 0.22511323988437654,
      "num_tokens": 93207927.0,
      "step": 53740
    },
    {
      "entropy": 5.3652387142181395,
      "epoch": 4.181521104843416,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033531615869113624,
      "loss": 4.771,
      "mean_token_accuracy": 0.2224264308810234,
      "num_tokens": 93217027.0,
      "step": 53745
    },
    {
      "entropy": 5.4007316589355465,
      "epoch": 4.1819101342151335,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033528945706608844,
      "loss": 4.8004,
      "mean_token_accuracy": 0.22059719413518905,
      "num_tokens": 93225415.0,
      "step": 53750
    },
    {
      "entropy": 5.421299934387207,
      "epoch": 4.182299163586851,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033526275452624407,
      "loss": 4.7843,
      "mean_token_accuracy": 0.21150245219469072,
      "num_tokens": 93234054.0,
      "step": 53755
    },
    {
      "entropy": 5.38250207901001,
      "epoch": 4.182688192958568,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033523605107200826,
      "loss": 4.7604,
      "mean_token_accuracy": 0.220413476228714,
      "num_tokens": 93242135.0,
      "step": 53760
    },
    {
      "entropy": 5.299374294281006,
      "epoch": 4.183077222330286,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033520934670378634,
      "loss": 4.6694,
      "mean_token_accuracy": 0.22988593727350234,
      "num_tokens": 93249946.0,
      "step": 53765
    },
    {
      "entropy": 5.472810077667236,
      "epoch": 4.183466251702003,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033518264142198335,
      "loss": 4.906,
      "mean_token_accuracy": 0.2101896196603775,
      "num_tokens": 93258692.0,
      "step": 53770
    },
    {
      "entropy": 5.408917045593261,
      "epoch": 4.183855281073721,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003351559352270047,
      "loss": 4.7715,
      "mean_token_accuracy": 0.2248375177383423,
      "num_tokens": 93267882.0,
      "step": 53775
    },
    {
      "entropy": 5.4407814025878904,
      "epoch": 4.184244310445439,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033512922811925535,
      "loss": 4.903,
      "mean_token_accuracy": 0.2104749083518982,
      "num_tokens": 93276922.0,
      "step": 53780
    },
    {
      "entropy": 5.440193557739258,
      "epoch": 4.1846333398171565,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003351025200991406,
      "loss": 4.8148,
      "mean_token_accuracy": 0.21636549532413482,
      "num_tokens": 93285918.0,
      "step": 53785
    },
    {
      "entropy": 5.4235235214233395,
      "epoch": 4.185022369188874,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003350758111670659,
      "loss": 4.7368,
      "mean_token_accuracy": 0.22986736744642258,
      "num_tokens": 93294116.0,
      "step": 53790
    },
    {
      "entropy": 5.234165143966675,
      "epoch": 4.185411398560591,
      "grad_norm": 1.71875,
      "learning_rate": 0.0003350491013234364,
      "loss": 4.5842,
      "mean_token_accuracy": 0.241888128221035,
      "num_tokens": 93303108.0,
      "step": 53795
    },
    {
      "entropy": 5.328868055343628,
      "epoch": 4.185800427932309,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003350223905686573,
      "loss": 4.722,
      "mean_token_accuracy": 0.22303550243377684,
      "num_tokens": 93311774.0,
      "step": 53800
    },
    {
      "entropy": 5.42891936302185,
      "epoch": 4.186189457304026,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000334995678903134,
      "loss": 4.8092,
      "mean_token_accuracy": 0.22451190054416656,
      "num_tokens": 93320312.0,
      "step": 53805
    },
    {
      "entropy": 5.3681875705719,
      "epoch": 4.186578486675744,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003349689663272718,
      "loss": 4.7244,
      "mean_token_accuracy": 0.22296006083488465,
      "num_tokens": 93328963.0,
      "step": 53810
    },
    {
      "entropy": 5.473629760742187,
      "epoch": 4.186967516047462,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003349422528414759,
      "loss": 4.8343,
      "mean_token_accuracy": 0.21576391011476517,
      "num_tokens": 93337323.0,
      "step": 53815
    },
    {
      "entropy": 5.311782026290894,
      "epoch": 4.1873565454191795,
      "grad_norm": 1.25,
      "learning_rate": 0.0003349155384461517,
      "loss": 4.7578,
      "mean_token_accuracy": 0.21739377379417418,
      "num_tokens": 93345571.0,
      "step": 53820
    },
    {
      "entropy": 5.357076501846313,
      "epoch": 4.187745574790897,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033488882314170463,
      "loss": 4.8018,
      "mean_token_accuracy": 0.21984760612249374,
      "num_tokens": 93355297.0,
      "step": 53825
    },
    {
      "entropy": 5.366396093368531,
      "epoch": 4.188134604162614,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033486210692854,
      "loss": 4.6997,
      "mean_token_accuracy": 0.2271649032831192,
      "num_tokens": 93363885.0,
      "step": 53830
    },
    {
      "entropy": 5.413512802124023,
      "epoch": 4.188523633534332,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003348353898070631,
      "loss": 4.8102,
      "mean_token_accuracy": 0.21603149622678758,
      "num_tokens": 93372231.0,
      "step": 53835
    },
    {
      "entropy": 5.301718807220459,
      "epoch": 4.188912662906049,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003348086717776795,
      "loss": 4.715,
      "mean_token_accuracy": 0.2264656290411949,
      "num_tokens": 93381391.0,
      "step": 53840
    },
    {
      "entropy": 5.403948831558227,
      "epoch": 4.189301692277767,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033478195284079447,
      "loss": 4.7943,
      "mean_token_accuracy": 0.22427505552768706,
      "num_tokens": 93390916.0,
      "step": 53845
    },
    {
      "entropy": 5.436588621139526,
      "epoch": 4.189690721649485,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003347552329968134,
      "loss": 4.8179,
      "mean_token_accuracy": 0.22005128860473633,
      "num_tokens": 93399318.0,
      "step": 53850
    },
    {
      "entropy": 5.409006929397583,
      "epoch": 4.1900797510212024,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003347285122461419,
      "loss": 4.872,
      "mean_token_accuracy": 0.21279677003622055,
      "num_tokens": 93408319.0,
      "step": 53855
    },
    {
      "entropy": 5.300862646102905,
      "epoch": 4.190468780392919,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003347017905891852,
      "loss": 4.7336,
      "mean_token_accuracy": 0.22485802322626114,
      "num_tokens": 93416765.0,
      "step": 53860
    },
    {
      "entropy": 5.402885246276855,
      "epoch": 4.190857809764637,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003346750680263488,
      "loss": 4.7662,
      "mean_token_accuracy": 0.21827685683965684,
      "num_tokens": 93425256.0,
      "step": 53865
    },
    {
      "entropy": 5.421969223022461,
      "epoch": 4.191246839136355,
      "grad_norm": 1.25,
      "learning_rate": 0.0003346483445580384,
      "loss": 4.7693,
      "mean_token_accuracy": 0.22109905183315276,
      "num_tokens": 93433513.0,
      "step": 53870
    },
    {
      "entropy": 5.449046802520752,
      "epoch": 4.191635868508072,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033462162018465917,
      "loss": 4.8902,
      "mean_token_accuracy": 0.2043241187930107,
      "num_tokens": 93441978.0,
      "step": 53875
    },
    {
      "entropy": 5.358885049819946,
      "epoch": 4.19202489787979,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033459489490661676,
      "loss": 4.7333,
      "mean_token_accuracy": 0.22785516381263732,
      "num_tokens": 93450943.0,
      "step": 53880
    },
    {
      "entropy": 5.402722501754761,
      "epoch": 4.192413927251508,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033456816872431673,
      "loss": 4.838,
      "mean_token_accuracy": 0.22170499116182327,
      "num_tokens": 93459408.0,
      "step": 53885
    },
    {
      "entropy": 5.4166364669799805,
      "epoch": 4.192802956623225,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003345414416381645,
      "loss": 4.8286,
      "mean_token_accuracy": 0.22364921271800994,
      "num_tokens": 93466959.0,
      "step": 53890
    },
    {
      "entropy": 5.431374263763428,
      "epoch": 4.193191985994942,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033451471364856565,
      "loss": 4.8368,
      "mean_token_accuracy": 0.21590153127908707,
      "num_tokens": 93475584.0,
      "step": 53895
    },
    {
      "entropy": 5.3815165042877195,
      "epoch": 4.19358101536666,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033448798475592585,
      "loss": 4.8888,
      "mean_token_accuracy": 0.22040918320417405,
      "num_tokens": 93484081.0,
      "step": 53900
    },
    {
      "entropy": 5.4178613185882565,
      "epoch": 4.193970044738378,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003344612549606505,
      "loss": 4.7908,
      "mean_token_accuracy": 0.2203831657767296,
      "num_tokens": 93492499.0,
      "step": 53905
    },
    {
      "entropy": 5.328008127212525,
      "epoch": 4.194359074110095,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003344345242631453,
      "loss": 4.7443,
      "mean_token_accuracy": 0.2275153651833534,
      "num_tokens": 93500676.0,
      "step": 53910
    },
    {
      "entropy": 5.357608032226563,
      "epoch": 4.194748103481813,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003344077926638157,
      "loss": 4.8141,
      "mean_token_accuracy": 0.2158311665058136,
      "num_tokens": 93509486.0,
      "step": 53915
    },
    {
      "entropy": 5.349962902069092,
      "epoch": 4.195137132853531,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033438106016306753,
      "loss": 4.6808,
      "mean_token_accuracy": 0.22268983572721482,
      "num_tokens": 93517864.0,
      "step": 53920
    },
    {
      "entropy": 5.382796955108643,
      "epoch": 4.195526162225248,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003343543267613062,
      "loss": 4.7904,
      "mean_token_accuracy": 0.21313904076814652,
      "num_tokens": 93526452.0,
      "step": 53925
    },
    {
      "entropy": 5.369426441192627,
      "epoch": 4.195915191596965,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003343275924589374,
      "loss": 4.7939,
      "mean_token_accuracy": 0.22036897391080856,
      "num_tokens": 93535114.0,
      "step": 53930
    },
    {
      "entropy": 5.291434717178345,
      "epoch": 4.196304220968683,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033430085725636685,
      "loss": 4.7738,
      "mean_token_accuracy": 0.2182529389858246,
      "num_tokens": 93543844.0,
      "step": 53935
    },
    {
      "entropy": 5.353103065490723,
      "epoch": 4.196693250340401,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003342741211540002,
      "loss": 4.7266,
      "mean_token_accuracy": 0.21847303956747055,
      "num_tokens": 93552439.0,
      "step": 53940
    },
    {
      "entropy": 5.377645492553711,
      "epoch": 4.197082279712118,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003342473841522431,
      "loss": 4.8568,
      "mean_token_accuracy": 0.21479688137769698,
      "num_tokens": 93561655.0,
      "step": 53945
    },
    {
      "entropy": 5.4018879413604735,
      "epoch": 4.197471309083836,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033422064625150124,
      "loss": 4.7929,
      "mean_token_accuracy": 0.21797087043523788,
      "num_tokens": 93570070.0,
      "step": 53950
    },
    {
      "entropy": 5.344179677963257,
      "epoch": 4.197860338455554,
      "grad_norm": 1.125,
      "learning_rate": 0.0003341939074521804,
      "loss": 4.7414,
      "mean_token_accuracy": 0.22746467739343643,
      "num_tokens": 93579052.0,
      "step": 53955
    },
    {
      "entropy": 5.374979066848755,
      "epoch": 4.1982493678272705,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003341671677546861,
      "loss": 4.7494,
      "mean_token_accuracy": 0.224351404607296,
      "num_tokens": 93588873.0,
      "step": 53960
    },
    {
      "entropy": 5.330835247039795,
      "epoch": 4.198638397198988,
      "grad_norm": 1.125,
      "learning_rate": 0.00033414042715942423,
      "loss": 4.7666,
      "mean_token_accuracy": 0.22554274648427963,
      "num_tokens": 93597132.0,
      "step": 53965
    },
    {
      "entropy": 5.388536882400513,
      "epoch": 4.199027426570706,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003341136856668005,
      "loss": 4.7338,
      "mean_token_accuracy": 0.22246773689985275,
      "num_tokens": 93604881.0,
      "step": 53970
    },
    {
      "entropy": 5.377799415588379,
      "epoch": 4.199416455942424,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003340869432772207,
      "loss": 4.7718,
      "mean_token_accuracy": 0.21984119713306427,
      "num_tokens": 93612935.0,
      "step": 53975
    },
    {
      "entropy": 5.399924802780151,
      "epoch": 4.199805485314141,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033406019999109065,
      "loss": 4.8131,
      "mean_token_accuracy": 0.22094771564006804,
      "num_tokens": 93621551.0,
      "step": 53980
    },
    {
      "entropy": 5.374249267578125,
      "epoch": 4.200194514685859,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033403345580881606,
      "loss": 4.7782,
      "mean_token_accuracy": 0.22279789447784423,
      "num_tokens": 93630275.0,
      "step": 53985
    },
    {
      "entropy": 5.370848178863525,
      "epoch": 4.200583544057577,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033400671073080274,
      "loss": 4.8331,
      "mean_token_accuracy": 0.2096216171979904,
      "num_tokens": 93639823.0,
      "step": 53990
    },
    {
      "entropy": 5.472932052612305,
      "epoch": 4.2009725734292935,
      "grad_norm": 1.25,
      "learning_rate": 0.0003339799647574564,
      "loss": 4.7637,
      "mean_token_accuracy": 0.2204767420887947,
      "num_tokens": 93647779.0,
      "step": 53995
    },
    {
      "entropy": 5.4163245677948,
      "epoch": 4.201361602801011,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003339532178891831,
      "loss": 4.7986,
      "mean_token_accuracy": 0.21846953332424163,
      "num_tokens": 93656282.0,
      "step": 54000
    },
    {
      "epoch": 4.201361602801011,
      "eval_entropy": 5.1061626067175325,
      "eval_loss": 4.952186584472656,
      "eval_mean_token_accuracy": 0.2207359118915958,
      "eval_num_tokens": 93656282.0,
      "eval_runtime": 28.8446,
      "eval_samples_per_second": 1440.757,
      "eval_steps_per_second": 180.103,
      "step": 54000
    },
    {
      "entropy": 5.413721227645874,
      "epoch": 4.201750632172729,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003339264701263885,
      "loss": 4.8783,
      "mean_token_accuracy": 0.21241841912269593,
      "num_tokens": 93664593.0,
      "step": 54005
    },
    {
      "entropy": 5.322483015060425,
      "epoch": 4.202139661544447,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003338997214694784,
      "loss": 4.6668,
      "mean_token_accuracy": 0.22761227041482926,
      "num_tokens": 93672608.0,
      "step": 54010
    },
    {
      "entropy": 5.394267129898071,
      "epoch": 4.202528690916164,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033387297191885896,
      "loss": 4.8586,
      "mean_token_accuracy": 0.21157996207475663,
      "num_tokens": 93681635.0,
      "step": 54015
    },
    {
      "entropy": 5.404514217376709,
      "epoch": 4.202917720287882,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003338462214749358,
      "loss": 4.787,
      "mean_token_accuracy": 0.2195822149515152,
      "num_tokens": 93690346.0,
      "step": 54020
    },
    {
      "entropy": 5.411816167831421,
      "epoch": 4.2033067496596,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003338194701381148,
      "loss": 4.8318,
      "mean_token_accuracy": 0.2175027996301651,
      "num_tokens": 93698629.0,
      "step": 54025
    },
    {
      "entropy": 5.4680705070495605,
      "epoch": 4.2036957790313165,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033379271790880203,
      "loss": 4.8792,
      "mean_token_accuracy": 0.2168081820011139,
      "num_tokens": 93707173.0,
      "step": 54030
    },
    {
      "entropy": 5.37090106010437,
      "epoch": 4.204084808403034,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003337659647874034,
      "loss": 4.7643,
      "mean_token_accuracy": 0.22579385042190553,
      "num_tokens": 93715999.0,
      "step": 54035
    },
    {
      "entropy": 5.406820631027221,
      "epoch": 4.204473837774752,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003337392107743248,
      "loss": 4.8112,
      "mean_token_accuracy": 0.22400469183921815,
      "num_tokens": 93725446.0,
      "step": 54040
    },
    {
      "entropy": 5.440329122543335,
      "epoch": 4.20486286714647,
      "grad_norm": 1.125,
      "learning_rate": 0.0003337124558699721,
      "loss": 4.7644,
      "mean_token_accuracy": 0.22213332056999208,
      "num_tokens": 93733652.0,
      "step": 54045
    },
    {
      "entropy": 5.381520223617554,
      "epoch": 4.205251896518187,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033368570007475133,
      "loss": 4.7646,
      "mean_token_accuracy": 0.2228358954191208,
      "num_tokens": 93742007.0,
      "step": 54050
    },
    {
      "entropy": 5.438037967681884,
      "epoch": 4.205640925889905,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003336589433890685,
      "loss": 4.889,
      "mean_token_accuracy": 0.21007149964570998,
      "num_tokens": 93751009.0,
      "step": 54055
    },
    {
      "entropy": 5.271495962142945,
      "epoch": 4.206029955261622,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033363218581332956,
      "loss": 4.7069,
      "mean_token_accuracy": 0.22716987133026123,
      "num_tokens": 93760412.0,
      "step": 54060
    },
    {
      "entropy": 5.3505784034729,
      "epoch": 4.2064189846333395,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003336054273479405,
      "loss": 4.7254,
      "mean_token_accuracy": 0.22015329152345658,
      "num_tokens": 93768786.0,
      "step": 54065
    },
    {
      "entropy": 5.4974394798278805,
      "epoch": 4.206808014005057,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033357866799330743,
      "loss": 4.9267,
      "mean_token_accuracy": 0.20547130405902864,
      "num_tokens": 93778537.0,
      "step": 54070
    },
    {
      "entropy": 5.311965751647949,
      "epoch": 4.207197043376775,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033355190774983626,
      "loss": 4.6907,
      "mean_token_accuracy": 0.22096313387155533,
      "num_tokens": 93786742.0,
      "step": 54075
    },
    {
      "entropy": 5.300377130508423,
      "epoch": 4.207586072748493,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033352514661793306,
      "loss": 4.7779,
      "mean_token_accuracy": 0.22542177885770798,
      "num_tokens": 93795837.0,
      "step": 54080
    },
    {
      "entropy": 5.310730075836181,
      "epoch": 4.20797510212021,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033349838459800394,
      "loss": 4.7743,
      "mean_token_accuracy": 0.22179348170757293,
      "num_tokens": 93805080.0,
      "step": 54085
    },
    {
      "entropy": 5.543870210647583,
      "epoch": 4.208364131491928,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000333471621690455,
      "loss": 4.8939,
      "mean_token_accuracy": 0.20702438205480575,
      "num_tokens": 93813934.0,
      "step": 54090
    },
    {
      "entropy": 5.437402629852295,
      "epoch": 4.208753160863645,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003334448578956921,
      "loss": 4.8325,
      "mean_token_accuracy": 0.21717978566884993,
      "num_tokens": 93822728.0,
      "step": 54095
    },
    {
      "entropy": 5.397933530807495,
      "epoch": 4.2091421902353625,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003334180932141217,
      "loss": 4.7437,
      "mean_token_accuracy": 0.2236942544579506,
      "num_tokens": 93831649.0,
      "step": 54100
    },
    {
      "entropy": 5.429098606109619,
      "epoch": 4.20953121960708,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003333913276461497,
      "loss": 4.8341,
      "mean_token_accuracy": 0.21753476560115814,
      "num_tokens": 93840634.0,
      "step": 54105
    },
    {
      "entropy": 5.432504415512085,
      "epoch": 4.209920248978798,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033336456119218214,
      "loss": 4.8095,
      "mean_token_accuracy": 0.22509506046772004,
      "num_tokens": 93848332.0,
      "step": 54110
    },
    {
      "entropy": 5.308867979049682,
      "epoch": 4.210309278350516,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033333779385262525,
      "loss": 4.7016,
      "mean_token_accuracy": 0.22646436393260955,
      "num_tokens": 93856867.0,
      "step": 54115
    },
    {
      "entropy": 5.271472501754761,
      "epoch": 4.210698307722233,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033331102562788527,
      "loss": 4.7271,
      "mean_token_accuracy": 0.22100264728069305,
      "num_tokens": 93864803.0,
      "step": 54120
    },
    {
      "entropy": 5.394699478149414,
      "epoch": 4.211087337093951,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003332842565183682,
      "loss": 4.817,
      "mean_token_accuracy": 0.21922930628061293,
      "num_tokens": 93872804.0,
      "step": 54125
    },
    {
      "entropy": 5.3162439346313475,
      "epoch": 4.211476366465668,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033325748652448037,
      "loss": 4.7067,
      "mean_token_accuracy": 0.22851113677024842,
      "num_tokens": 93881147.0,
      "step": 54130
    },
    {
      "entropy": 5.460040044784546,
      "epoch": 4.2118653958373855,
      "grad_norm": 1.125,
      "learning_rate": 0.00033323071564662796,
      "loss": 4.8435,
      "mean_token_accuracy": 0.21293179243803023,
      "num_tokens": 93890062.0,
      "step": 54135
    },
    {
      "entropy": 5.369645833969116,
      "epoch": 4.212254425209103,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033320394388521707,
      "loss": 4.7584,
      "mean_token_accuracy": 0.22452980428934097,
      "num_tokens": 93899032.0,
      "step": 54140
    },
    {
      "entropy": 5.433514785766602,
      "epoch": 4.212643454580821,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000333177171240654,
      "loss": 4.8225,
      "mean_token_accuracy": 0.22203367352485656,
      "num_tokens": 93907977.0,
      "step": 54145
    },
    {
      "entropy": 5.319333600997925,
      "epoch": 4.213032483952539,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033315039771334494,
      "loss": 4.7437,
      "mean_token_accuracy": 0.22303917557001113,
      "num_tokens": 93915818.0,
      "step": 54150
    },
    {
      "entropy": 5.351129674911499,
      "epoch": 4.213421513324256,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003331236233036961,
      "loss": 4.8073,
      "mean_token_accuracy": 0.2211129143834114,
      "num_tokens": 93924628.0,
      "step": 54155
    },
    {
      "entropy": 5.347161197662354,
      "epoch": 4.213810542695974,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003330968480121138,
      "loss": 4.7662,
      "mean_token_accuracy": 0.2211957648396492,
      "num_tokens": 93933173.0,
      "step": 54160
    },
    {
      "entropy": 5.313761472702026,
      "epoch": 4.214199572067691,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003330700718390043,
      "loss": 4.7138,
      "mean_token_accuracy": 0.23115722984075546,
      "num_tokens": 93941797.0,
      "step": 54165
    },
    {
      "entropy": 5.395924186706543,
      "epoch": 4.2145886014394085,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033304329478477395,
      "loss": 4.7917,
      "mean_token_accuracy": 0.22322095781564713,
      "num_tokens": 93950668.0,
      "step": 54170
    },
    {
      "entropy": 5.465241527557373,
      "epoch": 4.214977630811126,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033301651684982894,
      "loss": 4.8911,
      "mean_token_accuracy": 0.21785181760787964,
      "num_tokens": 93959508.0,
      "step": 54175
    },
    {
      "entropy": 5.393065547943115,
      "epoch": 4.215366660182844,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003329897380345757,
      "loss": 4.8347,
      "mean_token_accuracy": 0.22152369171380998,
      "num_tokens": 93967762.0,
      "step": 54180
    },
    {
      "entropy": 5.4191831111907955,
      "epoch": 4.215755689554562,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003329629583394204,
      "loss": 4.8411,
      "mean_token_accuracy": 0.21768186688423158,
      "num_tokens": 93976139.0,
      "step": 54185
    },
    {
      "entropy": 5.337548494338989,
      "epoch": 4.216144718926279,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003329361777647695,
      "loss": 4.7687,
      "mean_token_accuracy": 0.2307794988155365,
      "num_tokens": 93984833.0,
      "step": 54190
    },
    {
      "entropy": 5.388611316680908,
      "epoch": 4.216533748297996,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003329093963110293,
      "loss": 4.8239,
      "mean_token_accuracy": 0.21922485828399657,
      "num_tokens": 93992868.0,
      "step": 54195
    },
    {
      "entropy": 5.427775144577026,
      "epoch": 4.216922777669714,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033288261397860613,
      "loss": 4.7516,
      "mean_token_accuracy": 0.22263300120830537,
      "num_tokens": 94000832.0,
      "step": 54200
    },
    {
      "entropy": 5.436736345291138,
      "epoch": 4.2173118070414315,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033285583076790646,
      "loss": 4.8423,
      "mean_token_accuracy": 0.21735973805189132,
      "num_tokens": 94009511.0,
      "step": 54205
    },
    {
      "entropy": 5.334018039703369,
      "epoch": 4.217700836413149,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033282904667933675,
      "loss": 4.7058,
      "mean_token_accuracy": 0.22501945346593857,
      "num_tokens": 94017431.0,
      "step": 54210
    },
    {
      "entropy": 5.34884991645813,
      "epoch": 4.218089865784867,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003328022617133032,
      "loss": 4.8155,
      "mean_token_accuracy": 0.22382503300905227,
      "num_tokens": 94025818.0,
      "step": 54215
    },
    {
      "entropy": 5.419087648391724,
      "epoch": 4.218478895156585,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003327754758702124,
      "loss": 4.8293,
      "mean_token_accuracy": 0.2125893473625183,
      "num_tokens": 94033871.0,
      "step": 54220
    },
    {
      "entropy": 5.362149000167847,
      "epoch": 4.218867924528302,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033274868915047064,
      "loss": 4.8126,
      "mean_token_accuracy": 0.21389147490262986,
      "num_tokens": 94042377.0,
      "step": 54225
    },
    {
      "entropy": 5.440541410446167,
      "epoch": 4.219256953900019,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033272190155448446,
      "loss": 4.8686,
      "mean_token_accuracy": 0.20961861461400985,
      "num_tokens": 94051070.0,
      "step": 54230
    },
    {
      "entropy": 5.331201791763306,
      "epoch": 4.219645983271737,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003326951130826603,
      "loss": 4.7061,
      "mean_token_accuracy": 0.21504910290241241,
      "num_tokens": 94059239.0,
      "step": 54235
    },
    {
      "entropy": 5.331786155700684,
      "epoch": 4.2200350126434545,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033266832373540463,
      "loss": 4.6967,
      "mean_token_accuracy": 0.23223818391561507,
      "num_tokens": 94068104.0,
      "step": 54240
    },
    {
      "entropy": 5.326653337478637,
      "epoch": 4.220424042015172,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003326415335131239,
      "loss": 4.7567,
      "mean_token_accuracy": 0.22703139036893843,
      "num_tokens": 94077228.0,
      "step": 54245
    },
    {
      "entropy": 5.412949085235596,
      "epoch": 4.22081307138689,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033261474241622475,
      "loss": 4.8261,
      "mean_token_accuracy": 0.21173980981111526,
      "num_tokens": 94085504.0,
      "step": 54250
    },
    {
      "entropy": 5.411124801635742,
      "epoch": 4.221202100758608,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003325879504451135,
      "loss": 4.7354,
      "mean_token_accuracy": 0.22245051413774491,
      "num_tokens": 94094159.0,
      "step": 54255
    },
    {
      "entropy": 5.357756662368774,
      "epoch": 4.221591130130324,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033256115760019677,
      "loss": 4.8223,
      "mean_token_accuracy": 0.21488324254751207,
      "num_tokens": 94102936.0,
      "step": 54260
    },
    {
      "entropy": 5.390393924713135,
      "epoch": 4.221980159502042,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033253436388188106,
      "loss": 4.8609,
      "mean_token_accuracy": 0.21579848378896713,
      "num_tokens": 94112542.0,
      "step": 54265
    },
    {
      "entropy": 5.4438127994537355,
      "epoch": 4.22236918887376,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000332507569290573,
      "loss": 4.8349,
      "mean_token_accuracy": 0.21339676529169083,
      "num_tokens": 94121835.0,
      "step": 54270
    },
    {
      "entropy": 5.407558298110962,
      "epoch": 4.2227582182454775,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033248077382667913,
      "loss": 4.7645,
      "mean_token_accuracy": 0.21629995107650757,
      "num_tokens": 94130862.0,
      "step": 54275
    },
    {
      "entropy": 5.363882207870484,
      "epoch": 4.223147247617195,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000332453977490606,
      "loss": 4.6618,
      "mean_token_accuracy": 0.23720863759517669,
      "num_tokens": 94138824.0,
      "step": 54280
    },
    {
      "entropy": 5.327428245544434,
      "epoch": 4.223536276988913,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003324271802827602,
      "loss": 4.7693,
      "mean_token_accuracy": 0.21825953125953673,
      "num_tokens": 94147454.0,
      "step": 54285
    },
    {
      "entropy": 5.4311034202575685,
      "epoch": 4.223925306360631,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003324003822035483,
      "loss": 4.8164,
      "mean_token_accuracy": 0.2180324003100395,
      "num_tokens": 94155824.0,
      "step": 54290
    },
    {
      "entropy": 5.402398109436035,
      "epoch": 4.224314335732347,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003323735832533771,
      "loss": 4.7758,
      "mean_token_accuracy": 0.2224586397409439,
      "num_tokens": 94164698.0,
      "step": 54295
    },
    {
      "entropy": 5.294981861114502,
      "epoch": 4.224703365104065,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033234678343265297,
      "loss": 4.7065,
      "mean_token_accuracy": 0.2342843860387802,
      "num_tokens": 94172843.0,
      "step": 54300
    },
    {
      "entropy": 5.351537561416626,
      "epoch": 4.225092394475783,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003323199827417827,
      "loss": 4.7549,
      "mean_token_accuracy": 0.21461116671562194,
      "num_tokens": 94181318.0,
      "step": 54305
    },
    {
      "entropy": 5.427299642562867,
      "epoch": 4.2254814238475005,
      "grad_norm": 1.25,
      "learning_rate": 0.00033229318118117293,
      "loss": 4.8931,
      "mean_token_accuracy": 0.2248120576143265,
      "num_tokens": 94189704.0,
      "step": 54310
    },
    {
      "entropy": 5.422319555282593,
      "epoch": 4.225870453219218,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033226637875123026,
      "loss": 4.7451,
      "mean_token_accuracy": 0.2226056694984436,
      "num_tokens": 94197893.0,
      "step": 54315
    },
    {
      "entropy": 5.327775287628174,
      "epoch": 4.226259482590936,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033223957545236156,
      "loss": 4.7249,
      "mean_token_accuracy": 0.23564747720956802,
      "num_tokens": 94207120.0,
      "step": 54320
    },
    {
      "entropy": 5.465853881835938,
      "epoch": 4.226648511962654,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003322127712849733,
      "loss": 4.7893,
      "mean_token_accuracy": 0.21810625791549682,
      "num_tokens": 94215992.0,
      "step": 54325
    },
    {
      "entropy": 5.382071685791016,
      "epoch": 4.22703754133437,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003321859662494724,
      "loss": 4.7878,
      "mean_token_accuracy": 0.21572877764701842,
      "num_tokens": 94224807.0,
      "step": 54330
    },
    {
      "entropy": 5.398567152023316,
      "epoch": 4.227426570706088,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003321591603462654,
      "loss": 4.8435,
      "mean_token_accuracy": 0.21412058025598527,
      "num_tokens": 94233339.0,
      "step": 54335
    },
    {
      "entropy": 5.371603107452392,
      "epoch": 4.227815600077806,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003321323535757592,
      "loss": 4.693,
      "mean_token_accuracy": 0.22949099987745286,
      "num_tokens": 94241269.0,
      "step": 54340
    },
    {
      "entropy": 5.346029281616211,
      "epoch": 4.2282046294495235,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003321055459383604,
      "loss": 4.7916,
      "mean_token_accuracy": 0.21783567667007447,
      "num_tokens": 94250259.0,
      "step": 54345
    },
    {
      "entropy": 5.331884908676147,
      "epoch": 4.228593658821241,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033207873743447587,
      "loss": 4.6972,
      "mean_token_accuracy": 0.23068632930517197,
      "num_tokens": 94259021.0,
      "step": 54350
    },
    {
      "entropy": 5.450727558135986,
      "epoch": 4.228982688192959,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033205192806451227,
      "loss": 4.8254,
      "mean_token_accuracy": 0.2157275676727295,
      "num_tokens": 94267641.0,
      "step": 54355
    },
    {
      "entropy": 5.4195441722869875,
      "epoch": 4.229371717564677,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033202511782887655,
      "loss": 4.9035,
      "mean_token_accuracy": 0.2134892076253891,
      "num_tokens": 94276779.0,
      "step": 54360
    },
    {
      "entropy": 5.4301167011260985,
      "epoch": 4.229760746936393,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033199830672797545,
      "loss": 4.8605,
      "mean_token_accuracy": 0.21585453003644944,
      "num_tokens": 94285642.0,
      "step": 54365
    },
    {
      "entropy": 5.392042541503907,
      "epoch": 4.230149776308111,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003319714947622157,
      "loss": 4.7109,
      "mean_token_accuracy": 0.22836248874664306,
      "num_tokens": 94293994.0,
      "step": 54370
    },
    {
      "entropy": 5.318151903152466,
      "epoch": 4.230538805679829,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003319446819320043,
      "loss": 4.7167,
      "mean_token_accuracy": 0.22675875723361968,
      "num_tokens": 94302661.0,
      "step": 54375
    },
    {
      "entropy": 5.3777501583099365,
      "epoch": 4.2309278350515465,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003319178682377479,
      "loss": 4.8315,
      "mean_token_accuracy": 0.21500447243452073,
      "num_tokens": 94311996.0,
      "step": 54380
    },
    {
      "entropy": 5.349651050567627,
      "epoch": 4.231316864423264,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033189105367985343,
      "loss": 4.7115,
      "mean_token_accuracy": 0.22938310652971267,
      "num_tokens": 94320123.0,
      "step": 54385
    },
    {
      "entropy": 5.395993041992187,
      "epoch": 4.231705893794982,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033186423825872787,
      "loss": 4.7784,
      "mean_token_accuracy": 0.2277523994445801,
      "num_tokens": 94328027.0,
      "step": 54390
    },
    {
      "entropy": 5.412885236740112,
      "epoch": 4.232094923166699,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003318374219747779,
      "loss": 4.8642,
      "mean_token_accuracy": 0.21883898824453354,
      "num_tokens": 94337740.0,
      "step": 54395
    },
    {
      "entropy": 5.29658727645874,
      "epoch": 4.232483952538416,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033181060482841054,
      "loss": 4.6799,
      "mean_token_accuracy": 0.22856923788785935,
      "num_tokens": 94346203.0,
      "step": 54400
    },
    {
      "entropy": 5.31383261680603,
      "epoch": 4.232872981910134,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003317837868200327,
      "loss": 4.7529,
      "mean_token_accuracy": 0.22265704721212387,
      "num_tokens": 94354690.0,
      "step": 54405
    },
    {
      "entropy": 5.408491516113282,
      "epoch": 4.233262011281852,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033175696795005133,
      "loss": 4.838,
      "mean_token_accuracy": 0.21839593201875687,
      "num_tokens": 94363102.0,
      "step": 54410
    },
    {
      "entropy": 5.499411630630493,
      "epoch": 4.2336510406535695,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003317301482188732,
      "loss": 4.7604,
      "mean_token_accuracy": 0.2193394973874092,
      "num_tokens": 94370937.0,
      "step": 54415
    },
    {
      "entropy": 5.417542743682861,
      "epoch": 4.234040070025287,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003317033276269054,
      "loss": 4.8406,
      "mean_token_accuracy": 0.21558397859334946,
      "num_tokens": 94379746.0,
      "step": 54420
    },
    {
      "entropy": 5.269739294052124,
      "epoch": 4.234429099397005,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033167650617455484,
      "loss": 4.7195,
      "mean_token_accuracy": 0.23051675111055375,
      "num_tokens": 94388835.0,
      "step": 54425
    },
    {
      "entropy": 5.3575685024261475,
      "epoch": 4.234818128768722,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033164968386222856,
      "loss": 4.7005,
      "mean_token_accuracy": 0.22701139450073243,
      "num_tokens": 94397075.0,
      "step": 54430
    },
    {
      "entropy": 5.469365406036377,
      "epoch": 4.235207158140439,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003316228606903334,
      "loss": 4.8031,
      "mean_token_accuracy": 0.21975116729736327,
      "num_tokens": 94405554.0,
      "step": 54435
    },
    {
      "entropy": 5.362374973297119,
      "epoch": 4.235596187512157,
      "grad_norm": 1.25,
      "learning_rate": 0.00033159603665927655,
      "loss": 4.7746,
      "mean_token_accuracy": 0.21678227484226226,
      "num_tokens": 94414187.0,
      "step": 54440
    },
    {
      "entropy": 5.397217893600464,
      "epoch": 4.235985216883875,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033156921176946496,
      "loss": 4.8332,
      "mean_token_accuracy": 0.22247289419174193,
      "num_tokens": 94422756.0,
      "step": 54445
    },
    {
      "entropy": 5.355980634689331,
      "epoch": 4.2363742462555924,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003315423860213057,
      "loss": 4.7547,
      "mean_token_accuracy": 0.2213354751467705,
      "num_tokens": 94431562.0,
      "step": 54450
    },
    {
      "entropy": 5.424383544921875,
      "epoch": 4.23676327562731,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003315155594152055,
      "loss": 4.8548,
      "mean_token_accuracy": 0.215184023976326,
      "num_tokens": 94440452.0,
      "step": 54455
    },
    {
      "entropy": 5.423150539398193,
      "epoch": 4.237152304999028,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003314887319515717,
      "loss": 4.8323,
      "mean_token_accuracy": 0.21593841016292573,
      "num_tokens": 94448966.0,
      "step": 54460
    },
    {
      "entropy": 5.403658437728882,
      "epoch": 4.237541334370745,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003314619036308113,
      "loss": 4.7072,
      "mean_token_accuracy": 0.22892588824033738,
      "num_tokens": 94457190.0,
      "step": 54465
    },
    {
      "entropy": 5.349455976486206,
      "epoch": 4.237930363742462,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033143507445333145,
      "loss": 4.7612,
      "mean_token_accuracy": 0.22519470900297164,
      "num_tokens": 94465799.0,
      "step": 54470
    },
    {
      "entropy": 5.307358074188232,
      "epoch": 4.23831939311418,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033140824441953913,
      "loss": 4.7523,
      "mean_token_accuracy": 0.22900066077709197,
      "num_tokens": 94474544.0,
      "step": 54475
    },
    {
      "entropy": 5.456830167770386,
      "epoch": 4.238708422485898,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033138141352984145,
      "loss": 4.8812,
      "mean_token_accuracy": 0.22218739688396455,
      "num_tokens": 94482884.0,
      "step": 54480
    },
    {
      "entropy": 5.418743896484375,
      "epoch": 4.239097451857615,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033135458178464565,
      "loss": 4.7482,
      "mean_token_accuracy": 0.21810485869646073,
      "num_tokens": 94492096.0,
      "step": 54485
    },
    {
      "entropy": 5.36443338394165,
      "epoch": 4.239486481229333,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003313277491843587,
      "loss": 4.7302,
      "mean_token_accuracy": 0.2224389061331749,
      "num_tokens": 94499945.0,
      "step": 54490
    },
    {
      "entropy": 5.421720886230469,
      "epoch": 4.239875510601051,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033130091572938786,
      "loss": 4.8218,
      "mean_token_accuracy": 0.22085587978363036,
      "num_tokens": 94508200.0,
      "step": 54495
    },
    {
      "entropy": 5.327809000015259,
      "epoch": 4.240264539972768,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003312740814201402,
      "loss": 4.748,
      "mean_token_accuracy": 0.22709480673074722,
      "num_tokens": 94516876.0,
      "step": 54500
    },
    {
      "entropy": 5.350952768325806,
      "epoch": 4.240653569344485,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003312472462570229,
      "loss": 4.7004,
      "mean_token_accuracy": 0.2274550601840019,
      "num_tokens": 94525535.0,
      "step": 54505
    },
    {
      "entropy": 5.332509136199951,
      "epoch": 4.241042598716203,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033122041024044325,
      "loss": 4.7452,
      "mean_token_accuracy": 0.2222794249653816,
      "num_tokens": 94533596.0,
      "step": 54510
    },
    {
      "entropy": 5.373179626464844,
      "epoch": 4.241431628087921,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033119357337080827,
      "loss": 4.7521,
      "mean_token_accuracy": 0.2288521945476532,
      "num_tokens": 94542696.0,
      "step": 54515
    },
    {
      "entropy": 5.392169618606568,
      "epoch": 4.241820657459638,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003311667356485253,
      "loss": 4.7703,
      "mean_token_accuracy": 0.22014668732881545,
      "num_tokens": 94551724.0,
      "step": 54520
    },
    {
      "entropy": 5.330758762359619,
      "epoch": 4.242209686831356,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003311398970740015,
      "loss": 4.7103,
      "mean_token_accuracy": 0.2214592769742012,
      "num_tokens": 94559603.0,
      "step": 54525
    },
    {
      "entropy": 5.340587568283081,
      "epoch": 4.242598716203073,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003311130576476441,
      "loss": 4.7519,
      "mean_token_accuracy": 0.231382854282856,
      "num_tokens": 94568726.0,
      "step": 54530
    },
    {
      "entropy": 5.419903945922852,
      "epoch": 4.242987745574791,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003310862173698604,
      "loss": 4.7492,
      "mean_token_accuracy": 0.22232337296009064,
      "num_tokens": 94577224.0,
      "step": 54535
    },
    {
      "entropy": 5.3452558517456055,
      "epoch": 4.243376774946508,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003310593762410576,
      "loss": 4.785,
      "mean_token_accuracy": 0.2187017947435379,
      "num_tokens": 94586092.0,
      "step": 54540
    },
    {
      "entropy": 5.425574541091919,
      "epoch": 4.243765804318226,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000331032534261643,
      "loss": 4.9049,
      "mean_token_accuracy": 0.21400449872016908,
      "num_tokens": 94594874.0,
      "step": 54545
    },
    {
      "entropy": 5.376405239105225,
      "epoch": 4.244154833689944,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003310056914320239,
      "loss": 4.6793,
      "mean_token_accuracy": 0.23173997849225997,
      "num_tokens": 94604088.0,
      "step": 54550
    },
    {
      "entropy": 5.460483121871948,
      "epoch": 4.244543863061661,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033097884775260754,
      "loss": 4.8238,
      "mean_token_accuracy": 0.21691634505987167,
      "num_tokens": 94613046.0,
      "step": 54555
    },
    {
      "entropy": 5.379615116119385,
      "epoch": 4.244932892433379,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033095200322380124,
      "loss": 4.6577,
      "mean_token_accuracy": 0.23409704864025116,
      "num_tokens": 94621423.0,
      "step": 54560
    },
    {
      "entropy": 5.358602142333984,
      "epoch": 4.245321921805096,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033092515784601245,
      "loss": 4.7892,
      "mean_token_accuracy": 0.21795368641614915,
      "num_tokens": 94630499.0,
      "step": 54565
    },
    {
      "entropy": 5.430104112625122,
      "epoch": 4.245710951176814,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003308983116196483,
      "loss": 4.8234,
      "mean_token_accuracy": 0.21049887090921401,
      "num_tokens": 94639347.0,
      "step": 54570
    },
    {
      "entropy": 5.450504732131958,
      "epoch": 4.246099980548531,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003308714645451163,
      "loss": 4.803,
      "mean_token_accuracy": 0.22142817825078964,
      "num_tokens": 94648242.0,
      "step": 54575
    },
    {
      "entropy": 5.362428379058838,
      "epoch": 4.246489009920249,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003308446166228237,
      "loss": 4.7603,
      "mean_token_accuracy": 0.2240766316652298,
      "num_tokens": 94657534.0,
      "step": 54580
    },
    {
      "entropy": 5.375988149642945,
      "epoch": 4.246878039291967,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033081776785317803,
      "loss": 4.7821,
      "mean_token_accuracy": 0.2171188622713089,
      "num_tokens": 94666552.0,
      "step": 54585
    },
    {
      "entropy": 5.325899076461792,
      "epoch": 4.247267068663684,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033079091823658645,
      "loss": 4.6542,
      "mean_token_accuracy": 0.22637958079576492,
      "num_tokens": 94675577.0,
      "step": 54590
    },
    {
      "entropy": 5.3656309127807615,
      "epoch": 4.247656098035401,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003307640677734566,
      "loss": 4.7627,
      "mean_token_accuracy": 0.21637412756681443,
      "num_tokens": 94684119.0,
      "step": 54595
    },
    {
      "entropy": 5.317989921569824,
      "epoch": 4.248045127407119,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003307372164641957,
      "loss": 4.7354,
      "mean_token_accuracy": 0.22322663813829421,
      "num_tokens": 94692020.0,
      "step": 54600
    },
    {
      "entropy": 5.391620922088623,
      "epoch": 4.248434156778837,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033071036430921133,
      "loss": 4.8179,
      "mean_token_accuracy": 0.22123800814151764,
      "num_tokens": 94699966.0,
      "step": 54605
    },
    {
      "entropy": 5.411776399612426,
      "epoch": 4.248823186150554,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033068351130891076,
      "loss": 4.76,
      "mean_token_accuracy": 0.21453149020671844,
      "num_tokens": 94708033.0,
      "step": 54610
    },
    {
      "entropy": 5.4739244937896725,
      "epoch": 4.249212215522272,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003306566574637016,
      "loss": 4.883,
      "mean_token_accuracy": 0.2199149951338768,
      "num_tokens": 94716425.0,
      "step": 54615
    },
    {
      "entropy": 5.4653795719146725,
      "epoch": 4.24960124489399,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033062980277399125,
      "loss": 4.8342,
      "mean_token_accuracy": 0.21589039862155915,
      "num_tokens": 94724759.0,
      "step": 54620
    },
    {
      "entropy": 5.470993232727051,
      "epoch": 4.249990274265707,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003306029472401871,
      "loss": 4.852,
      "mean_token_accuracy": 0.2119984358549118,
      "num_tokens": 94734594.0,
      "step": 54625
    },
    {
      "entropy": 5.424247312545776,
      "epoch": 4.250379303637424,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003305760908626968,
      "loss": 4.7927,
      "mean_token_accuracy": 0.2198279857635498,
      "num_tokens": 94743601.0,
      "step": 54630
    },
    {
      "entropy": 5.427798652648926,
      "epoch": 4.250768333009142,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003305492336419277,
      "loss": 4.8406,
      "mean_token_accuracy": 0.21741432100534439,
      "num_tokens": 94752560.0,
      "step": 54635
    },
    {
      "entropy": 5.428102588653564,
      "epoch": 4.25115736238086,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033052237557828763,
      "loss": 4.7998,
      "mean_token_accuracy": 0.22330515831708908,
      "num_tokens": 94760533.0,
      "step": 54640
    },
    {
      "entropy": 5.381521320343017,
      "epoch": 4.251546391752577,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003304955166721837,
      "loss": 4.7864,
      "mean_token_accuracy": 0.2172737404704094,
      "num_tokens": 94768746.0,
      "step": 54645
    },
    {
      "entropy": 5.43982424736023,
      "epoch": 4.251935421124295,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003304686569240237,
      "loss": 4.8817,
      "mean_token_accuracy": 0.21746079623699188,
      "num_tokens": 94777844.0,
      "step": 54650
    },
    {
      "entropy": 5.422416830062867,
      "epoch": 4.252324450496013,
      "grad_norm": 1.234375,
      "learning_rate": 0.000330441796334215,
      "loss": 4.7552,
      "mean_token_accuracy": 0.21707698106765747,
      "num_tokens": 94786100.0,
      "step": 54655
    },
    {
      "entropy": 5.432130813598633,
      "epoch": 4.25271347986773,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003304149349031653,
      "loss": 4.7686,
      "mean_token_accuracy": 0.21576109379529954,
      "num_tokens": 94795262.0,
      "step": 54660
    },
    {
      "entropy": 5.438127040863037,
      "epoch": 4.253102509239447,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003303880726312822,
      "loss": 4.8735,
      "mean_token_accuracy": 0.2176504209637642,
      "num_tokens": 94804758.0,
      "step": 54665
    },
    {
      "entropy": 5.414253282546997,
      "epoch": 4.253491538611165,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033036120951897334,
      "loss": 4.7363,
      "mean_token_accuracy": 0.22886774092912673,
      "num_tokens": 94812704.0,
      "step": 54670
    },
    {
      "entropy": 5.322176790237426,
      "epoch": 4.253880567982883,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003303343455666462,
      "loss": 4.6686,
      "mean_token_accuracy": 0.23109240531921388,
      "num_tokens": 94821240.0,
      "step": 54675
    },
    {
      "entropy": 5.373547029495239,
      "epoch": 4.2542695973546,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033030748077470846,
      "loss": 4.8533,
      "mean_token_accuracy": 0.2141409084200859,
      "num_tokens": 94830369.0,
      "step": 54680
    },
    {
      "entropy": 5.46383957862854,
      "epoch": 4.254658626726318,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033028061514356766,
      "loss": 4.8227,
      "mean_token_accuracy": 0.21968912631273269,
      "num_tokens": 94838809.0,
      "step": 54685
    },
    {
      "entropy": 5.478913068771362,
      "epoch": 4.255047656098036,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003302537486736315,
      "loss": 4.797,
      "mean_token_accuracy": 0.22018113732337952,
      "num_tokens": 94847805.0,
      "step": 54690
    },
    {
      "entropy": 5.369990873336792,
      "epoch": 4.255436685469753,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003302268813653077,
      "loss": 4.7495,
      "mean_token_accuracy": 0.2225884735584259,
      "num_tokens": 94856703.0,
      "step": 54695
    },
    {
      "entropy": 5.452295255661011,
      "epoch": 4.25582571484147,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003302000132190039,
      "loss": 4.8254,
      "mean_token_accuracy": 0.21446521282196046,
      "num_tokens": 94865301.0,
      "step": 54700
    },
    {
      "entropy": 5.442586469650268,
      "epoch": 4.256214744213188,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003301731442351277,
      "loss": 4.7748,
      "mean_token_accuracy": 0.21602004617452622,
      "num_tokens": 94873269.0,
      "step": 54705
    },
    {
      "entropy": 5.418471193313598,
      "epoch": 4.256603773584906,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003301462744140869,
      "loss": 4.8134,
      "mean_token_accuracy": 0.2135033518075943,
      "num_tokens": 94882668.0,
      "step": 54710
    },
    {
      "entropy": 5.4302692890167235,
      "epoch": 4.256992802956623,
      "grad_norm": 1.125,
      "learning_rate": 0.0003301194037562892,
      "loss": 4.9118,
      "mean_token_accuracy": 0.20727849900722503,
      "num_tokens": 94892826.0,
      "step": 54715
    },
    {
      "entropy": 5.331305694580078,
      "epoch": 4.257381832328341,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003300925322621422,
      "loss": 4.7106,
      "mean_token_accuracy": 0.22634776085615158,
      "num_tokens": 94901854.0,
      "step": 54720
    },
    {
      "entropy": 5.397252988815308,
      "epoch": 4.257770861700059,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003300656599320537,
      "loss": 4.764,
      "mean_token_accuracy": 0.22191890478134155,
      "num_tokens": 94910573.0,
      "step": 54725
    },
    {
      "entropy": 5.529209136962891,
      "epoch": 4.2581598910717755,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003300387867664315,
      "loss": 5.0019,
      "mean_token_accuracy": 0.20054250210523605,
      "num_tokens": 94920019.0,
      "step": 54730
    },
    {
      "entropy": 5.323495769500733,
      "epoch": 4.258548920443493,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003300119127656833,
      "loss": 4.6639,
      "mean_token_accuracy": 0.23303650915622712,
      "num_tokens": 94928795.0,
      "step": 54735
    },
    {
      "entropy": 5.367957305908203,
      "epoch": 4.258937949815211,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032998503793021694,
      "loss": 4.7734,
      "mean_token_accuracy": 0.21868903785943986,
      "num_tokens": 94936585.0,
      "step": 54740
    },
    {
      "entropy": 5.429840469360352,
      "epoch": 4.259326979186929,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003299581622604401,
      "loss": 4.795,
      "mean_token_accuracy": 0.23060788512229918,
      "num_tokens": 94944480.0,
      "step": 54745
    },
    {
      "entropy": 5.433448934555054,
      "epoch": 4.259716008558646,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032993128575676064,
      "loss": 4.8581,
      "mean_token_accuracy": 0.21371012628078462,
      "num_tokens": 94954148.0,
      "step": 54750
    },
    {
      "entropy": 5.4394707679748535,
      "epoch": 4.260105037930364,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032990440841958635,
      "loss": 4.8534,
      "mean_token_accuracy": 0.21667168140411378,
      "num_tokens": 94962563.0,
      "step": 54755
    },
    {
      "entropy": 5.385912752151489,
      "epoch": 4.260494067302082,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000329877530249325,
      "loss": 4.8559,
      "mean_token_accuracy": 0.21686408966779708,
      "num_tokens": 94970737.0,
      "step": 54760
    },
    {
      "entropy": 5.382293796539306,
      "epoch": 4.2608830966737985,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003298506512463845,
      "loss": 4.7279,
      "mean_token_accuracy": 0.23097594678401948,
      "num_tokens": 94979418.0,
      "step": 54765
    },
    {
      "entropy": 5.363255262374878,
      "epoch": 4.261272126045516,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003298237714111727,
      "loss": 4.7628,
      "mean_token_accuracy": 0.2224505290389061,
      "num_tokens": 94988219.0,
      "step": 54770
    },
    {
      "entropy": 5.342290019989013,
      "epoch": 4.261661155417234,
      "grad_norm": 1.125,
      "learning_rate": 0.0003297968907440974,
      "loss": 4.7186,
      "mean_token_accuracy": 0.22236937135457993,
      "num_tokens": 94996013.0,
      "step": 54775
    },
    {
      "entropy": 5.430433225631714,
      "epoch": 4.262050184788952,
      "grad_norm": 1.125,
      "learning_rate": 0.0003297700092455665,
      "loss": 4.7946,
      "mean_token_accuracy": 0.21952625960111619,
      "num_tokens": 95004570.0,
      "step": 54780
    },
    {
      "entropy": 5.401397037506103,
      "epoch": 4.262439214160669,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032974312691598794,
      "loss": 4.7794,
      "mean_token_accuracy": 0.2282458171248436,
      "num_tokens": 95013460.0,
      "step": 54785
    },
    {
      "entropy": 5.341152048110962,
      "epoch": 4.262828243532387,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032971624375576956,
      "loss": 4.7329,
      "mean_token_accuracy": 0.22794912457466127,
      "num_tokens": 95022620.0,
      "step": 54790
    },
    {
      "entropy": 5.418966341018677,
      "epoch": 4.263217272904104,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003296893597653192,
      "loss": 4.8663,
      "mean_token_accuracy": 0.20991597026586534,
      "num_tokens": 95031402.0,
      "step": 54795
    },
    {
      "entropy": 5.397003841400147,
      "epoch": 4.2636063022758215,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003296624749450448,
      "loss": 4.7478,
      "mean_token_accuracy": 0.22213748544454576,
      "num_tokens": 95039421.0,
      "step": 54800
    },
    {
      "entropy": 5.352946376800537,
      "epoch": 4.263995331647539,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003296355892953545,
      "loss": 4.729,
      "mean_token_accuracy": 0.22223538011312485,
      "num_tokens": 95048238.0,
      "step": 54805
    },
    {
      "entropy": 5.374114942550659,
      "epoch": 4.264384361019257,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000329608702816656,
      "loss": 4.7693,
      "mean_token_accuracy": 0.2207953780889511,
      "num_tokens": 95057481.0,
      "step": 54810
    },
    {
      "entropy": 5.374326944351196,
      "epoch": 4.264773390390975,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003295818155093574,
      "loss": 4.7379,
      "mean_token_accuracy": 0.22741443961858748,
      "num_tokens": 95065754.0,
      "step": 54815
    },
    {
      "entropy": 5.4718602180480955,
      "epoch": 4.265162419762692,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032955492737386663,
      "loss": 4.9144,
      "mean_token_accuracy": 0.2132245972752571,
      "num_tokens": 95075045.0,
      "step": 54820
    },
    {
      "entropy": 5.4313380241394045,
      "epoch": 4.26555144913441,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003295280384105916,
      "loss": 4.7389,
      "mean_token_accuracy": 0.2324862450361252,
      "num_tokens": 95084021.0,
      "step": 54825
    },
    {
      "entropy": 5.518976068496704,
      "epoch": 4.265940478506128,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032950114861994047,
      "loss": 4.9641,
      "mean_token_accuracy": 0.20754128992557525,
      "num_tokens": 95093843.0,
      "step": 54830
    },
    {
      "entropy": 5.419281911849976,
      "epoch": 4.2663295078778445,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032947425800232107,
      "loss": 4.8156,
      "mean_token_accuracy": 0.21986821740865709,
      "num_tokens": 95103107.0,
      "step": 54835
    },
    {
      "entropy": 5.299954891204834,
      "epoch": 4.266718537249562,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003294473665581415,
      "loss": 4.646,
      "mean_token_accuracy": 0.2342312142252922,
      "num_tokens": 95111173.0,
      "step": 54840
    },
    {
      "entropy": 5.394480037689209,
      "epoch": 4.26710756662128,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032942047428780985,
      "loss": 4.8295,
      "mean_token_accuracy": 0.22379353493452073,
      "num_tokens": 95119999.0,
      "step": 54845
    },
    {
      "entropy": 5.430253648757935,
      "epoch": 4.267496595992998,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003293935811917342,
      "loss": 4.7969,
      "mean_token_accuracy": 0.21630345433950424,
      "num_tokens": 95128733.0,
      "step": 54850
    },
    {
      "entropy": 5.342151165008545,
      "epoch": 4.267885625364715,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003293666872703224,
      "loss": 4.8049,
      "mean_token_accuracy": 0.22271167784929274,
      "num_tokens": 95138000.0,
      "step": 54855
    },
    {
      "entropy": 5.352104902267456,
      "epoch": 4.268274654736433,
      "grad_norm": 1.25,
      "learning_rate": 0.0003293397925239827,
      "loss": 4.7371,
      "mean_token_accuracy": 0.22325584292411804,
      "num_tokens": 95146207.0,
      "step": 54860
    },
    {
      "entropy": 5.468888235092163,
      "epoch": 4.26866368410815,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003293128969531231,
      "loss": 4.8711,
      "mean_token_accuracy": 0.21632730215787888,
      "num_tokens": 95154607.0,
      "step": 54865
    },
    {
      "entropy": 5.397152757644653,
      "epoch": 4.2690527134798675,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003292860005581518,
      "loss": 4.8097,
      "mean_token_accuracy": 0.22666308730840684,
      "num_tokens": 95163441.0,
      "step": 54870
    },
    {
      "entropy": 5.372730398178101,
      "epoch": 4.269441742851585,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032925910333947694,
      "loss": 4.7975,
      "mean_token_accuracy": 0.21812900602817537,
      "num_tokens": 95172822.0,
      "step": 54875
    },
    {
      "entropy": 5.430889701843261,
      "epoch": 4.269830772223303,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003292322052975064,
      "loss": 4.8394,
      "mean_token_accuracy": 0.22389072328805923,
      "num_tokens": 95181538.0,
      "step": 54880
    },
    {
      "entropy": 5.433111000061035,
      "epoch": 4.270219801595021,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003292053064326485,
      "loss": 4.7905,
      "mean_token_accuracy": 0.2216987982392311,
      "num_tokens": 95190182.0,
      "step": 54885
    },
    {
      "entropy": 5.348947668075562,
      "epoch": 4.270608830966738,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003291784067453114,
      "loss": 4.7702,
      "mean_token_accuracy": 0.22684161365032196,
      "num_tokens": 95198757.0,
      "step": 54890
    },
    {
      "entropy": 5.407661867141724,
      "epoch": 4.270997860338456,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032915150623590304,
      "loss": 4.7641,
      "mean_token_accuracy": 0.22177619040012359,
      "num_tokens": 95207890.0,
      "step": 54895
    },
    {
      "entropy": 5.348844289779663,
      "epoch": 4.271386889710173,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032912460490483193,
      "loss": 4.7347,
      "mean_token_accuracy": 0.22969955056905747,
      "num_tokens": 95217011.0,
      "step": 54900
    },
    {
      "entropy": 5.3903872013092045,
      "epoch": 4.2717759190818905,
      "grad_norm": 1.125,
      "learning_rate": 0.0003290977027525061,
      "loss": 4.8119,
      "mean_token_accuracy": 0.2146987110376358,
      "num_tokens": 95226570.0,
      "step": 54905
    },
    {
      "entropy": 5.420358324050904,
      "epoch": 4.272164948453608,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032907079977933373,
      "loss": 4.82,
      "mean_token_accuracy": 0.22445839792490005,
      "num_tokens": 95235509.0,
      "step": 54910
    },
    {
      "entropy": 5.352297401428222,
      "epoch": 4.272553977825326,
      "grad_norm": 1.125,
      "learning_rate": 0.00032904389598572295,
      "loss": 4.6819,
      "mean_token_accuracy": 0.2313300922513008,
      "num_tokens": 95244347.0,
      "step": 54915
    },
    {
      "entropy": 5.372861719131469,
      "epoch": 4.272943007197044,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032901699137208215,
      "loss": 4.6732,
      "mean_token_accuracy": 0.23583931475877762,
      "num_tokens": 95252371.0,
      "step": 54920
    },
    {
      "entropy": 5.292420291900635,
      "epoch": 4.273332036568761,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003289900859388194,
      "loss": 4.788,
      "mean_token_accuracy": 0.22562117874622345,
      "num_tokens": 95261501.0,
      "step": 54925
    },
    {
      "entropy": 5.308648681640625,
      "epoch": 4.273721065940478,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032896317968634305,
      "loss": 4.7055,
      "mean_token_accuracy": 0.22571118026971818,
      "num_tokens": 95271101.0,
      "step": 54930
    },
    {
      "entropy": 5.416075897216797,
      "epoch": 4.274110095312196,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003289362726150614,
      "loss": 4.8079,
      "mean_token_accuracy": 0.21698296964168548,
      "num_tokens": 95279150.0,
      "step": 54935
    },
    {
      "entropy": 5.449749279022217,
      "epoch": 4.2744991246839135,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032890936472538255,
      "loss": 4.8066,
      "mean_token_accuracy": 0.21875184625387192,
      "num_tokens": 95287527.0,
      "step": 54940
    },
    {
      "entropy": 5.34789400100708,
      "epoch": 4.274888154055631,
      "grad_norm": 1.234375,
      "learning_rate": 0.000328882456017715,
      "loss": 4.7474,
      "mean_token_accuracy": 0.22220712900161743,
      "num_tokens": 95295941.0,
      "step": 54945
    },
    {
      "entropy": 5.292721605300903,
      "epoch": 4.275277183427349,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032885554649246697,
      "loss": 4.6849,
      "mean_token_accuracy": 0.23374254405498504,
      "num_tokens": 95304801.0,
      "step": 54950
    },
    {
      "entropy": 5.333906650543213,
      "epoch": 4.275666212799067,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003288286361500467,
      "loss": 4.6906,
      "mean_token_accuracy": 0.23225993812084197,
      "num_tokens": 95313097.0,
      "step": 54955
    },
    {
      "entropy": 5.379891681671142,
      "epoch": 4.276055242170784,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00032880172499086247,
      "loss": 4.7585,
      "mean_token_accuracy": 0.23272813856601715,
      "num_tokens": 95321326.0,
      "step": 54960
    },
    {
      "entropy": 5.446313142776489,
      "epoch": 4.276444271542501,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003287748130153228,
      "loss": 4.7743,
      "mean_token_accuracy": 0.22863819748163222,
      "num_tokens": 95331018.0,
      "step": 54965
    },
    {
      "entropy": 5.4627063274383545,
      "epoch": 4.276833300914219,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003287479002238359,
      "loss": 4.7804,
      "mean_token_accuracy": 0.2279689058661461,
      "num_tokens": 95339581.0,
      "step": 54970
    },
    {
      "entropy": 5.339817667007447,
      "epoch": 4.2772223302859365,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003287209866168102,
      "loss": 4.6883,
      "mean_token_accuracy": 0.22522517293691635,
      "num_tokens": 95348156.0,
      "step": 54975
    },
    {
      "entropy": 5.275270986557007,
      "epoch": 4.277611359657654,
      "grad_norm": 1.1875,
      "learning_rate": 0.000328694072194654,
      "loss": 4.7456,
      "mean_token_accuracy": 0.22237711846828462,
      "num_tokens": 95356777.0,
      "step": 54980
    },
    {
      "entropy": 5.343167686462403,
      "epoch": 4.278000389029372,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032866715695777565,
      "loss": 4.7286,
      "mean_token_accuracy": 0.22108532041311263,
      "num_tokens": 95365714.0,
      "step": 54985
    },
    {
      "entropy": 5.416306352615356,
      "epoch": 4.2783894184010896,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003286402409065837,
      "loss": 4.7943,
      "mean_token_accuracy": 0.22188184559345245,
      "num_tokens": 95374769.0,
      "step": 54990
    },
    {
      "entropy": 5.394586849212646,
      "epoch": 4.278778447772807,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032861332404148647,
      "loss": 4.7602,
      "mean_token_accuracy": 0.230273599922657,
      "num_tokens": 95383690.0,
      "step": 54995
    },
    {
      "entropy": 5.407264614105225,
      "epoch": 4.279167477144524,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032858640636289235,
      "loss": 4.8324,
      "mean_token_accuracy": 0.21614464670419692,
      "num_tokens": 95392351.0,
      "step": 55000
    },
    {
      "entropy": 5.379932022094726,
      "epoch": 4.279556506516242,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003285594878712098,
      "loss": 4.7379,
      "mean_token_accuracy": 0.21958334296941756,
      "num_tokens": 95401076.0,
      "step": 55005
    },
    {
      "entropy": 5.424181127548218,
      "epoch": 4.2799455358879595,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003285325685668473,
      "loss": 4.7376,
      "mean_token_accuracy": 0.22724032104015351,
      "num_tokens": 95409560.0,
      "step": 55010
    },
    {
      "entropy": 5.2767064571380615,
      "epoch": 4.280334565259677,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032850564845021327,
      "loss": 4.6379,
      "mean_token_accuracy": 0.2340892508625984,
      "num_tokens": 95418091.0,
      "step": 55015
    },
    {
      "entropy": 5.40285005569458,
      "epoch": 4.280723594631395,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032847872752171615,
      "loss": 4.9008,
      "mean_token_accuracy": 0.21234452724456787,
      "num_tokens": 95426396.0,
      "step": 55020
    },
    {
      "entropy": 5.3986128807067875,
      "epoch": 4.2811126240031125,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032845180578176456,
      "loss": 4.7588,
      "mean_token_accuracy": 0.22324058115482331,
      "num_tokens": 95435354.0,
      "step": 55025
    },
    {
      "entropy": 5.406235074996948,
      "epoch": 4.28150165337483,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032842488323076675,
      "loss": 4.8066,
      "mean_token_accuracy": 0.2182815507054329,
      "num_tokens": 95444341.0,
      "step": 55030
    },
    {
      "entropy": 5.331027412414551,
      "epoch": 4.281890682746547,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032839795986913146,
      "loss": 4.7688,
      "mean_token_accuracy": 0.21797224432229995,
      "num_tokens": 95453042.0,
      "step": 55035
    },
    {
      "entropy": 5.41580228805542,
      "epoch": 4.282279712118265,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032837103569726713,
      "loss": 4.808,
      "mean_token_accuracy": 0.21793263107538224,
      "num_tokens": 95461777.0,
      "step": 55040
    },
    {
      "entropy": 5.416983413696289,
      "epoch": 4.2826687414899824,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003283441107155822,
      "loss": 4.8221,
      "mean_token_accuracy": 0.2109811618924141,
      "num_tokens": 95469377.0,
      "step": 55045
    },
    {
      "entropy": 5.309278297424316,
      "epoch": 4.2830577708617,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003283171849244854,
      "loss": 4.6925,
      "mean_token_accuracy": 0.22799475938081742,
      "num_tokens": 95478224.0,
      "step": 55050
    },
    {
      "entropy": 5.419901418685913,
      "epoch": 4.283446800233418,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032829025832438507,
      "loss": 4.8116,
      "mean_token_accuracy": 0.21582454591989517,
      "num_tokens": 95487138.0,
      "step": 55055
    },
    {
      "entropy": 5.471138858795166,
      "epoch": 4.2838358296051355,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032826333091569,
      "loss": 4.7848,
      "mean_token_accuracy": 0.2206636980175972,
      "num_tokens": 95495801.0,
      "step": 55060
    },
    {
      "entropy": 5.329667425155639,
      "epoch": 4.284224858976852,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003282364026988087,
      "loss": 4.6992,
      "mean_token_accuracy": 0.23080913722515106,
      "num_tokens": 95504893.0,
      "step": 55065
    },
    {
      "entropy": 5.341891431808472,
      "epoch": 4.28461388834857,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032820947367414953,
      "loss": 4.7501,
      "mean_token_accuracy": 0.22822415679693223,
      "num_tokens": 95513282.0,
      "step": 55070
    },
    {
      "entropy": 5.339668846130371,
      "epoch": 4.285002917720288,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003281825438421215,
      "loss": 4.7576,
      "mean_token_accuracy": 0.22095125913619995,
      "num_tokens": 95521896.0,
      "step": 55075
    },
    {
      "entropy": 5.338019847869873,
      "epoch": 4.285391947092005,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032815561320313286,
      "loss": 4.649,
      "mean_token_accuracy": 0.22917956709861756,
      "num_tokens": 95529597.0,
      "step": 55080
    },
    {
      "entropy": 5.3562116622924805,
      "epoch": 4.285780976463723,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032812868175759245,
      "loss": 4.8331,
      "mean_token_accuracy": 0.21707769483327866,
      "num_tokens": 95538353.0,
      "step": 55085
    },
    {
      "entropy": 5.3809033870697025,
      "epoch": 4.286170005835441,
      "grad_norm": 1.125,
      "learning_rate": 0.0003281017495059088,
      "loss": 4.7874,
      "mean_token_accuracy": 0.2250244438648224,
      "num_tokens": 95546687.0,
      "step": 55090
    },
    {
      "entropy": 5.452684164047241,
      "epoch": 4.2865590352071585,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003280748164484907,
      "loss": 4.9072,
      "mean_token_accuracy": 0.2108091503381729,
      "num_tokens": 95556294.0,
      "step": 55095
    },
    {
      "entropy": 5.432414627075195,
      "epoch": 4.286948064578875,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003280478825857468,
      "loss": 4.8071,
      "mean_token_accuracy": 0.2165073722600937,
      "num_tokens": 95564421.0,
      "step": 55100
    },
    {
      "entropy": 5.351851224899292,
      "epoch": 4.287337093950593,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032802094791808565,
      "loss": 4.7286,
      "mean_token_accuracy": 0.23017796874046326,
      "num_tokens": 95572765.0,
      "step": 55105
    },
    {
      "entropy": 5.369527339935303,
      "epoch": 4.287726123322311,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000327994012445916,
      "loss": 4.759,
      "mean_token_accuracy": 0.22622842639684676,
      "num_tokens": 95581383.0,
      "step": 55110
    },
    {
      "entropy": 5.386286687850952,
      "epoch": 4.288115152694028,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003279670761696465,
      "loss": 4.7996,
      "mean_token_accuracy": 0.22207498997449876,
      "num_tokens": 95590536.0,
      "step": 55115
    },
    {
      "entropy": 5.404020500183106,
      "epoch": 4.288504182065746,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032794013908968593,
      "loss": 4.7657,
      "mean_token_accuracy": 0.2186480402946472,
      "num_tokens": 95599123.0,
      "step": 55120
    },
    {
      "entropy": 5.402689838409424,
      "epoch": 4.288893211437464,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003279132012064431,
      "loss": 4.7667,
      "mean_token_accuracy": 0.2230912536382675,
      "num_tokens": 95607875.0,
      "step": 55125
    },
    {
      "entropy": 5.3230818748474125,
      "epoch": 4.289282240809181,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003278862625203266,
      "loss": 4.6727,
      "mean_token_accuracy": 0.2332807660102844,
      "num_tokens": 95615168.0,
      "step": 55130
    },
    {
      "entropy": 5.437061548233032,
      "epoch": 4.289671270180898,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032785932303174524,
      "loss": 4.9046,
      "mean_token_accuracy": 0.20923303663730622,
      "num_tokens": 95623972.0,
      "step": 55135
    },
    {
      "entropy": 5.515171957015991,
      "epoch": 4.290060299552616,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032783238274110787,
      "loss": 4.9005,
      "mean_token_accuracy": 0.21681196838617325,
      "num_tokens": 95634162.0,
      "step": 55140
    },
    {
      "entropy": 5.469407606124878,
      "epoch": 4.290449328924334,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032780544164882317,
      "loss": 4.8274,
      "mean_token_accuracy": 0.21699175536632537,
      "num_tokens": 95642710.0,
      "step": 55145
    },
    {
      "entropy": 5.352968549728393,
      "epoch": 4.290838358296051,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003277784997552999,
      "loss": 4.6659,
      "mean_token_accuracy": 0.22392785549163818,
      "num_tokens": 95650457.0,
      "step": 55150
    },
    {
      "entropy": 5.445080947875977,
      "epoch": 4.291227387667769,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032775155706094684,
      "loss": 4.8486,
      "mean_token_accuracy": 0.22245485782623292,
      "num_tokens": 95658984.0,
      "step": 55155
    },
    {
      "entropy": 5.395202350616455,
      "epoch": 4.291616417039487,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003277246135661729,
      "loss": 4.8563,
      "mean_token_accuracy": 0.21954651027917862,
      "num_tokens": 95668014.0,
      "step": 55160
    },
    {
      "entropy": 5.488106918334961,
      "epoch": 4.2920054464112045,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003276976692713869,
      "loss": 4.932,
      "mean_token_accuracy": 0.2152042105793953,
      "num_tokens": 95676476.0,
      "step": 55165
    },
    {
      "entropy": 5.571859502792359,
      "epoch": 4.292394475782921,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003276707241769976,
      "loss": 4.8649,
      "mean_token_accuracy": 0.2157810151576996,
      "num_tokens": 95685100.0,
      "step": 55170
    },
    {
      "entropy": 5.451015138626099,
      "epoch": 4.292783505154639,
      "grad_norm": 1.15625,
      "learning_rate": 0.000327643778283414,
      "loss": 4.8091,
      "mean_token_accuracy": 0.21961962580680847,
      "num_tokens": 95693513.0,
      "step": 55175
    },
    {
      "entropy": 5.3230345249176025,
      "epoch": 4.293172534526357,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032761683159104475,
      "loss": 4.7756,
      "mean_token_accuracy": 0.22002748250961304,
      "num_tokens": 95701513.0,
      "step": 55180
    },
    {
      "entropy": 5.392961168289185,
      "epoch": 4.293561563898074,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003275898841002989,
      "loss": 4.7982,
      "mean_token_accuracy": 0.2207813322544098,
      "num_tokens": 95710193.0,
      "step": 55185
    },
    {
      "entropy": 5.410181665420533,
      "epoch": 4.293950593269792,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032756293581158517,
      "loss": 4.7562,
      "mean_token_accuracy": 0.23110752254724504,
      "num_tokens": 95718027.0,
      "step": 55190
    },
    {
      "entropy": 5.457648992538452,
      "epoch": 4.29433962264151,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003275359867253127,
      "loss": 4.8268,
      "mean_token_accuracy": 0.2198631316423416,
      "num_tokens": 95726708.0,
      "step": 55195
    },
    {
      "entropy": 5.349709701538086,
      "epoch": 4.294728652013227,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003275090368418901,
      "loss": 4.8018,
      "mean_token_accuracy": 0.21630767732858658,
      "num_tokens": 95735935.0,
      "step": 55200
    },
    {
      "entropy": 5.3614753723144535,
      "epoch": 4.295117681384944,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003274820861617265,
      "loss": 4.7608,
      "mean_token_accuracy": 0.22329528480768204,
      "num_tokens": 95745215.0,
      "step": 55205
    },
    {
      "entropy": 5.386175489425659,
      "epoch": 4.295506710756662,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003274551346852308,
      "loss": 4.7344,
      "mean_token_accuracy": 0.22658830434083937,
      "num_tokens": 95753815.0,
      "step": 55210
    },
    {
      "entropy": 5.330118560791016,
      "epoch": 4.29589574012838,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003274281824128119,
      "loss": 4.6828,
      "mean_token_accuracy": 0.21901508420705795,
      "num_tokens": 95763047.0,
      "step": 55215
    },
    {
      "entropy": 5.421271133422851,
      "epoch": 4.296284769500097,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032740122934487876,
      "loss": 4.6812,
      "mean_token_accuracy": 0.2286879003047943,
      "num_tokens": 95771925.0,
      "step": 55220
    },
    {
      "entropy": 5.365105009078979,
      "epoch": 4.296673798871815,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003273742754818405,
      "loss": 4.7025,
      "mean_token_accuracy": 0.22676587104797363,
      "num_tokens": 95780478.0,
      "step": 55225
    },
    {
      "entropy": 5.442048025131226,
      "epoch": 4.297062828243533,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003273473208241058,
      "loss": 4.8562,
      "mean_token_accuracy": 0.2161581888794899,
      "num_tokens": 95790122.0,
      "step": 55230
    },
    {
      "entropy": 5.350823593139649,
      "epoch": 4.29745185761525,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032732036537208386,
      "loss": 4.6822,
      "mean_token_accuracy": 0.22177618145942687,
      "num_tokens": 95798820.0,
      "step": 55235
    },
    {
      "entropy": 5.281313037872314,
      "epoch": 4.297840886986967,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003272934091261837,
      "loss": 4.6761,
      "mean_token_accuracy": 0.23342006504535676,
      "num_tokens": 95807789.0,
      "step": 55240
    },
    {
      "entropy": 5.28552474975586,
      "epoch": 4.298229916358685,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032726645208681427,
      "loss": 4.7281,
      "mean_token_accuracy": 0.22298945784568786,
      "num_tokens": 95815946.0,
      "step": 55245
    },
    {
      "entropy": 5.410597324371338,
      "epoch": 4.298618945730403,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032723949425438465,
      "loss": 4.794,
      "mean_token_accuracy": 0.21872412115335466,
      "num_tokens": 95824736.0,
      "step": 55250
    },
    {
      "entropy": 5.360561418533325,
      "epoch": 4.29900797510212,
      "grad_norm": 1.125,
      "learning_rate": 0.00032721253562930383,
      "loss": 4.6756,
      "mean_token_accuracy": 0.23224624395370483,
      "num_tokens": 95833130.0,
      "step": 55255
    },
    {
      "entropy": 5.444640207290649,
      "epoch": 4.299397004473838,
      "grad_norm": 1.125,
      "learning_rate": 0.00032718557621198087,
      "loss": 4.8644,
      "mean_token_accuracy": 0.2175348147749901,
      "num_tokens": 95841836.0,
      "step": 55260
    },
    {
      "entropy": 5.3505411624908445,
      "epoch": 4.299786033845555,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003271586160028249,
      "loss": 4.6922,
      "mean_token_accuracy": 0.2270873323082924,
      "num_tokens": 95850413.0,
      "step": 55265
    },
    {
      "entropy": 5.357793951034546,
      "epoch": 4.300175063217273,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003271316550022449,
      "loss": 4.8175,
      "mean_token_accuracy": 0.21754152625799178,
      "num_tokens": 95859177.0,
      "step": 55270
    },
    {
      "entropy": 5.402596569061279,
      "epoch": 4.30056409258899,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032710469321065,
      "loss": 4.8344,
      "mean_token_accuracy": 0.21562788635492325,
      "num_tokens": 95868835.0,
      "step": 55275
    },
    {
      "entropy": 5.389941310882568,
      "epoch": 4.300953121960708,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003270777306284493,
      "loss": 4.77,
      "mean_token_accuracy": 0.22567924857139587,
      "num_tokens": 95877425.0,
      "step": 55280
    },
    {
      "entropy": 5.4478906154632565,
      "epoch": 4.301342151332426,
      "grad_norm": 1.171875,
      "learning_rate": 0.000327050767256052,
      "loss": 4.8014,
      "mean_token_accuracy": 0.2183885544538498,
      "num_tokens": 95886816.0,
      "step": 55285
    },
    {
      "entropy": 5.422271394729615,
      "epoch": 4.301731180704143,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032702380309386714,
      "loss": 4.8976,
      "mean_token_accuracy": 0.21046326458454132,
      "num_tokens": 95896284.0,
      "step": 55290
    },
    {
      "entropy": 5.411538982391358,
      "epoch": 4.302120210075861,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032699683814230385,
      "loss": 4.808,
      "mean_token_accuracy": 0.21849098205566406,
      "num_tokens": 95904451.0,
      "step": 55295
    },
    {
      "entropy": 5.422853088378906,
      "epoch": 4.302509239447578,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003269698724017713,
      "loss": 4.7801,
      "mean_token_accuracy": 0.2241199567914009,
      "num_tokens": 95913190.0,
      "step": 55300
    },
    {
      "entropy": 5.4452223777771,
      "epoch": 4.302898268819296,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032694290587267867,
      "loss": 4.7952,
      "mean_token_accuracy": 0.22249393910169601,
      "num_tokens": 95921564.0,
      "step": 55305
    },
    {
      "entropy": 5.3123260021209715,
      "epoch": 4.303287298191013,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003269159385554352,
      "loss": 4.6703,
      "mean_token_accuracy": 0.22812095433473586,
      "num_tokens": 95930653.0,
      "step": 55310
    },
    {
      "entropy": 5.3373703956604,
      "epoch": 4.303676327562731,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003268889704504499,
      "loss": 4.7568,
      "mean_token_accuracy": 0.22075141668319703,
      "num_tokens": 95939125.0,
      "step": 55315
    },
    {
      "entropy": 5.335182332992554,
      "epoch": 4.304065356934449,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00032686200155813196,
      "loss": 4.7059,
      "mean_token_accuracy": 0.23106219470500947,
      "num_tokens": 95947792.0,
      "step": 55320
    },
    {
      "entropy": 5.389108324050904,
      "epoch": 4.304454386306166,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003268350318788908,
      "loss": 4.7799,
      "mean_token_accuracy": 0.21855859458446503,
      "num_tokens": 95955532.0,
      "step": 55325
    },
    {
      "entropy": 5.452947330474854,
      "epoch": 4.304843415677884,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003268080614131356,
      "loss": 4.831,
      "mean_token_accuracy": 0.21524378806352615,
      "num_tokens": 95964564.0,
      "step": 55330
    },
    {
      "entropy": 5.426184463500976,
      "epoch": 4.305232445049601,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003267810901612754,
      "loss": 4.8333,
      "mean_token_accuracy": 0.21540775299072265,
      "num_tokens": 95973122.0,
      "step": 55335
    },
    {
      "entropy": 5.395905160903931,
      "epoch": 4.305621474421319,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032675411812371975,
      "loss": 4.7798,
      "mean_token_accuracy": 0.22293207943439483,
      "num_tokens": 95980920.0,
      "step": 55340
    },
    {
      "entropy": 5.408984375,
      "epoch": 4.306010503793036,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032672714530087763,
      "loss": 4.7705,
      "mean_token_accuracy": 0.21723743677139282,
      "num_tokens": 95989372.0,
      "step": 55345
    },
    {
      "entropy": 5.4025181293487545,
      "epoch": 4.306399533164754,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032670017169315836,
      "loss": 4.7459,
      "mean_token_accuracy": 0.22096821069717407,
      "num_tokens": 95997737.0,
      "step": 55350
    },
    {
      "entropy": 5.336916208267212,
      "epoch": 4.306788562536472,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003266731973009713,
      "loss": 4.7562,
      "mean_token_accuracy": 0.2201735034584999,
      "num_tokens": 96006412.0,
      "step": 55355
    },
    {
      "entropy": 5.422471380233764,
      "epoch": 4.307177591908189,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003266462221247257,
      "loss": 4.8014,
      "mean_token_accuracy": 0.22605425715446473,
      "num_tokens": 96014986.0,
      "step": 55360
    },
    {
      "entropy": 5.384347629547119,
      "epoch": 4.307566621279907,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000326619246164831,
      "loss": 4.7488,
      "mean_token_accuracy": 0.21650106757879256,
      "num_tokens": 96023402.0,
      "step": 55365
    },
    {
      "entropy": 5.38533353805542,
      "epoch": 4.307955650651624,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003265922694216963,
      "loss": 4.8456,
      "mean_token_accuracy": 0.21930623799562454,
      "num_tokens": 96031836.0,
      "step": 55370
    },
    {
      "entropy": 5.356522417068481,
      "epoch": 4.308344680023342,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032656529189573107,
      "loss": 4.6803,
      "mean_token_accuracy": 0.22192334979772568,
      "num_tokens": 96040365.0,
      "step": 55375
    },
    {
      "entropy": 5.422477531433105,
      "epoch": 4.308733709395059,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003265383135873446,
      "loss": 4.8597,
      "mean_token_accuracy": 0.21980943977832795,
      "num_tokens": 96049000.0,
      "step": 55380
    },
    {
      "entropy": 5.476351022720337,
      "epoch": 4.309122738766777,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003265113344969462,
      "loss": 4.8069,
      "mean_token_accuracy": 0.220074363052845,
      "num_tokens": 96058141.0,
      "step": 55385
    },
    {
      "entropy": 5.445583248138428,
      "epoch": 4.309511768138495,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003264843546249453,
      "loss": 4.7531,
      "mean_token_accuracy": 0.23095903992652894,
      "num_tokens": 96066755.0,
      "step": 55390
    },
    {
      "entropy": 5.300701856613159,
      "epoch": 4.309900797510212,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032645737397175135,
      "loss": 4.6495,
      "mean_token_accuracy": 0.23668957352638245,
      "num_tokens": 96075266.0,
      "step": 55395
    },
    {
      "entropy": 5.305155515670776,
      "epoch": 4.310289826881929,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032643039253777356,
      "loss": 4.7441,
      "mean_token_accuracy": 0.21834357380867003,
      "num_tokens": 96083118.0,
      "step": 55400
    },
    {
      "entropy": 5.374486303329467,
      "epoch": 4.310678856253647,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032640341032342147,
      "loss": 4.7521,
      "mean_token_accuracy": 0.22649578452110292,
      "num_tokens": 96091958.0,
      "step": 55405
    },
    {
      "entropy": 5.391165494918823,
      "epoch": 4.311067885625365,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032637642732910447,
      "loss": 4.8096,
      "mean_token_accuracy": 0.218907231092453,
      "num_tokens": 96100789.0,
      "step": 55410
    },
    {
      "entropy": 5.398464155197144,
      "epoch": 4.311456914997082,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032634944355523196,
      "loss": 4.8105,
      "mean_token_accuracy": 0.22169774174690246,
      "num_tokens": 96109117.0,
      "step": 55415
    },
    {
      "entropy": 5.374966812133789,
      "epoch": 4.3118459443688,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003263224590022133,
      "loss": 4.8413,
      "mean_token_accuracy": 0.21848445236682892,
      "num_tokens": 96117676.0,
      "step": 55420
    },
    {
      "entropy": 5.389714765548706,
      "epoch": 4.312234973740518,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032629547367045805,
      "loss": 4.7729,
      "mean_token_accuracy": 0.22224421948194503,
      "num_tokens": 96127330.0,
      "step": 55425
    },
    {
      "entropy": 5.469387578964233,
      "epoch": 4.312624003112235,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032626848756037565,
      "loss": 4.93,
      "mean_token_accuracy": 0.21103732734918595,
      "num_tokens": 96136477.0,
      "step": 55430
    },
    {
      "entropy": 5.342172145843506,
      "epoch": 4.313013032483952,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032624150067237553,
      "loss": 4.7122,
      "mean_token_accuracy": 0.22278120070695878,
      "num_tokens": 96145719.0,
      "step": 55435
    },
    {
      "entropy": 5.305167961120605,
      "epoch": 4.31340206185567,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003262145130068672,
      "loss": 4.7367,
      "mean_token_accuracy": 0.22570502161979675,
      "num_tokens": 96155004.0,
      "step": 55440
    },
    {
      "entropy": 5.404322719573974,
      "epoch": 4.313791091227388,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032618752456426017,
      "loss": 4.8483,
      "mean_token_accuracy": 0.22060937732458114,
      "num_tokens": 96165544.0,
      "step": 55445
    },
    {
      "entropy": 5.387012338638305,
      "epoch": 4.314180120599105,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032616053534496387,
      "loss": 4.7295,
      "mean_token_accuracy": 0.23154066503047943,
      "num_tokens": 96173572.0,
      "step": 55450
    },
    {
      "entropy": 5.360696506500244,
      "epoch": 4.314569149970823,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003261335453493879,
      "loss": 4.8375,
      "mean_token_accuracy": 0.22176438719034194,
      "num_tokens": 96182275.0,
      "step": 55455
    },
    {
      "entropy": 5.420110368728638,
      "epoch": 4.314958179342541,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032610655457794184,
      "loss": 4.7913,
      "mean_token_accuracy": 0.2202710121870041,
      "num_tokens": 96190538.0,
      "step": 55460
    },
    {
      "entropy": 5.414336442947388,
      "epoch": 4.3153472087142575,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032607956303103516,
      "loss": 4.7479,
      "mean_token_accuracy": 0.22611221820116043,
      "num_tokens": 96198783.0,
      "step": 55465
    },
    {
      "entropy": 5.394220542907715,
      "epoch": 4.315736238085975,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032605257070907734,
      "loss": 4.7746,
      "mean_token_accuracy": 0.22512691766023635,
      "num_tokens": 96207291.0,
      "step": 55470
    },
    {
      "entropy": 5.368642807006836,
      "epoch": 4.316125267457693,
      "grad_norm": 1.25,
      "learning_rate": 0.00032602557761247805,
      "loss": 4.8471,
      "mean_token_accuracy": 0.2155747666954994,
      "num_tokens": 96215493.0,
      "step": 55475
    },
    {
      "entropy": 5.378756523132324,
      "epoch": 4.316514296829411,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003259985837416468,
      "loss": 4.7641,
      "mean_token_accuracy": 0.2223289802670479,
      "num_tokens": 96225095.0,
      "step": 55480
    },
    {
      "entropy": 5.408122539520264,
      "epoch": 4.316903326201128,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003259715890969932,
      "loss": 4.7061,
      "mean_token_accuracy": 0.2230312630534172,
      "num_tokens": 96232957.0,
      "step": 55485
    },
    {
      "entropy": 5.3922735214233395,
      "epoch": 4.317292355572846,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003259445936789269,
      "loss": 4.8539,
      "mean_token_accuracy": 0.21551900058984758,
      "num_tokens": 96242005.0,
      "step": 55490
    },
    {
      "entropy": 5.321115159988404,
      "epoch": 4.317681384944564,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032591759748785753,
      "loss": 4.73,
      "mean_token_accuracy": 0.22020068019628525,
      "num_tokens": 96251014.0,
      "step": 55495
    },
    {
      "entropy": 5.292581272125244,
      "epoch": 4.3180704143162805,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003258906005241946,
      "loss": 4.7395,
      "mean_token_accuracy": 0.22166526913642884,
      "num_tokens": 96259894.0,
      "step": 55500
    },
    {
      "entropy": 5.335675621032715,
      "epoch": 4.318459443687998,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003258636027883478,
      "loss": 4.6927,
      "mean_token_accuracy": 0.22346720546483995,
      "num_tokens": 96268257.0,
      "step": 55505
    },
    {
      "entropy": 5.3699757099151615,
      "epoch": 4.318848473059716,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003258366042807269,
      "loss": 4.7294,
      "mean_token_accuracy": 0.22322644293308258,
      "num_tokens": 96277366.0,
      "step": 55510
    },
    {
      "entropy": 5.358770084381104,
      "epoch": 4.319237502431434,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003258096050017413,
      "loss": 4.7837,
      "mean_token_accuracy": 0.2219999313354492,
      "num_tokens": 96286306.0,
      "step": 55515
    },
    {
      "entropy": 5.35988221168518,
      "epoch": 4.319626531803151,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003257826049518008,
      "loss": 4.739,
      "mean_token_accuracy": 0.21961160153150558,
      "num_tokens": 96295582.0,
      "step": 55520
    },
    {
      "entropy": 5.427718687057495,
      "epoch": 4.320015561174869,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003257556041313153,
      "loss": 4.8162,
      "mean_token_accuracy": 0.22021132707595825,
      "num_tokens": 96304540.0,
      "step": 55525
    },
    {
      "entropy": 5.2953760623931885,
      "epoch": 4.320404590546587,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003257286025406941,
      "loss": 4.6804,
      "mean_token_accuracy": 0.22188996970653535,
      "num_tokens": 96312872.0,
      "step": 55530
    },
    {
      "entropy": 5.424187231063843,
      "epoch": 4.3207936199183035,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032570160018034717,
      "loss": 4.7719,
      "mean_token_accuracy": 0.22222257554531097,
      "num_tokens": 96321378.0,
      "step": 55535
    },
    {
      "entropy": 5.352781200408936,
      "epoch": 4.321182649290021,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032567459705068415,
      "loss": 4.7454,
      "mean_token_accuracy": 0.228916797041893,
      "num_tokens": 96329751.0,
      "step": 55540
    },
    {
      "entropy": 5.409131622314453,
      "epoch": 4.321571678661739,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00032564759315211475,
      "loss": 4.8251,
      "mean_token_accuracy": 0.21533117443323135,
      "num_tokens": 96337928.0,
      "step": 55545
    },
    {
      "entropy": 5.4032628536224365,
      "epoch": 4.3219607080334566,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032562058848504877,
      "loss": 4.8452,
      "mean_token_accuracy": 0.21290225237607957,
      "num_tokens": 96347017.0,
      "step": 55550
    },
    {
      "entropy": 5.376733112335205,
      "epoch": 4.322349737405174,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003255935830498959,
      "loss": 4.7621,
      "mean_token_accuracy": 0.2229041799902916,
      "num_tokens": 96356195.0,
      "step": 55555
    },
    {
      "entropy": 5.488920211791992,
      "epoch": 4.322738766776892,
      "grad_norm": 1.125,
      "learning_rate": 0.000325566576847066,
      "loss": 4.9037,
      "mean_token_accuracy": 0.20501357614994048,
      "num_tokens": 96364980.0,
      "step": 55560
    },
    {
      "entropy": 5.467865943908691,
      "epoch": 4.32312779614861,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003255395698769687,
      "loss": 4.8805,
      "mean_token_accuracy": 0.21121373921632766,
      "num_tokens": 96372978.0,
      "step": 55565
    },
    {
      "entropy": 5.403638410568237,
      "epoch": 4.3235168255203265,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003255125621400139,
      "loss": 4.7538,
      "mean_token_accuracy": 0.21785937249660492,
      "num_tokens": 96381209.0,
      "step": 55570
    },
    {
      "entropy": 5.438088655471802,
      "epoch": 4.323905854892044,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003254855536366114,
      "loss": 4.8552,
      "mean_token_accuracy": 0.2178419291973114,
      "num_tokens": 96390625.0,
      "step": 55575
    },
    {
      "entropy": 5.484043312072754,
      "epoch": 4.324294884263762,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000325458544367171,
      "loss": 4.8188,
      "mean_token_accuracy": 0.22014171481132508,
      "num_tokens": 96399629.0,
      "step": 55580
    },
    {
      "entropy": 5.393654298782349,
      "epoch": 4.3246839136354795,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032543153433210243,
      "loss": 4.714,
      "mean_token_accuracy": 0.22539501786231994,
      "num_tokens": 96407718.0,
      "step": 55585
    },
    {
      "entropy": 5.365255117416382,
      "epoch": 4.325072943007197,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003254045235318156,
      "loss": 4.7092,
      "mean_token_accuracy": 0.23057196140289307,
      "num_tokens": 96416422.0,
      "step": 55590
    },
    {
      "entropy": 5.279518938064575,
      "epoch": 4.325461972378915,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032537751196672033,
      "loss": 4.6665,
      "mean_token_accuracy": 0.2265308678150177,
      "num_tokens": 96424582.0,
      "step": 55595
    },
    {
      "entropy": 5.533742523193359,
      "epoch": 4.325851001750632,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003253504996372266,
      "loss": 4.9855,
      "mean_token_accuracy": 0.20255263447761535,
      "num_tokens": 96433272.0,
      "step": 55600
    },
    {
      "entropy": 5.417198276519775,
      "epoch": 4.3262400311223494,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003253234865437442,
      "loss": 4.6997,
      "mean_token_accuracy": 0.23361830562353134,
      "num_tokens": 96442432.0,
      "step": 55605
    },
    {
      "entropy": 5.417437124252319,
      "epoch": 4.326629060494067,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003252964726866829,
      "loss": 4.8151,
      "mean_token_accuracy": 0.21862059384584426,
      "num_tokens": 96451233.0,
      "step": 55610
    },
    {
      "entropy": 5.363316392898559,
      "epoch": 4.327018089865785,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032526945806645265,
      "loss": 4.8629,
      "mean_token_accuracy": 0.21029772758483886,
      "num_tokens": 96460393.0,
      "step": 55615
    },
    {
      "entropy": 5.372008991241455,
      "epoch": 4.3274071192375025,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003252424426834634,
      "loss": 4.7087,
      "mean_token_accuracy": 0.22971729040145875,
      "num_tokens": 96467865.0,
      "step": 55620
    },
    {
      "entropy": 5.33195013999939,
      "epoch": 4.32779614860922,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003252154265381251,
      "loss": 4.689,
      "mean_token_accuracy": 0.22645098268985747,
      "num_tokens": 96475931.0,
      "step": 55625
    },
    {
      "entropy": 5.381135368347168,
      "epoch": 4.328185177980938,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003251884096308476,
      "loss": 4.7752,
      "mean_token_accuracy": 0.22809132635593415,
      "num_tokens": 96484532.0,
      "step": 55630
    },
    {
      "entropy": 5.367303943634033,
      "epoch": 4.328574207352655,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032516139196204087,
      "loss": 4.757,
      "mean_token_accuracy": 0.23212565779685973,
      "num_tokens": 96492182.0,
      "step": 55635
    },
    {
      "entropy": 5.418198680877685,
      "epoch": 4.328963236724372,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032513437353211483,
      "loss": 4.771,
      "mean_token_accuracy": 0.23258078396320342,
      "num_tokens": 96500146.0,
      "step": 55640
    },
    {
      "entropy": 5.419314336776734,
      "epoch": 4.32935226609609,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003251073543414796,
      "loss": 4.8295,
      "mean_token_accuracy": 0.21825287640094757,
      "num_tokens": 96509116.0,
      "step": 55645
    },
    {
      "entropy": 5.448482275009155,
      "epoch": 4.329741295467808,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032508033439054493,
      "loss": 4.8405,
      "mean_token_accuracy": 0.21710984259843827,
      "num_tokens": 96517846.0,
      "step": 55650
    },
    {
      "entropy": 5.348165798187256,
      "epoch": 4.3301303248395255,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003250533136797208,
      "loss": 4.7077,
      "mean_token_accuracy": 0.22872622013092042,
      "num_tokens": 96526639.0,
      "step": 55655
    },
    {
      "entropy": 5.4071722507476805,
      "epoch": 4.330519354211243,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032502629220941743,
      "loss": 4.8547,
      "mean_token_accuracy": 0.21390550285577775,
      "num_tokens": 96535689.0,
      "step": 55660
    },
    {
      "entropy": 5.392814588546753,
      "epoch": 4.33090838358296,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032499926998004464,
      "loss": 4.7854,
      "mean_token_accuracy": 0.22436124384403228,
      "num_tokens": 96545071.0,
      "step": 55665
    },
    {
      "entropy": 5.403608703613282,
      "epoch": 4.331297412954678,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003249722469920126,
      "loss": 4.7334,
      "mean_token_accuracy": 0.22524525076150895,
      "num_tokens": 96553077.0,
      "step": 55670
    },
    {
      "entropy": 5.329644823074341,
      "epoch": 4.331686442326395,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003249452232457312,
      "loss": 4.7115,
      "mean_token_accuracy": 0.2247196242213249,
      "num_tokens": 96561928.0,
      "step": 55675
    },
    {
      "entropy": 5.395749759674072,
      "epoch": 4.332075471698113,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032491819874161046,
      "loss": 4.8615,
      "mean_token_accuracy": 0.22728123515844345,
      "num_tokens": 96570695.0,
      "step": 55680
    },
    {
      "entropy": 5.403173589706421,
      "epoch": 4.332464501069831,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003248911734800606,
      "loss": 4.7574,
      "mean_token_accuracy": 0.22319168150424956,
      "num_tokens": 96578727.0,
      "step": 55685
    },
    {
      "entropy": 5.490252304077148,
      "epoch": 4.3328535304415485,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032486414746149154,
      "loss": 4.8686,
      "mean_token_accuracy": 0.20879190564155578,
      "num_tokens": 96587759.0,
      "step": 55690
    },
    {
      "entropy": 5.40419487953186,
      "epoch": 4.333242559813266,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032483712068631345,
      "loss": 4.8014,
      "mean_token_accuracy": 0.22317678183317186,
      "num_tokens": 96596456.0,
      "step": 55695
    },
    {
      "entropy": 5.377751350402832,
      "epoch": 4.333631589184984,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003248100931549364,
      "loss": 4.7725,
      "mean_token_accuracy": 0.22799796611070633,
      "num_tokens": 96605877.0,
      "step": 55700
    },
    {
      "entropy": 5.464987707138062,
      "epoch": 4.334020618556701,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032478306486777037,
      "loss": 4.7394,
      "mean_token_accuracy": 0.22770272493362426,
      "num_tokens": 96614327.0,
      "step": 55705
    },
    {
      "entropy": 5.371473264694214,
      "epoch": 4.334409647928418,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003247560358252256,
      "loss": 4.721,
      "mean_token_accuracy": 0.22613715529441833,
      "num_tokens": 96622826.0,
      "step": 55710
    },
    {
      "entropy": 5.309128522872925,
      "epoch": 4.334798677300136,
      "grad_norm": 1.25,
      "learning_rate": 0.00032472900602771224,
      "loss": 4.7589,
      "mean_token_accuracy": 0.22518249750137329,
      "num_tokens": 96631639.0,
      "step": 55715
    },
    {
      "entropy": 5.433979797363281,
      "epoch": 4.335187706671854,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032470197547564036,
      "loss": 4.8487,
      "mean_token_accuracy": 0.21633694916963578,
      "num_tokens": 96639711.0,
      "step": 55720
    },
    {
      "entropy": 5.433394050598144,
      "epoch": 4.3355767360435715,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032467494416942006,
      "loss": 4.7524,
      "mean_token_accuracy": 0.224424247443676,
      "num_tokens": 96648046.0,
      "step": 55725
    },
    {
      "entropy": 5.357724189758301,
      "epoch": 4.335965765415289,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003246479121094616,
      "loss": 4.6462,
      "mean_token_accuracy": 0.23440519124269485,
      "num_tokens": 96657794.0,
      "step": 55730
    },
    {
      "entropy": 5.353157567977905,
      "epoch": 4.336354794787006,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000324620879296175,
      "loss": 4.7885,
      "mean_token_accuracy": 0.2139231637120247,
      "num_tokens": 96666282.0,
      "step": 55735
    },
    {
      "entropy": 5.455154275894165,
      "epoch": 4.336743824158724,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003245938457299706,
      "loss": 4.854,
      "mean_token_accuracy": 0.2165993332862854,
      "num_tokens": 96675189.0,
      "step": 55740
    },
    {
      "entropy": 5.43734073638916,
      "epoch": 4.337132853530441,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032456681141125853,
      "loss": 4.8026,
      "mean_token_accuracy": 0.2215997964143753,
      "num_tokens": 96683637.0,
      "step": 55745
    },
    {
      "entropy": 5.399583911895752,
      "epoch": 4.337521882902159,
      "grad_norm": 1.15625,
      "learning_rate": 0.000324539776340449,
      "loss": 4.7807,
      "mean_token_accuracy": 0.2228444203734398,
      "num_tokens": 96693000.0,
      "step": 55750
    },
    {
      "entropy": 5.328756809234619,
      "epoch": 4.337910912273877,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003245127405179522,
      "loss": 4.7423,
      "mean_token_accuracy": 0.22194072753190994,
      "num_tokens": 96702088.0,
      "step": 55755
    },
    {
      "entropy": 5.390795087814331,
      "epoch": 4.3382999416455945,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003244857039441784,
      "loss": 4.8095,
      "mean_token_accuracy": 0.2224502071738243,
      "num_tokens": 96710497.0,
      "step": 55760
    },
    {
      "entropy": 5.488256645202637,
      "epoch": 4.338688971017312,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032445866661953774,
      "loss": 4.8482,
      "mean_token_accuracy": 0.21836442053318023,
      "num_tokens": 96720175.0,
      "step": 55765
    },
    {
      "entropy": 5.404442930221558,
      "epoch": 4.339078000389029,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032443162854444064,
      "loss": 4.8185,
      "mean_token_accuracy": 0.2181785762310028,
      "num_tokens": 96728760.0,
      "step": 55770
    },
    {
      "entropy": 5.4370355129241945,
      "epoch": 4.339467029760747,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003244045897192972,
      "loss": 4.868,
      "mean_token_accuracy": 0.21257621049880981,
      "num_tokens": 96737105.0,
      "step": 55775
    },
    {
      "entropy": 5.428970623016357,
      "epoch": 4.339856059132464,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032437755014451777,
      "loss": 4.8931,
      "mean_token_accuracy": 0.21427603513002397,
      "num_tokens": 96745881.0,
      "step": 55780
    },
    {
      "entropy": 5.491665601730347,
      "epoch": 4.340245088504182,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032435050982051246,
      "loss": 4.8208,
      "mean_token_accuracy": 0.22051716297864915,
      "num_tokens": 96754427.0,
      "step": 55785
    },
    {
      "entropy": 5.402885866165161,
      "epoch": 4.3406341178759,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032432346874769187,
      "loss": 4.81,
      "mean_token_accuracy": 0.2187369093298912,
      "num_tokens": 96763517.0,
      "step": 55790
    },
    {
      "entropy": 5.417707872390747,
      "epoch": 4.3410231472476175,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032429642692646614,
      "loss": 4.8399,
      "mean_token_accuracy": 0.21524917036294938,
      "num_tokens": 96772872.0,
      "step": 55795
    },
    {
      "entropy": 5.471979141235352,
      "epoch": 4.341412176619334,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032426938435724554,
      "loss": 4.7752,
      "mean_token_accuracy": 0.21772324442863464,
      "num_tokens": 96781249.0,
      "step": 55800
    },
    {
      "entropy": 5.365498685836792,
      "epoch": 4.341801205991052,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032424234104044055,
      "loss": 4.7462,
      "mean_token_accuracy": 0.22384527921676636,
      "num_tokens": 96789769.0,
      "step": 55805
    },
    {
      "entropy": 5.39996337890625,
      "epoch": 4.34219023536277,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003242152969764613,
      "loss": 4.8796,
      "mean_token_accuracy": 0.21005725711584092,
      "num_tokens": 96798859.0,
      "step": 55810
    },
    {
      "entropy": 5.405779027938843,
      "epoch": 4.342579264734487,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003241882521657183,
      "loss": 4.7363,
      "mean_token_accuracy": 0.22263825982809066,
      "num_tokens": 96807301.0,
      "step": 55815
    },
    {
      "entropy": 5.398718786239624,
      "epoch": 4.342968294106205,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032416120660862187,
      "loss": 4.7427,
      "mean_token_accuracy": 0.22943371683359146,
      "num_tokens": 96815302.0,
      "step": 55820
    },
    {
      "entropy": 5.361724615097046,
      "epoch": 4.343357323477923,
      "grad_norm": 1.25,
      "learning_rate": 0.0003241341603055823,
      "loss": 4.7743,
      "mean_token_accuracy": 0.2285906746983528,
      "num_tokens": 96823206.0,
      "step": 55825
    },
    {
      "entropy": 5.4169398784637455,
      "epoch": 4.3437463528496405,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003241071132570101,
      "loss": 4.7902,
      "mean_token_accuracy": 0.22267747521400452,
      "num_tokens": 96831843.0,
      "step": 55830
    },
    {
      "entropy": 5.488817644119263,
      "epoch": 4.344135382221357,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003240800654633157,
      "loss": 4.8871,
      "mean_token_accuracy": 0.2118205487728119,
      "num_tokens": 96841221.0,
      "step": 55835
    },
    {
      "entropy": 5.428368139266968,
      "epoch": 4.344524411593075,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003240530169249094,
      "loss": 4.8347,
      "mean_token_accuracy": 0.21994154155254364,
      "num_tokens": 96849970.0,
      "step": 55840
    },
    {
      "entropy": 5.418295955657959,
      "epoch": 4.344913440964793,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003240259676422016,
      "loss": 4.7728,
      "mean_token_accuracy": 0.2156427890062332,
      "num_tokens": 96858970.0,
      "step": 55845
    },
    {
      "entropy": 5.324556446075439,
      "epoch": 4.34530247033651,
      "grad_norm": 1.125,
      "learning_rate": 0.00032399891761560274,
      "loss": 4.6808,
      "mean_token_accuracy": 0.23872142434120178,
      "num_tokens": 96867546.0,
      "step": 55850
    },
    {
      "entropy": 5.287373495101929,
      "epoch": 4.345691499708228,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003239718668455233,
      "loss": 4.6828,
      "mean_token_accuracy": 0.23273766934871673,
      "num_tokens": 96877038.0,
      "step": 55855
    },
    {
      "entropy": 5.460830640792847,
      "epoch": 4.346080529079946,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003239448153323737,
      "loss": 4.8989,
      "mean_token_accuracy": 0.20918368697166442,
      "num_tokens": 96886502.0,
      "step": 55860
    },
    {
      "entropy": 5.491157007217407,
      "epoch": 4.3464695584516635,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032391776307656456,
      "loss": 4.8605,
      "mean_token_accuracy": 0.21206588298082352,
      "num_tokens": 96895629.0,
      "step": 55865
    },
    {
      "entropy": 5.442399978637695,
      "epoch": 4.34685858782338,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032389071007850616,
      "loss": 4.8014,
      "mean_token_accuracy": 0.21925843954086305,
      "num_tokens": 96903907.0,
      "step": 55870
    },
    {
      "entropy": 5.350760889053345,
      "epoch": 4.347247617195098,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032386365633860914,
      "loss": 4.8005,
      "mean_token_accuracy": 0.2209215611219406,
      "num_tokens": 96911860.0,
      "step": 55875
    },
    {
      "entropy": 5.267929124832153,
      "epoch": 4.347636646566816,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003238366018572838,
      "loss": 4.6943,
      "mean_token_accuracy": 0.22725993394851685,
      "num_tokens": 96920449.0,
      "step": 55880
    },
    {
      "entropy": 5.365199899673462,
      "epoch": 4.348025675938533,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003238095466349407,
      "loss": 4.6974,
      "mean_token_accuracy": 0.22886136919260025,
      "num_tokens": 96929256.0,
      "step": 55885
    },
    {
      "entropy": 5.357767391204834,
      "epoch": 4.348414705310251,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003237824906719905,
      "loss": 4.7499,
      "mean_token_accuracy": 0.21974247545003892,
      "num_tokens": 96937945.0,
      "step": 55890
    },
    {
      "entropy": 5.419110679626465,
      "epoch": 4.348803734681969,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032375543396884367,
      "loss": 4.8698,
      "mean_token_accuracy": 0.22124201208353042,
      "num_tokens": 96948296.0,
      "step": 55895
    },
    {
      "entropy": 5.310656309127808,
      "epoch": 4.3491927640536865,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003237283765259107,
      "loss": 4.5903,
      "mean_token_accuracy": 0.2321484535932541,
      "num_tokens": 96957515.0,
      "step": 55900
    },
    {
      "entropy": 5.447336673736572,
      "epoch": 4.349581793425403,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032370131834360215,
      "loss": 4.8715,
      "mean_token_accuracy": 0.21613024473190307,
      "num_tokens": 96966051.0,
      "step": 55905
    },
    {
      "entropy": 5.368315839767456,
      "epoch": 4.349970822797121,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003236742594223287,
      "loss": 4.7261,
      "mean_token_accuracy": 0.22677332758903504,
      "num_tokens": 96973963.0,
      "step": 55910
    },
    {
      "entropy": 5.31332802772522,
      "epoch": 4.350359852168839,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003236471997625008,
      "loss": 4.7507,
      "mean_token_accuracy": 0.22661279290914535,
      "num_tokens": 96983207.0,
      "step": 55915
    },
    {
      "entropy": 5.37418007850647,
      "epoch": 4.350748881540556,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003236201393645291,
      "loss": 4.7451,
      "mean_token_accuracy": 0.23111149072647094,
      "num_tokens": 96992542.0,
      "step": 55920
    },
    {
      "entropy": 5.386187934875489,
      "epoch": 4.351137910912274,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032359307822882403,
      "loss": 4.7993,
      "mean_token_accuracy": 0.2202426239848137,
      "num_tokens": 97000678.0,
      "step": 55925
    },
    {
      "entropy": 5.4081987857818605,
      "epoch": 4.351526940283992,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032356601635579645,
      "loss": 4.7592,
      "mean_token_accuracy": 0.22231945842504502,
      "num_tokens": 97009120.0,
      "step": 55930
    },
    {
      "entropy": 5.338505506515503,
      "epoch": 4.351915969655709,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003235389537458569,
      "loss": 4.7365,
      "mean_token_accuracy": 0.22983648777008056,
      "num_tokens": 97018340.0,
      "step": 55935
    },
    {
      "entropy": 5.3908055305480955,
      "epoch": 4.352304999027426,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032351189039941594,
      "loss": 4.7553,
      "mean_token_accuracy": 0.23117996752262115,
      "num_tokens": 97027250.0,
      "step": 55940
    },
    {
      "entropy": 5.387021017074585,
      "epoch": 4.352694028399144,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032348482631688424,
      "loss": 4.7528,
      "mean_token_accuracy": 0.22428754717111588,
      "num_tokens": 97035335.0,
      "step": 55945
    },
    {
      "entropy": 5.338643789291382,
      "epoch": 4.353083057770862,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003234577614986725,
      "loss": 4.7725,
      "mean_token_accuracy": 0.21736646741628646,
      "num_tokens": 97043851.0,
      "step": 55950
    },
    {
      "entropy": 5.342312955856324,
      "epoch": 4.353472087142579,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003234306959451914,
      "loss": 4.7856,
      "mean_token_accuracy": 0.22453118860721588,
      "num_tokens": 97052155.0,
      "step": 55955
    },
    {
      "entropy": 5.385831356048584,
      "epoch": 4.353861116514297,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032340362965685145,
      "loss": 4.7188,
      "mean_token_accuracy": 0.22177911996841432,
      "num_tokens": 97061173.0,
      "step": 55960
    },
    {
      "entropy": 5.4687456607818605,
      "epoch": 4.354250145886015,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003233765626340636,
      "loss": 4.8427,
      "mean_token_accuracy": 0.2141509473323822,
      "num_tokens": 97069686.0,
      "step": 55965
    },
    {
      "entropy": 5.422771167755127,
      "epoch": 4.354639175257732,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003233494948772384,
      "loss": 4.8427,
      "mean_token_accuracy": 0.2132783442735672,
      "num_tokens": 97079066.0,
      "step": 55970
    },
    {
      "entropy": 5.4230124950408936,
      "epoch": 4.355028204629449,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032332242638678647,
      "loss": 4.8234,
      "mean_token_accuracy": 0.21738970875740052,
      "num_tokens": 97089238.0,
      "step": 55975
    },
    {
      "entropy": 5.416968154907226,
      "epoch": 4.355417234001167,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003232953571631187,
      "loss": 4.7778,
      "mean_token_accuracy": 0.22134379893541337,
      "num_tokens": 97098042.0,
      "step": 55980
    },
    {
      "entropy": 5.511040830612183,
      "epoch": 4.355806263372885,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003232682872066457,
      "loss": 4.8173,
      "mean_token_accuracy": 0.21987354904413223,
      "num_tokens": 97107173.0,
      "step": 55985
    },
    {
      "entropy": 5.335352802276612,
      "epoch": 4.356195292744602,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003232412165177783,
      "loss": 4.6467,
      "mean_token_accuracy": 0.22969443649053572,
      "num_tokens": 97115813.0,
      "step": 55990
    },
    {
      "entropy": 5.461706113815308,
      "epoch": 4.35658432211632,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032321414509692726,
      "loss": 4.8587,
      "mean_token_accuracy": 0.21749609112739562,
      "num_tokens": 97124106.0,
      "step": 55995
    },
    {
      "entropy": 5.372380208969116,
      "epoch": 4.356973351488037,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032318707294450326,
      "loss": 4.7833,
      "mean_token_accuracy": 0.22204227298498153,
      "num_tokens": 97133703.0,
      "step": 56000
    },
    {
      "entropy": 5.405798768997192,
      "epoch": 4.357362380859755,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003231600000609171,
      "loss": 4.7399,
      "mean_token_accuracy": 0.22545213550329207,
      "num_tokens": 97142826.0,
      "step": 56005
    },
    {
      "entropy": 5.368091583251953,
      "epoch": 4.357751410231472,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032313292644657967,
      "loss": 4.7327,
      "mean_token_accuracy": 0.22490902841091157,
      "num_tokens": 97151408.0,
      "step": 56010
    },
    {
      "entropy": 5.393712759017944,
      "epoch": 4.35814043960319,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003231058521019017,
      "loss": 4.8536,
      "mean_token_accuracy": 0.21505127251148223,
      "num_tokens": 97160210.0,
      "step": 56015
    },
    {
      "entropy": 5.505608749389649,
      "epoch": 4.358529468974908,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003230787770272939,
      "loss": 4.8429,
      "mean_token_accuracy": 0.22239350378513337,
      "num_tokens": 97168618.0,
      "step": 56020
    },
    {
      "entropy": 5.472224473953247,
      "epoch": 4.358918498346625,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003230517012231673,
      "loss": 4.822,
      "mean_token_accuracy": 0.22635573595762254,
      "num_tokens": 97176668.0,
      "step": 56025
    },
    {
      "entropy": 5.265771102905274,
      "epoch": 4.359307527718343,
      "grad_norm": 1.25,
      "learning_rate": 0.00032302462468993253,
      "loss": 4.7331,
      "mean_token_accuracy": 0.2277931347489357,
      "num_tokens": 97184195.0,
      "step": 56030
    },
    {
      "entropy": 5.381201553344726,
      "epoch": 4.359696557090061,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003229975474280006,
      "loss": 4.8734,
      "mean_token_accuracy": 0.21844094395637512,
      "num_tokens": 97192535.0,
      "step": 56035
    },
    {
      "entropy": 5.400267457962036,
      "epoch": 4.360085586461778,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032297046943778216,
      "loss": 4.7268,
      "mean_token_accuracy": 0.22568205147981643,
      "num_tokens": 97201334.0,
      "step": 56040
    },
    {
      "entropy": 5.3981232166290285,
      "epoch": 4.360474615833495,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032294339071968833,
      "loss": 4.7555,
      "mean_token_accuracy": 0.22293631136417388,
      "num_tokens": 97209826.0,
      "step": 56045
    },
    {
      "entropy": 5.353059148788452,
      "epoch": 4.360863645205213,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032291631127412976,
      "loss": 4.7772,
      "mean_token_accuracy": 0.21820842772722243,
      "num_tokens": 97218469.0,
      "step": 56050
    },
    {
      "entropy": 5.395062255859375,
      "epoch": 4.361252674576931,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003228892311015174,
      "loss": 4.7633,
      "mean_token_accuracy": 0.21846552789211274,
      "num_tokens": 97227341.0,
      "step": 56055
    },
    {
      "entropy": 5.456218576431274,
      "epoch": 4.361641703948648,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032286215020226224,
      "loss": 4.8193,
      "mean_token_accuracy": 0.22159145623445511,
      "num_tokens": 97236582.0,
      "step": 56060
    },
    {
      "entropy": 5.451393365859985,
      "epoch": 4.362030733320366,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032283506857677517,
      "loss": 4.8346,
      "mean_token_accuracy": 0.2176318049430847,
      "num_tokens": 97244390.0,
      "step": 56065
    },
    {
      "entropy": 5.3915690898895265,
      "epoch": 4.362419762692083,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003228079862254671,
      "loss": 4.7573,
      "mean_token_accuracy": 0.23124109506607055,
      "num_tokens": 97252889.0,
      "step": 56070
    },
    {
      "entropy": 5.335072946548462,
      "epoch": 4.362808792063801,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032278090314874887,
      "loss": 4.7587,
      "mean_token_accuracy": 0.22069679498672484,
      "num_tokens": 97260928.0,
      "step": 56075
    },
    {
      "entropy": 5.359328269958496,
      "epoch": 4.363197821435518,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032275381934703147,
      "loss": 4.7381,
      "mean_token_accuracy": 0.22488827258348465,
      "num_tokens": 97268654.0,
      "step": 56080
    },
    {
      "entropy": 5.336133289337158,
      "epoch": 4.363586850807236,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003227267348207258,
      "loss": 4.7534,
      "mean_token_accuracy": 0.21835248321294784,
      "num_tokens": 97277233.0,
      "step": 56085
    },
    {
      "entropy": 5.420194387435913,
      "epoch": 4.363975880178954,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032269964957024305,
      "loss": 4.8162,
      "mean_token_accuracy": 0.22223170548677446,
      "num_tokens": 97285955.0,
      "step": 56090
    },
    {
      "entropy": 5.423932027816773,
      "epoch": 4.364364909550671,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032267256359599396,
      "loss": 4.7992,
      "mean_token_accuracy": 0.2197403132915497,
      "num_tokens": 97294524.0,
      "step": 56095
    },
    {
      "entropy": 5.3944244384765625,
      "epoch": 4.364753938922389,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032264547689838963,
      "loss": 4.7236,
      "mean_token_accuracy": 0.2224283277988434,
      "num_tokens": 97303325.0,
      "step": 56100
    },
    {
      "entropy": 5.372149133682251,
      "epoch": 4.365142968294106,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032261838947784103,
      "loss": 4.7336,
      "mean_token_accuracy": 0.2247239261865616,
      "num_tokens": 97311512.0,
      "step": 56105
    },
    {
      "entropy": 5.313755655288697,
      "epoch": 4.365531997665824,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003225913013347591,
      "loss": 4.7779,
      "mean_token_accuracy": 0.22200810462236403,
      "num_tokens": 97320381.0,
      "step": 56110
    },
    {
      "entropy": 5.37011833190918,
      "epoch": 4.365921027037541,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032256421246955504,
      "loss": 4.7345,
      "mean_token_accuracy": 0.22982241809368134,
      "num_tokens": 97329021.0,
      "step": 56115
    },
    {
      "entropy": 5.341499376296997,
      "epoch": 4.366310056409259,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003225371228826396,
      "loss": 4.6795,
      "mean_token_accuracy": 0.2306458443403244,
      "num_tokens": 97337307.0,
      "step": 56120
    },
    {
      "entropy": 5.403027629852295,
      "epoch": 4.366699085780977,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003225100325744241,
      "loss": 4.8289,
      "mean_token_accuracy": 0.2215326502919197,
      "num_tokens": 97346538.0,
      "step": 56125
    },
    {
      "entropy": 5.444408559799195,
      "epoch": 4.367088115152694,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003224829415453194,
      "loss": 4.7464,
      "mean_token_accuracy": 0.22983803153038024,
      "num_tokens": 97355135.0,
      "step": 56130
    },
    {
      "entropy": 5.360068655014038,
      "epoch": 4.367477144524411,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003224558497957367,
      "loss": 4.7362,
      "mean_token_accuracy": 0.22589855790138244,
      "num_tokens": 97364800.0,
      "step": 56135
    },
    {
      "entropy": 5.412807893753052,
      "epoch": 4.367866173896129,
      "grad_norm": 1.25,
      "learning_rate": 0.00032242875732608697,
      "loss": 4.856,
      "mean_token_accuracy": 0.21397557556629182,
      "num_tokens": 97373860.0,
      "step": 56140
    },
    {
      "entropy": 5.447054386138916,
      "epoch": 4.3682552032678466,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003224016641367814,
      "loss": 4.7934,
      "mean_token_accuracy": 0.21940065324306487,
      "num_tokens": 97382232.0,
      "step": 56145
    },
    {
      "entropy": 5.4790791988372805,
      "epoch": 4.368644232639564,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032237457022823095,
      "loss": 4.848,
      "mean_token_accuracy": 0.22616732716560364,
      "num_tokens": 97391785.0,
      "step": 56150
    },
    {
      "entropy": 5.3922828197479244,
      "epoch": 4.369033262011282,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032234747560084675,
      "loss": 4.7833,
      "mean_token_accuracy": 0.21653473228216172,
      "num_tokens": 97400179.0,
      "step": 56155
    },
    {
      "entropy": 5.381615209579468,
      "epoch": 4.369422291383,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032232038025504015,
      "loss": 4.8218,
      "mean_token_accuracy": 0.22236624211072922,
      "num_tokens": 97409108.0,
      "step": 56160
    },
    {
      "entropy": 5.403571367263794,
      "epoch": 4.369811320754717,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032229328419122196,
      "loss": 4.7624,
      "mean_token_accuracy": 0.22837934494018555,
      "num_tokens": 97417578.0,
      "step": 56165
    },
    {
      "entropy": 5.34228024482727,
      "epoch": 4.370200350126434,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003222661874098035,
      "loss": 4.7372,
      "mean_token_accuracy": 0.22372948229312897,
      "num_tokens": 97426147.0,
      "step": 56170
    },
    {
      "entropy": 5.366935110092163,
      "epoch": 4.370589379498152,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003222390899111959,
      "loss": 4.7846,
      "mean_token_accuracy": 0.22217825055122375,
      "num_tokens": 97435551.0,
      "step": 56175
    },
    {
      "entropy": 5.366704511642456,
      "epoch": 4.3709784088698695,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003222119916958103,
      "loss": 4.7337,
      "mean_token_accuracy": 0.22579503804445267,
      "num_tokens": 97444245.0,
      "step": 56180
    },
    {
      "entropy": 5.370033073425293,
      "epoch": 4.371367438241587,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032218489276405777,
      "loss": 4.7337,
      "mean_token_accuracy": 0.22305500209331514,
      "num_tokens": 97452756.0,
      "step": 56185
    },
    {
      "entropy": 5.352962589263916,
      "epoch": 4.371756467613305,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032215779311634967,
      "loss": 4.6913,
      "mean_token_accuracy": 0.23692746758460997,
      "num_tokens": 97461151.0,
      "step": 56190
    },
    {
      "entropy": 5.3922467708587645,
      "epoch": 4.372145496985023,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032213069275309714,
      "loss": 4.7677,
      "mean_token_accuracy": 0.22473716884851455,
      "num_tokens": 97470310.0,
      "step": 56195
    },
    {
      "entropy": 5.3499339580535885,
      "epoch": 4.37253452635674,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032210359167471136,
      "loss": 4.6815,
      "mean_token_accuracy": 0.2313734546303749,
      "num_tokens": 97479037.0,
      "step": 56200
    },
    {
      "entropy": 5.344713878631592,
      "epoch": 4.372923555728457,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003220764898816035,
      "loss": 4.7797,
      "mean_token_accuracy": 0.22320988029241562,
      "num_tokens": 97487418.0,
      "step": 56205
    },
    {
      "entropy": 5.342199373245239,
      "epoch": 4.373312585100175,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032204938737418495,
      "loss": 4.8065,
      "mean_token_accuracy": 0.2276156023144722,
      "num_tokens": 97495871.0,
      "step": 56210
    },
    {
      "entropy": 5.423250484466553,
      "epoch": 4.3737016144718925,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032202228415286664,
      "loss": 4.7608,
      "mean_token_accuracy": 0.22285333126783372,
      "num_tokens": 97504109.0,
      "step": 56215
    },
    {
      "entropy": 5.365264558792115,
      "epoch": 4.37409064384361,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003219951802180602,
      "loss": 4.7668,
      "mean_token_accuracy": 0.22619300484657287,
      "num_tokens": 97513882.0,
      "step": 56220
    },
    {
      "entropy": 5.414183712005615,
      "epoch": 4.374479673215328,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003219680755701766,
      "loss": 4.8224,
      "mean_token_accuracy": 0.22421636879444123,
      "num_tokens": 97523130.0,
      "step": 56225
    },
    {
      "entropy": 5.4847901344299315,
      "epoch": 4.374868702587046,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032194097020962726,
      "loss": 4.9092,
      "mean_token_accuracy": 0.21692904680967331,
      "num_tokens": 97531179.0,
      "step": 56230
    },
    {
      "entropy": 5.381154108047485,
      "epoch": 4.375257731958763,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003219138641368234,
      "loss": 4.7495,
      "mean_token_accuracy": 0.22420942038297653,
      "num_tokens": 97540131.0,
      "step": 56235
    },
    {
      "entropy": 5.336398696899414,
      "epoch": 4.37564676133048,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032188675735217627,
      "loss": 4.7282,
      "mean_token_accuracy": 0.22687490731477739,
      "num_tokens": 97548592.0,
      "step": 56240
    },
    {
      "entropy": 5.446547031402588,
      "epoch": 4.376035790702198,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003218596498560973,
      "loss": 4.7684,
      "mean_token_accuracy": 0.2272908478975296,
      "num_tokens": 97557589.0,
      "step": 56245
    },
    {
      "entropy": 5.331271457672119,
      "epoch": 4.3764248200739155,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003218325416489976,
      "loss": 4.7387,
      "mean_token_accuracy": 0.22876322865486146,
      "num_tokens": 97565676.0,
      "step": 56250
    },
    {
      "entropy": 5.371689081192017,
      "epoch": 4.376813849445633,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003218054327312888,
      "loss": 4.6903,
      "mean_token_accuracy": 0.22380103170871735,
      "num_tokens": 97573368.0,
      "step": 56255
    },
    {
      "entropy": 5.301996088027954,
      "epoch": 4.377202878817351,
      "grad_norm": 1.125,
      "learning_rate": 0.00032177832310338193,
      "loss": 4.7143,
      "mean_token_accuracy": 0.23484406918287276,
      "num_tokens": 97582424.0,
      "step": 56260
    },
    {
      "entropy": 5.351057052612305,
      "epoch": 4.377591908189069,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032175121276568853,
      "loss": 4.7653,
      "mean_token_accuracy": 0.22647581547498702,
      "num_tokens": 97591402.0,
      "step": 56265
    },
    {
      "entropy": 5.396567726135254,
      "epoch": 4.377980937560785,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032172410171861987,
      "loss": 4.8025,
      "mean_token_accuracy": 0.21755914390087128,
      "num_tokens": 97600465.0,
      "step": 56270
    },
    {
      "entropy": 5.374724912643432,
      "epoch": 4.378369966932503,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032169698996258733,
      "loss": 4.6914,
      "mean_token_accuracy": 0.225983364880085,
      "num_tokens": 97608248.0,
      "step": 56275
    },
    {
      "entropy": 5.507989025115966,
      "epoch": 4.378758996304221,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003216698774980023,
      "loss": 4.9155,
      "mean_token_accuracy": 0.21362703293561935,
      "num_tokens": 97617035.0,
      "step": 56280
    },
    {
      "entropy": 5.3750029563903805,
      "epoch": 4.3791480256759385,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032164276432527605,
      "loss": 4.7571,
      "mean_token_accuracy": 0.22198230773210526,
      "num_tokens": 97625264.0,
      "step": 56285
    },
    {
      "entropy": 5.4048809051513675,
      "epoch": 4.379537055047656,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003216156504448202,
      "loss": 4.8275,
      "mean_token_accuracy": 0.2204088792204857,
      "num_tokens": 97633446.0,
      "step": 56290
    },
    {
      "entropy": 5.458734798431396,
      "epoch": 4.379926084419374,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003215885358570461,
      "loss": 4.8007,
      "mean_token_accuracy": 0.22244305461645125,
      "num_tokens": 97641440.0,
      "step": 56295
    },
    {
      "entropy": 5.373304033279419,
      "epoch": 4.380315113791092,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003215614205623651,
      "loss": 4.712,
      "mean_token_accuracy": 0.22935060560703277,
      "num_tokens": 97649396.0,
      "step": 56300
    },
    {
      "entropy": 5.384699058532715,
      "epoch": 4.380704143162808,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003215343045611886,
      "loss": 4.8024,
      "mean_token_accuracy": 0.22392283976078034,
      "num_tokens": 97658126.0,
      "step": 56305
    },
    {
      "entropy": 5.450688219070434,
      "epoch": 4.381093172534526,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003215071878539281,
      "loss": 4.7634,
      "mean_token_accuracy": 0.21931175887584686,
      "num_tokens": 97667177.0,
      "step": 56310
    },
    {
      "entropy": 5.390646314620971,
      "epoch": 4.381482201906244,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032148007044099507,
      "loss": 4.7663,
      "mean_token_accuracy": 0.22120947241783143,
      "num_tokens": 97675548.0,
      "step": 56315
    },
    {
      "entropy": 5.332130384445191,
      "epoch": 4.3818712312779615,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003214529523228009,
      "loss": 4.6842,
      "mean_token_accuracy": 0.2317298099398613,
      "num_tokens": 97683580.0,
      "step": 56320
    },
    {
      "entropy": 5.455280065536499,
      "epoch": 4.382260260649679,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003214258334997572,
      "loss": 4.849,
      "mean_token_accuracy": 0.21339709758758546,
      "num_tokens": 97692179.0,
      "step": 56325
    },
    {
      "entropy": 5.348147392272949,
      "epoch": 4.382649290021397,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003213987139722753,
      "loss": 4.7321,
      "mean_token_accuracy": 0.2255592241883278,
      "num_tokens": 97700989.0,
      "step": 56330
    },
    {
      "entropy": 5.368076324462891,
      "epoch": 4.383038319393114,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032137159374076693,
      "loss": 4.7595,
      "mean_token_accuracy": 0.2170540601015091,
      "num_tokens": 97709445.0,
      "step": 56335
    },
    {
      "entropy": 5.447507047653199,
      "epoch": 4.383427348764831,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003213444728056432,
      "loss": 4.8422,
      "mean_token_accuracy": 0.21912931650877,
      "num_tokens": 97719561.0,
      "step": 56340
    },
    {
      "entropy": 5.307501554489136,
      "epoch": 4.383816378136549,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032131735116731604,
      "loss": 4.6445,
      "mean_token_accuracy": 0.23574916273355484,
      "num_tokens": 97728015.0,
      "step": 56345
    },
    {
      "entropy": 5.453648042678833,
      "epoch": 4.384205407508267,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003212902288261966,
      "loss": 4.9269,
      "mean_token_accuracy": 0.2149940699338913,
      "num_tokens": 97737135.0,
      "step": 56350
    },
    {
      "entropy": 5.353962230682373,
      "epoch": 4.3845944368799845,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003212631057826968,
      "loss": 4.7233,
      "mean_token_accuracy": 0.2235427737236023,
      "num_tokens": 97745744.0,
      "step": 56355
    },
    {
      "entropy": 5.4424351215362545,
      "epoch": 4.384983466251702,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032123598203722786,
      "loss": 4.8007,
      "mean_token_accuracy": 0.2105320230126381,
      "num_tokens": 97754778.0,
      "step": 56360
    },
    {
      "entropy": 5.46599440574646,
      "epoch": 4.38537249562342,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003212088575902016,
      "loss": 4.8706,
      "mean_token_accuracy": 0.21225354820489883,
      "num_tokens": 97763988.0,
      "step": 56365
    },
    {
      "entropy": 5.400659227371216,
      "epoch": 4.385761524995137,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032118173244202947,
      "loss": 4.8208,
      "mean_token_accuracy": 0.21859641075134278,
      "num_tokens": 97773303.0,
      "step": 56370
    },
    {
      "entropy": 5.343986225128174,
      "epoch": 4.386150554366854,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032115460659312304,
      "loss": 4.7568,
      "mean_token_accuracy": 0.22588880658149718,
      "num_tokens": 97781756.0,
      "step": 56375
    },
    {
      "entropy": 5.318846654891968,
      "epoch": 4.386539583738572,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003211274800438939,
      "loss": 4.6645,
      "mean_token_accuracy": 0.2343151107430458,
      "num_tokens": 97789703.0,
      "step": 56380
    },
    {
      "entropy": 5.340959548950195,
      "epoch": 4.38692861311029,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003211003527947537,
      "loss": 4.7578,
      "mean_token_accuracy": 0.2216591417789459,
      "num_tokens": 97797407.0,
      "step": 56385
    },
    {
      "entropy": 5.354788970947266,
      "epoch": 4.3873176424820075,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032107322484611396,
      "loss": 4.7637,
      "mean_token_accuracy": 0.22668472975492476,
      "num_tokens": 97805779.0,
      "step": 56390
    },
    {
      "entropy": 5.376725196838379,
      "epoch": 4.387706671853725,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032104609619838635,
      "loss": 4.7474,
      "mean_token_accuracy": 0.22612350136041642,
      "num_tokens": 97813735.0,
      "step": 56395
    },
    {
      "entropy": 5.332343339920044,
      "epoch": 4.388095701225443,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003210189668519826,
      "loss": 4.7512,
      "mean_token_accuracy": 0.22069841027259826,
      "num_tokens": 97822548.0,
      "step": 56400
    },
    {
      "entropy": 5.364507627487183,
      "epoch": 4.38848473059716,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003209918368073143,
      "loss": 4.74,
      "mean_token_accuracy": 0.22095053791999816,
      "num_tokens": 97830784.0,
      "step": 56405
    },
    {
      "entropy": 5.3569377899169925,
      "epoch": 4.388873759968877,
      "grad_norm": 1.21875,
      "learning_rate": 0.000320964706064793,
      "loss": 4.7963,
      "mean_token_accuracy": 0.22228676974773406,
      "num_tokens": 97839292.0,
      "step": 56410
    },
    {
      "entropy": 5.396024608612061,
      "epoch": 4.389262789340595,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032093757462483055,
      "loss": 4.7964,
      "mean_token_accuracy": 0.2156072199344635,
      "num_tokens": 97847427.0,
      "step": 56415
    },
    {
      "entropy": 5.402605676651001,
      "epoch": 4.389651818712313,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032091044248783835,
      "loss": 4.7211,
      "mean_token_accuracy": 0.2312668666243553,
      "num_tokens": 97855240.0,
      "step": 56420
    },
    {
      "entropy": 5.464087009429932,
      "epoch": 4.3900408480840305,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032088330965422836,
      "loss": 4.7871,
      "mean_token_accuracy": 0.223911052942276,
      "num_tokens": 97862620.0,
      "step": 56425
    },
    {
      "entropy": 5.336995840072632,
      "epoch": 4.390429877455748,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032085617612441215,
      "loss": 4.6857,
      "mean_token_accuracy": 0.23047897666692735,
      "num_tokens": 97871835.0,
      "step": 56430
    },
    {
      "entropy": 5.423362588882446,
      "epoch": 4.390818906827466,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003208290418988015,
      "loss": 4.7609,
      "mean_token_accuracy": 0.23372793048620225,
      "num_tokens": 97880690.0,
      "step": 56435
    },
    {
      "entropy": 5.404538679122925,
      "epoch": 4.391207936199183,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000320801906977808,
      "loss": 4.7964,
      "mean_token_accuracy": 0.22281818985939025,
      "num_tokens": 97889604.0,
      "step": 56440
    },
    {
      "entropy": 5.302058553695678,
      "epoch": 4.3915969655709,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032077477136184356,
      "loss": 4.727,
      "mean_token_accuracy": 0.224255833029747,
      "num_tokens": 97897606.0,
      "step": 56445
    },
    {
      "entropy": 5.4425006866455075,
      "epoch": 4.391985994942618,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003207476350513198,
      "loss": 4.8741,
      "mean_token_accuracy": 0.2114188253879547,
      "num_tokens": 97905886.0,
      "step": 56450
    },
    {
      "entropy": 5.3370747566223145,
      "epoch": 4.392375024314336,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003207204980466485,
      "loss": 4.7013,
      "mean_token_accuracy": 0.23470528274774552,
      "num_tokens": 97914141.0,
      "step": 56455
    },
    {
      "entropy": 5.429505825042725,
      "epoch": 4.3927640536860535,
      "grad_norm": 1.25,
      "learning_rate": 0.00032069336034824136,
      "loss": 4.856,
      "mean_token_accuracy": 0.213875350356102,
      "num_tokens": 97922321.0,
      "step": 56460
    },
    {
      "entropy": 5.4151349544525145,
      "epoch": 4.393153083057771,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032066622195651024,
      "loss": 4.7978,
      "mean_token_accuracy": 0.22905352115631103,
      "num_tokens": 97931045.0,
      "step": 56465
    },
    {
      "entropy": 5.327529239654541,
      "epoch": 4.393542112429488,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032063908287186687,
      "loss": 4.6957,
      "mean_token_accuracy": 0.22773615270853043,
      "num_tokens": 97940998.0,
      "step": 56470
    },
    {
      "entropy": 5.437001752853393,
      "epoch": 4.393931141801206,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032061194309472314,
      "loss": 4.8447,
      "mean_token_accuracy": 0.2172312006354332,
      "num_tokens": 97949530.0,
      "step": 56475
    },
    {
      "entropy": 5.424756813049316,
      "epoch": 4.394320171172923,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032058480262549065,
      "loss": 4.8013,
      "mean_token_accuracy": 0.22273593097925187,
      "num_tokens": 97958291.0,
      "step": 56480
    },
    {
      "entropy": 5.401621961593628,
      "epoch": 4.394709200544641,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003205576614645814,
      "loss": 4.8132,
      "mean_token_accuracy": 0.22325518280267714,
      "num_tokens": 97966452.0,
      "step": 56485
    },
    {
      "entropy": 5.3582805633544925,
      "epoch": 4.395098229916359,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032053051961240716,
      "loss": 4.7058,
      "mean_token_accuracy": 0.2339236244559288,
      "num_tokens": 97974586.0,
      "step": 56490
    },
    {
      "entropy": 5.369989585876465,
      "epoch": 4.3954872592880765,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003205033770693797,
      "loss": 4.7432,
      "mean_token_accuracy": 0.22153111547231674,
      "num_tokens": 97983725.0,
      "step": 56495
    },
    {
      "entropy": 5.392746782302856,
      "epoch": 4.395876288659794,
      "grad_norm": 1.15625,
      "learning_rate": 0.000320476233835911,
      "loss": 4.7421,
      "mean_token_accuracy": 0.22314399480819702,
      "num_tokens": 97992241.0,
      "step": 56500
    },
    {
      "entropy": 5.385956764221191,
      "epoch": 4.396265318031511,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032044908991241274,
      "loss": 4.7528,
      "mean_token_accuracy": 0.2258208692073822,
      "num_tokens": 98001297.0,
      "step": 56505
    },
    {
      "entropy": 5.355286693572998,
      "epoch": 4.396654347403229,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032042194529929694,
      "loss": 4.7085,
      "mean_token_accuracy": 0.2205297753214836,
      "num_tokens": 98009897.0,
      "step": 56510
    },
    {
      "entropy": 5.291087055206299,
      "epoch": 4.397043376774946,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032039479999697545,
      "loss": 4.6946,
      "mean_token_accuracy": 0.23060269206762313,
      "num_tokens": 98018390.0,
      "step": 56515
    },
    {
      "entropy": 5.399711322784424,
      "epoch": 4.397432406146664,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032036765400586005,
      "loss": 4.7892,
      "mean_token_accuracy": 0.22231783270835875,
      "num_tokens": 98026885.0,
      "step": 56520
    },
    {
      "entropy": 5.469669342041016,
      "epoch": 4.397821435518382,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032034050732636274,
      "loss": 4.816,
      "mean_token_accuracy": 0.2219847947359085,
      "num_tokens": 98035432.0,
      "step": 56525
    },
    {
      "entropy": 5.428636598587036,
      "epoch": 4.3982104648900995,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003203133599588954,
      "loss": 4.8457,
      "mean_token_accuracy": 0.21817982345819473,
      "num_tokens": 98044849.0,
      "step": 56530
    },
    {
      "entropy": 5.406703329086303,
      "epoch": 4.398599494261817,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032028621190386997,
      "loss": 4.797,
      "mean_token_accuracy": 0.22317960113286972,
      "num_tokens": 98053258.0,
      "step": 56535
    },
    {
      "entropy": 5.315113067626953,
      "epoch": 4.398988523633534,
      "grad_norm": 1.25,
      "learning_rate": 0.00032025906316169835,
      "loss": 4.7622,
      "mean_token_accuracy": 0.22249849885702133,
      "num_tokens": 98062446.0,
      "step": 56540
    },
    {
      "entropy": 5.280483150482178,
      "epoch": 4.399377553005252,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003202319137327924,
      "loss": 4.788,
      "mean_token_accuracy": 0.22238420844078063,
      "num_tokens": 98070965.0,
      "step": 56545
    },
    {
      "entropy": 5.407123470306397,
      "epoch": 4.399766582376969,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032020476361756424,
      "loss": 4.8271,
      "mean_token_accuracy": 0.22398722767829896,
      "num_tokens": 98078818.0,
      "step": 56550
    },
    {
      "entropy": 5.406821060180664,
      "epoch": 4.400155611748687,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032017761281642564,
      "loss": 4.8241,
      "mean_token_accuracy": 0.22521023452281952,
      "num_tokens": 98088841.0,
      "step": 56555
    },
    {
      "entropy": 5.3922301769256595,
      "epoch": 4.400544641120405,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003201504613297888,
      "loss": 4.7175,
      "mean_token_accuracy": 0.22613370716571807,
      "num_tokens": 98098046.0,
      "step": 56560
    },
    {
      "entropy": 5.327527713775635,
      "epoch": 4.4009336704921225,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032012330915806553,
      "loss": 4.7071,
      "mean_token_accuracy": 0.22706825733184816,
      "num_tokens": 98106411.0,
      "step": 56565
    },
    {
      "entropy": 5.410301542282104,
      "epoch": 4.40132269986384,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032009615630166786,
      "loss": 4.8536,
      "mean_token_accuracy": 0.21100362688302993,
      "num_tokens": 98115519.0,
      "step": 56570
    },
    {
      "entropy": 5.394170093536377,
      "epoch": 4.401711729235557,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003200690027610078,
      "loss": 4.7487,
      "mean_token_accuracy": 0.22798893004655837,
      "num_tokens": 98124988.0,
      "step": 56575
    },
    {
      "entropy": 5.440633296966553,
      "epoch": 4.402100758607275,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003200418485364974,
      "loss": 4.7373,
      "mean_token_accuracy": 0.22475164383649826,
      "num_tokens": 98133712.0,
      "step": 56580
    },
    {
      "entropy": 5.444111728668213,
      "epoch": 4.402489787978992,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003200146936285485,
      "loss": 4.9087,
      "mean_token_accuracy": 0.21857898384332658,
      "num_tokens": 98142022.0,
      "step": 56585
    },
    {
      "entropy": 5.333130264282227,
      "epoch": 4.40287881735071,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003199875380375734,
      "loss": 4.7008,
      "mean_token_accuracy": 0.2301072284579277,
      "num_tokens": 98150482.0,
      "step": 56590
    },
    {
      "entropy": 5.4020861148834225,
      "epoch": 4.403267846722428,
      "grad_norm": 1.171875,
      "learning_rate": 0.000319960381763984,
      "loss": 4.8187,
      "mean_token_accuracy": 0.21445932388305664,
      "num_tokens": 98158546.0,
      "step": 56595
    },
    {
      "entropy": 5.455578756332398,
      "epoch": 4.4036568760941455,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003199332248081923,
      "loss": 4.8342,
      "mean_token_accuracy": 0.21237119287252426,
      "num_tokens": 98166703.0,
      "step": 56600
    },
    {
      "entropy": 5.41999397277832,
      "epoch": 4.404045905465862,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031990606717061055,
      "loss": 4.8138,
      "mean_token_accuracy": 0.22752324044704436,
      "num_tokens": 98176210.0,
      "step": 56605
    },
    {
      "entropy": 5.448330593109131,
      "epoch": 4.40443493483758,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003198789088516506,
      "loss": 4.8882,
      "mean_token_accuracy": 0.21632859259843826,
      "num_tokens": 98185144.0,
      "step": 56610
    },
    {
      "entropy": 5.391887569427491,
      "epoch": 4.404823964209298,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031985174985172467,
      "loss": 4.7851,
      "mean_token_accuracy": 0.22056968063116072,
      "num_tokens": 98194274.0,
      "step": 56615
    },
    {
      "entropy": 5.324222612380981,
      "epoch": 4.405212993581015,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003198245901712449,
      "loss": 4.761,
      "mean_token_accuracy": 0.22647663503885268,
      "num_tokens": 98202790.0,
      "step": 56620
    },
    {
      "entropy": 5.4077253341674805,
      "epoch": 4.405602022952733,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003197974298106232,
      "loss": 4.8581,
      "mean_token_accuracy": 0.2165965810418129,
      "num_tokens": 98211491.0,
      "step": 56625
    },
    {
      "entropy": 5.454235982894898,
      "epoch": 4.405991052324451,
      "grad_norm": 1.125,
      "learning_rate": 0.00031977026877027186,
      "loss": 4.792,
      "mean_token_accuracy": 0.22515439242124557,
      "num_tokens": 98221506.0,
      "step": 56630
    },
    {
      "entropy": 5.384983921051026,
      "epoch": 4.4063800816961685,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031974310705060297,
      "loss": 4.7857,
      "mean_token_accuracy": 0.22442274987697602,
      "num_tokens": 98229555.0,
      "step": 56635
    },
    {
      "entropy": 5.353589630126953,
      "epoch": 4.406769111067885,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003197159446520286,
      "loss": 4.8008,
      "mean_token_accuracy": 0.21323480159044267,
      "num_tokens": 98238239.0,
      "step": 56640
    },
    {
      "entropy": 5.459129047393799,
      "epoch": 4.407158140439603,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031968878157496105,
      "loss": 4.7925,
      "mean_token_accuracy": 0.22283149212598802,
      "num_tokens": 98247437.0,
      "step": 56645
    },
    {
      "entropy": 5.423033285140991,
      "epoch": 4.407547169811321,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031966161781981224,
      "loss": 4.8475,
      "mean_token_accuracy": 0.21883999109268187,
      "num_tokens": 98256244.0,
      "step": 56650
    },
    {
      "entropy": 5.286911296844482,
      "epoch": 4.407936199183038,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031963445338699446,
      "loss": 4.5891,
      "mean_token_accuracy": 0.23325791209936142,
      "num_tokens": 98264331.0,
      "step": 56655
    },
    {
      "entropy": 5.377865505218506,
      "epoch": 4.408325228554756,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031960728827692,
      "loss": 4.7523,
      "mean_token_accuracy": 0.2225775197148323,
      "num_tokens": 98272753.0,
      "step": 56660
    },
    {
      "entropy": 5.368564653396606,
      "epoch": 4.408714257926474,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003195801224900009,
      "loss": 4.7619,
      "mean_token_accuracy": 0.22827353328466415,
      "num_tokens": 98281664.0,
      "step": 56665
    },
    {
      "entropy": 5.389024543762207,
      "epoch": 4.409103287298191,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031955295602664936,
      "loss": 4.8584,
      "mean_token_accuracy": 0.21355960071086882,
      "num_tokens": 98290576.0,
      "step": 56670
    },
    {
      "entropy": 5.468315076828003,
      "epoch": 4.409492316669908,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031952578888727763,
      "loss": 4.8822,
      "mean_token_accuracy": 0.214592607319355,
      "num_tokens": 98300108.0,
      "step": 56675
    },
    {
      "entropy": 5.5010699272155765,
      "epoch": 4.409881346041626,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031949862107229796,
      "loss": 4.8444,
      "mean_token_accuracy": 0.21984684020280837,
      "num_tokens": 98308071.0,
      "step": 56680
    },
    {
      "entropy": 5.476771354675293,
      "epoch": 4.410270375413344,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003194714525821225,
      "loss": 4.8786,
      "mean_token_accuracy": 0.21113119572401046,
      "num_tokens": 98316614.0,
      "step": 56685
    },
    {
      "entropy": 5.434524917602539,
      "epoch": 4.410659404785061,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003194442834171636,
      "loss": 4.804,
      "mean_token_accuracy": 0.2225937142968178,
      "num_tokens": 98325464.0,
      "step": 56690
    },
    {
      "entropy": 5.387014579772949,
      "epoch": 4.411048434156779,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003194171135778334,
      "loss": 4.7843,
      "mean_token_accuracy": 0.22121630162000655,
      "num_tokens": 98334559.0,
      "step": 56695
    },
    {
      "entropy": 5.350517892837525,
      "epoch": 4.411437463528497,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003193899430645442,
      "loss": 4.7282,
      "mean_token_accuracy": 0.22359798848628998,
      "num_tokens": 98342853.0,
      "step": 56700
    },
    {
      "entropy": 5.368137311935425,
      "epoch": 4.411826492900214,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031936277187770824,
      "loss": 4.7528,
      "mean_token_accuracy": 0.2278421014547348,
      "num_tokens": 98351600.0,
      "step": 56705
    },
    {
      "entropy": 5.403238201141358,
      "epoch": 4.412215522271931,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003193356000177378,
      "loss": 4.7753,
      "mean_token_accuracy": 0.22316981554031373,
      "num_tokens": 98360045.0,
      "step": 56710
    },
    {
      "entropy": 5.4411109447479244,
      "epoch": 4.412604551643649,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003193084274850453,
      "loss": 4.8565,
      "mean_token_accuracy": 0.21883872002363206,
      "num_tokens": 98367721.0,
      "step": 56715
    },
    {
      "entropy": 5.4236836433410645,
      "epoch": 4.412993581015367,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031928125428004286,
      "loss": 4.8996,
      "mean_token_accuracy": 0.21294945925474168,
      "num_tokens": 98376314.0,
      "step": 56720
    },
    {
      "entropy": 5.37960901260376,
      "epoch": 4.413382610387084,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003192540804031429,
      "loss": 4.8468,
      "mean_token_accuracy": 0.2191078320145607,
      "num_tokens": 98385369.0,
      "step": 56725
    },
    {
      "entropy": 5.361766815185547,
      "epoch": 4.413771639758802,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003192269058547577,
      "loss": 4.7107,
      "mean_token_accuracy": 0.22600883096456528,
      "num_tokens": 98393733.0,
      "step": 56730
    },
    {
      "entropy": 5.415708208084107,
      "epoch": 4.41416066913052,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003191997306352996,
      "loss": 4.7988,
      "mean_token_accuracy": 0.23034971952438354,
      "num_tokens": 98403188.0,
      "step": 56735
    },
    {
      "entropy": 5.398107814788818,
      "epoch": 4.4145496985022366,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000319172554745181,
      "loss": 4.7877,
      "mean_token_accuracy": 0.2256931781768799,
      "num_tokens": 98411839.0,
      "step": 56740
    },
    {
      "entropy": 5.482608222961426,
      "epoch": 4.414938727873954,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031914537818481404,
      "loss": 4.9543,
      "mean_token_accuracy": 0.21508198529481887,
      "num_tokens": 98420333.0,
      "step": 56745
    },
    {
      "entropy": 5.483489942550659,
      "epoch": 4.415327757245672,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003191182009546113,
      "loss": 4.8328,
      "mean_token_accuracy": 0.22132343649864197,
      "num_tokens": 98428903.0,
      "step": 56750
    },
    {
      "entropy": 5.368740653991699,
      "epoch": 4.41571678661739,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031909102305498514,
      "loss": 4.7631,
      "mean_token_accuracy": 0.22474606484174728,
      "num_tokens": 98437088.0,
      "step": 56755
    },
    {
      "entropy": 5.435957765579223,
      "epoch": 4.416105815989107,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031906384448634784,
      "loss": 4.8394,
      "mean_token_accuracy": 0.21721961796283723,
      "num_tokens": 98447124.0,
      "step": 56760
    },
    {
      "entropy": 5.349113988876343,
      "epoch": 4.416494845360825,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031903666524911185,
      "loss": 4.7171,
      "mean_token_accuracy": 0.22939716130495072,
      "num_tokens": 98456144.0,
      "step": 56765
    },
    {
      "entropy": 5.440991067886353,
      "epoch": 4.416883874732543,
      "grad_norm": 1.125,
      "learning_rate": 0.00031900948534368954,
      "loss": 4.8542,
      "mean_token_accuracy": 0.21945908516645432,
      "num_tokens": 98466011.0,
      "step": 56770
    },
    {
      "entropy": 5.360636806488037,
      "epoch": 4.4172729041042595,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031898230477049334,
      "loss": 4.7138,
      "mean_token_accuracy": 0.22861912548542024,
      "num_tokens": 98474633.0,
      "step": 56775
    },
    {
      "entropy": 5.333235263824463,
      "epoch": 4.417661933475977,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003189551235299357,
      "loss": 4.6929,
      "mean_token_accuracy": 0.23141441941261293,
      "num_tokens": 98483217.0,
      "step": 56780
    },
    {
      "entropy": 5.424423122406006,
      "epoch": 4.418050962847695,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003189279416224289,
      "loss": 4.8822,
      "mean_token_accuracy": 0.2162069112062454,
      "num_tokens": 98491797.0,
      "step": 56785
    },
    {
      "entropy": 5.347007083892822,
      "epoch": 4.418439992219413,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031890075904838556,
      "loss": 4.7911,
      "mean_token_accuracy": 0.21958240866661072,
      "num_tokens": 98501051.0,
      "step": 56790
    },
    {
      "entropy": 5.452200078964234,
      "epoch": 4.41882902159113,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031887357580821817,
      "loss": 4.8638,
      "mean_token_accuracy": 0.21950167715549468,
      "num_tokens": 98509130.0,
      "step": 56795
    },
    {
      "entropy": 5.424875545501709,
      "epoch": 4.419218050962848,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031884639190233905,
      "loss": 4.7645,
      "mean_token_accuracy": 0.21998487263917924,
      "num_tokens": 98517750.0,
      "step": 56800
    },
    {
      "entropy": 5.340837717056274,
      "epoch": 4.419607080334565,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003188192073311608,
      "loss": 4.7523,
      "mean_token_accuracy": 0.2215457633137703,
      "num_tokens": 98526515.0,
      "step": 56805
    },
    {
      "entropy": 5.319750499725342,
      "epoch": 4.4199961097062825,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031879202209509564,
      "loss": 4.7366,
      "mean_token_accuracy": 0.23112548142671585,
      "num_tokens": 98535067.0,
      "step": 56810
    },
    {
      "entropy": 5.381924152374268,
      "epoch": 4.420385139078,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031876483619455633,
      "loss": 4.7171,
      "mean_token_accuracy": 0.23041563630104064,
      "num_tokens": 98543040.0,
      "step": 56815
    },
    {
      "entropy": 5.303400564193725,
      "epoch": 4.420774168449718,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003187376496299553,
      "loss": 4.7548,
      "mean_token_accuracy": 0.22649939805269242,
      "num_tokens": 98551988.0,
      "step": 56820
    },
    {
      "entropy": 5.320533561706543,
      "epoch": 4.421163197821436,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000318710462401705,
      "loss": 4.6877,
      "mean_token_accuracy": 0.22183310985565186,
      "num_tokens": 98560866.0,
      "step": 56825
    },
    {
      "entropy": 5.363276720046997,
      "epoch": 4.421552227193153,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031868327451021816,
      "loss": 4.804,
      "mean_token_accuracy": 0.22242223471403122,
      "num_tokens": 98569258.0,
      "step": 56830
    },
    {
      "entropy": 5.394886589050293,
      "epoch": 4.421941256564871,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031865608595590706,
      "loss": 4.8254,
      "mean_token_accuracy": 0.21386872082948685,
      "num_tokens": 98577152.0,
      "step": 56835
    },
    {
      "entropy": 5.386197900772094,
      "epoch": 4.422330285936588,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031862889673918437,
      "loss": 4.7846,
      "mean_token_accuracy": 0.21988025456666946,
      "num_tokens": 98585677.0,
      "step": 56840
    },
    {
      "entropy": 5.430694103240967,
      "epoch": 4.4227193153083055,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003186017068604626,
      "loss": 4.7901,
      "mean_token_accuracy": 0.22707925438880922,
      "num_tokens": 98595425.0,
      "step": 56845
    },
    {
      "entropy": 5.474516725540161,
      "epoch": 4.423108344680023,
      "grad_norm": 1.125,
      "learning_rate": 0.0003185745163201543,
      "loss": 4.8459,
      "mean_token_accuracy": 0.21679288893938065,
      "num_tokens": 98603738.0,
      "step": 56850
    },
    {
      "entropy": 5.339152908325195,
      "epoch": 4.423497374051741,
      "grad_norm": 1.125,
      "learning_rate": 0.0003185473251186721,
      "loss": 4.6818,
      "mean_token_accuracy": 0.2299037054181099,
      "num_tokens": 98612144.0,
      "step": 56855
    },
    {
      "entropy": 5.403916692733764,
      "epoch": 4.423886403423459,
      "grad_norm": 1.25,
      "learning_rate": 0.0003185201332564285,
      "loss": 4.7806,
      "mean_token_accuracy": 0.2258450523018837,
      "num_tokens": 98619841.0,
      "step": 56860
    },
    {
      "entropy": 5.346432065963745,
      "epoch": 4.424275432795176,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003184929407338362,
      "loss": 4.7236,
      "mean_token_accuracy": 0.2239163488149643,
      "num_tokens": 98628136.0,
      "step": 56865
    },
    {
      "entropy": 5.477701044082641,
      "epoch": 4.424664462166893,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031846574755130773,
      "loss": 4.8458,
      "mean_token_accuracy": 0.2170452892780304,
      "num_tokens": 98636965.0,
      "step": 56870
    },
    {
      "entropy": 5.379834747314453,
      "epoch": 4.425053491538611,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031843855370925574,
      "loss": 4.79,
      "mean_token_accuracy": 0.21598914861679078,
      "num_tokens": 98646527.0,
      "step": 56875
    },
    {
      "entropy": 5.369980382919311,
      "epoch": 4.4254425209103285,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031841135920809285,
      "loss": 4.7787,
      "mean_token_accuracy": 0.2245008021593094,
      "num_tokens": 98655489.0,
      "step": 56880
    },
    {
      "entropy": 5.442750930786133,
      "epoch": 4.425831550282046,
      "grad_norm": 1.125,
      "learning_rate": 0.0003183841640482316,
      "loss": 4.7801,
      "mean_token_accuracy": 0.222225821018219,
      "num_tokens": 98664134.0,
      "step": 56885
    },
    {
      "entropy": 5.391176223754883,
      "epoch": 4.426220579653764,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031835696823008476,
      "loss": 4.7053,
      "mean_token_accuracy": 0.22763820439577104,
      "num_tokens": 98672307.0,
      "step": 56890
    },
    {
      "entropy": 5.426659488677979,
      "epoch": 4.426609609025482,
      "grad_norm": 1.25,
      "learning_rate": 0.00031832977175406496,
      "loss": 4.8923,
      "mean_token_accuracy": 0.22413357049226762,
      "num_tokens": 98680550.0,
      "step": 56895
    },
    {
      "entropy": 5.412976408004761,
      "epoch": 4.426998638397199,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003183025746205848,
      "loss": 4.7715,
      "mean_token_accuracy": 0.23026934266090393,
      "num_tokens": 98689313.0,
      "step": 56900
    },
    {
      "entropy": 5.4550799369812015,
      "epoch": 4.427387667768917,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003182753768300571,
      "loss": 4.8268,
      "mean_token_accuracy": 0.22053991556167601,
      "num_tokens": 98698658.0,
      "step": 56905
    },
    {
      "entropy": 5.461801481246948,
      "epoch": 4.427776697140634,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003182481783828943,
      "loss": 4.8187,
      "mean_token_accuracy": 0.2197228491306305,
      "num_tokens": 98707583.0,
      "step": 56910
    },
    {
      "entropy": 5.412831163406372,
      "epoch": 4.4281657265123515,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031822097927950934,
      "loss": 4.7736,
      "mean_token_accuracy": 0.22417523711919785,
      "num_tokens": 98716306.0,
      "step": 56915
    },
    {
      "entropy": 5.435632133483887,
      "epoch": 4.428554755884069,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031819377952031476,
      "loss": 4.9035,
      "mean_token_accuracy": 0.2131044462323189,
      "num_tokens": 98724394.0,
      "step": 56920
    },
    {
      "entropy": 5.427605438232422,
      "epoch": 4.428943785255787,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031816657910572327,
      "loss": 4.8331,
      "mean_token_accuracy": 0.21631375551223755,
      "num_tokens": 98733537.0,
      "step": 56925
    },
    {
      "entropy": 5.4513959884643555,
      "epoch": 4.429332814627505,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031813937803614774,
      "loss": 4.8292,
      "mean_token_accuracy": 0.2173965334892273,
      "num_tokens": 98742373.0,
      "step": 56930
    },
    {
      "entropy": 5.408585023880005,
      "epoch": 4.429721843999222,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003181121763120008,
      "loss": 4.7986,
      "mean_token_accuracy": 0.21813859343528746,
      "num_tokens": 98751184.0,
      "step": 56935
    },
    {
      "entropy": 5.505879831314087,
      "epoch": 4.430110873370939,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003180849739336952,
      "loss": 4.8606,
      "mean_token_accuracy": 0.217446368932724,
      "num_tokens": 98760589.0,
      "step": 56940
    },
    {
      "entropy": 5.449555158615112,
      "epoch": 4.430499902742657,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031805777090164364,
      "loss": 4.751,
      "mean_token_accuracy": 0.21862780898809434,
      "num_tokens": 98768929.0,
      "step": 56945
    },
    {
      "entropy": 5.367335844039917,
      "epoch": 4.4308889321143745,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000318030567216259,
      "loss": 4.6631,
      "mean_token_accuracy": 0.22898049503564835,
      "num_tokens": 98777102.0,
      "step": 56950
    },
    {
      "entropy": 5.4575944423675535,
      "epoch": 4.431277961486092,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031800336287795395,
      "loss": 4.8348,
      "mean_token_accuracy": 0.21976025700569152,
      "num_tokens": 98785315.0,
      "step": 56955
    },
    {
      "entropy": 5.424097871780395,
      "epoch": 4.43166699085781,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031797615788714133,
      "loss": 4.7563,
      "mean_token_accuracy": 0.21845750212669374,
      "num_tokens": 98793689.0,
      "step": 56960
    },
    {
      "entropy": 5.441212558746338,
      "epoch": 4.432056020229528,
      "grad_norm": 1.25,
      "learning_rate": 0.00031794895224423396,
      "loss": 4.8752,
      "mean_token_accuracy": 0.2216723769903183,
      "num_tokens": 98802628.0,
      "step": 56965
    },
    {
      "entropy": 5.420555591583252,
      "epoch": 4.432445049601245,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003179217459496446,
      "loss": 4.7524,
      "mean_token_accuracy": 0.22800272554159165,
      "num_tokens": 98810722.0,
      "step": 56970
    },
    {
      "entropy": 5.458615684509278,
      "epoch": 4.432834078972962,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000317894539003786,
      "loss": 4.7912,
      "mean_token_accuracy": 0.2256120651960373,
      "num_tokens": 98819398.0,
      "step": 56975
    },
    {
      "entropy": 5.426866817474365,
      "epoch": 4.43322310834468,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000317867331407071,
      "loss": 4.8428,
      "mean_token_accuracy": 0.22009177953004838,
      "num_tokens": 98827724.0,
      "step": 56980
    },
    {
      "entropy": 5.438529109954834,
      "epoch": 4.4336121377163975,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003178401231599126,
      "loss": 4.7777,
      "mean_token_accuracy": 0.22300432473421097,
      "num_tokens": 98836680.0,
      "step": 56985
    },
    {
      "entropy": 5.2925543785095215,
      "epoch": 4.434001167088115,
      "grad_norm": 1.125,
      "learning_rate": 0.00031781291426272347,
      "loss": 4.6747,
      "mean_token_accuracy": 0.23232204467058182,
      "num_tokens": 98845746.0,
      "step": 56990
    },
    {
      "entropy": 5.4284608364105225,
      "epoch": 4.434390196459833,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003177857047159165,
      "loss": 4.7995,
      "mean_token_accuracy": 0.21908573806285858,
      "num_tokens": 98854293.0,
      "step": 56995
    },
    {
      "entropy": 5.396098947525024,
      "epoch": 4.434779225831551,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003177584945199046,
      "loss": 4.7598,
      "mean_token_accuracy": 0.21752981543540956,
      "num_tokens": 98862747.0,
      "step": 57000
    },
    {
      "epoch": 4.434779225831551,
      "eval_entropy": 5.132750196108116,
      "eval_loss": 4.937704563140869,
      "eval_mean_token_accuracy": 0.22257486845692037,
      "eval_num_tokens": 98862747.0,
      "eval_runtime": 28.6287,
      "eval_samples_per_second": 1451.62,
      "eval_steps_per_second": 181.461,
      "step": 57000
    },
    {
      "entropy": 5.3947896480560305,
      "epoch": 4.435168255203267,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003177312836751005,
      "loss": 4.8259,
      "mean_token_accuracy": 0.21537525057792664,
      "num_tokens": 98871867.0,
      "step": 57005
    },
    {
      "entropy": 5.435635566711426,
      "epoch": 4.435557284574985,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031770407218191727,
      "loss": 4.787,
      "mean_token_accuracy": 0.2182581678032875,
      "num_tokens": 98880780.0,
      "step": 57010
    },
    {
      "entropy": 5.334830284118652,
      "epoch": 4.435946313946703,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003176768600407677,
      "loss": 4.7652,
      "mean_token_accuracy": 0.23272209167480468,
      "num_tokens": 98890265.0,
      "step": 57015
    },
    {
      "entropy": 5.41881217956543,
      "epoch": 4.4363353433184205,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003176496472520647,
      "loss": 4.7641,
      "mean_token_accuracy": 0.21972244381904601,
      "num_tokens": 98899406.0,
      "step": 57020
    },
    {
      "entropy": 5.402377367019653,
      "epoch": 4.436724372690138,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003176224338162212,
      "loss": 4.8138,
      "mean_token_accuracy": 0.22147863060235978,
      "num_tokens": 98908492.0,
      "step": 57025
    },
    {
      "entropy": 5.313281869888305,
      "epoch": 4.437113402061856,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031759521973365006,
      "loss": 4.716,
      "mean_token_accuracy": 0.23108238726854324,
      "num_tokens": 98917339.0,
      "step": 57030
    },
    {
      "entropy": 5.362147665023803,
      "epoch": 4.437502431433574,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031756800500476435,
      "loss": 4.7478,
      "mean_token_accuracy": 0.2226343870162964,
      "num_tokens": 98926350.0,
      "step": 57035
    },
    {
      "entropy": 5.3626714706420895,
      "epoch": 4.43789146080529,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031754078962997686,
      "loss": 4.7903,
      "mean_token_accuracy": 0.22022322565317154,
      "num_tokens": 98936009.0,
      "step": 57040
    },
    {
      "entropy": 5.35137038230896,
      "epoch": 4.438280490177008,
      "grad_norm": 1.125,
      "learning_rate": 0.0003175135736097006,
      "loss": 4.7574,
      "mean_token_accuracy": 0.2237767830491066,
      "num_tokens": 98944350.0,
      "step": 57045
    },
    {
      "entropy": 5.409536218643188,
      "epoch": 4.438669519548726,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031748635694434844,
      "loss": 4.8169,
      "mean_token_accuracy": 0.22062480300664902,
      "num_tokens": 98953611.0,
      "step": 57050
    },
    {
      "entropy": 5.373613023757935,
      "epoch": 4.4390585489204435,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003174591396343336,
      "loss": 4.6692,
      "mean_token_accuracy": 0.23507629632949828,
      "num_tokens": 98962084.0,
      "step": 57055
    },
    {
      "entropy": 5.334583473205567,
      "epoch": 4.439447578292161,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003174319216800688,
      "loss": 4.7664,
      "mean_token_accuracy": 0.22832910865545272,
      "num_tokens": 98971278.0,
      "step": 57060
    },
    {
      "entropy": 5.328181648254395,
      "epoch": 4.439836607663879,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031740470308196724,
      "loss": 4.6921,
      "mean_token_accuracy": 0.23726969212293625,
      "num_tokens": 98979597.0,
      "step": 57065
    },
    {
      "entropy": 5.419636964797974,
      "epoch": 4.440225637035597,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003173774838404417,
      "loss": 4.7352,
      "mean_token_accuracy": 0.22396946996450423,
      "num_tokens": 98988326.0,
      "step": 57070
    },
    {
      "entropy": 5.4224522590637205,
      "epoch": 4.440614666407313,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031735026395590535,
      "loss": 4.7752,
      "mean_token_accuracy": 0.22019947469234466,
      "num_tokens": 98996809.0,
      "step": 57075
    },
    {
      "entropy": 5.398856115341187,
      "epoch": 4.441003695779031,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000317323043428771,
      "loss": 4.7432,
      "mean_token_accuracy": 0.22793539315462114,
      "num_tokens": 99005934.0,
      "step": 57080
    },
    {
      "entropy": 5.366226053237915,
      "epoch": 4.441392725150749,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003172958222594519,
      "loss": 4.8199,
      "mean_token_accuracy": 0.2224924236536026,
      "num_tokens": 99014940.0,
      "step": 57085
    },
    {
      "entropy": 5.394231510162354,
      "epoch": 4.4417817545224665,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003172686004483611,
      "loss": 4.8296,
      "mean_token_accuracy": 0.22607219517230986,
      "num_tokens": 99023609.0,
      "step": 57090
    },
    {
      "entropy": 5.419731330871582,
      "epoch": 4.442170783894184,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031724137799591154,
      "loss": 4.7948,
      "mean_token_accuracy": 0.22202716916799545,
      "num_tokens": 99032018.0,
      "step": 57095
    },
    {
      "entropy": 5.460139799118042,
      "epoch": 4.442559813265902,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031721415490251634,
      "loss": 4.8845,
      "mean_token_accuracy": 0.208880278468132,
      "num_tokens": 99040605.0,
      "step": 57100
    },
    {
      "entropy": 5.462630939483643,
      "epoch": 4.44294884263762,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003171869311685885,
      "loss": 4.8028,
      "mean_token_accuracy": 0.22756650149822236,
      "num_tokens": 99048540.0,
      "step": 57105
    },
    {
      "entropy": 5.311354541778565,
      "epoch": 4.443337872009336,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003171597067945411,
      "loss": 4.6696,
      "mean_token_accuracy": 0.23524145931005477,
      "num_tokens": 99057923.0,
      "step": 57110
    },
    {
      "entropy": 5.388445854187012,
      "epoch": 4.443726901381054,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031713248178078717,
      "loss": 4.7299,
      "mean_token_accuracy": 0.22386448830366135,
      "num_tokens": 99066537.0,
      "step": 57115
    },
    {
      "entropy": 5.4314391136169435,
      "epoch": 4.444115930752772,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031710525612774,
      "loss": 4.8302,
      "mean_token_accuracy": 0.2184027150273323,
      "num_tokens": 99074780.0,
      "step": 57120
    },
    {
      "entropy": 5.362411022186279,
      "epoch": 4.4445049601244895,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003170780298358126,
      "loss": 4.7372,
      "mean_token_accuracy": 0.2222791314125061,
      "num_tokens": 99082883.0,
      "step": 57125
    },
    {
      "entropy": 5.389003324508667,
      "epoch": 4.444893989496207,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031705080290541797,
      "loss": 4.7825,
      "mean_token_accuracy": 0.22626209259033203,
      "num_tokens": 99091877.0,
      "step": 57130
    },
    {
      "entropy": 5.426477766036987,
      "epoch": 4.445283018867925,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031702357533696946,
      "loss": 4.7005,
      "mean_token_accuracy": 0.23964035660028457,
      "num_tokens": 99099860.0,
      "step": 57135
    },
    {
      "entropy": 5.357645845413208,
      "epoch": 4.445672048239642,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031699634713088,
      "loss": 4.7174,
      "mean_token_accuracy": 0.23085438907146455,
      "num_tokens": 99107940.0,
      "step": 57140
    },
    {
      "entropy": 5.381301975250244,
      "epoch": 4.446061077611359,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003169691182875628,
      "loss": 4.8088,
      "mean_token_accuracy": 0.2232929363846779,
      "num_tokens": 99116786.0,
      "step": 57145
    },
    {
      "entropy": 5.399720287322998,
      "epoch": 4.446450106983077,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003169418888074311,
      "loss": 4.7248,
      "mean_token_accuracy": 0.22199890464544297,
      "num_tokens": 99126444.0,
      "step": 57150
    },
    {
      "entropy": 5.4557818412780765,
      "epoch": 4.446839136354795,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031691465869089795,
      "loss": 4.8164,
      "mean_token_accuracy": 0.21623874008655547,
      "num_tokens": 99134537.0,
      "step": 57155
    },
    {
      "entropy": 5.3911937236785885,
      "epoch": 4.4472281657265125,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003168874279383766,
      "loss": 4.7609,
      "mean_token_accuracy": 0.22416484355926514,
      "num_tokens": 99142871.0,
      "step": 57160
    },
    {
      "entropy": 5.333127498626709,
      "epoch": 4.44761719509823,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003168601965502802,
      "loss": 4.6995,
      "mean_token_accuracy": 0.22946427166461944,
      "num_tokens": 99150444.0,
      "step": 57165
    },
    {
      "entropy": 5.342962980270386,
      "epoch": 4.448006224469948,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003168329645270219,
      "loss": 4.7509,
      "mean_token_accuracy": 0.2350730910897255,
      "num_tokens": 99159332.0,
      "step": 57170
    },
    {
      "entropy": 5.435609483718872,
      "epoch": 4.448395253841665,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000316805731869015,
      "loss": 4.8977,
      "mean_token_accuracy": 0.21870713531970978,
      "num_tokens": 99168927.0,
      "step": 57175
    },
    {
      "entropy": 5.389696788787842,
      "epoch": 4.448784283213382,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031677849857667266,
      "loss": 4.7949,
      "mean_token_accuracy": 0.2257964640855789,
      "num_tokens": 99177491.0,
      "step": 57180
    },
    {
      "entropy": 5.4416810989379885,
      "epoch": 4.4491733125851,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003167512646504081,
      "loss": 4.8012,
      "mean_token_accuracy": 0.21793338507413865,
      "num_tokens": 99186392.0,
      "step": 57185
    },
    {
      "entropy": 5.392916822433472,
      "epoch": 4.449562341956818,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003167240300906345,
      "loss": 4.8399,
      "mean_token_accuracy": 0.22277383655309677,
      "num_tokens": 99195242.0,
      "step": 57190
    },
    {
      "entropy": 5.449929428100586,
      "epoch": 4.4499513713285355,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003166967948977652,
      "loss": 4.7966,
      "mean_token_accuracy": 0.2263351112604141,
      "num_tokens": 99203167.0,
      "step": 57195
    },
    {
      "entropy": 5.413981819152832,
      "epoch": 4.450340400700253,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003166695590722134,
      "loss": 4.7073,
      "mean_token_accuracy": 0.2333373725414276,
      "num_tokens": 99211417.0,
      "step": 57200
    },
    {
      "entropy": 5.378722095489502,
      "epoch": 4.45072943007197,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031664232261439235,
      "loss": 4.7502,
      "mean_token_accuracy": 0.2209574967622757,
      "num_tokens": 99219658.0,
      "step": 57205
    },
    {
      "entropy": 5.31967306137085,
      "epoch": 4.451118459443688,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003166150855247153,
      "loss": 4.7495,
      "mean_token_accuracy": 0.22316686809062958,
      "num_tokens": 99228335.0,
      "step": 57210
    },
    {
      "entropy": 5.40112419128418,
      "epoch": 4.451507488815405,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003165878478035956,
      "loss": 4.7662,
      "mean_token_accuracy": 0.2244488388299942,
      "num_tokens": 99236500.0,
      "step": 57215
    },
    {
      "entropy": 5.468771314620971,
      "epoch": 4.451896518187123,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003165606094514465,
      "loss": 4.8312,
      "mean_token_accuracy": 0.21763159483671188,
      "num_tokens": 99246834.0,
      "step": 57220
    },
    {
      "entropy": 5.383902454376221,
      "epoch": 4.452285547558841,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031653337046868137,
      "loss": 4.7558,
      "mean_token_accuracy": 0.2263956770300865,
      "num_tokens": 99255387.0,
      "step": 57225
    },
    {
      "entropy": 5.330097293853759,
      "epoch": 4.4526745769305585,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031650613085571334,
      "loss": 4.6032,
      "mean_token_accuracy": 0.23518190979957582,
      "num_tokens": 99263403.0,
      "step": 57230
    },
    {
      "entropy": 5.340671110153198,
      "epoch": 4.453063606302276,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003164788906129559,
      "loss": 4.7459,
      "mean_token_accuracy": 0.23140041083097457,
      "num_tokens": 99271563.0,
      "step": 57235
    },
    {
      "entropy": 5.330229234695435,
      "epoch": 4.453452635673994,
      "grad_norm": 1.125,
      "learning_rate": 0.0003164516497408223,
      "loss": 4.6761,
      "mean_token_accuracy": 0.22964081913232803,
      "num_tokens": 99280075.0,
      "step": 57240
    },
    {
      "entropy": 5.3949438571929935,
      "epoch": 4.453841665045711,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003164244082397259,
      "loss": 4.7703,
      "mean_token_accuracy": 0.22245306968688966,
      "num_tokens": 99288907.0,
      "step": 57245
    },
    {
      "entropy": 5.39880862236023,
      "epoch": 4.454230694417428,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031639716611008,
      "loss": 4.7508,
      "mean_token_accuracy": 0.22721540331840515,
      "num_tokens": 99297468.0,
      "step": 57250
    },
    {
      "entropy": 5.372001647949219,
      "epoch": 4.454619723789146,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003163699233522981,
      "loss": 4.7322,
      "mean_token_accuracy": 0.23018756359815598,
      "num_tokens": 99306161.0,
      "step": 57255
    },
    {
      "entropy": 5.422857475280762,
      "epoch": 4.455008753160864,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003163426799667934,
      "loss": 4.7314,
      "mean_token_accuracy": 0.23585088104009627,
      "num_tokens": 99315531.0,
      "step": 57260
    },
    {
      "entropy": 5.3585618019104,
      "epoch": 4.4553977825325815,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003163154359539793,
      "loss": 4.7975,
      "mean_token_accuracy": 0.22313887625932693,
      "num_tokens": 99324263.0,
      "step": 57265
    },
    {
      "entropy": 5.370906686782837,
      "epoch": 4.455786811904299,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031628819131426924,
      "loss": 4.799,
      "mean_token_accuracy": 0.2234163150191307,
      "num_tokens": 99333542.0,
      "step": 57270
    },
    {
      "entropy": 5.421927881240845,
      "epoch": 4.456175841276016,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003162609460480766,
      "loss": 4.7789,
      "mean_token_accuracy": 0.2176763042807579,
      "num_tokens": 99341860.0,
      "step": 57275
    },
    {
      "entropy": 5.431547689437866,
      "epoch": 4.456564870647734,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003162337001558147,
      "loss": 4.8458,
      "mean_token_accuracy": 0.21868441104888917,
      "num_tokens": 99351060.0,
      "step": 57280
    },
    {
      "entropy": 5.4300354480743405,
      "epoch": 4.456953900019451,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031620645363789713,
      "loss": 4.7882,
      "mean_token_accuracy": 0.22376328557729722,
      "num_tokens": 99359318.0,
      "step": 57285
    },
    {
      "entropy": 5.385527420043945,
      "epoch": 4.457342929391169,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003161792064947372,
      "loss": 4.6957,
      "mean_token_accuracy": 0.23153160512447357,
      "num_tokens": 99367538.0,
      "step": 57290
    },
    {
      "entropy": 5.324669981002808,
      "epoch": 4.457731958762887,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031615195872674836,
      "loss": 4.7005,
      "mean_token_accuracy": 0.22496049702167512,
      "num_tokens": 99375649.0,
      "step": 57295
    },
    {
      "entropy": 5.378129148483277,
      "epoch": 4.4581209881346044,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031612471033434406,
      "loss": 4.7699,
      "mean_token_accuracy": 0.22402811497449876,
      "num_tokens": 99384557.0,
      "step": 57300
    },
    {
      "entropy": 5.384200620651245,
      "epoch": 4.458510017506322,
      "grad_norm": 1.25,
      "learning_rate": 0.0003160974613179377,
      "loss": 4.7326,
      "mean_token_accuracy": 0.23265922963619232,
      "num_tokens": 99392752.0,
      "step": 57305
    },
    {
      "entropy": 5.430941820144653,
      "epoch": 4.458899046878039,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003160702116779428,
      "loss": 4.8545,
      "mean_token_accuracy": 0.2132594957947731,
      "num_tokens": 99402060.0,
      "step": 57310
    },
    {
      "entropy": 5.452469825744629,
      "epoch": 4.459288076249757,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003160429614147727,
      "loss": 4.8459,
      "mean_token_accuracy": 0.21793683767318725,
      "num_tokens": 99410992.0,
      "step": 57315
    },
    {
      "entropy": 5.423767232894898,
      "epoch": 4.459677105621474,
      "grad_norm": 1.125,
      "learning_rate": 0.00031601571052884117,
      "loss": 4.8159,
      "mean_token_accuracy": 0.21756284236907958,
      "num_tokens": 99419703.0,
      "step": 57320
    },
    {
      "entropy": 5.469342136383057,
      "epoch": 4.460066134993192,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003159884590205613,
      "loss": 4.8221,
      "mean_token_accuracy": 0.22593525499105455,
      "num_tokens": 99428305.0,
      "step": 57325
    },
    {
      "entropy": 5.361087656021118,
      "epoch": 4.46045516436491,
      "grad_norm": 1.140625,
      "learning_rate": 0.000315961206890347,
      "loss": 4.7743,
      "mean_token_accuracy": 0.2220757097005844,
      "num_tokens": 99437056.0,
      "step": 57330
    },
    {
      "entropy": 5.448331737518311,
      "epoch": 4.460844193736627,
      "grad_norm": 1.125,
      "learning_rate": 0.0003159339541386115,
      "loss": 4.9536,
      "mean_token_accuracy": 0.21096302717924117,
      "num_tokens": 99446122.0,
      "step": 57335
    },
    {
      "entropy": 5.524873495101929,
      "epoch": 4.461233223108344,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031590670076576846,
      "loss": 4.8352,
      "mean_token_accuracy": 0.2186709001660347,
      "num_tokens": 99454064.0,
      "step": 57340
    },
    {
      "entropy": 5.3967857837677,
      "epoch": 4.461622252480062,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031587944677223133,
      "loss": 4.7924,
      "mean_token_accuracy": 0.22580429017543793,
      "num_tokens": 99462871.0,
      "step": 57345
    },
    {
      "entropy": 5.377498006820678,
      "epoch": 4.46201128185178,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003158521921584136,
      "loss": 4.739,
      "mean_token_accuracy": 0.22834757566452027,
      "num_tokens": 99470251.0,
      "step": 57350
    },
    {
      "entropy": 5.311630296707153,
      "epoch": 4.462400311223497,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031582493692472886,
      "loss": 4.7806,
      "mean_token_accuracy": 0.22358567267656326,
      "num_tokens": 99478761.0,
      "step": 57355
    },
    {
      "entropy": 5.386785173416138,
      "epoch": 4.462789340595215,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003157976810715907,
      "loss": 4.7772,
      "mean_token_accuracy": 0.22461158484220506,
      "num_tokens": 99487753.0,
      "step": 57360
    },
    {
      "entropy": 5.422971153259278,
      "epoch": 4.463178369966933,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031577042459941275,
      "loss": 4.7418,
      "mean_token_accuracy": 0.23000783771276473,
      "num_tokens": 99496566.0,
      "step": 57365
    },
    {
      "entropy": 5.338047695159912,
      "epoch": 4.46356739933865,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031574316750860844,
      "loss": 4.7624,
      "mean_token_accuracy": 0.2265368342399597,
      "num_tokens": 99504384.0,
      "step": 57370
    },
    {
      "entropy": 5.356696939468383,
      "epoch": 4.463956428710367,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003157159097995914,
      "loss": 4.7902,
      "mean_token_accuracy": 0.2200967401266098,
      "num_tokens": 99512945.0,
      "step": 57375
    },
    {
      "entropy": 5.456883716583252,
      "epoch": 4.464345458082085,
      "grad_norm": 1.25,
      "learning_rate": 0.00031568865147277525,
      "loss": 4.8372,
      "mean_token_accuracy": 0.22017484456300734,
      "num_tokens": 99521188.0,
      "step": 57380
    },
    {
      "entropy": 5.449392318725586,
      "epoch": 4.464734487453803,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00031566139252857357,
      "loss": 4.7622,
      "mean_token_accuracy": 0.22014689445495605,
      "num_tokens": 99528938.0,
      "step": 57385
    },
    {
      "entropy": 5.45880618095398,
      "epoch": 4.46512351682552,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003156341329674,
      "loss": 4.8512,
      "mean_token_accuracy": 0.21641667634248735,
      "num_tokens": 99538418.0,
      "step": 57390
    },
    {
      "entropy": 5.34148097038269,
      "epoch": 4.465512546197238,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003156068727896681,
      "loss": 4.6977,
      "mean_token_accuracy": 0.22954367101192474,
      "num_tokens": 99546513.0,
      "step": 57395
    },
    {
      "entropy": 5.417756080627441,
      "epoch": 4.465901575568956,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003155796119957915,
      "loss": 4.8703,
      "mean_token_accuracy": 0.20931911170482637,
      "num_tokens": 99554665.0,
      "step": 57400
    },
    {
      "entropy": 5.4124205112457275,
      "epoch": 4.466290604940673,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031555235058618403,
      "loss": 4.8282,
      "mean_token_accuracy": 0.2141205370426178,
      "num_tokens": 99563953.0,
      "step": 57405
    },
    {
      "entropy": 5.465652799606323,
      "epoch": 4.46667963431239,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031552508856125906,
      "loss": 4.8644,
      "mean_token_accuracy": 0.22333737909793855,
      "num_tokens": 99572809.0,
      "step": 57410
    },
    {
      "entropy": 5.406736421585083,
      "epoch": 4.467068663684108,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003154978259214304,
      "loss": 4.7083,
      "mean_token_accuracy": 0.22489060312509537,
      "num_tokens": 99581642.0,
      "step": 57415
    },
    {
      "entropy": 5.328680610656738,
      "epoch": 4.467457693055826,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031547056266711173,
      "loss": 4.7105,
      "mean_token_accuracy": 0.2273159831762314,
      "num_tokens": 99590261.0,
      "step": 57420
    },
    {
      "entropy": 5.37578501701355,
      "epoch": 4.467846722427543,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003154432987987166,
      "loss": 4.7857,
      "mean_token_accuracy": 0.2249089926481247,
      "num_tokens": 99598866.0,
      "step": 57425
    },
    {
      "entropy": 5.415650224685669,
      "epoch": 4.468235751799261,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003154160343166589,
      "loss": 4.7439,
      "mean_token_accuracy": 0.22042347341775895,
      "num_tokens": 99607198.0,
      "step": 57430
    },
    {
      "entropy": 5.4388586521148685,
      "epoch": 4.468624781170979,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031538876922135214,
      "loss": 4.913,
      "mean_token_accuracy": 0.20446106940507888,
      "num_tokens": 99615776.0,
      "step": 57435
    },
    {
      "entropy": 5.408769512176514,
      "epoch": 4.469013810542696,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003153615035132101,
      "loss": 4.7987,
      "mean_token_accuracy": 0.22756477296352387,
      "num_tokens": 99624240.0,
      "step": 57440
    },
    {
      "entropy": 5.380187511444092,
      "epoch": 4.469402839914413,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031533423719264645,
      "loss": 4.6908,
      "mean_token_accuracy": 0.2310074970126152,
      "num_tokens": 99632547.0,
      "step": 57445
    },
    {
      "entropy": 5.374131107330323,
      "epoch": 4.469791869286131,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000315306970260075,
      "loss": 4.8044,
      "mean_token_accuracy": 0.21942253708839415,
      "num_tokens": 99641678.0,
      "step": 57450
    },
    {
      "entropy": 5.4187358856201175,
      "epoch": 4.470180898657849,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003152797027159094,
      "loss": 4.8647,
      "mean_token_accuracy": 0.22007633745670319,
      "num_tokens": 99650090.0,
      "step": 57455
    },
    {
      "entropy": 5.413580274581909,
      "epoch": 4.470569928029566,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031525243456056345,
      "loss": 4.7582,
      "mean_token_accuracy": 0.22532665580511094,
      "num_tokens": 99659469.0,
      "step": 57460
    },
    {
      "entropy": 5.452715253829956,
      "epoch": 4.470958957401284,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003152251657944509,
      "loss": 4.8164,
      "mean_token_accuracy": 0.22347097098827362,
      "num_tokens": 99668056.0,
      "step": 57465
    },
    {
      "entropy": 5.348944664001465,
      "epoch": 4.471347986773002,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003151978964179854,
      "loss": 4.6819,
      "mean_token_accuracy": 0.23364021331071855,
      "num_tokens": 99676905.0,
      "step": 57470
    },
    {
      "entropy": 5.354820442199707,
      "epoch": 4.4717370161447185,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031517062643158085,
      "loss": 4.7585,
      "mean_token_accuracy": 0.22438254207372665,
      "num_tokens": 99684518.0,
      "step": 57475
    },
    {
      "entropy": 5.365765571594238,
      "epoch": 4.472126045516436,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031514335583565097,
      "loss": 4.7779,
      "mean_token_accuracy": 0.22294967770576476,
      "num_tokens": 99692541.0,
      "step": 57480
    },
    {
      "entropy": 5.417402648925782,
      "epoch": 4.472515074888154,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031511608463060954,
      "loss": 4.8297,
      "mean_token_accuracy": 0.21504515260457993,
      "num_tokens": 99701866.0,
      "step": 57485
    },
    {
      "entropy": 5.400264120101928,
      "epoch": 4.472904104259872,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031508881281687044,
      "loss": 4.7705,
      "mean_token_accuracy": 0.2200484722852707,
      "num_tokens": 99711825.0,
      "step": 57490
    },
    {
      "entropy": 5.37727108001709,
      "epoch": 4.473293133631589,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031506154039484736,
      "loss": 4.785,
      "mean_token_accuracy": 0.2261470824480057,
      "num_tokens": 99720635.0,
      "step": 57495
    },
    {
      "entropy": 5.429576396942139,
      "epoch": 4.473682163003307,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031503426736495424,
      "loss": 4.7831,
      "mean_token_accuracy": 0.21691728979349137,
      "num_tokens": 99728938.0,
      "step": 57500
    },
    {
      "entropy": 5.380933666229248,
      "epoch": 4.474071192375025,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031500699372760476,
      "loss": 4.7485,
      "mean_token_accuracy": 0.226787032186985,
      "num_tokens": 99737403.0,
      "step": 57505
    },
    {
      "entropy": 5.358431386947632,
      "epoch": 4.4744602217467415,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031497971948321287,
      "loss": 4.7239,
      "mean_token_accuracy": 0.22793036997318267,
      "num_tokens": 99746004.0,
      "step": 57510
    },
    {
      "entropy": 5.419115018844605,
      "epoch": 4.474849251118459,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003149524446321923,
      "loss": 4.7782,
      "mean_token_accuracy": 0.22354136556386947,
      "num_tokens": 99755337.0,
      "step": 57515
    },
    {
      "entropy": 5.412204551696777,
      "epoch": 4.475238280490177,
      "grad_norm": 1.203125,
      "learning_rate": 0.000314925169174957,
      "loss": 4.7958,
      "mean_token_accuracy": 0.22437187284231186,
      "num_tokens": 99763428.0,
      "step": 57520
    },
    {
      "entropy": 5.405033683776855,
      "epoch": 4.475627309861895,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031489789311192086,
      "loss": 4.8248,
      "mean_token_accuracy": 0.22013988494873046,
      "num_tokens": 99771939.0,
      "step": 57525
    },
    {
      "entropy": 5.416330194473266,
      "epoch": 4.476016339233612,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003148706164434977,
      "loss": 4.7824,
      "mean_token_accuracy": 0.21838113218545913,
      "num_tokens": 99780877.0,
      "step": 57530
    },
    {
      "entropy": 5.40864143371582,
      "epoch": 4.47640536860533,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003148433391701014,
      "loss": 4.7325,
      "mean_token_accuracy": 0.22760019749403,
      "num_tokens": 99789309.0,
      "step": 57535
    },
    {
      "entropy": 5.354705333709717,
      "epoch": 4.476794397977047,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003148160612921457,
      "loss": 4.7636,
      "mean_token_accuracy": 0.2218502104282379,
      "num_tokens": 99798124.0,
      "step": 57540
    },
    {
      "entropy": 5.369080114364624,
      "epoch": 4.4771834273487645,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031478878281004474,
      "loss": 4.8875,
      "mean_token_accuracy": 0.2189884975552559,
      "num_tokens": 99806773.0,
      "step": 57545
    },
    {
      "entropy": 5.326339960098267,
      "epoch": 4.477572456720482,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031476150372421234,
      "loss": 4.7376,
      "mean_token_accuracy": 0.21797392815351485,
      "num_tokens": 99815825.0,
      "step": 57550
    },
    {
      "entropy": 5.395896339416504,
      "epoch": 4.4779614860922,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003147342240350625,
      "loss": 4.7739,
      "mean_token_accuracy": 0.22774604707956314,
      "num_tokens": 99823737.0,
      "step": 57555
    },
    {
      "entropy": 5.43360390663147,
      "epoch": 4.478350515463918,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003147069437430089,
      "loss": 4.8318,
      "mean_token_accuracy": 0.22052946984767913,
      "num_tokens": 99832601.0,
      "step": 57560
    },
    {
      "entropy": 5.3546778678894045,
      "epoch": 4.478739544835635,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031467966284846573,
      "loss": 4.7248,
      "mean_token_accuracy": 0.2260730728507042,
      "num_tokens": 99840817.0,
      "step": 57565
    },
    {
      "entropy": 5.451509618759156,
      "epoch": 4.479128574207353,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003146523813518468,
      "loss": 4.7729,
      "mean_token_accuracy": 0.2280184507369995,
      "num_tokens": 99848525.0,
      "step": 57570
    },
    {
      "entropy": 5.338032913208008,
      "epoch": 4.47951760357907,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031462509925356616,
      "loss": 4.7861,
      "mean_token_accuracy": 0.22469336539506912,
      "num_tokens": 99857756.0,
      "step": 57575
    },
    {
      "entropy": 5.3849865913391115,
      "epoch": 4.4799066329507875,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031459781655403756,
      "loss": 4.7247,
      "mean_token_accuracy": 0.22604879438877107,
      "num_tokens": 99866171.0,
      "step": 57580
    },
    {
      "entropy": 5.364056777954102,
      "epoch": 4.480295662322505,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003145705332536752,
      "loss": 4.76,
      "mean_token_accuracy": 0.21728412806987762,
      "num_tokens": 99874842.0,
      "step": 57585
    },
    {
      "entropy": 5.393701648712158,
      "epoch": 4.480684691694223,
      "grad_norm": 1.21875,
      "learning_rate": 0.000314543249352893,
      "loss": 4.8051,
      "mean_token_accuracy": 0.22341737449169158,
      "num_tokens": 99882840.0,
      "step": 57590
    },
    {
      "entropy": 5.454952621459961,
      "epoch": 4.481073721065941,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031451596485210486,
      "loss": 4.7868,
      "mean_token_accuracy": 0.21950131356716157,
      "num_tokens": 99891633.0,
      "step": 57595
    },
    {
      "entropy": 5.473793172836304,
      "epoch": 4.481462750437658,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031448867975172493,
      "loss": 4.7985,
      "mean_token_accuracy": 0.2265819415450096,
      "num_tokens": 99900181.0,
      "step": 57600
    },
    {
      "entropy": 5.447542667388916,
      "epoch": 4.481851779809376,
      "grad_norm": 1.125,
      "learning_rate": 0.00031446139405216713,
      "loss": 4.8181,
      "mean_token_accuracy": 0.22350174337625503,
      "num_tokens": 99909333.0,
      "step": 57605
    },
    {
      "entropy": 5.387915515899659,
      "epoch": 4.482240809181093,
      "grad_norm": 1.25,
      "learning_rate": 0.0003144341077538455,
      "loss": 4.8402,
      "mean_token_accuracy": 0.21899865120649337,
      "num_tokens": 99918639.0,
      "step": 57610
    },
    {
      "entropy": 5.407594108581543,
      "epoch": 4.4826298385528105,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000314406820857174,
      "loss": 4.8041,
      "mean_token_accuracy": 0.22440048158168793,
      "num_tokens": 99927693.0,
      "step": 57615
    },
    {
      "entropy": 5.37995753288269,
      "epoch": 4.483018867924528,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031437953336256667,
      "loss": 4.6935,
      "mean_token_accuracy": 0.23045378923416138,
      "num_tokens": 99935885.0,
      "step": 57620
    },
    {
      "entropy": 5.474698114395141,
      "epoch": 4.483407897296246,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003143522452704377,
      "loss": 4.8845,
      "mean_token_accuracy": 0.21683274656534196,
      "num_tokens": 99944472.0,
      "step": 57625
    },
    {
      "entropy": 5.39496340751648,
      "epoch": 4.483796926667964,
      "grad_norm": 1.15625,
      "learning_rate": 0.000314324956581201,
      "loss": 4.7603,
      "mean_token_accuracy": 0.22516266703605653,
      "num_tokens": 99952376.0,
      "step": 57630
    },
    {
      "entropy": 5.397978782653809,
      "epoch": 4.484185956039681,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003142976672952707,
      "loss": 4.7984,
      "mean_token_accuracy": 0.21861375123262405,
      "num_tokens": 99961277.0,
      "step": 57635
    },
    {
      "entropy": 5.48889422416687,
      "epoch": 4.484574985411399,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031427037741306083,
      "loss": 4.8832,
      "mean_token_accuracy": 0.21433400511741638,
      "num_tokens": 99970195.0,
      "step": 57640
    },
    {
      "entropy": 5.3329792499542235,
      "epoch": 4.484964014783116,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003142430869349855,
      "loss": 4.6956,
      "mean_token_accuracy": 0.22559254318475724,
      "num_tokens": 99978580.0,
      "step": 57645
    },
    {
      "entropy": 5.348759031295776,
      "epoch": 4.4853530441548335,
      "grad_norm": 1.125,
      "learning_rate": 0.0003142157958614588,
      "loss": 4.8135,
      "mean_token_accuracy": 0.2236206516623497,
      "num_tokens": 99986848.0,
      "step": 57650
    },
    {
      "entropy": 5.416388511657715,
      "epoch": 4.485742073526551,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003141885041928948,
      "loss": 4.7886,
      "mean_token_accuracy": 0.21941682696342468,
      "num_tokens": 99995664.0,
      "step": 57655
    },
    {
      "entropy": 5.4966721534729,
      "epoch": 4.486131102898269,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031416121192970754,
      "loss": 4.9302,
      "mean_token_accuracy": 0.21860788762569427,
      "num_tokens": 100003750.0,
      "step": 57660
    },
    {
      "entropy": 5.410041570663452,
      "epoch": 4.486520132269987,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031413391907231133,
      "loss": 4.7819,
      "mean_token_accuracy": 0.22354753613471984,
      "num_tokens": 100012074.0,
      "step": 57665
    },
    {
      "entropy": 5.2828134059906,
      "epoch": 4.486909161641704,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031410662562112015,
      "loss": 4.6907,
      "mean_token_accuracy": 0.2316438302397728,
      "num_tokens": 100020666.0,
      "step": 57670
    },
    {
      "entropy": 5.416921472549438,
      "epoch": 4.487298191013421,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031407933157654816,
      "loss": 4.8543,
      "mean_token_accuracy": 0.21581071615219116,
      "num_tokens": 100028973.0,
      "step": 57675
    },
    {
      "entropy": 5.340580606460572,
      "epoch": 4.487687220385139,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003140520369390095,
      "loss": 4.6583,
      "mean_token_accuracy": 0.23521411865949632,
      "num_tokens": 100037324.0,
      "step": 57680
    },
    {
      "entropy": 5.426223659515381,
      "epoch": 4.4880762497568565,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003140247417089184,
      "loss": 4.8836,
      "mean_token_accuracy": 0.21098316907882692,
      "num_tokens": 100046383.0,
      "step": 57685
    },
    {
      "entropy": 5.277834939956665,
      "epoch": 4.488465279128574,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003139974458866889,
      "loss": 4.6452,
      "mean_token_accuracy": 0.2338574454188347,
      "num_tokens": 100054857.0,
      "step": 57690
    },
    {
      "entropy": 5.407357263565063,
      "epoch": 4.488854308500292,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003139701494727353,
      "loss": 4.8486,
      "mean_token_accuracy": 0.21710706502199173,
      "num_tokens": 100064228.0,
      "step": 57695
    },
    {
      "entropy": 5.439156007766724,
      "epoch": 4.48924333787201,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003139428524674716,
      "loss": 4.7672,
      "mean_token_accuracy": 0.2210906371474266,
      "num_tokens": 100072580.0,
      "step": 57700
    },
    {
      "entropy": 5.337649011611939,
      "epoch": 4.489632367243727,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003139155548713122,
      "loss": 4.7051,
      "mean_token_accuracy": 0.22351212054491043,
      "num_tokens": 100081119.0,
      "step": 57705
    },
    {
      "entropy": 5.343809461593628,
      "epoch": 4.490021396615444,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031388825668467115,
      "loss": 4.7599,
      "mean_token_accuracy": 0.22047473937273027,
      "num_tokens": 100089814.0,
      "step": 57710
    },
    {
      "entropy": 5.407919597625733,
      "epoch": 4.490410425987162,
      "grad_norm": 1.296875,
      "learning_rate": 0.00031386095790796265,
      "loss": 4.8152,
      "mean_token_accuracy": 0.22950376719236373,
      "num_tokens": 100098311.0,
      "step": 57715
    },
    {
      "entropy": 5.4462785720825195,
      "epoch": 4.4907994553588795,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031383365854160104,
      "loss": 4.8049,
      "mean_token_accuracy": 0.22170085906982423,
      "num_tokens": 100106028.0,
      "step": 57720
    },
    {
      "entropy": 5.35135817527771,
      "epoch": 4.491188484730597,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003138063585860005,
      "loss": 4.7207,
      "mean_token_accuracy": 0.2340254992246628,
      "num_tokens": 100114227.0,
      "step": 57725
    },
    {
      "entropy": 5.30558443069458,
      "epoch": 4.491577514102315,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003137790580415752,
      "loss": 4.7254,
      "mean_token_accuracy": 0.22943395227193833,
      "num_tokens": 100123026.0,
      "step": 57730
    },
    {
      "entropy": 5.407859849929809,
      "epoch": 4.491966543474033,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00031375175690873943,
      "loss": 4.791,
      "mean_token_accuracy": 0.22014037221670152,
      "num_tokens": 100132018.0,
      "step": 57735
    },
    {
      "entropy": 5.48375973701477,
      "epoch": 4.49235557284575,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031372445518790737,
      "loss": 4.906,
      "mean_token_accuracy": 0.20902176797389985,
      "num_tokens": 100140142.0,
      "step": 57740
    },
    {
      "entropy": 5.3409325122833256,
      "epoch": 4.492744602217467,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031369715287949334,
      "loss": 4.7682,
      "mean_token_accuracy": 0.2285037085413933,
      "num_tokens": 100148666.0,
      "step": 57745
    },
    {
      "entropy": 5.43045859336853,
      "epoch": 4.493133631589185,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031366984998391163,
      "loss": 4.8053,
      "mean_token_accuracy": 0.2134558767080307,
      "num_tokens": 100157149.0,
      "step": 57750
    },
    {
      "entropy": 5.320727348327637,
      "epoch": 4.4935226609609025,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031364254650157646,
      "loss": 4.7285,
      "mean_token_accuracy": 0.22789667397737504,
      "num_tokens": 100165742.0,
      "step": 57755
    },
    {
      "entropy": 5.426425170898438,
      "epoch": 4.49391169033262,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003136152424329023,
      "loss": 4.7729,
      "mean_token_accuracy": 0.21458699852228164,
      "num_tokens": 100174208.0,
      "step": 57760
    },
    {
      "entropy": 5.496550989151001,
      "epoch": 4.494300719704338,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003135879377783032,
      "loss": 4.8894,
      "mean_token_accuracy": 0.20936702489852904,
      "num_tokens": 100183356.0,
      "step": 57765
    },
    {
      "entropy": 5.386419200897217,
      "epoch": 4.494689749076056,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031356063253819354,
      "loss": 4.82,
      "mean_token_accuracy": 0.21511639207601546,
      "num_tokens": 100192136.0,
      "step": 57770
    },
    {
      "entropy": 5.437817907333374,
      "epoch": 4.495078778447773,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003135333267129877,
      "loss": 4.8153,
      "mean_token_accuracy": 0.21417911499738693,
      "num_tokens": 100200274.0,
      "step": 57775
    },
    {
      "entropy": 5.3432456970214846,
      "epoch": 4.49546780781949,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003135060203030999,
      "loss": 4.697,
      "mean_token_accuracy": 0.23537123054265977,
      "num_tokens": 100208510.0,
      "step": 57780
    },
    {
      "entropy": 5.335248565673828,
      "epoch": 4.495856837191208,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031347871330894464,
      "loss": 4.7187,
      "mean_token_accuracy": 0.2289268285036087,
      "num_tokens": 100216725.0,
      "step": 57785
    },
    {
      "entropy": 5.308446979522705,
      "epoch": 4.4962458665629255,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000313451405730936,
      "loss": 4.6819,
      "mean_token_accuracy": 0.22728558480739594,
      "num_tokens": 100225283.0,
      "step": 57790
    },
    {
      "entropy": 5.3715136528015135,
      "epoch": 4.496634895934643,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003134240975694886,
      "loss": 4.7758,
      "mean_token_accuracy": 0.22343828678131103,
      "num_tokens": 100234598.0,
      "step": 57795
    },
    {
      "entropy": 5.455445623397827,
      "epoch": 4.497023925306361,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031339678882501667,
      "loss": 4.7797,
      "mean_token_accuracy": 0.22201773077249526,
      "num_tokens": 100243291.0,
      "step": 57800
    },
    {
      "entropy": 5.399333143234253,
      "epoch": 4.497412954678079,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031336947949793455,
      "loss": 4.8026,
      "mean_token_accuracy": 0.21889971047639847,
      "num_tokens": 100251137.0,
      "step": 57805
    },
    {
      "entropy": 5.3522453784942625,
      "epoch": 4.497801984049795,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031334216958865663,
      "loss": 4.7435,
      "mean_token_accuracy": 0.22608824223279952,
      "num_tokens": 100259458.0,
      "step": 57810
    },
    {
      "entropy": 5.352729320526123,
      "epoch": 4.498191013421513,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031331485909759734,
      "loss": 4.7188,
      "mean_token_accuracy": 0.22311104238033294,
      "num_tokens": 100267806.0,
      "step": 57815
    },
    {
      "entropy": 5.378627824783325,
      "epoch": 4.498580042793231,
      "grad_norm": 1.109375,
      "learning_rate": 0.000313287548025171,
      "loss": 4.793,
      "mean_token_accuracy": 0.22081533670425416,
      "num_tokens": 100276391.0,
      "step": 57820
    },
    {
      "entropy": 5.3758782863616945,
      "epoch": 4.4989690721649485,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003132602363717921,
      "loss": 4.7413,
      "mean_token_accuracy": 0.23077651858329773,
      "num_tokens": 100284437.0,
      "step": 57825
    },
    {
      "entropy": 5.360558414459229,
      "epoch": 4.499358101536666,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031323292413787504,
      "loss": 4.7926,
      "mean_token_accuracy": 0.22399294078350068,
      "num_tokens": 100293463.0,
      "step": 57830
    },
    {
      "entropy": 5.401128053665161,
      "epoch": 4.499747130908384,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003132056113238341,
      "loss": 4.7818,
      "mean_token_accuracy": 0.22694533467292785,
      "num_tokens": 100302053.0,
      "step": 57835
    },
    {
      "entropy": 5.475760841369629,
      "epoch": 4.5001361602801015,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031317829793008403,
      "loss": 4.8815,
      "mean_token_accuracy": 0.21565925180912018,
      "num_tokens": 100309746.0,
      "step": 57840
    },
    {
      "entropy": 5.419690084457398,
      "epoch": 4.500525189651818,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003131509839570389,
      "loss": 4.7554,
      "mean_token_accuracy": 0.22166290879249573,
      "num_tokens": 100318297.0,
      "step": 57845
    },
    {
      "entropy": 5.511315631866455,
      "epoch": 4.500914219023536,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031312366940511327,
      "loss": 4.9056,
      "mean_token_accuracy": 0.2157449260354042,
      "num_tokens": 100327300.0,
      "step": 57850
    },
    {
      "entropy": 5.40597620010376,
      "epoch": 4.501303248395254,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003130963542747217,
      "loss": 4.7508,
      "mean_token_accuracy": 0.2217692106962204,
      "num_tokens": 100335479.0,
      "step": 57855
    },
    {
      "entropy": 5.416769695281983,
      "epoch": 4.5016922777669715,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031306903856627866,
      "loss": 4.7842,
      "mean_token_accuracy": 0.22478729337453843,
      "num_tokens": 100344225.0,
      "step": 57860
    },
    {
      "entropy": 5.4443529605865475,
      "epoch": 4.502081307138689,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031304172228019845,
      "loss": 4.918,
      "mean_token_accuracy": 0.20887303054332734,
      "num_tokens": 100353005.0,
      "step": 57865
    },
    {
      "entropy": 5.466445350646973,
      "epoch": 4.502470336510407,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031301440541689577,
      "loss": 4.8318,
      "mean_token_accuracy": 0.213143952190876,
      "num_tokens": 100361716.0,
      "step": 57870
    },
    {
      "entropy": 5.395691347122193,
      "epoch": 4.502859365882124,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003129870879767849,
      "loss": 4.6863,
      "mean_token_accuracy": 0.23259471356868744,
      "num_tokens": 100370305.0,
      "step": 57875
    },
    {
      "entropy": 5.31770887374878,
      "epoch": 4.503248395253841,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003129597699602804,
      "loss": 4.6673,
      "mean_token_accuracy": 0.23455159217119217,
      "num_tokens": 100378616.0,
      "step": 57880
    },
    {
      "entropy": 5.396580362319947,
      "epoch": 4.503637424625559,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003129324513677969,
      "loss": 4.8601,
      "mean_token_accuracy": 0.21577189713716508,
      "num_tokens": 100388303.0,
      "step": 57885
    },
    {
      "entropy": 5.4653116226196286,
      "epoch": 4.504026453997277,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031290513219974875,
      "loss": 4.7804,
      "mean_token_accuracy": 0.22436352968215942,
      "num_tokens": 100396331.0,
      "step": 57890
    },
    {
      "entropy": 5.422437524795532,
      "epoch": 4.504415483368994,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031287781245655064,
      "loss": 4.7809,
      "mean_token_accuracy": 0.22076030522584916,
      "num_tokens": 100405441.0,
      "step": 57895
    },
    {
      "entropy": 5.373691987991333,
      "epoch": 4.504804512740712,
      "grad_norm": 1.1875,
      "learning_rate": 0.000312850492138617,
      "loss": 4.7428,
      "mean_token_accuracy": 0.22496703267097473,
      "num_tokens": 100413255.0,
      "step": 57900
    },
    {
      "entropy": 5.381462335586548,
      "epoch": 4.50519354211243,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003128231712463623,
      "loss": 4.8336,
      "mean_token_accuracy": 0.21518344432115555,
      "num_tokens": 100422641.0,
      "step": 57905
    },
    {
      "entropy": 5.427124500274658,
      "epoch": 4.5055825714841475,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003127958497802012,
      "loss": 4.691,
      "mean_token_accuracy": 0.22952450960874557,
      "num_tokens": 100431634.0,
      "step": 57910
    },
    {
      "entropy": 5.477061605453491,
      "epoch": 4.505971600855864,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003127685277405483,
      "loss": 4.8263,
      "mean_token_accuracy": 0.21461721062660216,
      "num_tokens": 100439977.0,
      "step": 57915
    },
    {
      "entropy": 5.323012256622315,
      "epoch": 4.506360630227582,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031274120512781806,
      "loss": 4.7689,
      "mean_token_accuracy": 0.22612911015748977,
      "num_tokens": 100449697.0,
      "step": 57920
    },
    {
      "entropy": 5.4165356159210205,
      "epoch": 4.5067496595993,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003127138819424252,
      "loss": 4.8238,
      "mean_token_accuracy": 0.221655935049057,
      "num_tokens": 100458510.0,
      "step": 57925
    },
    {
      "entropy": 5.512553310394287,
      "epoch": 4.507138688971017,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003126865581847841,
      "loss": 4.8451,
      "mean_token_accuracy": 0.22541442066431044,
      "num_tokens": 100466169.0,
      "step": 57930
    },
    {
      "entropy": 5.458947324752808,
      "epoch": 4.507527718342735,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031265923385530943,
      "loss": 4.8102,
      "mean_token_accuracy": 0.2168567255139351,
      "num_tokens": 100473503.0,
      "step": 57935
    },
    {
      "entropy": 5.299635314941407,
      "epoch": 4.507916747714452,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003126319089544159,
      "loss": 4.7127,
      "mean_token_accuracy": 0.22135361582040786,
      "num_tokens": 100482437.0,
      "step": 57940
    },
    {
      "entropy": 5.404043102264405,
      "epoch": 4.50830577708617,
      "grad_norm": 1.25,
      "learning_rate": 0.000312604583482518,
      "loss": 4.7666,
      "mean_token_accuracy": 0.22112592905759812,
      "num_tokens": 100491705.0,
      "step": 57945
    },
    {
      "entropy": 5.450191307067871,
      "epoch": 4.508694806457887,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003125772574400305,
      "loss": 4.797,
      "mean_token_accuracy": 0.23065756410360336,
      "num_tokens": 100499566.0,
      "step": 57950
    },
    {
      "entropy": 5.390372562408447,
      "epoch": 4.509083835829605,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003125499308273679,
      "loss": 4.7766,
      "mean_token_accuracy": 0.22196501195430757,
      "num_tokens": 100508326.0,
      "step": 57955
    },
    {
      "entropy": 5.408315277099609,
      "epoch": 4.509472865201323,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031252260364494477,
      "loss": 4.7473,
      "mean_token_accuracy": 0.22300993502140046,
      "num_tokens": 100517334.0,
      "step": 57960
    },
    {
      "entropy": 5.3416139602661135,
      "epoch": 4.50986189457304,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031249527589317596,
      "loss": 4.7968,
      "mean_token_accuracy": 0.22325329184532167,
      "num_tokens": 100526326.0,
      "step": 57965
    },
    {
      "entropy": 5.388043832778931,
      "epoch": 4.510250923944758,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003124679475724759,
      "loss": 4.7183,
      "mean_token_accuracy": 0.2335127428174019,
      "num_tokens": 100534865.0,
      "step": 57970
    },
    {
      "entropy": 5.400729990005493,
      "epoch": 4.510639953316476,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003124406186832595,
      "loss": 4.7669,
      "mean_token_accuracy": 0.22801960110664368,
      "num_tokens": 100543320.0,
      "step": 57975
    },
    {
      "entropy": 5.3139852523803714,
      "epoch": 4.511028982688193,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003124132892259412,
      "loss": 4.7229,
      "mean_token_accuracy": 0.22526452392339708,
      "num_tokens": 100552924.0,
      "step": 57980
    },
    {
      "entropy": 5.386468982696533,
      "epoch": 4.51141801205991,
      "grad_norm": 1.25,
      "learning_rate": 0.00031238595920093584,
      "loss": 4.7415,
      "mean_token_accuracy": 0.228739832341671,
      "num_tokens": 100561350.0,
      "step": 57985
    },
    {
      "entropy": 5.375076532363892,
      "epoch": 4.511807041431628,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031235862860865807,
      "loss": 4.7755,
      "mean_token_accuracy": 0.23037723153829576,
      "num_tokens": 100569843.0,
      "step": 57990
    },
    {
      "entropy": 5.345647144317627,
      "epoch": 4.512196070803346,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003123312974495226,
      "loss": 4.7781,
      "mean_token_accuracy": 0.2167186915874481,
      "num_tokens": 100578286.0,
      "step": 57995
    },
    {
      "entropy": 5.352966260910034,
      "epoch": 4.512585100175063,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031230396572394406,
      "loss": 4.7436,
      "mean_token_accuracy": 0.22409137040376664,
      "num_tokens": 100587035.0,
      "step": 58000
    },
    {
      "entropy": 5.372035789489746,
      "epoch": 4.512974129546781,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003122766334323372,
      "loss": 4.7726,
      "mean_token_accuracy": 0.22840918600559235,
      "num_tokens": 100594959.0,
      "step": 58005
    },
    {
      "entropy": 5.331689548492432,
      "epoch": 4.513363158918498,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031224930057511673,
      "loss": 4.7238,
      "mean_token_accuracy": 0.2250116065144539,
      "num_tokens": 100603368.0,
      "step": 58010
    },
    {
      "entropy": 5.328898620605469,
      "epoch": 4.513752188290216,
      "grad_norm": 1.125,
      "learning_rate": 0.0003122219671526974,
      "loss": 4.7263,
      "mean_token_accuracy": 0.22797801792621614,
      "num_tokens": 100612254.0,
      "step": 58015
    },
    {
      "entropy": 5.48079891204834,
      "epoch": 4.514141217661933,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031219463316549406,
      "loss": 4.8278,
      "mean_token_accuracy": 0.22416602224111556,
      "num_tokens": 100620227.0,
      "step": 58020
    },
    {
      "entropy": 5.464543676376342,
      "epoch": 4.514530247033651,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031216729861392137,
      "loss": 4.9239,
      "mean_token_accuracy": 0.21303140074014665,
      "num_tokens": 100628740.0,
      "step": 58025
    },
    {
      "entropy": 5.393756008148193,
      "epoch": 4.514919276405369,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003121399634983941,
      "loss": 4.7761,
      "mean_token_accuracy": 0.2287479594349861,
      "num_tokens": 100637671.0,
      "step": 58030
    },
    {
      "entropy": 5.389538240432739,
      "epoch": 4.515308305777086,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031211262781932693,
      "loss": 4.7572,
      "mean_token_accuracy": 0.21432053595781325,
      "num_tokens": 100645681.0,
      "step": 58035
    },
    {
      "entropy": 5.4296613216400145,
      "epoch": 4.515697335148804,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031208529157713465,
      "loss": 4.7991,
      "mean_token_accuracy": 0.22633033692836763,
      "num_tokens": 100654477.0,
      "step": 58040
    },
    {
      "entropy": 5.4257865905761715,
      "epoch": 4.516086364520521,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031205795477223217,
      "loss": 4.8685,
      "mean_token_accuracy": 0.21169258952140807,
      "num_tokens": 100662919.0,
      "step": 58045
    },
    {
      "entropy": 5.382138729095459,
      "epoch": 4.516475393892239,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031203061740503414,
      "loss": 4.7564,
      "mean_token_accuracy": 0.231247578561306,
      "num_tokens": 100671795.0,
      "step": 58050
    },
    {
      "entropy": 5.442680883407593,
      "epoch": 4.516864423263956,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003120032794759555,
      "loss": 4.8112,
      "mean_token_accuracy": 0.21712231636047363,
      "num_tokens": 100680330.0,
      "step": 58055
    },
    {
      "entropy": 5.407100486755371,
      "epoch": 4.517253452635674,
      "grad_norm": 1.25,
      "learning_rate": 0.000311975940985411,
      "loss": 4.6936,
      "mean_token_accuracy": 0.23344154357910157,
      "num_tokens": 100688286.0,
      "step": 58060
    },
    {
      "entropy": 5.338248062133789,
      "epoch": 4.517642482007392,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003119486019338154,
      "loss": 4.7934,
      "mean_token_accuracy": 0.22439491003751755,
      "num_tokens": 100698379.0,
      "step": 58065
    },
    {
      "entropy": 5.458255434036255,
      "epoch": 4.518031511379109,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031192126232158367,
      "loss": 4.899,
      "mean_token_accuracy": 0.21628058403730394,
      "num_tokens": 100707946.0,
      "step": 58070
    },
    {
      "entropy": 5.476126718521118,
      "epoch": 4.518420540750826,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003118939221491305,
      "loss": 4.8828,
      "mean_token_accuracy": 0.21490413099527358,
      "num_tokens": 100716180.0,
      "step": 58075
    },
    {
      "entropy": 5.3755530834198,
      "epoch": 4.518809570122544,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003118665814168707,
      "loss": 4.7191,
      "mean_token_accuracy": 0.23380044996738433,
      "num_tokens": 100724033.0,
      "step": 58080
    },
    {
      "entropy": 5.259496879577637,
      "epoch": 4.519198599494262,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031183924012521925,
      "loss": 4.6055,
      "mean_token_accuracy": 0.23686518371105195,
      "num_tokens": 100732749.0,
      "step": 58085
    },
    {
      "entropy": 5.427208042144775,
      "epoch": 4.519587628865979,
      "grad_norm": 1.1875,
      "learning_rate": 0.000311811898274591,
      "loss": 4.7975,
      "mean_token_accuracy": 0.22220032811164855,
      "num_tokens": 100742250.0,
      "step": 58090
    },
    {
      "entropy": 5.315772819519043,
      "epoch": 4.519976658237697,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031178455586540076,
      "loss": 4.7348,
      "mean_token_accuracy": 0.22659672945737838,
      "num_tokens": 100751390.0,
      "step": 58095
    },
    {
      "entropy": 5.4787829399108885,
      "epoch": 4.520365687609415,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031175721289806345,
      "loss": 4.8263,
      "mean_token_accuracy": 0.21820921897888185,
      "num_tokens": 100760796.0,
      "step": 58100
    },
    {
      "entropy": 5.338122367858887,
      "epoch": 4.520754716981132,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031172986937299396,
      "loss": 4.6653,
      "mean_token_accuracy": 0.23375268578529357,
      "num_tokens": 100769303.0,
      "step": 58105
    },
    {
      "entropy": 5.361729955673217,
      "epoch": 4.52114374635285,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031170252529060715,
      "loss": 4.7757,
      "mean_token_accuracy": 0.22059105783700944,
      "num_tokens": 100777588.0,
      "step": 58110
    },
    {
      "entropy": 5.368405342102051,
      "epoch": 4.521532775724567,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003116751806513179,
      "loss": 4.7195,
      "mean_token_accuracy": 0.22722605764865875,
      "num_tokens": 100785908.0,
      "step": 58115
    },
    {
      "entropy": 5.42820782661438,
      "epoch": 4.521921805096285,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003116478354555412,
      "loss": 4.736,
      "mean_token_accuracy": 0.22856306731700898,
      "num_tokens": 100794440.0,
      "step": 58120
    },
    {
      "entropy": 5.388200187683106,
      "epoch": 4.522310834468002,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031162048970369196,
      "loss": 4.8058,
      "mean_token_accuracy": 0.21930038779973984,
      "num_tokens": 100803100.0,
      "step": 58125
    },
    {
      "entropy": 5.40218596458435,
      "epoch": 4.52269986383972,
      "grad_norm": 1.140625,
      "learning_rate": 0.000311593143396185,
      "loss": 4.8732,
      "mean_token_accuracy": 0.21446277648210527,
      "num_tokens": 100811734.0,
      "step": 58130
    },
    {
      "entropy": 5.478357744216919,
      "epoch": 4.523088893211438,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031156579653343543,
      "loss": 4.7903,
      "mean_token_accuracy": 0.22012398391962051,
      "num_tokens": 100819690.0,
      "step": 58135
    },
    {
      "entropy": 5.424999189376831,
      "epoch": 4.523477922583155,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000311538449115858,
      "loss": 4.7237,
      "mean_token_accuracy": 0.22409381568431855,
      "num_tokens": 100827786.0,
      "step": 58140
    },
    {
      "entropy": 5.375665950775146,
      "epoch": 4.523866951954872,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003115111011438678,
      "loss": 4.7693,
      "mean_token_accuracy": 0.22417058795690536,
      "num_tokens": 100836527.0,
      "step": 58145
    },
    {
      "entropy": 5.453056573867798,
      "epoch": 4.52425598132659,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003114837526178797,
      "loss": 4.7485,
      "mean_token_accuracy": 0.22356841564178467,
      "num_tokens": 100845341.0,
      "step": 58150
    },
    {
      "entropy": 5.418053531646729,
      "epoch": 4.524645010698308,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003114564035383088,
      "loss": 4.8128,
      "mean_token_accuracy": 0.21992308050394058,
      "num_tokens": 100853895.0,
      "step": 58155
    },
    {
      "entropy": 5.35748176574707,
      "epoch": 4.525034040070025,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031142905390556995,
      "loss": 4.7911,
      "mean_token_accuracy": 0.21773683726787568,
      "num_tokens": 100862196.0,
      "step": 58160
    },
    {
      "entropy": 5.462615871429444,
      "epoch": 4.525423069441743,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003114017037200782,
      "loss": 4.8468,
      "mean_token_accuracy": 0.2246977612376213,
      "num_tokens": 100870393.0,
      "step": 58165
    },
    {
      "entropy": 5.418280792236328,
      "epoch": 4.525812098813461,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031137435298224863,
      "loss": 4.7524,
      "mean_token_accuracy": 0.2236534282565117,
      "num_tokens": 100880326.0,
      "step": 58170
    },
    {
      "entropy": 5.400307750701904,
      "epoch": 4.526201128185178,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003113470016924961,
      "loss": 4.7441,
      "mean_token_accuracy": 0.22486046701669693,
      "num_tokens": 100888738.0,
      "step": 58175
    },
    {
      "entropy": 5.431799268722534,
      "epoch": 4.526590157556895,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031131964985123555,
      "loss": 4.779,
      "mean_token_accuracy": 0.22227229475975036,
      "num_tokens": 100896903.0,
      "step": 58180
    },
    {
      "entropy": 5.378903865814209,
      "epoch": 4.526979186928613,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003112922974588822,
      "loss": 4.7351,
      "mean_token_accuracy": 0.22840025275945663,
      "num_tokens": 100906051.0,
      "step": 58185
    },
    {
      "entropy": 5.40859432220459,
      "epoch": 4.527368216300331,
      "grad_norm": 1.1875,
      "learning_rate": 0.000311264944515851,
      "loss": 4.8233,
      "mean_token_accuracy": 0.2159008651971817,
      "num_tokens": 100914523.0,
      "step": 58190
    },
    {
      "entropy": 5.378433418273926,
      "epoch": 4.527757245672048,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031123759102255714,
      "loss": 4.7271,
      "mean_token_accuracy": 0.22145889550447465,
      "num_tokens": 100922670.0,
      "step": 58195
    },
    {
      "entropy": 5.359075927734375,
      "epoch": 4.528146275043766,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003112102369794153,
      "loss": 4.7895,
      "mean_token_accuracy": 0.22025570571422576,
      "num_tokens": 100932246.0,
      "step": 58200
    },
    {
      "entropy": 5.37579574584961,
      "epoch": 4.528535304415484,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031118288238684086,
      "loss": 4.7883,
      "mean_token_accuracy": 0.2188274383544922,
      "num_tokens": 100941517.0,
      "step": 58205
    },
    {
      "entropy": 5.35895266532898,
      "epoch": 4.5289243337872005,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003111555272452486,
      "loss": 4.7565,
      "mean_token_accuracy": 0.22843674570322037,
      "num_tokens": 100949387.0,
      "step": 58210
    },
    {
      "entropy": 5.311302995681762,
      "epoch": 4.529313363158918,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00031112817155505394,
      "loss": 4.7089,
      "mean_token_accuracy": 0.2280023068189621,
      "num_tokens": 100958488.0,
      "step": 58215
    },
    {
      "entropy": 5.401602697372437,
      "epoch": 4.529702392530636,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031110081531667184,
      "loss": 4.8155,
      "mean_token_accuracy": 0.22135763466358185,
      "num_tokens": 100966709.0,
      "step": 58220
    },
    {
      "entropy": 5.3420384407043455,
      "epoch": 4.530091421902354,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031107345853051714,
      "loss": 4.6867,
      "mean_token_accuracy": 0.22972368746995925,
      "num_tokens": 100975230.0,
      "step": 58225
    },
    {
      "entropy": 5.408716535568237,
      "epoch": 4.530480451274071,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031104610119700525,
      "loss": 4.8045,
      "mean_token_accuracy": 0.22842177748680115,
      "num_tokens": 100983845.0,
      "step": 58230
    },
    {
      "entropy": 5.378239440917969,
      "epoch": 4.530869480645789,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003110187433165511,
      "loss": 4.7761,
      "mean_token_accuracy": 0.22688053995370866,
      "num_tokens": 100993189.0,
      "step": 58235
    },
    {
      "entropy": 5.410401439666748,
      "epoch": 4.531258510017507,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003109913848895699,
      "loss": 4.7672,
      "mean_token_accuracy": 0.22082017958164216,
      "num_tokens": 101001738.0,
      "step": 58240
    },
    {
      "entropy": 5.4208934783935545,
      "epoch": 4.531647539389224,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003109640259164766,
      "loss": 4.7267,
      "mean_token_accuracy": 0.22999136298894882,
      "num_tokens": 101009614.0,
      "step": 58245
    },
    {
      "entropy": 5.418928527832032,
      "epoch": 4.532036568760941,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031093666639768654,
      "loss": 4.8201,
      "mean_token_accuracy": 0.2128514438867569,
      "num_tokens": 101017953.0,
      "step": 58250
    },
    {
      "entropy": 5.3761701583862305,
      "epoch": 4.532425598132659,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003109093063336147,
      "loss": 4.765,
      "mean_token_accuracy": 0.2330994799733162,
      "num_tokens": 101026214.0,
      "step": 58255
    },
    {
      "entropy": 5.366291618347168,
      "epoch": 4.532814627504377,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003108819457246764,
      "loss": 4.7578,
      "mean_token_accuracy": 0.22493016570806504,
      "num_tokens": 101035323.0,
      "step": 58260
    },
    {
      "entropy": 5.417659139633178,
      "epoch": 4.533203656876094,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031085458457128667,
      "loss": 4.8118,
      "mean_token_accuracy": 0.21742163002490997,
      "num_tokens": 101044349.0,
      "step": 58265
    },
    {
      "entropy": 5.339498138427734,
      "epoch": 4.533592686247812,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031082722287386065,
      "loss": 4.7184,
      "mean_token_accuracy": 0.2315596118569374,
      "num_tokens": 101052964.0,
      "step": 58270
    },
    {
      "entropy": 5.394572973251343,
      "epoch": 4.533981715619529,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003107998606328135,
      "loss": 4.7836,
      "mean_token_accuracy": 0.21632648557424544,
      "num_tokens": 101061696.0,
      "step": 58275
    },
    {
      "entropy": 5.35782265663147,
      "epoch": 4.5343707449912465,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031077249784856057,
      "loss": 4.7667,
      "mean_token_accuracy": 0.2224101424217224,
      "num_tokens": 101070651.0,
      "step": 58280
    },
    {
      "entropy": 5.410215377807617,
      "epoch": 4.534759774362964,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031074513452151683,
      "loss": 4.8694,
      "mean_token_accuracy": 0.22474585473537445,
      "num_tokens": 101079770.0,
      "step": 58285
    },
    {
      "entropy": 5.462438535690308,
      "epoch": 4.535148803734682,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003107177706520975,
      "loss": 4.768,
      "mean_token_accuracy": 0.2227836385369301,
      "num_tokens": 101088262.0,
      "step": 58290
    },
    {
      "entropy": 5.382482194900513,
      "epoch": 4.5355378331064,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000310690406240718,
      "loss": 4.8018,
      "mean_token_accuracy": 0.21591911017894744,
      "num_tokens": 101097448.0,
      "step": 58295
    },
    {
      "entropy": 5.400859594345093,
      "epoch": 4.535926862478117,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031066304128779336,
      "loss": 4.8555,
      "mean_token_accuracy": 0.2132144570350647,
      "num_tokens": 101106327.0,
      "step": 58300
    },
    {
      "entropy": 5.492597150802612,
      "epoch": 4.536315891849835,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031063567579373876,
      "loss": 4.9104,
      "mean_token_accuracy": 0.20826180875301362,
      "num_tokens": 101115183.0,
      "step": 58305
    },
    {
      "entropy": 5.417735052108765,
      "epoch": 4.536704921221553,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031060830975896954,
      "loss": 4.7157,
      "mean_token_accuracy": 0.23095286786556243,
      "num_tokens": 101123664.0,
      "step": 58310
    },
    {
      "entropy": 5.351444673538208,
      "epoch": 4.5370939505932695,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031058094318390085,
      "loss": 4.825,
      "mean_token_accuracy": 0.22114546746015548,
      "num_tokens": 101133277.0,
      "step": 58315
    },
    {
      "entropy": 5.4151404857635494,
      "epoch": 4.537482979964987,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031055357606894806,
      "loss": 4.8194,
      "mean_token_accuracy": 0.22099029123783112,
      "num_tokens": 101141564.0,
      "step": 58320
    },
    {
      "entropy": 5.342559671401977,
      "epoch": 4.537872009336705,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003105262084145263,
      "loss": 4.6724,
      "mean_token_accuracy": 0.23135567605495452,
      "num_tokens": 101150373.0,
      "step": 58325
    },
    {
      "entropy": 5.401526212692261,
      "epoch": 4.538261038708423,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003104988402210509,
      "loss": 4.805,
      "mean_token_accuracy": 0.22245359569787979,
      "num_tokens": 101159402.0,
      "step": 58330
    },
    {
      "entropy": 5.3302545070648195,
      "epoch": 4.53865006808014,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003104714714889371,
      "loss": 4.6842,
      "mean_token_accuracy": 0.23220265954732894,
      "num_tokens": 101167990.0,
      "step": 58335
    },
    {
      "entropy": 5.405477285385132,
      "epoch": 4.539039097451858,
      "grad_norm": 1.25,
      "learning_rate": 0.00031044410221860017,
      "loss": 4.7996,
      "mean_token_accuracy": 0.2277314320206642,
      "num_tokens": 101176528.0,
      "step": 58340
    },
    {
      "entropy": 5.375904941558838,
      "epoch": 4.539428126823575,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003104167324104553,
      "loss": 4.7696,
      "mean_token_accuracy": 0.23114805519580842,
      "num_tokens": 101185364.0,
      "step": 58345
    },
    {
      "entropy": 5.4028733253479,
      "epoch": 4.5398171561952925,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031038936206491803,
      "loss": 4.8213,
      "mean_token_accuracy": 0.22057699114084245,
      "num_tokens": 101193873.0,
      "step": 58350
    },
    {
      "entropy": 5.484736680984497,
      "epoch": 4.54020618556701,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003103619911824035,
      "loss": 4.8819,
      "mean_token_accuracy": 0.21697301417589188,
      "num_tokens": 101202491.0,
      "step": 58355
    },
    {
      "entropy": 5.393422937393188,
      "epoch": 4.540595214938728,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000310334619763327,
      "loss": 4.748,
      "mean_token_accuracy": 0.22058592289686202,
      "num_tokens": 101210011.0,
      "step": 58360
    },
    {
      "entropy": 5.372115230560302,
      "epoch": 4.540984244310446,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031030724780810385,
      "loss": 4.7386,
      "mean_token_accuracy": 0.22441744804382324,
      "num_tokens": 101219027.0,
      "step": 58365
    },
    {
      "entropy": 5.3576085567474365,
      "epoch": 4.541373273682163,
      "grad_norm": 1.25,
      "learning_rate": 0.00031027987531714944,
      "loss": 4.8127,
      "mean_token_accuracy": 0.22582671493291856,
      "num_tokens": 101227553.0,
      "step": 58370
    },
    {
      "entropy": 5.336372804641724,
      "epoch": 4.541762303053881,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031025250229087915,
      "loss": 4.7286,
      "mean_token_accuracy": 0.23139597922563554,
      "num_tokens": 101235859.0,
      "step": 58375
    },
    {
      "entropy": 5.306879663467408,
      "epoch": 4.542151332425598,
      "grad_norm": 1.125,
      "learning_rate": 0.00031022512872970823,
      "loss": 4.6496,
      "mean_token_accuracy": 0.23559586703777313,
      "num_tokens": 101244514.0,
      "step": 58380
    },
    {
      "entropy": 5.374550724029541,
      "epoch": 4.5425403617973155,
      "grad_norm": 1.125,
      "learning_rate": 0.000310197754634052,
      "loss": 4.7234,
      "mean_token_accuracy": 0.23451983630657197,
      "num_tokens": 101253148.0,
      "step": 58385
    },
    {
      "entropy": 5.35664963722229,
      "epoch": 4.542929391169033,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003101703800043258,
      "loss": 4.7556,
      "mean_token_accuracy": 0.21944353580474854,
      "num_tokens": 101262372.0,
      "step": 58390
    },
    {
      "entropy": 5.404948568344116,
      "epoch": 4.543318420540751,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031014300484094516,
      "loss": 4.7846,
      "mean_token_accuracy": 0.22497855573892594,
      "num_tokens": 101271560.0,
      "step": 58395
    },
    {
      "entropy": 5.409999132156372,
      "epoch": 4.5437074499124686,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003101156291443254,
      "loss": 4.7952,
      "mean_token_accuracy": 0.23000595569610596,
      "num_tokens": 101279546.0,
      "step": 58400
    },
    {
      "entropy": 5.491427755355835,
      "epoch": 4.544096479284186,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003100882529148817,
      "loss": 4.9136,
      "mean_token_accuracy": 0.21337536871433258,
      "num_tokens": 101288197.0,
      "step": 58405
    },
    {
      "entropy": 5.425955772399902,
      "epoch": 4.544485508655903,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031006087615302976,
      "loss": 4.8656,
      "mean_token_accuracy": 0.21235856860876084,
      "num_tokens": 101297503.0,
      "step": 58410
    },
    {
      "entropy": 5.375863838195801,
      "epoch": 4.544874538027621,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031003349885918473,
      "loss": 4.7743,
      "mean_token_accuracy": 0.2260446399450302,
      "num_tokens": 101306629.0,
      "step": 58415
    },
    {
      "entropy": 5.333800983428955,
      "epoch": 4.5452635673993385,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031000612103376226,
      "loss": 4.7391,
      "mean_token_accuracy": 0.22388123869895935,
      "num_tokens": 101314556.0,
      "step": 58420
    },
    {
      "entropy": 5.440681457519531,
      "epoch": 4.545652596771056,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003099787426771775,
      "loss": 4.8037,
      "mean_token_accuracy": 0.22522311210632323,
      "num_tokens": 101322641.0,
      "step": 58425
    },
    {
      "entropy": 5.462031745910645,
      "epoch": 4.546041626142774,
      "grad_norm": 1.21875,
      "learning_rate": 0.000309951363789846,
      "loss": 4.8264,
      "mean_token_accuracy": 0.2190435454249382,
      "num_tokens": 101330941.0,
      "step": 58430
    },
    {
      "entropy": 5.293946075439453,
      "epoch": 4.5464306555144915,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030992398437218314,
      "loss": 4.7219,
      "mean_token_accuracy": 0.22934822887182235,
      "num_tokens": 101339898.0,
      "step": 58435
    },
    {
      "entropy": 5.371078395843506,
      "epoch": 4.546819684886209,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003098966044246045,
      "loss": 4.749,
      "mean_token_accuracy": 0.22495688498020172,
      "num_tokens": 101348964.0,
      "step": 58440
    },
    {
      "entropy": 5.43597378730774,
      "epoch": 4.547208714257927,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030986922394752534,
      "loss": 4.8647,
      "mean_token_accuracy": 0.21258753687143325,
      "num_tokens": 101358550.0,
      "step": 58445
    },
    {
      "entropy": 5.363087320327759,
      "epoch": 4.547597743629644,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030984184294136134,
      "loss": 4.7369,
      "mean_token_accuracy": 0.23031437247991562,
      "num_tokens": 101367306.0,
      "step": 58450
    },
    {
      "entropy": 5.3149699687957765,
      "epoch": 4.5479867730013614,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030981446140652773,
      "loss": 4.6904,
      "mean_token_accuracy": 0.22936820536851882,
      "num_tokens": 101375166.0,
      "step": 58455
    },
    {
      "entropy": 5.29676570892334,
      "epoch": 4.548375802373079,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030978707934344014,
      "loss": 4.6955,
      "mean_token_accuracy": 0.22530278712511062,
      "num_tokens": 101383257.0,
      "step": 58460
    },
    {
      "entropy": 5.347620868682862,
      "epoch": 4.548764831744797,
      "grad_norm": 1.25,
      "learning_rate": 0.000309759696752514,
      "loss": 4.762,
      "mean_token_accuracy": 0.21836528927087784,
      "num_tokens": 101391645.0,
      "step": 58465
    },
    {
      "entropy": 5.339993667602539,
      "epoch": 4.5491538611165145,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003097323136341647,
      "loss": 4.6664,
      "mean_token_accuracy": 0.2246425434947014,
      "num_tokens": 101400439.0,
      "step": 58470
    },
    {
      "entropy": 5.349112129211425,
      "epoch": 4.549542890488232,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003097049299888078,
      "loss": 4.7712,
      "mean_token_accuracy": 0.22329528927803038,
      "num_tokens": 101408197.0,
      "step": 58475
    },
    {
      "entropy": 5.372624254226684,
      "epoch": 4.549931919859949,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003096775458168589,
      "loss": 4.7805,
      "mean_token_accuracy": 0.2206585094332695,
      "num_tokens": 101416537.0,
      "step": 58480
    },
    {
      "entropy": 5.344941711425781,
      "epoch": 4.550320949231667,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030965016111873345,
      "loss": 4.7497,
      "mean_token_accuracy": 0.22840164303779603,
      "num_tokens": 101425215.0,
      "step": 58485
    },
    {
      "entropy": 5.384683656692505,
      "epoch": 4.550709978603384,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030962277589484697,
      "loss": 4.8012,
      "mean_token_accuracy": 0.21754027605056764,
      "num_tokens": 101434177.0,
      "step": 58490
    },
    {
      "entropy": 5.358398532867431,
      "epoch": 4.551099007975102,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030959539014561494,
      "loss": 4.7942,
      "mean_token_accuracy": 0.22111717909574508,
      "num_tokens": 101443270.0,
      "step": 58495
    },
    {
      "entropy": 5.4394614696502686,
      "epoch": 4.55148803734682,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030956800387145294,
      "loss": 4.8082,
      "mean_token_accuracy": 0.22035863250494003,
      "num_tokens": 101451860.0,
      "step": 58500
    },
    {
      "entropy": 5.375517416000366,
      "epoch": 4.5518770667185375,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030954061707277647,
      "loss": 4.6996,
      "mean_token_accuracy": 0.233867147564888,
      "num_tokens": 101460566.0,
      "step": 58505
    },
    {
      "entropy": 5.361736154556274,
      "epoch": 4.552266096090255,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003095132297500011,
      "loss": 4.8488,
      "mean_token_accuracy": 0.2174581378698349,
      "num_tokens": 101469650.0,
      "step": 58510
    },
    {
      "entropy": 5.384230470657348,
      "epoch": 4.552655125461972,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030948584190354246,
      "loss": 4.7019,
      "mean_token_accuracy": 0.22824659794569016,
      "num_tokens": 101477840.0,
      "step": 58515
    },
    {
      "entropy": 5.43159441947937,
      "epoch": 4.55304415483369,
      "grad_norm": 1.1875,
      "learning_rate": 0.000309458453533816,
      "loss": 4.8281,
      "mean_token_accuracy": 0.22008295655250548,
      "num_tokens": 101486174.0,
      "step": 58520
    },
    {
      "entropy": 5.371632146835327,
      "epoch": 4.553433184205407,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030943106464123744,
      "loss": 4.7674,
      "mean_token_accuracy": 0.2267102539539337,
      "num_tokens": 101495108.0,
      "step": 58525
    },
    {
      "entropy": 5.326167917251587,
      "epoch": 4.553822213577125,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030940367522622224,
      "loss": 4.7503,
      "mean_token_accuracy": 0.22753020673990249,
      "num_tokens": 101503690.0,
      "step": 58530
    },
    {
      "entropy": 5.439635753631592,
      "epoch": 4.554211242948843,
      "grad_norm": 1.234375,
      "learning_rate": 0.000309376285289186,
      "loss": 4.8356,
      "mean_token_accuracy": 0.21505067944526673,
      "num_tokens": 101512227.0,
      "step": 58535
    },
    {
      "entropy": 5.408129739761352,
      "epoch": 4.5546002723205605,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030934889483054433,
      "loss": 4.771,
      "mean_token_accuracy": 0.21732620000839234,
      "num_tokens": 101521258.0,
      "step": 58540
    },
    {
      "entropy": 5.415787982940674,
      "epoch": 4.554989301692277,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003093215038507129,
      "loss": 4.8183,
      "mean_token_accuracy": 0.21869875192642213,
      "num_tokens": 101530327.0,
      "step": 58545
    },
    {
      "entropy": 5.451783084869385,
      "epoch": 4.555378331063995,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003092941123501073,
      "loss": 4.8865,
      "mean_token_accuracy": 0.21311419159173967,
      "num_tokens": 101538969.0,
      "step": 58550
    },
    {
      "entropy": 5.473888540267945,
      "epoch": 4.555767360435713,
      "grad_norm": 1.25,
      "learning_rate": 0.00030926672032914306,
      "loss": 4.8781,
      "mean_token_accuracy": 0.20768835693597792,
      "num_tokens": 101547945.0,
      "step": 58555
    },
    {
      "entropy": 5.44500412940979,
      "epoch": 4.55615638980743,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003092393277882359,
      "loss": 4.7755,
      "mean_token_accuracy": 0.22230335921049119,
      "num_tokens": 101556080.0,
      "step": 58560
    },
    {
      "entropy": 5.351024007797241,
      "epoch": 4.556545419179148,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030921193472780144,
      "loss": 4.7393,
      "mean_token_accuracy": 0.22503097951412201,
      "num_tokens": 101564301.0,
      "step": 58565
    },
    {
      "entropy": 5.339399242401123,
      "epoch": 4.556934448550866,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003091845411482554,
      "loss": 4.7955,
      "mean_token_accuracy": 0.23197293281555176,
      "num_tokens": 101573458.0,
      "step": 58570
    },
    {
      "entropy": 5.381864452362061,
      "epoch": 4.5573234779225835,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030915714705001323,
      "loss": 4.7427,
      "mean_token_accuracy": 0.22622648924589156,
      "num_tokens": 101582507.0,
      "step": 58575
    },
    {
      "entropy": 5.334383630752564,
      "epoch": 4.557712507294301,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003091297524334908,
      "loss": 4.7229,
      "mean_token_accuracy": 0.22965901345014572,
      "num_tokens": 101591341.0,
      "step": 58580
    },
    {
      "entropy": 5.415007781982422,
      "epoch": 4.558101536666018,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003091023572991036,
      "loss": 4.8827,
      "mean_token_accuracy": 0.2142559051513672,
      "num_tokens": 101599945.0,
      "step": 58585
    },
    {
      "entropy": 5.426520442962646,
      "epoch": 4.558490566037736,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003090749616472675,
      "loss": 4.8172,
      "mean_token_accuracy": 0.2246808663010597,
      "num_tokens": 101608186.0,
      "step": 58590
    },
    {
      "entropy": 5.39422721862793,
      "epoch": 4.558879595409453,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003090475654783981,
      "loss": 4.7907,
      "mean_token_accuracy": 0.2310628205537796,
      "num_tokens": 101616591.0,
      "step": 58595
    },
    {
      "entropy": 5.420369100570679,
      "epoch": 4.559268624781171,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003090201687929111,
      "loss": 4.8788,
      "mean_token_accuracy": 0.22207229286432267,
      "num_tokens": 101625410.0,
      "step": 58600
    },
    {
      "entropy": 5.421879386901855,
      "epoch": 4.559657654152889,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030899277159122213,
      "loss": 4.7687,
      "mean_token_accuracy": 0.21609336286783218,
      "num_tokens": 101633320.0,
      "step": 58605
    },
    {
      "entropy": 5.419233512878418,
      "epoch": 4.560046683524606,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030896537387374703,
      "loss": 4.8402,
      "mean_token_accuracy": 0.22356653213500977,
      "num_tokens": 101642043.0,
      "step": 58610
    },
    {
      "entropy": 5.434904527664185,
      "epoch": 4.560435712896323,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030893797564090136,
      "loss": 4.8486,
      "mean_token_accuracy": 0.21413152366876603,
      "num_tokens": 101650479.0,
      "step": 58615
    },
    {
      "entropy": 5.407833433151245,
      "epoch": 4.560824742268041,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030891057689310097,
      "loss": 4.7708,
      "mean_token_accuracy": 0.22305862009525299,
      "num_tokens": 101660412.0,
      "step": 58620
    },
    {
      "entropy": 5.497016477584839,
      "epoch": 4.561213771639759,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003088831776307616,
      "loss": 4.9228,
      "mean_token_accuracy": 0.21400541812181473,
      "num_tokens": 101668845.0,
      "step": 58625
    },
    {
      "entropy": 5.428175926208496,
      "epoch": 4.561602801011476,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030885577785429885,
      "loss": 4.7964,
      "mean_token_accuracy": 0.22050926834344864,
      "num_tokens": 101677443.0,
      "step": 58630
    },
    {
      "entropy": 5.382854032516479,
      "epoch": 4.561991830383194,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003088283775641286,
      "loss": 4.7696,
      "mean_token_accuracy": 0.2241586998105049,
      "num_tokens": 101685421.0,
      "step": 58635
    },
    {
      "entropy": 5.372738456726074,
      "epoch": 4.562380859754912,
      "grad_norm": 1.25,
      "learning_rate": 0.00030880097676066646,
      "loss": 4.7493,
      "mean_token_accuracy": 0.2297380656003952,
      "num_tokens": 101694362.0,
      "step": 58640
    },
    {
      "entropy": 5.397185325622559,
      "epoch": 4.5627698891266295,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003087735754443285,
      "loss": 4.8125,
      "mean_token_accuracy": 0.22512245178222656,
      "num_tokens": 101703504.0,
      "step": 58645
    },
    {
      "entropy": 5.359457778930664,
      "epoch": 4.563158918498346,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003087461736155301,
      "loss": 4.6197,
      "mean_token_accuracy": 0.2305836036801338,
      "num_tokens": 101712044.0,
      "step": 58650
    },
    {
      "entropy": 5.344888496398926,
      "epoch": 4.563547947870064,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003087187712746873,
      "loss": 4.6392,
      "mean_token_accuracy": 0.23510736227035522,
      "num_tokens": 101720592.0,
      "step": 58655
    },
    {
      "entropy": 5.377733993530273,
      "epoch": 4.563936977241782,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003086913684222159,
      "loss": 4.8336,
      "mean_token_accuracy": 0.22147910594940184,
      "num_tokens": 101729286.0,
      "step": 58660
    },
    {
      "entropy": 5.433221101760864,
      "epoch": 4.564326006613499,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003086639650585315,
      "loss": 4.8306,
      "mean_token_accuracy": 0.22062344700098038,
      "num_tokens": 101738356.0,
      "step": 58665
    },
    {
      "entropy": 5.3957380771636965,
      "epoch": 4.564715035985217,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030863656118405007,
      "loss": 4.7896,
      "mean_token_accuracy": 0.21831643730401992,
      "num_tokens": 101746610.0,
      "step": 58670
    },
    {
      "entropy": 5.412343311309814,
      "epoch": 4.565104065356935,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030860915679918724,
      "loss": 4.7961,
      "mean_token_accuracy": 0.222471784055233,
      "num_tokens": 101755541.0,
      "step": 58675
    },
    {
      "entropy": 5.437191200256348,
      "epoch": 4.565493094728652,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030858175190435913,
      "loss": 4.8482,
      "mean_token_accuracy": 0.2153368577361107,
      "num_tokens": 101764573.0,
      "step": 58680
    },
    {
      "entropy": 5.465342330932617,
      "epoch": 4.565882124100369,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003085543464999813,
      "loss": 4.8106,
      "mean_token_accuracy": 0.22372996658086777,
      "num_tokens": 101773261.0,
      "step": 58685
    },
    {
      "entropy": 5.42166428565979,
      "epoch": 4.566271153472087,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003085269405864697,
      "loss": 4.7997,
      "mean_token_accuracy": 0.22234020084142686,
      "num_tokens": 101781579.0,
      "step": 58690
    },
    {
      "entropy": 5.405756139755249,
      "epoch": 4.566660182843805,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003084995341642401,
      "loss": 4.7692,
      "mean_token_accuracy": 0.22910643368959427,
      "num_tokens": 101790976.0,
      "step": 58695
    },
    {
      "entropy": 5.435870885848999,
      "epoch": 4.567049212215522,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003084721272337084,
      "loss": 4.8183,
      "mean_token_accuracy": 0.21318754851818084,
      "num_tokens": 101799276.0,
      "step": 58700
    },
    {
      "entropy": 5.405096101760864,
      "epoch": 4.56743824158724,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003084447197952904,
      "loss": 4.8419,
      "mean_token_accuracy": 0.22381850630044936,
      "num_tokens": 101807621.0,
      "step": 58705
    },
    {
      "entropy": 5.401069402694702,
      "epoch": 4.567827270958958,
      "grad_norm": 1.25,
      "learning_rate": 0.00030841731184940205,
      "loss": 4.7855,
      "mean_token_accuracy": 0.21901678442955017,
      "num_tokens": 101816355.0,
      "step": 58710
    },
    {
      "entropy": 5.374290990829468,
      "epoch": 4.568216300330675,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030838990339645915,
      "loss": 4.6567,
      "mean_token_accuracy": 0.23098761290311814,
      "num_tokens": 101824509.0,
      "step": 58715
    },
    {
      "entropy": 5.326511335372925,
      "epoch": 4.568605329702392,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030836249443687774,
      "loss": 4.7342,
      "mean_token_accuracy": 0.23187603503465654,
      "num_tokens": 101832558.0,
      "step": 58720
    },
    {
      "entropy": 5.387884330749512,
      "epoch": 4.56899435907411,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030833508497107344,
      "loss": 4.7613,
      "mean_token_accuracy": 0.23114885985851288,
      "num_tokens": 101840719.0,
      "step": 58725
    },
    {
      "entropy": 5.454538345336914,
      "epoch": 4.569383388445828,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003083076749994623,
      "loss": 4.8613,
      "mean_token_accuracy": 0.21562547236680984,
      "num_tokens": 101848996.0,
      "step": 58730
    },
    {
      "entropy": 5.443039846420288,
      "epoch": 4.569772417817545,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003082802645224603,
      "loss": 4.7928,
      "mean_token_accuracy": 0.2177319809794426,
      "num_tokens": 101857831.0,
      "step": 58735
    },
    {
      "entropy": 5.416802644729614,
      "epoch": 4.570161447189263,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003082528535404832,
      "loss": 4.8477,
      "mean_token_accuracy": 0.22361548990011215,
      "num_tokens": 101866321.0,
      "step": 58740
    },
    {
      "entropy": 5.381209087371826,
      "epoch": 4.57055047656098,
      "grad_norm": 1.25,
      "learning_rate": 0.00030822544205394686,
      "loss": 4.8359,
      "mean_token_accuracy": 0.21768156588077545,
      "num_tokens": 101875711.0,
      "step": 58745
    },
    {
      "entropy": 5.443686008453369,
      "epoch": 4.570939505932698,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030819803006326745,
      "loss": 4.8062,
      "mean_token_accuracy": 0.22209957391023635,
      "num_tokens": 101884631.0,
      "step": 58750
    },
    {
      "entropy": 5.3167304515838625,
      "epoch": 4.571328535304415,
      "grad_norm": 1.125,
      "learning_rate": 0.00030817061756886077,
      "loss": 4.6176,
      "mean_token_accuracy": 0.22973824441432952,
      "num_tokens": 101893195.0,
      "step": 58755
    },
    {
      "entropy": 5.418706607818604,
      "epoch": 4.571717564676133,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030814320457114283,
      "loss": 4.8079,
      "mean_token_accuracy": 0.22279595583677292,
      "num_tokens": 101901886.0,
      "step": 58760
    },
    {
      "entropy": 5.406677675247193,
      "epoch": 4.572106594047851,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030811579107052937,
      "loss": 4.8012,
      "mean_token_accuracy": 0.22246210873126984,
      "num_tokens": 101910432.0,
      "step": 58765
    },
    {
      "entropy": 5.347889518737793,
      "epoch": 4.572495623419568,
      "grad_norm": 1.125,
      "learning_rate": 0.00030808837706743666,
      "loss": 4.7201,
      "mean_token_accuracy": 0.2269265204668045,
      "num_tokens": 101919271.0,
      "step": 58770
    },
    {
      "entropy": 5.395749807357788,
      "epoch": 4.572884652791286,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003080609625622803,
      "loss": 4.7677,
      "mean_token_accuracy": 0.2205295369029045,
      "num_tokens": 101927684.0,
      "step": 58775
    },
    {
      "entropy": 5.342611360549927,
      "epoch": 4.573273682163004,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030803354755547653,
      "loss": 4.7301,
      "mean_token_accuracy": 0.23668060898780824,
      "num_tokens": 101936286.0,
      "step": 58780
    },
    {
      "entropy": 5.397234535217285,
      "epoch": 4.573662711534721,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003080061320474412,
      "loss": 4.7735,
      "mean_token_accuracy": 0.22345242500305176,
      "num_tokens": 101944884.0,
      "step": 58785
    },
    {
      "entropy": 5.370807075500489,
      "epoch": 4.574051740906438,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003079787160385905,
      "loss": 4.7191,
      "mean_token_accuracy": 0.223422609269619,
      "num_tokens": 101952717.0,
      "step": 58790
    },
    {
      "entropy": 5.368356704711914,
      "epoch": 4.574440770278156,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003079512995293402,
      "loss": 4.7403,
      "mean_token_accuracy": 0.23128392547369003,
      "num_tokens": 101961520.0,
      "step": 58795
    },
    {
      "entropy": 5.381726408004761,
      "epoch": 4.574829799649874,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003079238825201064,
      "loss": 4.8285,
      "mean_token_accuracy": 0.22042268812656401,
      "num_tokens": 101970327.0,
      "step": 58800
    },
    {
      "entropy": 5.4600725173950195,
      "epoch": 4.575218829021591,
      "grad_norm": 1.1875,
      "learning_rate": 0.000307896465011305,
      "loss": 4.848,
      "mean_token_accuracy": 0.2149516835808754,
      "num_tokens": 101978788.0,
      "step": 58805
    },
    {
      "entropy": 5.330757331848145,
      "epoch": 4.575607858393309,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030786904700335217,
      "loss": 4.6571,
      "mean_token_accuracy": 0.232748444378376,
      "num_tokens": 101986815.0,
      "step": 58810
    },
    {
      "entropy": 5.450636529922486,
      "epoch": 4.575996887765026,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003078416284966639,
      "loss": 4.8957,
      "mean_token_accuracy": 0.2180438295006752,
      "num_tokens": 101996441.0,
      "step": 58815
    },
    {
      "entropy": 5.407009220123291,
      "epoch": 4.576385917136744,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030781420949165607,
      "loss": 4.7719,
      "mean_token_accuracy": 0.2268631175160408,
      "num_tokens": 102004711.0,
      "step": 58820
    },
    {
      "entropy": 5.380636930465698,
      "epoch": 4.576774946508461,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030778678998874494,
      "loss": 4.7145,
      "mean_token_accuracy": 0.23312707096338273,
      "num_tokens": 102013190.0,
      "step": 58825
    },
    {
      "entropy": 5.38257327079773,
      "epoch": 4.577163975880179,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003077593699883465,
      "loss": 4.7584,
      "mean_token_accuracy": 0.21515025049448014,
      "num_tokens": 102022330.0,
      "step": 58830
    },
    {
      "entropy": 5.361949634552002,
      "epoch": 4.577553005251897,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030773194949087665,
      "loss": 4.7573,
      "mean_token_accuracy": 0.22889523953199387,
      "num_tokens": 102030689.0,
      "step": 58835
    },
    {
      "entropy": 5.438253593444824,
      "epoch": 4.577942034623614,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003077045284967516,
      "loss": 4.8654,
      "mean_token_accuracy": 0.21740335673093797,
      "num_tokens": 102039496.0,
      "step": 58840
    },
    {
      "entropy": 5.39219069480896,
      "epoch": 4.578331063995332,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003076771070063874,
      "loss": 4.7693,
      "mean_token_accuracy": 0.2261998772621155,
      "num_tokens": 102048406.0,
      "step": 58845
    },
    {
      "entropy": 5.41925163269043,
      "epoch": 4.578720093367049,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003076496850202001,
      "loss": 4.7101,
      "mean_token_accuracy": 0.23082437515258789,
      "num_tokens": 102057188.0,
      "step": 58850
    },
    {
      "entropy": 5.327297306060791,
      "epoch": 4.579109122738767,
      "grad_norm": 1.125,
      "learning_rate": 0.00030762226253860584,
      "loss": 4.7263,
      "mean_token_accuracy": 0.22330184131860734,
      "num_tokens": 102066231.0,
      "step": 58855
    },
    {
      "entropy": 5.4057886600494385,
      "epoch": 4.579498152110484,
      "grad_norm": 1.25,
      "learning_rate": 0.0003075948395620206,
      "loss": 4.7619,
      "mean_token_accuracy": 0.22150541841983795,
      "num_tokens": 102074345.0,
      "step": 58860
    },
    {
      "entropy": 5.441884756088257,
      "epoch": 4.579887181482202,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030756741609086057,
      "loss": 4.73,
      "mean_token_accuracy": 0.2274385869503021,
      "num_tokens": 102083887.0,
      "step": 58865
    },
    {
      "entropy": 5.470286083221436,
      "epoch": 4.58027621085392,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030753999212554183,
      "loss": 4.8535,
      "mean_token_accuracy": 0.2109972208738327,
      "num_tokens": 102092625.0,
      "step": 58870
    },
    {
      "entropy": 5.375290966033935,
      "epoch": 4.580665240225637,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003075125676664806,
      "loss": 4.7604,
      "mean_token_accuracy": 0.22446558028459548,
      "num_tokens": 102101904.0,
      "step": 58875
    },
    {
      "entropy": 5.306251478195191,
      "epoch": 4.581054269597354,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003074851427140928,
      "loss": 4.762,
      "mean_token_accuracy": 0.21854794919490814,
      "num_tokens": 102110407.0,
      "step": 58880
    },
    {
      "entropy": 5.359929990768433,
      "epoch": 4.581443298969072,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030745771726879467,
      "loss": 4.7329,
      "mean_token_accuracy": 0.22457651942968368,
      "num_tokens": 102119046.0,
      "step": 58885
    },
    {
      "entropy": 5.453454637527466,
      "epoch": 4.58183232834079,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003074302913310024,
      "loss": 4.8461,
      "mean_token_accuracy": 0.21435781568288803,
      "num_tokens": 102128262.0,
      "step": 58890
    },
    {
      "entropy": 5.400657272338867,
      "epoch": 4.582221357712507,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000307402864901132,
      "loss": 4.8039,
      "mean_token_accuracy": 0.2247469574213028,
      "num_tokens": 102136660.0,
      "step": 58895
    },
    {
      "entropy": 5.356222009658813,
      "epoch": 4.582610387084225,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003073754379795998,
      "loss": 4.7814,
      "mean_token_accuracy": 0.22670048773288726,
      "num_tokens": 102146039.0,
      "step": 58900
    },
    {
      "entropy": 5.382696580886841,
      "epoch": 4.582999416455943,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030734801056682176,
      "loss": 4.7613,
      "mean_token_accuracy": 0.2263263449072838,
      "num_tokens": 102154561.0,
      "step": 58905
    },
    {
      "entropy": 5.466377115249633,
      "epoch": 4.58338844582766,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030732058266321415,
      "loss": 4.8468,
      "mean_token_accuracy": 0.21772095412015915,
      "num_tokens": 102163652.0,
      "step": 58910
    },
    {
      "entropy": 5.505028104782104,
      "epoch": 4.583777475199377,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003072931542691932,
      "loss": 4.7687,
      "mean_token_accuracy": 0.22600067853927613,
      "num_tokens": 102172557.0,
      "step": 58915
    },
    {
      "entropy": 5.417150259017944,
      "epoch": 4.584166504571095,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000307265725385175,
      "loss": 4.7907,
      "mean_token_accuracy": 0.22122683823108674,
      "num_tokens": 102181793.0,
      "step": 58920
    },
    {
      "entropy": 5.354289627075195,
      "epoch": 4.584555533942813,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003072382960115758,
      "loss": 4.7516,
      "mean_token_accuracy": 0.2292405441403389,
      "num_tokens": 102190988.0,
      "step": 58925
    },
    {
      "entropy": 5.353587055206299,
      "epoch": 4.58494456331453,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030721086614881183,
      "loss": 4.7147,
      "mean_token_accuracy": 0.22879569381475448,
      "num_tokens": 102200217.0,
      "step": 58930
    },
    {
      "entropy": 5.376117277145386,
      "epoch": 4.585333592686248,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003071834357972991,
      "loss": 4.6702,
      "mean_token_accuracy": 0.23176377713680268,
      "num_tokens": 102209322.0,
      "step": 58935
    },
    {
      "entropy": 5.385023832321167,
      "epoch": 4.585722622057966,
      "grad_norm": 1.140625,
      "learning_rate": 0.000307156004957454,
      "loss": 4.7244,
      "mean_token_accuracy": 0.2259310245513916,
      "num_tokens": 102218943.0,
      "step": 58940
    },
    {
      "entropy": 5.375421237945557,
      "epoch": 4.5861116514296825,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003071285736296928,
      "loss": 4.7198,
      "mean_token_accuracy": 0.23241844028234482,
      "num_tokens": 102227769.0,
      "step": 58945
    },
    {
      "entropy": 5.3893256187438965,
      "epoch": 4.5865006808014,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003071011418144315,
      "loss": 4.7705,
      "mean_token_accuracy": 0.22214339971542357,
      "num_tokens": 102235885.0,
      "step": 58950
    },
    {
      "entropy": 5.359109163284302,
      "epoch": 4.586889710173118,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030707370951208655,
      "loss": 4.7633,
      "mean_token_accuracy": 0.2270162045955658,
      "num_tokens": 102244854.0,
      "step": 58955
    },
    {
      "entropy": 5.40849666595459,
      "epoch": 4.587278739544836,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003070462767230741,
      "loss": 4.7744,
      "mean_token_accuracy": 0.22259124517440795,
      "num_tokens": 102253576.0,
      "step": 58960
    },
    {
      "entropy": 5.439527177810669,
      "epoch": 4.587667768916553,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003070188434478104,
      "loss": 4.8589,
      "mean_token_accuracy": 0.21885799914598464,
      "num_tokens": 102261649.0,
      "step": 58965
    },
    {
      "entropy": 5.367828130722046,
      "epoch": 4.588056798288271,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003069914096867117,
      "loss": 4.6932,
      "mean_token_accuracy": 0.22745512723922728,
      "num_tokens": 102270138.0,
      "step": 58970
    },
    {
      "entropy": 5.34058895111084,
      "epoch": 4.588445827659989,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003069639754401942,
      "loss": 4.7532,
      "mean_token_accuracy": 0.22583162933588027,
      "num_tokens": 102278056.0,
      "step": 58975
    },
    {
      "entropy": 5.42393651008606,
      "epoch": 4.588834857031706,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003069365407086744,
      "loss": 4.8319,
      "mean_token_accuracy": 0.21723568737506865,
      "num_tokens": 102287037.0,
      "step": 58980
    },
    {
      "entropy": 5.48359522819519,
      "epoch": 4.589223886403423,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030690910549256836,
      "loss": 4.808,
      "mean_token_accuracy": 0.22290144115686417,
      "num_tokens": 102295435.0,
      "step": 58985
    },
    {
      "entropy": 5.4370780944824215,
      "epoch": 4.589612915775141,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003068816697922925,
      "loss": 4.8065,
      "mean_token_accuracy": 0.2203659862279892,
      "num_tokens": 102304153.0,
      "step": 58990
    },
    {
      "entropy": 5.406552696228028,
      "epoch": 4.5900019451468586,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003068542336082629,
      "loss": 4.8237,
      "mean_token_accuracy": 0.22332260608673096,
      "num_tokens": 102312514.0,
      "step": 58995
    },
    {
      "entropy": 5.398936080932617,
      "epoch": 4.590390974518576,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030682679694089615,
      "loss": 4.7755,
      "mean_token_accuracy": 0.22708785831928252,
      "num_tokens": 102321480.0,
      "step": 59000
    },
    {
      "entropy": 5.474066257476807,
      "epoch": 4.590780003890294,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003067993597906083,
      "loss": 4.866,
      "mean_token_accuracy": 0.21117934584617615,
      "num_tokens": 102329451.0,
      "step": 59005
    },
    {
      "entropy": 5.413908958435059,
      "epoch": 4.591169033262012,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003067719221578158,
      "loss": 4.803,
      "mean_token_accuracy": 0.22612608671188356,
      "num_tokens": 102338020.0,
      "step": 59010
    },
    {
      "entropy": 5.3281008243560795,
      "epoch": 4.5915580626337285,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000306744484042935,
      "loss": 4.6907,
      "mean_token_accuracy": 0.2314828872680664,
      "num_tokens": 102346749.0,
      "step": 59015
    },
    {
      "entropy": 5.460558462142944,
      "epoch": 4.591947092005446,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003067170454463822,
      "loss": 4.8693,
      "mean_token_accuracy": 0.2255280002951622,
      "num_tokens": 102354915.0,
      "step": 59020
    },
    {
      "entropy": 5.4052471160888675,
      "epoch": 4.592336121377164,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030668960636857364,
      "loss": 4.7286,
      "mean_token_accuracy": 0.22246203422546387,
      "num_tokens": 102363682.0,
      "step": 59025
    },
    {
      "entropy": 5.426143550872803,
      "epoch": 4.5927251507488815,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003066621668099258,
      "loss": 4.7356,
      "mean_token_accuracy": 0.2227473884820938,
      "num_tokens": 102371421.0,
      "step": 59030
    },
    {
      "entropy": 5.333470964431763,
      "epoch": 4.593114180120599,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030663472677085494,
      "loss": 4.7461,
      "mean_token_accuracy": 0.22326613664627076,
      "num_tokens": 102380204.0,
      "step": 59035
    },
    {
      "entropy": 5.306757402420044,
      "epoch": 4.593503209492317,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030660728625177746,
      "loss": 4.6638,
      "mean_token_accuracy": 0.22726166844367982,
      "num_tokens": 102388877.0,
      "step": 59040
    },
    {
      "entropy": 5.40672869682312,
      "epoch": 4.593892238864035,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003065798452531097,
      "loss": 4.8394,
      "mean_token_accuracy": 0.21499379128217697,
      "num_tokens": 102397608.0,
      "step": 59045
    },
    {
      "entropy": 5.464392566680909,
      "epoch": 4.5942812682357514,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00030655240377526813,
      "loss": 4.8597,
      "mean_token_accuracy": 0.2227764144539833,
      "num_tokens": 102406458.0,
      "step": 59050
    },
    {
      "entropy": 5.393937826156616,
      "epoch": 4.594670297607469,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000306524961818669,
      "loss": 4.7471,
      "mean_token_accuracy": 0.2202335849404335,
      "num_tokens": 102415351.0,
      "step": 59055
    },
    {
      "entropy": 5.4614198207855225,
      "epoch": 4.595059326979187,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003064975193837287,
      "loss": 4.8174,
      "mean_token_accuracy": 0.2166559636592865,
      "num_tokens": 102423442.0,
      "step": 59060
    },
    {
      "entropy": 5.516652011871338,
      "epoch": 4.5954483563509045,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003064700764708638,
      "loss": 4.8601,
      "mean_token_accuracy": 0.21895045787096024,
      "num_tokens": 102431780.0,
      "step": 59065
    },
    {
      "entropy": 5.441657543182373,
      "epoch": 4.595837385722622,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030644263308049046,
      "loss": 4.8162,
      "mean_token_accuracy": 0.22027312517166137,
      "num_tokens": 102440298.0,
      "step": 59070
    },
    {
      "entropy": 5.430720710754395,
      "epoch": 4.59622641509434,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003064151892130252,
      "loss": 4.7657,
      "mean_token_accuracy": 0.22761312872171402,
      "num_tokens": 102449002.0,
      "step": 59075
    },
    {
      "entropy": 5.447211599349975,
      "epoch": 4.596615444466057,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003063877448688846,
      "loss": 4.7641,
      "mean_token_accuracy": 0.22692374289035797,
      "num_tokens": 102457171.0,
      "step": 59080
    },
    {
      "entropy": 5.460038089752198,
      "epoch": 4.597004473837774,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003063603000484848,
      "loss": 4.7914,
      "mean_token_accuracy": 0.22247145026922227,
      "num_tokens": 102465866.0,
      "step": 59085
    },
    {
      "entropy": 5.300097274780273,
      "epoch": 4.597393503209492,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030633285475224244,
      "loss": 4.6827,
      "mean_token_accuracy": 0.22482237368822097,
      "num_tokens": 102474026.0,
      "step": 59090
    },
    {
      "entropy": 5.436284589767456,
      "epoch": 4.59778253258121,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030630540898057384,
      "loss": 4.9477,
      "mean_token_accuracy": 0.20357305854558944,
      "num_tokens": 102482327.0,
      "step": 59095
    },
    {
      "entropy": 5.432371044158936,
      "epoch": 4.5981715619529275,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030627796273389546,
      "loss": 4.8684,
      "mean_token_accuracy": 0.21978526711463928,
      "num_tokens": 102490618.0,
      "step": 59100
    },
    {
      "entropy": 5.468838739395141,
      "epoch": 4.598560591324645,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030625051601262384,
      "loss": 4.7784,
      "mean_token_accuracy": 0.22034766972064973,
      "num_tokens": 102498976.0,
      "step": 59105
    },
    {
      "entropy": 5.355074024200439,
      "epoch": 4.598949620696363,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003062230688171753,
      "loss": 4.7119,
      "mean_token_accuracy": 0.22980451732873916,
      "num_tokens": 102508118.0,
      "step": 59110
    },
    {
      "entropy": 5.313330745697021,
      "epoch": 4.599338650068081,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003061956211479665,
      "loss": 4.7794,
      "mean_token_accuracy": 0.22101232558488845,
      "num_tokens": 102516575.0,
      "step": 59115
    },
    {
      "entropy": 5.417083978652954,
      "epoch": 4.599727679439797,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003061681730054137,
      "loss": 4.8203,
      "mean_token_accuracy": 0.21364429146051406,
      "num_tokens": 102526207.0,
      "step": 59120
    },
    {
      "entropy": 5.486366415023804,
      "epoch": 4.600116708811515,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003061407243899335,
      "loss": 4.8515,
      "mean_token_accuracy": 0.22342208176851272,
      "num_tokens": 102536135.0,
      "step": 59125
    },
    {
      "entropy": 5.534539031982422,
      "epoch": 4.600505738183233,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00030611327530194243,
      "loss": 4.8359,
      "mean_token_accuracy": 0.21975705474615098,
      "num_tokens": 102545294.0,
      "step": 59130
    },
    {
      "entropy": 5.425827312469482,
      "epoch": 4.6008947675549505,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003060858257418569,
      "loss": 4.7919,
      "mean_token_accuracy": 0.2206922397017479,
      "num_tokens": 102554128.0,
      "step": 59135
    },
    {
      "entropy": 5.316024208068848,
      "epoch": 4.601283796926668,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003060583757100934,
      "loss": 4.75,
      "mean_token_accuracy": 0.2240620583295822,
      "num_tokens": 102562530.0,
      "step": 59140
    },
    {
      "entropy": 5.463724660873413,
      "epoch": 4.601672826298385,
      "grad_norm": 1.25,
      "learning_rate": 0.0003060309252070685,
      "loss": 4.9222,
      "mean_token_accuracy": 0.21036737859249116,
      "num_tokens": 102571002.0,
      "step": 59145
    },
    {
      "entropy": 5.454732465744018,
      "epoch": 4.602061855670103,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003060034742331987,
      "loss": 4.7366,
      "mean_token_accuracy": 0.22658173441886903,
      "num_tokens": 102579462.0,
      "step": 59150
    },
    {
      "entropy": 5.462719058990478,
      "epoch": 4.60245088504182,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003059760227889006,
      "loss": 4.8221,
      "mean_token_accuracy": 0.21942254155874252,
      "num_tokens": 102587968.0,
      "step": 59155
    },
    {
      "entropy": 5.353231143951416,
      "epoch": 4.602839914413538,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003059485708745906,
      "loss": 4.7798,
      "mean_token_accuracy": 0.22457939982414246,
      "num_tokens": 102597126.0,
      "step": 59160
    },
    {
      "entropy": 5.368075275421143,
      "epoch": 4.603228943785256,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003059211184906853,
      "loss": 4.8156,
      "mean_token_accuracy": 0.2187693491578102,
      "num_tokens": 102606742.0,
      "step": 59165
    },
    {
      "entropy": 5.423300504684448,
      "epoch": 4.6036179731569735,
      "grad_norm": 1.09375,
      "learning_rate": 0.00030589366563760125,
      "loss": 4.8129,
      "mean_token_accuracy": 0.22047675102949144,
      "num_tokens": 102616849.0,
      "step": 59170
    },
    {
      "entropy": 5.434229040145874,
      "epoch": 4.604007002528691,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030586621231575494,
      "loss": 4.8328,
      "mean_token_accuracy": 0.2177261933684349,
      "num_tokens": 102625810.0,
      "step": 59175
    },
    {
      "entropy": 5.375242710113525,
      "epoch": 4.604396031900409,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000305838758525563,
      "loss": 4.7552,
      "mean_token_accuracy": 0.2240580901503563,
      "num_tokens": 102634672.0,
      "step": 59180
    },
    {
      "entropy": 5.425861549377442,
      "epoch": 4.604785061272126,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003058113042674421,
      "loss": 4.8819,
      "mean_token_accuracy": 0.2118202567100525,
      "num_tokens": 102644500.0,
      "step": 59185
    },
    {
      "entropy": 5.485645914077759,
      "epoch": 4.605174090643843,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030578384954180866,
      "loss": 4.8468,
      "mean_token_accuracy": 0.2232686385512352,
      "num_tokens": 102653453.0,
      "step": 59190
    },
    {
      "entropy": 5.462964868545532,
      "epoch": 4.605563120015561,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003057563943490792,
      "loss": 4.8128,
      "mean_token_accuracy": 0.22442420274019242,
      "num_tokens": 102662203.0,
      "step": 59195
    },
    {
      "entropy": 5.361122989654541,
      "epoch": 4.605952149387279,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030572893868967044,
      "loss": 4.6935,
      "mean_token_accuracy": 0.23159783631563186,
      "num_tokens": 102670837.0,
      "step": 59200
    },
    {
      "entropy": 5.362486505508423,
      "epoch": 4.6063411787589965,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000305701482563999,
      "loss": 4.7904,
      "mean_token_accuracy": 0.22988068908452988,
      "num_tokens": 102678941.0,
      "step": 59205
    },
    {
      "entropy": 5.371567869186402,
      "epoch": 4.606730208130714,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003056740259724814,
      "loss": 4.7045,
      "mean_token_accuracy": 0.22389051765203477,
      "num_tokens": 102687246.0,
      "step": 59210
    },
    {
      "entropy": 5.41153473854065,
      "epoch": 4.607119237502431,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003056465689155342,
      "loss": 4.8195,
      "mean_token_accuracy": 0.2231117993593216,
      "num_tokens": 102695966.0,
      "step": 59215
    },
    {
      "entropy": 5.385826110839844,
      "epoch": 4.607508266874149,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030561911139357424,
      "loss": 4.7223,
      "mean_token_accuracy": 0.22547744810581208,
      "num_tokens": 102705517.0,
      "step": 59220
    },
    {
      "entropy": 5.419586229324341,
      "epoch": 4.607897296245866,
      "grad_norm": 1.03125,
      "learning_rate": 0.00030559165340701797,
      "loss": 4.8144,
      "mean_token_accuracy": 0.2193841814994812,
      "num_tokens": 102715451.0,
      "step": 59225
    },
    {
      "entropy": 5.367695045471192,
      "epoch": 4.608286325617584,
      "grad_norm": 1.21875,
      "learning_rate": 0.000305564194956282,
      "loss": 4.6657,
      "mean_token_accuracy": 0.22789881080389024,
      "num_tokens": 102724171.0,
      "step": 59230
    },
    {
      "entropy": 5.33506064414978,
      "epoch": 4.608675354989302,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030553673604178305,
      "loss": 4.7471,
      "mean_token_accuracy": 0.23264963179826736,
      "num_tokens": 102732752.0,
      "step": 59235
    },
    {
      "entropy": 5.359577465057373,
      "epoch": 4.6090643843610195,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030550927666393767,
      "loss": 4.7783,
      "mean_token_accuracy": 0.22995106279850006,
      "num_tokens": 102740838.0,
      "step": 59240
    },
    {
      "entropy": 5.395800161361694,
      "epoch": 4.609453413732737,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030548181682316263,
      "loss": 4.7573,
      "mean_token_accuracy": 0.22783261090517043,
      "num_tokens": 102748670.0,
      "step": 59245
    },
    {
      "entropy": 5.451660633087158,
      "epoch": 4.609842443104454,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030545435651987453,
      "loss": 4.9499,
      "mean_token_accuracy": 0.20696998983621598,
      "num_tokens": 102757649.0,
      "step": 59250
    },
    {
      "entropy": 5.417370700836182,
      "epoch": 4.610231472476172,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003054268957544901,
      "loss": 4.7965,
      "mean_token_accuracy": 0.22761355191469193,
      "num_tokens": 102765836.0,
      "step": 59255
    },
    {
      "entropy": 5.3944920063018795,
      "epoch": 4.610620501847889,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030539943452742585,
      "loss": 4.7952,
      "mean_token_accuracy": 0.2207413613796234,
      "num_tokens": 102774193.0,
      "step": 59260
    },
    {
      "entropy": 5.3830560684204105,
      "epoch": 4.611009531219607,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030537197283909875,
      "loss": 4.7868,
      "mean_token_accuracy": 0.22193364799022675,
      "num_tokens": 102782991.0,
      "step": 59265
    },
    {
      "entropy": 5.396105194091797,
      "epoch": 4.611398560591325,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030534451068992506,
      "loss": 4.8035,
      "mean_token_accuracy": 0.2198040172457695,
      "num_tokens": 102791736.0,
      "step": 59270
    },
    {
      "entropy": 5.447607517242432,
      "epoch": 4.6117875899630425,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030531704808032183,
      "loss": 4.8163,
      "mean_token_accuracy": 0.21752910166978837,
      "num_tokens": 102799909.0,
      "step": 59275
    },
    {
      "entropy": 5.406945514678955,
      "epoch": 4.612176619334759,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003052895850107057,
      "loss": 4.7496,
      "mean_token_accuracy": 0.22497616559267045,
      "num_tokens": 102808465.0,
      "step": 59280
    },
    {
      "entropy": 5.3510266780853275,
      "epoch": 4.612565648706477,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030526212148149326,
      "loss": 4.6659,
      "mean_token_accuracy": 0.23037125617265702,
      "num_tokens": 102816868.0,
      "step": 59285
    },
    {
      "entropy": 5.337236404418945,
      "epoch": 4.612954678078195,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003052346574931013,
      "loss": 4.8242,
      "mean_token_accuracy": 0.2219412788748741,
      "num_tokens": 102825235.0,
      "step": 59290
    },
    {
      "entropy": 5.45988130569458,
      "epoch": 4.613343707449912,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003052071930459465,
      "loss": 4.8389,
      "mean_token_accuracy": 0.213229401409626,
      "num_tokens": 102833543.0,
      "step": 59295
    },
    {
      "entropy": 5.455262231826782,
      "epoch": 4.61373273682163,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003051797281404457,
      "loss": 4.8379,
      "mean_token_accuracy": 0.21948667168617247,
      "num_tokens": 102842825.0,
      "step": 59300
    },
    {
      "entropy": 5.406916999816895,
      "epoch": 4.614121766193348,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030515226277701553,
      "loss": 4.7651,
      "mean_token_accuracy": 0.22625656723976134,
      "num_tokens": 102850644.0,
      "step": 59305
    },
    {
      "entropy": 5.355059862136841,
      "epoch": 4.6145107955650655,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003051247969560728,
      "loss": 4.7853,
      "mean_token_accuracy": 0.2188876137137413,
      "num_tokens": 102859134.0,
      "step": 59310
    },
    {
      "entropy": 5.294962596893311,
      "epoch": 4.614899824936783,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003050973306780342,
      "loss": 4.6932,
      "mean_token_accuracy": 0.22890383154153823,
      "num_tokens": 102867780.0,
      "step": 59315
    },
    {
      "entropy": 5.365436172485351,
      "epoch": 4.6152888543085,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030506986394331656,
      "loss": 4.6467,
      "mean_token_accuracy": 0.23533979207277297,
      "num_tokens": 102876331.0,
      "step": 59320
    },
    {
      "entropy": 5.406923246383667,
      "epoch": 4.615677883680218,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030504239675233655,
      "loss": 4.8013,
      "mean_token_accuracy": 0.2200856477022171,
      "num_tokens": 102885373.0,
      "step": 59325
    },
    {
      "entropy": 5.312823724746704,
      "epoch": 4.616066913051935,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030501492910551094,
      "loss": 4.7425,
      "mean_token_accuracy": 0.22548244744539261,
      "num_tokens": 102894235.0,
      "step": 59330
    },
    {
      "entropy": 5.398810863494873,
      "epoch": 4.616455942423653,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00030498746100325656,
      "loss": 4.7593,
      "mean_token_accuracy": 0.22378096580505372,
      "num_tokens": 102903017.0,
      "step": 59335
    },
    {
      "entropy": 5.428119087219239,
      "epoch": 4.616844971795371,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003049599924459903,
      "loss": 4.7907,
      "mean_token_accuracy": 0.22568943500518798,
      "num_tokens": 102911979.0,
      "step": 59340
    },
    {
      "entropy": 5.285250902175903,
      "epoch": 4.6172340011670885,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003049325234341287,
      "loss": 4.6342,
      "mean_token_accuracy": 0.23229753226041794,
      "num_tokens": 102919829.0,
      "step": 59345
    },
    {
      "entropy": 5.351491832733155,
      "epoch": 4.617623030538805,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030490505396808874,
      "loss": 4.8229,
      "mean_token_accuracy": 0.22476918995380402,
      "num_tokens": 102928742.0,
      "step": 59350
    },
    {
      "entropy": 5.43349552154541,
      "epoch": 4.618012059910523,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003048775840482873,
      "loss": 4.8147,
      "mean_token_accuracy": 0.22231560796499253,
      "num_tokens": 102937113.0,
      "step": 59355
    },
    {
      "entropy": 5.404096698760986,
      "epoch": 4.618401089282241,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000304850113675141,
      "loss": 4.745,
      "mean_token_accuracy": 0.22368726134300232,
      "num_tokens": 102945551.0,
      "step": 59360
    },
    {
      "entropy": 5.421985769271851,
      "epoch": 4.618790118653958,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003048226428490666,
      "loss": 4.8708,
      "mean_token_accuracy": 0.21244501918554307,
      "num_tokens": 102954150.0,
      "step": 59365
    },
    {
      "entropy": 5.381055736541748,
      "epoch": 4.619179148025676,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030479517157048125,
      "loss": 4.7329,
      "mean_token_accuracy": 0.22612216025590898,
      "num_tokens": 102962673.0,
      "step": 59370
    },
    {
      "entropy": 5.455684089660645,
      "epoch": 4.619568177397394,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003047676998398015,
      "loss": 4.7956,
      "mean_token_accuracy": 0.22503697574138642,
      "num_tokens": 102971063.0,
      "step": 59375
    },
    {
      "entropy": 5.454899883270263,
      "epoch": 4.6199572067691115,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030474022765744434,
      "loss": 4.8132,
      "mean_token_accuracy": 0.22912015914916992,
      "num_tokens": 102979618.0,
      "step": 59380
    },
    {
      "entropy": 5.457771301269531,
      "epoch": 4.620346236140828,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030471275502382656,
      "loss": 4.7697,
      "mean_token_accuracy": 0.21079721599817275,
      "num_tokens": 102988627.0,
      "step": 59385
    },
    {
      "entropy": 5.428489828109742,
      "epoch": 4.620735265512546,
      "grad_norm": 1.21875,
      "learning_rate": 0.000304685281939365,
      "loss": 4.7882,
      "mean_token_accuracy": 0.22390297949314117,
      "num_tokens": 102998086.0,
      "step": 59390
    },
    {
      "entropy": 5.473314619064331,
      "epoch": 4.621124294884264,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003046578084044765,
      "loss": 4.7782,
      "mean_token_accuracy": 0.22288812398910524,
      "num_tokens": 103006617.0,
      "step": 59395
    },
    {
      "entropy": 5.300986623764038,
      "epoch": 4.621513324255981,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000304630334419578,
      "loss": 4.6246,
      "mean_token_accuracy": 0.23913885951042174,
      "num_tokens": 103015536.0,
      "step": 59400
    },
    {
      "entropy": 5.29842848777771,
      "epoch": 4.621902353627699,
      "grad_norm": 1.125,
      "learning_rate": 0.0003046028599850863,
      "loss": 4.7215,
      "mean_token_accuracy": 0.2285516619682312,
      "num_tokens": 103025201.0,
      "step": 59405
    },
    {
      "entropy": 5.358543729782104,
      "epoch": 4.622291382999417,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003045753851014184,
      "loss": 4.687,
      "mean_token_accuracy": 0.23344453424215317,
      "num_tokens": 103034806.0,
      "step": 59410
    },
    {
      "entropy": 5.506098937988281,
      "epoch": 4.622680412371134,
      "grad_norm": 1.25,
      "learning_rate": 0.00030454790976899105,
      "loss": 4.9411,
      "mean_token_accuracy": 0.21753142774105072,
      "num_tokens": 103043591.0,
      "step": 59415
    },
    {
      "entropy": 5.356209325790405,
      "epoch": 4.623069441742851,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003045204339882212,
      "loss": 4.7054,
      "mean_token_accuracy": 0.22725777328014374,
      "num_tokens": 103051807.0,
      "step": 59420
    },
    {
      "entropy": 5.367528438568115,
      "epoch": 4.623458471114569,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003044929577595257,
      "loss": 4.7934,
      "mean_token_accuracy": 0.21963736563920974,
      "num_tokens": 103060626.0,
      "step": 59425
    },
    {
      "entropy": 5.4177892208099365,
      "epoch": 4.623847500486287,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030446548108332164,
      "loss": 4.8443,
      "mean_token_accuracy": 0.22301000356674194,
      "num_tokens": 103069154.0,
      "step": 59430
    },
    {
      "entropy": 5.429731559753418,
      "epoch": 4.624236529858004,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030443800396002575,
      "loss": 4.8027,
      "mean_token_accuracy": 0.2201646164059639,
      "num_tokens": 103078715.0,
      "step": 59435
    },
    {
      "entropy": 5.433872079849243,
      "epoch": 4.624625559229722,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003044105263900549,
      "loss": 4.8255,
      "mean_token_accuracy": 0.21911799907684326,
      "num_tokens": 103087258.0,
      "step": 59440
    },
    {
      "entropy": 5.3350742816925045,
      "epoch": 4.62501458860144,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003043830483738262,
      "loss": 4.6662,
      "mean_token_accuracy": 0.22980394661426545,
      "num_tokens": 103096593.0,
      "step": 59445
    },
    {
      "entropy": 5.365390491485596,
      "epoch": 4.6254036179731575,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003043555699117565,
      "loss": 4.7328,
      "mean_token_accuracy": 0.22823262512683867,
      "num_tokens": 103104894.0,
      "step": 59450
    },
    {
      "entropy": 5.376445865631103,
      "epoch": 4.625792647344874,
      "grad_norm": 1.25,
      "learning_rate": 0.00030432809100426286,
      "loss": 4.7692,
      "mean_token_accuracy": 0.21576579064130783,
      "num_tokens": 103113252.0,
      "step": 59455
    },
    {
      "entropy": 5.3977968215942385,
      "epoch": 4.626181676716592,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003043006116517619,
      "loss": 4.8299,
      "mean_token_accuracy": 0.21935662925243377,
      "num_tokens": 103123103.0,
      "step": 59460
    },
    {
      "entropy": 5.441963052749633,
      "epoch": 4.62657070608831,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030427313185467104,
      "loss": 4.8148,
      "mean_token_accuracy": 0.22613126486539842,
      "num_tokens": 103131492.0,
      "step": 59465
    },
    {
      "entropy": 5.364855813980102,
      "epoch": 4.626959735460027,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003042456516134068,
      "loss": 4.761,
      "mean_token_accuracy": 0.22040342837572097,
      "num_tokens": 103140160.0,
      "step": 59470
    },
    {
      "entropy": 5.405936765670776,
      "epoch": 4.627348764831745,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030421817092838644,
      "loss": 4.7667,
      "mean_token_accuracy": 0.2189886152744293,
      "num_tokens": 103148532.0,
      "step": 59475
    },
    {
      "entropy": 5.335491037368774,
      "epoch": 4.627737794203462,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003041906898000268,
      "loss": 4.7144,
      "mean_token_accuracy": 0.22896186709403993,
      "num_tokens": 103156981.0,
      "step": 59480
    },
    {
      "entropy": 5.503323554992676,
      "epoch": 4.62812682357518,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003041632082287449,
      "loss": 4.8865,
      "mean_token_accuracy": 0.20630740821361543,
      "num_tokens": 103166308.0,
      "step": 59485
    },
    {
      "entropy": 5.441832780838013,
      "epoch": 4.628515852946897,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003041357262149577,
      "loss": 4.6996,
      "mean_token_accuracy": 0.22959533035755159,
      "num_tokens": 103174495.0,
      "step": 59490
    },
    {
      "entropy": 5.344913721084595,
      "epoch": 4.628904882318615,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003041082437590824,
      "loss": 4.7596,
      "mean_token_accuracy": 0.2203319936990738,
      "num_tokens": 103184670.0,
      "step": 59495
    },
    {
      "entropy": 5.386821031570435,
      "epoch": 4.629293911690333,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030408076086153575,
      "loss": 4.7835,
      "mean_token_accuracy": 0.22132351547479628,
      "num_tokens": 103192630.0,
      "step": 59500
    },
    {
      "entropy": 5.420797634124756,
      "epoch": 4.62968294106205,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003040532775227347,
      "loss": 4.7796,
      "mean_token_accuracy": 0.21238777935504913,
      "num_tokens": 103201125.0,
      "step": 59505
    },
    {
      "entropy": 5.497645473480224,
      "epoch": 4.630071970433768,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030402579374309656,
      "loss": 4.9178,
      "mean_token_accuracy": 0.21161684393882751,
      "num_tokens": 103209806.0,
      "step": 59510
    },
    {
      "entropy": 5.449560737609863,
      "epoch": 4.630460999805486,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030399830952303815,
      "loss": 4.7124,
      "mean_token_accuracy": 0.22842631340026856,
      "num_tokens": 103219514.0,
      "step": 59515
    },
    {
      "entropy": 5.4249976634979244,
      "epoch": 4.630850029177203,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030397082486297657,
      "loss": 4.8244,
      "mean_token_accuracy": 0.2244926944375038,
      "num_tokens": 103228777.0,
      "step": 59520
    },
    {
      "entropy": 5.29830117225647,
      "epoch": 4.63123905854892,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003039433397633288,
      "loss": 4.6932,
      "mean_token_accuracy": 0.221771302819252,
      "num_tokens": 103237844.0,
      "step": 59525
    },
    {
      "entropy": 5.363972139358521,
      "epoch": 4.631628087920638,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030391585422451196,
      "loss": 4.7127,
      "mean_token_accuracy": 0.2308679535984993,
      "num_tokens": 103246038.0,
      "step": 59530
    },
    {
      "entropy": 5.3501287460327145,
      "epoch": 4.632017117292356,
      "grad_norm": 1.25,
      "learning_rate": 0.00030388836824694307,
      "loss": 4.7349,
      "mean_token_accuracy": 0.22470701336860657,
      "num_tokens": 103254383.0,
      "step": 59535
    },
    {
      "entropy": 5.439510631561279,
      "epoch": 4.632406146664073,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003038608818310391,
      "loss": 4.8126,
      "mean_token_accuracy": 0.22790600955486298,
      "num_tokens": 103262110.0,
      "step": 59540
    },
    {
      "entropy": 5.365524339675903,
      "epoch": 4.632795176035791,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030383339497721726,
      "loss": 4.7199,
      "mean_token_accuracy": 0.2330671012401581,
      "num_tokens": 103270206.0,
      "step": 59545
    },
    {
      "entropy": 5.396467781066894,
      "epoch": 4.633184205407508,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003038059076858944,
      "loss": 4.7717,
      "mean_token_accuracy": 0.22478654384613037,
      "num_tokens": 103279764.0,
      "step": 59550
    },
    {
      "entropy": 5.467930459976197,
      "epoch": 4.633573234779226,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003037784199574879,
      "loss": 4.8357,
      "mean_token_accuracy": 0.22245290428400039,
      "num_tokens": 103288848.0,
      "step": 59555
    },
    {
      "entropy": 5.428844404220581,
      "epoch": 4.633962264150943,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030375093179241464,
      "loss": 4.8204,
      "mean_token_accuracy": 0.22605613768100738,
      "num_tokens": 103297517.0,
      "step": 59560
    },
    {
      "entropy": 5.363326930999756,
      "epoch": 4.634351293522661,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030372344319109173,
      "loss": 4.7957,
      "mean_token_accuracy": 0.22415723353624345,
      "num_tokens": 103306075.0,
      "step": 59565
    },
    {
      "entropy": 5.402189683914185,
      "epoch": 4.634740322894379,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003036959541539363,
      "loss": 4.7746,
      "mean_token_accuracy": 0.22701624929904937,
      "num_tokens": 103315857.0,
      "step": 59570
    },
    {
      "entropy": 5.408189153671264,
      "epoch": 4.635129352266096,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030366846468136543,
      "loss": 4.7663,
      "mean_token_accuracy": 0.22891604155302048,
      "num_tokens": 103325458.0,
      "step": 59575
    },
    {
      "entropy": 5.4254982471466064,
      "epoch": 4.635518381637814,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003036409747737963,
      "loss": 4.8975,
      "mean_token_accuracy": 0.21704998314380647,
      "num_tokens": 103334933.0,
      "step": 59580
    },
    {
      "entropy": 5.387787342071533,
      "epoch": 4.635907411009531,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030361348443164583,
      "loss": 4.7231,
      "mean_token_accuracy": 0.22778742462396623,
      "num_tokens": 103343624.0,
      "step": 59585
    },
    {
      "entropy": 5.348550605773926,
      "epoch": 4.6362964403812486,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003035859936553314,
      "loss": 4.667,
      "mean_token_accuracy": 0.23422258645296096,
      "num_tokens": 103352510.0,
      "step": 59590
    },
    {
      "entropy": 5.411761808395386,
      "epoch": 4.636685469752966,
      "grad_norm": 1.09375,
      "learning_rate": 0.00030355850244527,
      "loss": 4.8493,
      "mean_token_accuracy": 0.22011567801237106,
      "num_tokens": 103361360.0,
      "step": 59595
    },
    {
      "entropy": 5.305577230453491,
      "epoch": 4.637074499124684,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003035310108018787,
      "loss": 4.7297,
      "mean_token_accuracy": 0.22526883333921432,
      "num_tokens": 103370891.0,
      "step": 59600
    },
    {
      "entropy": 5.42244668006897,
      "epoch": 4.637463528496402,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003035035187255748,
      "loss": 4.6382,
      "mean_token_accuracy": 0.23618705868721007,
      "num_tokens": 103378615.0,
      "step": 59605
    },
    {
      "entropy": 5.46828327178955,
      "epoch": 4.637852557868119,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030347602621677534,
      "loss": 4.9186,
      "mean_token_accuracy": 0.2191417396068573,
      "num_tokens": 103388114.0,
      "step": 59610
    },
    {
      "entropy": 5.410670804977417,
      "epoch": 4.638241587239836,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003034485332758975,
      "loss": 4.8405,
      "mean_token_accuracy": 0.2116706922650337,
      "num_tokens": 103396124.0,
      "step": 59615
    },
    {
      "entropy": 5.479038906097412,
      "epoch": 4.638630616611554,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030342103990335846,
      "loss": 4.8331,
      "mean_token_accuracy": 0.2161700189113617,
      "num_tokens": 103404454.0,
      "step": 59620
    },
    {
      "entropy": 5.348663187026977,
      "epoch": 4.6390196459832715,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030339354609957545,
      "loss": 4.7019,
      "mean_token_accuracy": 0.22383491843938827,
      "num_tokens": 103412453.0,
      "step": 59625
    },
    {
      "entropy": 5.324551677703857,
      "epoch": 4.639408675354989,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030336605186496546,
      "loss": 4.7131,
      "mean_token_accuracy": 0.22779101133346558,
      "num_tokens": 103420432.0,
      "step": 59630
    },
    {
      "entropy": 5.460262155532837,
      "epoch": 4.639797704726707,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030333855719994585,
      "loss": 4.875,
      "mean_token_accuracy": 0.21412641704082488,
      "num_tokens": 103429973.0,
      "step": 59635
    },
    {
      "entropy": 5.469717931747437,
      "epoch": 4.640186734098425,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003033110621049336,
      "loss": 4.8373,
      "mean_token_accuracy": 0.21623394787311553,
      "num_tokens": 103439809.0,
      "step": 59640
    },
    {
      "entropy": 5.470744371414185,
      "epoch": 4.640575763470142,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030328356658034623,
      "loss": 4.7125,
      "mean_token_accuracy": 0.21844536811113358,
      "num_tokens": 103448364.0,
      "step": 59645
    },
    {
      "entropy": 5.418440151214599,
      "epoch": 4.64096479284186,
      "grad_norm": 1.125,
      "learning_rate": 0.00030325607062660066,
      "loss": 4.7846,
      "mean_token_accuracy": 0.2236788973212242,
      "num_tokens": 103457513.0,
      "step": 59650
    },
    {
      "entropy": 5.393074607849121,
      "epoch": 4.641353822213577,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003032285742441143,
      "loss": 4.7818,
      "mean_token_accuracy": 0.22212940007448195,
      "num_tokens": 103465898.0,
      "step": 59655
    },
    {
      "entropy": 5.334016895294189,
      "epoch": 4.6417428515852945,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003032010774333041,
      "loss": 4.7224,
      "mean_token_accuracy": 0.23126882761716844,
      "num_tokens": 103474106.0,
      "step": 59660
    },
    {
      "entropy": 5.317964935302735,
      "epoch": 4.642131880957012,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030317358019458753,
      "loss": 4.674,
      "mean_token_accuracy": 0.22896128743886948,
      "num_tokens": 103482843.0,
      "step": 59665
    },
    {
      "entropy": 5.431198072433472,
      "epoch": 4.64252091032873,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003031460825283817,
      "loss": 4.7495,
      "mean_token_accuracy": 0.22788007706403732,
      "num_tokens": 103492305.0,
      "step": 59670
    },
    {
      "entropy": 5.427869272232056,
      "epoch": 4.642909939700448,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030311858443510383,
      "loss": 4.7968,
      "mean_token_accuracy": 0.22938809543848038,
      "num_tokens": 103500304.0,
      "step": 59675
    },
    {
      "entropy": 5.354594898223877,
      "epoch": 4.643298969072165,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003030910859151712,
      "loss": 4.6972,
      "mean_token_accuracy": 0.22967077046632767,
      "num_tokens": 103508743.0,
      "step": 59680
    },
    {
      "entropy": 5.321511840820312,
      "epoch": 4.643687998443882,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030306358696900114,
      "loss": 4.6669,
      "mean_token_accuracy": 0.23482585698366165,
      "num_tokens": 103517400.0,
      "step": 59685
    },
    {
      "entropy": 5.435042381286621,
      "epoch": 4.6440770278156,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00030303608759701077,
      "loss": 4.7809,
      "mean_token_accuracy": 0.2281481683254242,
      "num_tokens": 103525240.0,
      "step": 59690
    },
    {
      "entropy": 5.420337724685669,
      "epoch": 4.6444660571873175,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003030085877996174,
      "loss": 4.8619,
      "mean_token_accuracy": 0.21933754980564119,
      "num_tokens": 103533668.0,
      "step": 59695
    },
    {
      "entropy": 5.378507804870606,
      "epoch": 4.644855086559035,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030298108757723823,
      "loss": 4.7108,
      "mean_token_accuracy": 0.23457382023334503,
      "num_tokens": 103542436.0,
      "step": 59700
    },
    {
      "entropy": 5.319800662994385,
      "epoch": 4.645244115930753,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003029535869302906,
      "loss": 4.6478,
      "mean_token_accuracy": 0.22903958410024644,
      "num_tokens": 103550289.0,
      "step": 59705
    },
    {
      "entropy": 5.424223804473877,
      "epoch": 4.645633145302471,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030292608585919185,
      "loss": 4.8036,
      "mean_token_accuracy": 0.2269161492586136,
      "num_tokens": 103558606.0,
      "step": 59710
    },
    {
      "entropy": 5.446231889724731,
      "epoch": 4.646022174674188,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030289858436435916,
      "loss": 4.7944,
      "mean_token_accuracy": 0.21791634410619737,
      "num_tokens": 103567481.0,
      "step": 59715
    },
    {
      "entropy": 5.469941425323486,
      "epoch": 4.646411204045905,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030287108244620993,
      "loss": 4.8445,
      "mean_token_accuracy": 0.21667399108409882,
      "num_tokens": 103575460.0,
      "step": 59720
    },
    {
      "entropy": 5.434540462493897,
      "epoch": 4.646800233417623,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030284358010516135,
      "loss": 4.7999,
      "mean_token_accuracy": 0.2228355199098587,
      "num_tokens": 103583913.0,
      "step": 59725
    },
    {
      "entropy": 5.328849649429321,
      "epoch": 4.6471892627893405,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003028160773416307,
      "loss": 4.7702,
      "mean_token_accuracy": 0.223815481364727,
      "num_tokens": 103592711.0,
      "step": 59730
    },
    {
      "entropy": 5.396610975265503,
      "epoch": 4.647578292161058,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030278857415603534,
      "loss": 4.804,
      "mean_token_accuracy": 0.22200295180082322,
      "num_tokens": 103600886.0,
      "step": 59735
    },
    {
      "entropy": 5.381673288345337,
      "epoch": 4.647967321532776,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030276107054879263,
      "loss": 4.7712,
      "mean_token_accuracy": 0.2297120586037636,
      "num_tokens": 103610099.0,
      "step": 59740
    },
    {
      "entropy": 5.478265857696533,
      "epoch": 4.648356350904494,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030273356652031993,
      "loss": 4.8983,
      "mean_token_accuracy": 0.21882153004407884,
      "num_tokens": 103619009.0,
      "step": 59745
    },
    {
      "entropy": 5.439951658248901,
      "epoch": 4.64874538027621,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003027060620710344,
      "loss": 4.8136,
      "mean_token_accuracy": 0.22527068853378296,
      "num_tokens": 103626831.0,
      "step": 59750
    },
    {
      "entropy": 5.400478935241699,
      "epoch": 4.649134409647928,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030267855720135343,
      "loss": 4.7694,
      "mean_token_accuracy": 0.2236724704504013,
      "num_tokens": 103636907.0,
      "step": 59755
    },
    {
      "entropy": 5.3751373291015625,
      "epoch": 4.649523439019646,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003026510519116946,
      "loss": 4.7631,
      "mean_token_accuracy": 0.22680717557668686,
      "num_tokens": 103646517.0,
      "step": 59760
    },
    {
      "entropy": 5.485721778869629,
      "epoch": 4.6499124683913635,
      "grad_norm": 1.3125,
      "learning_rate": 0.000302623546202475,
      "loss": 4.8462,
      "mean_token_accuracy": 0.2155029892921448,
      "num_tokens": 103655050.0,
      "step": 59765
    },
    {
      "entropy": 5.591313457489013,
      "epoch": 4.650301497763081,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003025960400741119,
      "loss": 4.9849,
      "mean_token_accuracy": 0.2018994376063347,
      "num_tokens": 103664505.0,
      "step": 59770
    },
    {
      "entropy": 5.434939622879028,
      "epoch": 4.650690527134799,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000302568533527023,
      "loss": 4.7712,
      "mean_token_accuracy": 0.22444880306720733,
      "num_tokens": 103673176.0,
      "step": 59775
    },
    {
      "entropy": 5.353394460678101,
      "epoch": 4.651079556506517,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003025410265616254,
      "loss": 4.7117,
      "mean_token_accuracy": 0.2253936305642128,
      "num_tokens": 103681219.0,
      "step": 59780
    },
    {
      "entropy": 5.356380796432495,
      "epoch": 4.651468585878233,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003025135191783366,
      "loss": 4.7248,
      "mean_token_accuracy": 0.22505516260862352,
      "num_tokens": 103688792.0,
      "step": 59785
    },
    {
      "entropy": 5.343975400924682,
      "epoch": 4.651857615249951,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030248601137757394,
      "loss": 4.6822,
      "mean_token_accuracy": 0.2284650757908821,
      "num_tokens": 103697242.0,
      "step": 59790
    },
    {
      "entropy": 5.452278709411621,
      "epoch": 4.652246644621669,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003024585031597547,
      "loss": 4.7435,
      "mean_token_accuracy": 0.2308778077363968,
      "num_tokens": 103706277.0,
      "step": 59795
    },
    {
      "entropy": 5.443483638763428,
      "epoch": 4.6526356739933865,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003024309945252965,
      "loss": 4.8642,
      "mean_token_accuracy": 0.21820130795240403,
      "num_tokens": 103714924.0,
      "step": 59800
    },
    {
      "entropy": 5.392613267898559,
      "epoch": 4.653024703365104,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030240348547461653,
      "loss": 4.8088,
      "mean_token_accuracy": 0.2163159132003784,
      "num_tokens": 103722982.0,
      "step": 59805
    },
    {
      "entropy": 5.379876375198364,
      "epoch": 4.653413732736822,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030237597600813235,
      "loss": 4.7831,
      "mean_token_accuracy": 0.22688182890415193,
      "num_tokens": 103731838.0,
      "step": 59810
    },
    {
      "entropy": 5.42807412147522,
      "epoch": 4.653802762108539,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003023484661262613,
      "loss": 4.7391,
      "mean_token_accuracy": 0.22742576003074647,
      "num_tokens": 103740310.0,
      "step": 59815
    },
    {
      "entropy": 5.345364427566528,
      "epoch": 4.654191791480256,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003023209558294208,
      "loss": 4.6733,
      "mean_token_accuracy": 0.2352745532989502,
      "num_tokens": 103748981.0,
      "step": 59820
    },
    {
      "entropy": 5.361807250976563,
      "epoch": 4.654580820851974,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030229344511802824,
      "loss": 4.7104,
      "mean_token_accuracy": 0.23067983835935593,
      "num_tokens": 103757443.0,
      "step": 59825
    },
    {
      "entropy": 5.3469208717346195,
      "epoch": 4.654969850223692,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030226593399250114,
      "loss": 4.7617,
      "mean_token_accuracy": 0.2221172034740448,
      "num_tokens": 103767323.0,
      "step": 59830
    },
    {
      "entropy": 5.358058881759644,
      "epoch": 4.6553588795954095,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030223842245325687,
      "loss": 4.7436,
      "mean_token_accuracy": 0.2205578938126564,
      "num_tokens": 103776513.0,
      "step": 59835
    },
    {
      "entropy": 5.43656177520752,
      "epoch": 4.655747908967127,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030221091050071293,
      "loss": 4.8327,
      "mean_token_accuracy": 0.2160925328731537,
      "num_tokens": 103785043.0,
      "step": 59840
    },
    {
      "entropy": 5.376711082458496,
      "epoch": 4.656136938338845,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00030218339813528665,
      "loss": 4.7113,
      "mean_token_accuracy": 0.22611784487962722,
      "num_tokens": 103794015.0,
      "step": 59845
    },
    {
      "entropy": 5.345044803619385,
      "epoch": 4.656525967710563,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003021558853573957,
      "loss": 4.7825,
      "mean_token_accuracy": 0.22772388458251952,
      "num_tokens": 103803687.0,
      "step": 59850
    },
    {
      "entropy": 5.439833784103394,
      "epoch": 4.656914997082279,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003021283721674573,
      "loss": 4.8595,
      "mean_token_accuracy": 0.22572697699069977,
      "num_tokens": 103812355.0,
      "step": 59855
    },
    {
      "entropy": 5.361315393447876,
      "epoch": 4.657304026453997,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003021008585658891,
      "loss": 4.7169,
      "mean_token_accuracy": 0.22691328823566437,
      "num_tokens": 103820363.0,
      "step": 59860
    },
    {
      "entropy": 5.376403188705444,
      "epoch": 4.657693055825715,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030207334455310846,
      "loss": 4.7966,
      "mean_token_accuracy": 0.22476357072591782,
      "num_tokens": 103828665.0,
      "step": 59865
    },
    {
      "entropy": 5.46757869720459,
      "epoch": 4.6580820851974325,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003020458301295329,
      "loss": 4.8249,
      "mean_token_accuracy": 0.23431562185287474,
      "num_tokens": 103838495.0,
      "step": 59870
    },
    {
      "entropy": 5.376807355880738,
      "epoch": 4.65847111456915,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030201831529557994,
      "loss": 4.7439,
      "mean_token_accuracy": 0.21982259452342987,
      "num_tokens": 103846476.0,
      "step": 59875
    },
    {
      "entropy": 5.436042356491089,
      "epoch": 4.658860143940868,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003019908000516671,
      "loss": 4.86,
      "mean_token_accuracy": 0.21401695609092714,
      "num_tokens": 103854926.0,
      "step": 59880
    },
    {
      "entropy": 5.421493291854858,
      "epoch": 4.659249173312585,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003019632843982117,
      "loss": 4.8431,
      "mean_token_accuracy": 0.22191568613052368,
      "num_tokens": 103863372.0,
      "step": 59885
    },
    {
      "entropy": 5.398956203460694,
      "epoch": 4.659638202684302,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030193576833563153,
      "loss": 4.8125,
      "mean_token_accuracy": 0.2237105131149292,
      "num_tokens": 103871830.0,
      "step": 59890
    },
    {
      "entropy": 5.439985036849976,
      "epoch": 4.66002723205602,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003019082518643438,
      "loss": 4.7875,
      "mean_token_accuracy": 0.21817277371883392,
      "num_tokens": 103879976.0,
      "step": 59895
    },
    {
      "entropy": 5.431389951705933,
      "epoch": 4.660416261427738,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030188073498476633,
      "loss": 4.7836,
      "mean_token_accuracy": 0.22518945336341858,
      "num_tokens": 103888362.0,
      "step": 59900
    },
    {
      "entropy": 5.47089147567749,
      "epoch": 4.6608052907994555,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003018532176973163,
      "loss": 4.8609,
      "mean_token_accuracy": 0.2085472524166107,
      "num_tokens": 103897526.0,
      "step": 59905
    },
    {
      "entropy": 5.364666366577149,
      "epoch": 4.661194320171173,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030182570000241154,
      "loss": 4.7372,
      "mean_token_accuracy": 0.2255645364522934,
      "num_tokens": 103906292.0,
      "step": 59910
    },
    {
      "entropy": 5.347701787948608,
      "epoch": 4.661583349542891,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003017981819004696,
      "loss": 4.7219,
      "mean_token_accuracy": 0.2291928917169571,
      "num_tokens": 103915688.0,
      "step": 59915
    },
    {
      "entropy": 5.361847543716431,
      "epoch": 4.661972378914608,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003017706633919078,
      "loss": 4.8118,
      "mean_token_accuracy": 0.2193927437067032,
      "num_tokens": 103924749.0,
      "step": 59920
    },
    {
      "entropy": 5.338251066207886,
      "epoch": 4.662361408286325,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003017431444771437,
      "loss": 4.6762,
      "mean_token_accuracy": 0.23218585252761842,
      "num_tokens": 103932731.0,
      "step": 59925
    },
    {
      "entropy": 5.4523766040802,
      "epoch": 4.662750437658043,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030171562515659504,
      "loss": 4.9115,
      "mean_token_accuracy": 0.21043730676174163,
      "num_tokens": 103941284.0,
      "step": 59930
    },
    {
      "entropy": 5.3361071109771725,
      "epoch": 4.663139467029761,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003016881054306793,
      "loss": 4.7422,
      "mean_token_accuracy": 0.22682922780513765,
      "num_tokens": 103950304.0,
      "step": 59935
    },
    {
      "entropy": 5.382370662689209,
      "epoch": 4.6635284964014785,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003016605852998139,
      "loss": 4.7551,
      "mean_token_accuracy": 0.22328772097826005,
      "num_tokens": 103958472.0,
      "step": 59940
    },
    {
      "entropy": 5.4095337867736815,
      "epoch": 4.663917525773196,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030163306476441675,
      "loss": 4.7747,
      "mean_token_accuracy": 0.2267397537827492,
      "num_tokens": 103967068.0,
      "step": 59945
    },
    {
      "entropy": 5.485649299621582,
      "epoch": 4.664306555144913,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003016055438249051,
      "loss": 4.9259,
      "mean_token_accuracy": 0.21011196374893187,
      "num_tokens": 103975615.0,
      "step": 59950
    },
    {
      "entropy": 5.523872041702271,
      "epoch": 4.664695584516631,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003015780224816968,
      "loss": 4.9018,
      "mean_token_accuracy": 0.2133241131901741,
      "num_tokens": 103984524.0,
      "step": 59955
    },
    {
      "entropy": 5.411612129211425,
      "epoch": 4.665084613888348,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003015505007352093,
      "loss": 4.7405,
      "mean_token_accuracy": 0.23036046028137208,
      "num_tokens": 103993217.0,
      "step": 59960
    },
    {
      "entropy": 5.373046016693115,
      "epoch": 4.665473643260066,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030152297858586016,
      "loss": 4.8103,
      "mean_token_accuracy": 0.2201085388660431,
      "num_tokens": 104001623.0,
      "step": 59965
    },
    {
      "entropy": 5.311851692199707,
      "epoch": 4.665862672631784,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030149545603406694,
      "loss": 4.773,
      "mean_token_accuracy": 0.21999589651823043,
      "num_tokens": 104010794.0,
      "step": 59970
    },
    {
      "entropy": 5.395993566513061,
      "epoch": 4.6662517020035015,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003014679330802474,
      "loss": 4.7713,
      "mean_token_accuracy": 0.21992125660181044,
      "num_tokens": 104019118.0,
      "step": 59975
    },
    {
      "entropy": 5.537430667877198,
      "epoch": 4.666640731375219,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030144040972481916,
      "loss": 4.8951,
      "mean_token_accuracy": 0.21607646495103836,
      "num_tokens": 104027823.0,
      "step": 59980
    },
    {
      "entropy": 5.480741214752197,
      "epoch": 4.667029760746937,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030141288596819977,
      "loss": 4.9219,
      "mean_token_accuracy": 0.20840731710195542,
      "num_tokens": 104036995.0,
      "step": 59985
    },
    {
      "entropy": 5.428569412231445,
      "epoch": 4.667418790118654,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030138536181080683,
      "loss": 4.8005,
      "mean_token_accuracy": 0.227342689037323,
      "num_tokens": 104045998.0,
      "step": 59990
    },
    {
      "entropy": 5.379879093170166,
      "epoch": 4.667807819490371,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030135783725305815,
      "loss": 4.6564,
      "mean_token_accuracy": 0.229261215031147,
      "num_tokens": 104054380.0,
      "step": 59995
    },
    {
      "entropy": 5.287859058380127,
      "epoch": 4.668196848862089,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003013303122953712,
      "loss": 4.6829,
      "mean_token_accuracy": 0.23285471200942992,
      "num_tokens": 104063672.0,
      "step": 60000
    },
    {
      "epoch": 4.668196848862089,
      "eval_entropy": 5.151966845564709,
      "eval_loss": 4.928664207458496,
      "eval_mean_token_accuracy": 0.22482802388011777,
      "eval_num_tokens": 104063672.0,
      "eval_runtime": 28.6068,
      "eval_samples_per_second": 1452.729,
      "eval_steps_per_second": 181.6,
      "step": 60000
    },
    {
      "entropy": 5.410231733322144,
      "epoch": 4.668585878233807,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003013027869381635,
      "loss": 4.7895,
      "mean_token_accuracy": 0.2236448273062706,
      "num_tokens": 104073128.0,
      "step": 60005
    },
    {
      "entropy": 5.406169033050537,
      "epoch": 4.6689749076055245,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000301275261181853,
      "loss": 4.738,
      "mean_token_accuracy": 0.21737637668848037,
      "num_tokens": 104081957.0,
      "step": 60010
    },
    {
      "entropy": 5.390545845031738,
      "epoch": 4.669363936977242,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003012477350268572,
      "loss": 4.8253,
      "mean_token_accuracy": 0.22602433413267137,
      "num_tokens": 104090678.0,
      "step": 60015
    },
    {
      "entropy": 5.429363489151001,
      "epoch": 4.669752966348959,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030122020847359385,
      "loss": 4.7009,
      "mean_token_accuracy": 0.22935028076171876,
      "num_tokens": 104099729.0,
      "step": 60020
    },
    {
      "entropy": 5.4134115219116214,
      "epoch": 4.670141995720677,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003011926815224805,
      "loss": 4.7333,
      "mean_token_accuracy": 0.22642356008291245,
      "num_tokens": 104108343.0,
      "step": 60025
    },
    {
      "entropy": 5.396613597869873,
      "epoch": 4.670531025092394,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030116515417393495,
      "loss": 4.7769,
      "mean_token_accuracy": 0.2232011675834656,
      "num_tokens": 104117347.0,
      "step": 60030
    },
    {
      "entropy": 5.340514516830444,
      "epoch": 4.670920054464112,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030113762642837487,
      "loss": 4.7837,
      "mean_token_accuracy": 0.22642913907766343,
      "num_tokens": 104126193.0,
      "step": 60035
    },
    {
      "entropy": 5.409164667129517,
      "epoch": 4.67130908383583,
      "grad_norm": 1.09375,
      "learning_rate": 0.00030111009828621786,
      "loss": 4.7417,
      "mean_token_accuracy": 0.21985155791044236,
      "num_tokens": 104134806.0,
      "step": 60040
    },
    {
      "entropy": 5.345441293716431,
      "epoch": 4.6716981132075475,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003010825697478817,
      "loss": 4.7187,
      "mean_token_accuracy": 0.22869377136230468,
      "num_tokens": 104143092.0,
      "step": 60045
    },
    {
      "entropy": 5.403673505783081,
      "epoch": 4.672087142579265,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000301055040813784,
      "loss": 4.7734,
      "mean_token_accuracy": 0.22186497896909713,
      "num_tokens": 104151561.0,
      "step": 60050
    },
    {
      "entropy": 5.421459054946899,
      "epoch": 4.672476171950982,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030102751148434255,
      "loss": 4.7488,
      "mean_token_accuracy": 0.2309314489364624,
      "num_tokens": 104159468.0,
      "step": 60055
    },
    {
      "entropy": 5.320933723449707,
      "epoch": 4.6728652013227,
      "grad_norm": 1.21875,
      "learning_rate": 0.000300999981759975,
      "loss": 4.6787,
      "mean_token_accuracy": 0.23188885897397996,
      "num_tokens": 104167452.0,
      "step": 60060
    },
    {
      "entropy": 5.278049468994141,
      "epoch": 4.673254230694417,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003009724516410992,
      "loss": 4.6384,
      "mean_token_accuracy": 0.23615677505731583,
      "num_tokens": 104175821.0,
      "step": 60065
    },
    {
      "entropy": 5.210593843460083,
      "epoch": 4.673643260066135,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003009449211281328,
      "loss": 4.6804,
      "mean_token_accuracy": 0.23629230558872222,
      "num_tokens": 104184221.0,
      "step": 60070
    },
    {
      "entropy": 5.399457836151123,
      "epoch": 4.674032289437853,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003009173902214934,
      "loss": 4.7612,
      "mean_token_accuracy": 0.21897516697645186,
      "num_tokens": 104192991.0,
      "step": 60075
    },
    {
      "entropy": 5.388724946975708,
      "epoch": 4.6744213188095705,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003008898589215989,
      "loss": 4.7079,
      "mean_token_accuracy": 0.22104968279600143,
      "num_tokens": 104201262.0,
      "step": 60080
    },
    {
      "entropy": 5.4223753929138185,
      "epoch": 4.674810348181287,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030086232722886704,
      "loss": 4.9469,
      "mean_token_accuracy": 0.20952533334493637,
      "num_tokens": 104209794.0,
      "step": 60085
    },
    {
      "entropy": 5.506208276748657,
      "epoch": 4.675199377553005,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00030083479514371545,
      "loss": 4.9284,
      "mean_token_accuracy": 0.20567905008792878,
      "num_tokens": 104217941.0,
      "step": 60090
    },
    {
      "entropy": 5.42679877281189,
      "epoch": 4.675588406924723,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030080726266656205,
      "loss": 4.8363,
      "mean_token_accuracy": 0.21523621827363967,
      "num_tokens": 104227077.0,
      "step": 60095
    },
    {
      "entropy": 5.440836429595947,
      "epoch": 4.67597743629644,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003007797297978245,
      "loss": 4.7659,
      "mean_token_accuracy": 0.22364491671323777,
      "num_tokens": 104236015.0,
      "step": 60100
    },
    {
      "entropy": 5.41000599861145,
      "epoch": 4.676366465668158,
      "grad_norm": 1.25,
      "learning_rate": 0.0003007521965379206,
      "loss": 4.7287,
      "mean_token_accuracy": 0.225263774394989,
      "num_tokens": 104244390.0,
      "step": 60105
    },
    {
      "entropy": 5.419080543518066,
      "epoch": 4.676755495039876,
      "grad_norm": 1.28125,
      "learning_rate": 0.000300724662887268,
      "loss": 4.8213,
      "mean_token_accuracy": 0.21655511111021042,
      "num_tokens": 104253307.0,
      "step": 60110
    },
    {
      "entropy": 5.378915500640869,
      "epoch": 4.6771445244115935,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030069712884628477,
      "loss": 4.7425,
      "mean_token_accuracy": 0.23066506236791612,
      "num_tokens": 104261378.0,
      "step": 60115
    },
    {
      "entropy": 5.339394855499267,
      "epoch": 4.67753355378331,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003006695944153883,
      "loss": 4.6844,
      "mean_token_accuracy": 0.23073328882455826,
      "num_tokens": 104270008.0,
      "step": 60120
    },
    {
      "entropy": 5.404970407485962,
      "epoch": 4.677922583155028,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003006420595949967,
      "loss": 4.8144,
      "mean_token_accuracy": 0.22418383955955506,
      "num_tokens": 104278495.0,
      "step": 60125
    },
    {
      "entropy": 5.402882146835327,
      "epoch": 4.678311612526746,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030061452438552767,
      "loss": 4.7452,
      "mean_token_accuracy": 0.22842894792556762,
      "num_tokens": 104286227.0,
      "step": 60130
    },
    {
      "entropy": 5.401876974105835,
      "epoch": 4.678700641898463,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000300586988787399,
      "loss": 4.82,
      "mean_token_accuracy": 0.225905741751194,
      "num_tokens": 104295279.0,
      "step": 60135
    },
    {
      "entropy": 5.388350772857666,
      "epoch": 4.679089671270181,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030055945280102844,
      "loss": 4.7488,
      "mean_token_accuracy": 0.22071312814950944,
      "num_tokens": 104304551.0,
      "step": 60140
    },
    {
      "entropy": 5.374317026138305,
      "epoch": 4.679478700641899,
      "grad_norm": 1.125,
      "learning_rate": 0.00030053191642683393,
      "loss": 4.7138,
      "mean_token_accuracy": 0.22631774693727494,
      "num_tokens": 104313072.0,
      "step": 60145
    },
    {
      "entropy": 5.467886638641358,
      "epoch": 4.679867730013616,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003005043796652332,
      "loss": 4.8496,
      "mean_token_accuracy": 0.21799646317958832,
      "num_tokens": 104321556.0,
      "step": 60150
    },
    {
      "entropy": 5.420023632049561,
      "epoch": 4.680256759385333,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030047684251664417,
      "loss": 4.8225,
      "mean_token_accuracy": 0.22314722537994386,
      "num_tokens": 104331076.0,
      "step": 60155
    },
    {
      "entropy": 5.349316501617432,
      "epoch": 4.680645788757051,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003004493049814845,
      "loss": 4.7845,
      "mean_token_accuracy": 0.22418000549077988,
      "num_tokens": 104340550.0,
      "step": 60160
    },
    {
      "entropy": 5.386523914337158,
      "epoch": 4.681034818128769,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030042176706017215,
      "loss": 4.7413,
      "mean_token_accuracy": 0.22461476027965546,
      "num_tokens": 104349189.0,
      "step": 60165
    },
    {
      "entropy": 5.425446653366089,
      "epoch": 4.681423847500486,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030039422875312495,
      "loss": 4.7864,
      "mean_token_accuracy": 0.2179361790418625,
      "num_tokens": 104357272.0,
      "step": 60170
    },
    {
      "entropy": 5.396423578262329,
      "epoch": 4.681812876872204,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030036669006076075,
      "loss": 4.8112,
      "mean_token_accuracy": 0.22226592153310776,
      "num_tokens": 104365876.0,
      "step": 60175
    },
    {
      "entropy": 5.286186933517456,
      "epoch": 4.682201906243922,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030033915098349746,
      "loss": 4.6998,
      "mean_token_accuracy": 0.23372445851564408,
      "num_tokens": 104374819.0,
      "step": 60180
    },
    {
      "entropy": 5.3681684017181395,
      "epoch": 4.682590935615639,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003003116115217529,
      "loss": 4.7737,
      "mean_token_accuracy": 0.22794780135154724,
      "num_tokens": 104383108.0,
      "step": 60185
    },
    {
      "entropy": 5.3533162593841555,
      "epoch": 4.682979964987356,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030028407167594486,
      "loss": 4.7958,
      "mean_token_accuracy": 0.22355110794305802,
      "num_tokens": 104391765.0,
      "step": 60190
    },
    {
      "entropy": 5.441995143890381,
      "epoch": 4.683368994359074,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003002565314464913,
      "loss": 4.8266,
      "mean_token_accuracy": 0.21691330671310424,
      "num_tokens": 104400148.0,
      "step": 60195
    },
    {
      "entropy": 5.388433933258057,
      "epoch": 4.683758023730792,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030022899083381006,
      "loss": 4.7043,
      "mean_token_accuracy": 0.2246743142604828,
      "num_tokens": 104409071.0,
      "step": 60200
    },
    {
      "entropy": 5.3608098983764645,
      "epoch": 4.684147053102509,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000300201449838319,
      "loss": 4.7096,
      "mean_token_accuracy": 0.2297302395105362,
      "num_tokens": 104418009.0,
      "step": 60205
    },
    {
      "entropy": 5.346059799194336,
      "epoch": 4.684536082474227,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003001739084604361,
      "loss": 4.7892,
      "mean_token_accuracy": 0.23155113607645034,
      "num_tokens": 104426705.0,
      "step": 60210
    },
    {
      "entropy": 5.450073766708374,
      "epoch": 4.684925111845945,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030014636670057913,
      "loss": 4.8621,
      "mean_token_accuracy": 0.2165462151169777,
      "num_tokens": 104435366.0,
      "step": 60215
    },
    {
      "entropy": 5.455569744110107,
      "epoch": 4.6853141412176615,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003001188245591662,
      "loss": 4.902,
      "mean_token_accuracy": 0.21186133623123168,
      "num_tokens": 104443441.0,
      "step": 60220
    },
    {
      "entropy": 5.414336442947388,
      "epoch": 4.685703170589379,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003000912820366151,
      "loss": 4.7343,
      "mean_token_accuracy": 0.2263411447405815,
      "num_tokens": 104452397.0,
      "step": 60225
    },
    {
      "entropy": 5.4703319549560545,
      "epoch": 4.686092199961097,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030006373913334366,
      "loss": 4.895,
      "mean_token_accuracy": 0.21586623340845107,
      "num_tokens": 104461819.0,
      "step": 60230
    },
    {
      "entropy": 5.3592884063720705,
      "epoch": 4.686481229332815,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003000361958497699,
      "loss": 4.6695,
      "mean_token_accuracy": 0.22960598021745682,
      "num_tokens": 104469654.0,
      "step": 60235
    },
    {
      "entropy": 5.365390300750732,
      "epoch": 4.686870258704532,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003000086521863116,
      "loss": 4.742,
      "mean_token_accuracy": 0.23242188096046448,
      "num_tokens": 104477909.0,
      "step": 60240
    },
    {
      "entropy": 5.354189205169678,
      "epoch": 4.68725928807625,
      "grad_norm": 1.203125,
      "learning_rate": 0.000299981108143387,
      "loss": 4.7268,
      "mean_token_accuracy": 0.22358058243989945,
      "num_tokens": 104486849.0,
      "step": 60245
    },
    {
      "entropy": 5.360296821594238,
      "epoch": 4.687648317447968,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029995356372141367,
      "loss": 4.7764,
      "mean_token_accuracy": 0.2258102849125862,
      "num_tokens": 104496036.0,
      "step": 60250
    },
    {
      "entropy": 5.290598392486572,
      "epoch": 4.6880373468196845,
      "grad_norm": 1.125,
      "learning_rate": 0.0002999260189208098,
      "loss": 4.6637,
      "mean_token_accuracy": 0.2367432877421379,
      "num_tokens": 104504755.0,
      "step": 60255
    },
    {
      "entropy": 5.404932117462158,
      "epoch": 4.688426376191402,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029989847374199333,
      "loss": 4.747,
      "mean_token_accuracy": 0.2247995063662529,
      "num_tokens": 104512375.0,
      "step": 60260
    },
    {
      "entropy": 5.387936449050903,
      "epoch": 4.68881540556312,
      "grad_norm": 1.0625,
      "learning_rate": 0.00029987092818538217,
      "loss": 4.7109,
      "mean_token_accuracy": 0.22521898299455642,
      "num_tokens": 104521351.0,
      "step": 60265
    },
    {
      "entropy": 5.338318586349487,
      "epoch": 4.689204434934838,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029984338225139405,
      "loss": 4.7487,
      "mean_token_accuracy": 0.229100102186203,
      "num_tokens": 104529854.0,
      "step": 60270
    },
    {
      "entropy": 5.401294183731079,
      "epoch": 4.689593464306555,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002998158359404473,
      "loss": 4.8573,
      "mean_token_accuracy": 0.21474246084690093,
      "num_tokens": 104538417.0,
      "step": 60275
    },
    {
      "entropy": 5.407567930221558,
      "epoch": 4.689982493678273,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002997882892529597,
      "loss": 4.8497,
      "mean_token_accuracy": 0.22154248952865602,
      "num_tokens": 104547572.0,
      "step": 60280
    },
    {
      "entropy": 5.416397619247436,
      "epoch": 4.69037152304999,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002997607421893493,
      "loss": 4.7737,
      "mean_token_accuracy": 0.2249545469880104,
      "num_tokens": 104556529.0,
      "step": 60285
    },
    {
      "entropy": 5.374614429473877,
      "epoch": 4.6907605524217075,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000299733194750034,
      "loss": 4.6591,
      "mean_token_accuracy": 0.22901337146759032,
      "num_tokens": 104564810.0,
      "step": 60290
    },
    {
      "entropy": 5.45711932182312,
      "epoch": 4.691149581793425,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002997056469354319,
      "loss": 4.8869,
      "mean_token_accuracy": 0.21583421975374223,
      "num_tokens": 104573983.0,
      "step": 60295
    },
    {
      "entropy": 5.45582218170166,
      "epoch": 4.691538611165143,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002996780987459609,
      "loss": 4.8235,
      "mean_token_accuracy": 0.2177221342921257,
      "num_tokens": 104582026.0,
      "step": 60300
    },
    {
      "entropy": 5.439003562927246,
      "epoch": 4.691927640536861,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002996505501820391,
      "loss": 4.8528,
      "mean_token_accuracy": 0.218086639046669,
      "num_tokens": 104590455.0,
      "step": 60305
    },
    {
      "entropy": 5.352758884429932,
      "epoch": 4.692316669908578,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002996230012440843,
      "loss": 4.7658,
      "mean_token_accuracy": 0.22782424092292786,
      "num_tokens": 104597970.0,
      "step": 60310
    },
    {
      "entropy": 5.323047542572022,
      "epoch": 4.692705699280296,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029959545193251473,
      "loss": 4.7055,
      "mean_token_accuracy": 0.2321253687143326,
      "num_tokens": 104606633.0,
      "step": 60315
    },
    {
      "entropy": 5.420656681060791,
      "epoch": 4.693094728652014,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029956790224774834,
      "loss": 4.8286,
      "mean_token_accuracy": 0.21114153563976287,
      "num_tokens": 104614943.0,
      "step": 60320
    },
    {
      "entropy": 5.413617897033691,
      "epoch": 4.6934837580237305,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029954035219020305,
      "loss": 4.8269,
      "mean_token_accuracy": 0.2167972892522812,
      "num_tokens": 104624031.0,
      "step": 60325
    },
    {
      "entropy": 5.384133958816529,
      "epoch": 4.693872787395448,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029951280176029703,
      "loss": 4.7219,
      "mean_token_accuracy": 0.22871862649917601,
      "num_tokens": 104632923.0,
      "step": 60330
    },
    {
      "entropy": 5.399416303634643,
      "epoch": 4.694261816767166,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029948525095844825,
      "loss": 4.7763,
      "mean_token_accuracy": 0.21840869784355163,
      "num_tokens": 104641551.0,
      "step": 60335
    },
    {
      "entropy": 5.303564500808716,
      "epoch": 4.694650846138884,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002994576997850748,
      "loss": 4.6789,
      "mean_token_accuracy": 0.23053375482559205,
      "num_tokens": 104649851.0,
      "step": 60340
    },
    {
      "entropy": 5.413760328292847,
      "epoch": 4.695039875510601,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002994301482405946,
      "loss": 4.8525,
      "mean_token_accuracy": 0.21939559876918793,
      "num_tokens": 104658388.0,
      "step": 60345
    },
    {
      "entropy": 5.386781930923462,
      "epoch": 4.695428904882318,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002994025963254259,
      "loss": 4.7528,
      "mean_token_accuracy": 0.22399701327085494,
      "num_tokens": 104667117.0,
      "step": 60350
    },
    {
      "entropy": 5.377438116073608,
      "epoch": 4.695817934254036,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002993750440399866,
      "loss": 4.7269,
      "mean_token_accuracy": 0.22762708216905594,
      "num_tokens": 104675877.0,
      "step": 60355
    },
    {
      "entropy": 5.479869985580445,
      "epoch": 4.6962069636257535,
      "grad_norm": 1.25,
      "learning_rate": 0.0002993474913846948,
      "loss": 4.949,
      "mean_token_accuracy": 0.21404290348291397,
      "num_tokens": 104684872.0,
      "step": 60360
    },
    {
      "entropy": 5.477115774154663,
      "epoch": 4.696595992997471,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002993199383599685,
      "loss": 4.8385,
      "mean_token_accuracy": 0.2203797683119774,
      "num_tokens": 104694015.0,
      "step": 60365
    },
    {
      "entropy": 5.578516483306885,
      "epoch": 4.696985022369189,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002992923849662259,
      "loss": 4.9793,
      "mean_token_accuracy": 0.21035704761743546,
      "num_tokens": 104702627.0,
      "step": 60370
    },
    {
      "entropy": 5.486182689666748,
      "epoch": 4.697374051740907,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029926483120388506,
      "loss": 4.8378,
      "mean_token_accuracy": 0.2162863075733185,
      "num_tokens": 104711738.0,
      "step": 60375
    },
    {
      "entropy": 5.44186692237854,
      "epoch": 4.697763081112624,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000299237277073364,
      "loss": 4.7411,
      "mean_token_accuracy": 0.22660953551530838,
      "num_tokens": 104721066.0,
      "step": 60380
    },
    {
      "entropy": 5.391833877563476,
      "epoch": 4.698152110484342,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002992097225750808,
      "loss": 4.7871,
      "mean_token_accuracy": 0.2217494234442711,
      "num_tokens": 104729747.0,
      "step": 60385
    },
    {
      "entropy": 5.351077222824097,
      "epoch": 4.698541139856059,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029918216770945363,
      "loss": 4.6236,
      "mean_token_accuracy": 0.23450458645820618,
      "num_tokens": 104738630.0,
      "step": 60390
    },
    {
      "entropy": 5.271799373626709,
      "epoch": 4.6989301692277765,
      "grad_norm": 1.25,
      "learning_rate": 0.0002991546124769005,
      "loss": 4.6957,
      "mean_token_accuracy": 0.2337660387158394,
      "num_tokens": 104746564.0,
      "step": 60395
    },
    {
      "entropy": 5.247576951980591,
      "epoch": 4.699319198599494,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002991270568778396,
      "loss": 4.6706,
      "mean_token_accuracy": 0.23339837044477463,
      "num_tokens": 104755589.0,
      "step": 60400
    },
    {
      "entropy": 5.3798693180084225,
      "epoch": 4.699708227971212,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002990995009126891,
      "loss": 4.849,
      "mean_token_accuracy": 0.21535152941942215,
      "num_tokens": 104764608.0,
      "step": 60405
    },
    {
      "entropy": 5.442778825759888,
      "epoch": 4.70009725734293,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002990719445818669,
      "loss": 4.8127,
      "mean_token_accuracy": 0.22578059434890746,
      "num_tokens": 104772363.0,
      "step": 60410
    },
    {
      "entropy": 5.371535015106201,
      "epoch": 4.700486286714647,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002990443878857914,
      "loss": 4.7557,
      "mean_token_accuracy": 0.23090972900390624,
      "num_tokens": 104781461.0,
      "step": 60415
    },
    {
      "entropy": 5.25320839881897,
      "epoch": 4.700875316086364,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029901683082488045,
      "loss": 4.6448,
      "mean_token_accuracy": 0.23372883051633836,
      "num_tokens": 104789966.0,
      "step": 60420
    },
    {
      "entropy": 5.322723817825318,
      "epoch": 4.701264345458082,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002989892733995524,
      "loss": 4.7448,
      "mean_token_accuracy": 0.22545391619205474,
      "num_tokens": 104798289.0,
      "step": 60425
    },
    {
      "entropy": 5.4582967281341555,
      "epoch": 4.7016533748297995,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002989617156102253,
      "loss": 4.8655,
      "mean_token_accuracy": 0.22121088206768036,
      "num_tokens": 104807189.0,
      "step": 60430
    },
    {
      "entropy": 5.369121456146241,
      "epoch": 4.702042404201517,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002989341574573172,
      "loss": 4.672,
      "mean_token_accuracy": 0.22906961292028427,
      "num_tokens": 104816150.0,
      "step": 60435
    },
    {
      "entropy": 5.318262195587158,
      "epoch": 4.702431433573235,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029890659894124647,
      "loss": 4.6514,
      "mean_token_accuracy": 0.22709394842386246,
      "num_tokens": 104824323.0,
      "step": 60440
    },
    {
      "entropy": 5.32560248374939,
      "epoch": 4.702820462944953,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002988790400624312,
      "loss": 4.7901,
      "mean_token_accuracy": 0.2174514651298523,
      "num_tokens": 104833341.0,
      "step": 60445
    },
    {
      "entropy": 5.411396837234497,
      "epoch": 4.70320949231667,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002988514808212894,
      "loss": 4.7711,
      "mean_token_accuracy": 0.22721890807151796,
      "num_tokens": 104841570.0,
      "step": 60450
    },
    {
      "entropy": 5.518634796142578,
      "epoch": 4.703598521688387,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002988239212182394,
      "loss": 4.8852,
      "mean_token_accuracy": 0.21117112338542937,
      "num_tokens": 104851673.0,
      "step": 60455
    },
    {
      "entropy": 5.419036436080932,
      "epoch": 4.703987551060105,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002987963612536993,
      "loss": 4.7759,
      "mean_token_accuracy": 0.22906404733657837,
      "num_tokens": 104860198.0,
      "step": 60460
    },
    {
      "entropy": 5.369667959213257,
      "epoch": 4.7043765804318225,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029876880092808724,
      "loss": 4.7221,
      "mean_token_accuracy": 0.22666166424751283,
      "num_tokens": 104868243.0,
      "step": 60465
    },
    {
      "entropy": 5.377785158157349,
      "epoch": 4.70476560980354,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002987412402418215,
      "loss": 4.818,
      "mean_token_accuracy": 0.23084015697240828,
      "num_tokens": 104877473.0,
      "step": 60470
    },
    {
      "entropy": 5.361766386032104,
      "epoch": 4.705154639175258,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029871367919532014,
      "loss": 4.6882,
      "mean_token_accuracy": 0.23672889471054076,
      "num_tokens": 104885942.0,
      "step": 60475
    },
    {
      "entropy": 5.479023551940918,
      "epoch": 4.705543668546976,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002986861177890016,
      "loss": 4.7813,
      "mean_token_accuracy": 0.227249351143837,
      "num_tokens": 104894299.0,
      "step": 60480
    },
    {
      "entropy": 5.313975715637207,
      "epoch": 4.705932697918692,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002986585560232838,
      "loss": 4.6812,
      "mean_token_accuracy": 0.2312782123684883,
      "num_tokens": 104902658.0,
      "step": 60485
    },
    {
      "entropy": 5.277635812759399,
      "epoch": 4.70632172729041,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029863099389858516,
      "loss": 4.6515,
      "mean_token_accuracy": 0.22762805074453354,
      "num_tokens": 104911351.0,
      "step": 60490
    },
    {
      "entropy": 5.472627115249634,
      "epoch": 4.706710756662128,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002986034314153238,
      "loss": 4.8532,
      "mean_token_accuracy": 0.2159953758120537,
      "num_tokens": 104921064.0,
      "step": 60495
    },
    {
      "entropy": 5.450775289535523,
      "epoch": 4.7070997860338455,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029857586857391796,
      "loss": 4.7667,
      "mean_token_accuracy": 0.2128722682595253,
      "num_tokens": 104929723.0,
      "step": 60500
    },
    {
      "entropy": 5.39575161933899,
      "epoch": 4.707488815405563,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029854830537478567,
      "loss": 4.8334,
      "mean_token_accuracy": 0.2261297360062599,
      "num_tokens": 104939215.0,
      "step": 60505
    },
    {
      "entropy": 5.413983583450317,
      "epoch": 4.707877844777281,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029852074181834545,
      "loss": 4.7898,
      "mean_token_accuracy": 0.22414161264896393,
      "num_tokens": 104948062.0,
      "step": 60510
    },
    {
      "entropy": 5.460183715820312,
      "epoch": 4.708266874148999,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029849317790501536,
      "loss": 4.7875,
      "mean_token_accuracy": 0.22126740366220474,
      "num_tokens": 104957495.0,
      "step": 60515
    },
    {
      "entropy": 5.413878107070923,
      "epoch": 4.708655903520716,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029846561363521375,
      "loss": 4.9029,
      "mean_token_accuracy": 0.21352116912603378,
      "num_tokens": 104966049.0,
      "step": 60520
    },
    {
      "entropy": 5.362669801712036,
      "epoch": 4.709044932892433,
      "grad_norm": 1.25,
      "learning_rate": 0.0002984380490093588,
      "loss": 4.6737,
      "mean_token_accuracy": 0.23406160920858382,
      "num_tokens": 104973957.0,
      "step": 60525
    },
    {
      "entropy": 5.410939264297485,
      "epoch": 4.709433962264151,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002984104840278687,
      "loss": 4.8374,
      "mean_token_accuracy": 0.2175553858280182,
      "num_tokens": 104982960.0,
      "step": 60530
    },
    {
      "entropy": 5.385405778884888,
      "epoch": 4.7098229916358685,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002983829186911618,
      "loss": 4.845,
      "mean_token_accuracy": 0.21627107411623,
      "num_tokens": 104992251.0,
      "step": 60535
    },
    {
      "entropy": 5.374772548675537,
      "epoch": 4.710212021007586,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029835535299965635,
      "loss": 4.7867,
      "mean_token_accuracy": 0.22377547919750213,
      "num_tokens": 105000304.0,
      "step": 60540
    },
    {
      "entropy": 5.474846220016479,
      "epoch": 4.710601050379304,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029832778695377057,
      "loss": 4.8227,
      "mean_token_accuracy": 0.21919759064912797,
      "num_tokens": 105008643.0,
      "step": 60545
    },
    {
      "entropy": 5.31832480430603,
      "epoch": 4.710990079751022,
      "grad_norm": 1.125,
      "learning_rate": 0.00029830022055392277,
      "loss": 4.6869,
      "mean_token_accuracy": 0.2293759897351265,
      "num_tokens": 105017865.0,
      "step": 60550
    },
    {
      "entropy": 5.381431007385254,
      "epoch": 4.711379109122738,
      "grad_norm": 1.25,
      "learning_rate": 0.00029827265380053116,
      "loss": 4.7726,
      "mean_token_accuracy": 0.2261948898434639,
      "num_tokens": 105025638.0,
      "step": 60555
    },
    {
      "entropy": 5.46579008102417,
      "epoch": 4.711768138494456,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002982450866940141,
      "loss": 4.9033,
      "mean_token_accuracy": 0.2154247835278511,
      "num_tokens": 105034526.0,
      "step": 60560
    },
    {
      "entropy": 5.3371203422546385,
      "epoch": 4.712157167866174,
      "grad_norm": 1.25,
      "learning_rate": 0.0002982175192347899,
      "loss": 4.645,
      "mean_token_accuracy": 0.23753733783960343,
      "num_tokens": 105043059.0,
      "step": 60565
    },
    {
      "entropy": 5.415519332885742,
      "epoch": 4.7125461972378915,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002981899514232767,
      "loss": 4.8439,
      "mean_token_accuracy": 0.22339830547571182,
      "num_tokens": 105051937.0,
      "step": 60570
    },
    {
      "entropy": 5.328790140151978,
      "epoch": 4.712935226609609,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000298162383259893,
      "loss": 4.7305,
      "mean_token_accuracy": 0.2258306249976158,
      "num_tokens": 105061304.0,
      "step": 60575
    },
    {
      "entropy": 5.345498752593994,
      "epoch": 4.713324255981327,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029813481474505693,
      "loss": 4.7145,
      "mean_token_accuracy": 0.22235454022884368,
      "num_tokens": 105069871.0,
      "step": 60580
    },
    {
      "entropy": 5.417628955841065,
      "epoch": 4.713713285353045,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029810724587918684,
      "loss": 4.7601,
      "mean_token_accuracy": 0.22741216868162156,
      "num_tokens": 105078616.0,
      "step": 60585
    },
    {
      "entropy": 5.332662153244018,
      "epoch": 4.714102314724761,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002980796766627012,
      "loss": 4.6432,
      "mean_token_accuracy": 0.23454943150281907,
      "num_tokens": 105086135.0,
      "step": 60590
    },
    {
      "entropy": 5.344504499435425,
      "epoch": 4.714491344096479,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002980521070960181,
      "loss": 4.765,
      "mean_token_accuracy": 0.2268511325120926,
      "num_tokens": 105094933.0,
      "step": 60595
    },
    {
      "entropy": 5.422547769546509,
      "epoch": 4.714880373468197,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029802453717955597,
      "loss": 4.8481,
      "mean_token_accuracy": 0.22450993210077286,
      "num_tokens": 105104198.0,
      "step": 60600
    },
    {
      "entropy": 5.361999273300171,
      "epoch": 4.7152694028399145,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002979969669137332,
      "loss": 4.7538,
      "mean_token_accuracy": 0.2271420270204544,
      "num_tokens": 105112733.0,
      "step": 60605
    },
    {
      "entropy": 5.464924478530884,
      "epoch": 4.715658432211632,
      "grad_norm": 1.1875,
      "learning_rate": 0.000297969396298968,
      "loss": 4.8138,
      "mean_token_accuracy": 0.22097700834274292,
      "num_tokens": 105121296.0,
      "step": 60610
    },
    {
      "entropy": 5.3705606937408445,
      "epoch": 4.71604746158335,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002979418253356788,
      "loss": 4.7698,
      "mean_token_accuracy": 0.22860443890094756,
      "num_tokens": 105130056.0,
      "step": 60615
    },
    {
      "entropy": 5.40941596031189,
      "epoch": 4.716436490955067,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002979142540242839,
      "loss": 4.7048,
      "mean_token_accuracy": 0.23447709083557128,
      "num_tokens": 105138400.0,
      "step": 60620
    },
    {
      "entropy": 5.348446226119995,
      "epoch": 4.716825520326784,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029788668236520164,
      "loss": 4.8352,
      "mean_token_accuracy": 0.22121359258890153,
      "num_tokens": 105146544.0,
      "step": 60625
    },
    {
      "entropy": 5.497742319107056,
      "epoch": 4.717214549698502,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029785911035885046,
      "loss": 4.8976,
      "mean_token_accuracy": 0.2159128487110138,
      "num_tokens": 105155404.0,
      "step": 60630
    },
    {
      "entropy": 5.479377555847168,
      "epoch": 4.71760357907022,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029783153800564863,
      "loss": 4.8624,
      "mean_token_accuracy": 0.2220592603087425,
      "num_tokens": 105165292.0,
      "step": 60635
    },
    {
      "entropy": 5.470874881744384,
      "epoch": 4.7179926084419375,
      "grad_norm": 1.25,
      "learning_rate": 0.00029780396530601456,
      "loss": 4.8766,
      "mean_token_accuracy": 0.21533263027667998,
      "num_tokens": 105174149.0,
      "step": 60640
    },
    {
      "entropy": 5.4024450302124025,
      "epoch": 4.718381637813655,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029777639226036663,
      "loss": 4.7966,
      "mean_token_accuracy": 0.21946212649345398,
      "num_tokens": 105183277.0,
      "step": 60645
    },
    {
      "entropy": 5.506145906448364,
      "epoch": 4.718770667185373,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002977488188691232,
      "loss": 4.9637,
      "mean_token_accuracy": 0.21099164783954621,
      "num_tokens": 105191789.0,
      "step": 60650
    },
    {
      "entropy": 5.45927882194519,
      "epoch": 4.7191596965570906,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002977212451327026,
      "loss": 4.7724,
      "mean_token_accuracy": 0.22717005610466004,
      "num_tokens": 105200619.0,
      "step": 60655
    },
    {
      "entropy": 5.364016914367676,
      "epoch": 4.719548725928807,
      "grad_norm": 1.125,
      "learning_rate": 0.0002976936710515233,
      "loss": 4.8542,
      "mean_token_accuracy": 0.21881923973560333,
      "num_tokens": 105209657.0,
      "step": 60660
    },
    {
      "entropy": 5.411705112457275,
      "epoch": 4.719937755300525,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029766609662600367,
      "loss": 4.8079,
      "mean_token_accuracy": 0.2255818858742714,
      "num_tokens": 105218788.0,
      "step": 60665
    },
    {
      "entropy": 5.394647169113159,
      "epoch": 4.720326784672243,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029763852185656206,
      "loss": 4.7456,
      "mean_token_accuracy": 0.22670212984085084,
      "num_tokens": 105227185.0,
      "step": 60670
    },
    {
      "entropy": 5.3910784244537355,
      "epoch": 4.7207158140439605,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029761094674361695,
      "loss": 4.7964,
      "mean_token_accuracy": 0.22627199292182923,
      "num_tokens": 105235594.0,
      "step": 60675
    },
    {
      "entropy": 5.426207828521728,
      "epoch": 4.721104843415678,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029758337128758675,
      "loss": 4.8422,
      "mean_token_accuracy": 0.2224229410290718,
      "num_tokens": 105244281.0,
      "step": 60680
    },
    {
      "entropy": 5.384992074966431,
      "epoch": 4.721493872787395,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002975557954888898,
      "loss": 4.7245,
      "mean_token_accuracy": 0.22402421087026597,
      "num_tokens": 105252212.0,
      "step": 60685
    },
    {
      "entropy": 5.468262338638306,
      "epoch": 4.721882902159113,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029752821934794457,
      "loss": 4.9707,
      "mean_token_accuracy": 0.2129919484257698,
      "num_tokens": 105260704.0,
      "step": 60690
    },
    {
      "entropy": 5.395763921737671,
      "epoch": 4.72227193153083,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029750064286516943,
      "loss": 4.7595,
      "mean_token_accuracy": 0.22723571956157684,
      "num_tokens": 105270262.0,
      "step": 60695
    },
    {
      "entropy": 5.463924980163574,
      "epoch": 4.722660960902548,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002974730660409828,
      "loss": 4.8356,
      "mean_token_accuracy": 0.21521886438131332,
      "num_tokens": 105278647.0,
      "step": 60700
    },
    {
      "entropy": 5.3655601978302006,
      "epoch": 4.723049990274266,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029744548887580325,
      "loss": 4.7586,
      "mean_token_accuracy": 0.22643764764070512,
      "num_tokens": 105287239.0,
      "step": 60705
    },
    {
      "entropy": 5.401728391647339,
      "epoch": 4.7234390196459834,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029741791137004916,
      "loss": 4.7576,
      "mean_token_accuracy": 0.22341520041227342,
      "num_tokens": 105295636.0,
      "step": 60710
    },
    {
      "entropy": 5.407438898086548,
      "epoch": 4.723828049017701,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029739033352413883,
      "loss": 4.8147,
      "mean_token_accuracy": 0.22686165571212769,
      "num_tokens": 105304370.0,
      "step": 60715
    },
    {
      "entropy": 5.385833883285523,
      "epoch": 4.724217078389419,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029736275533849094,
      "loss": 4.8682,
      "mean_token_accuracy": 0.2145046263933182,
      "num_tokens": 105313293.0,
      "step": 60720
    },
    {
      "entropy": 5.452903366088867,
      "epoch": 4.724606107761136,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002973351768135239,
      "loss": 4.7791,
      "mean_token_accuracy": 0.22150887697935104,
      "num_tokens": 105320902.0,
      "step": 60725
    },
    {
      "entropy": 5.437490367889405,
      "epoch": 4.724995137132853,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029730759794965595,
      "loss": 4.8871,
      "mean_token_accuracy": 0.21357990503311158,
      "num_tokens": 105329869.0,
      "step": 60730
    },
    {
      "entropy": 5.445147132873535,
      "epoch": 4.725384166504571,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002972800187473057,
      "loss": 4.831,
      "mean_token_accuracy": 0.21722299307584764,
      "num_tokens": 105338310.0,
      "step": 60735
    },
    {
      "entropy": 5.426887083053589,
      "epoch": 4.725773195876289,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029725243920689163,
      "loss": 4.7823,
      "mean_token_accuracy": 0.21699030548334122,
      "num_tokens": 105347265.0,
      "step": 60740
    },
    {
      "entropy": 5.402003526687622,
      "epoch": 4.726162225248006,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029722485932883233,
      "loss": 4.7716,
      "mean_token_accuracy": 0.2251127690076828,
      "num_tokens": 105355368.0,
      "step": 60745
    },
    {
      "entropy": 5.388609790802002,
      "epoch": 4.726551254619724,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029719727911354604,
      "loss": 4.8186,
      "mean_token_accuracy": 0.21800445765256882,
      "num_tokens": 105364977.0,
      "step": 60750
    },
    {
      "entropy": 5.481606769561767,
      "epoch": 4.726940283991441,
      "grad_norm": 1.25,
      "learning_rate": 0.00029716969856145143,
      "loss": 4.827,
      "mean_token_accuracy": 0.2246081590652466,
      "num_tokens": 105374088.0,
      "step": 60755
    },
    {
      "entropy": 5.386549949645996,
      "epoch": 4.727329313363159,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029714211767296693,
      "loss": 4.6793,
      "mean_token_accuracy": 0.23195631802082062,
      "num_tokens": 105382925.0,
      "step": 60760
    },
    {
      "entropy": 5.4242490291595455,
      "epoch": 4.727718342734876,
      "grad_norm": 1.125,
      "learning_rate": 0.0002971145364485111,
      "loss": 4.74,
      "mean_token_accuracy": 0.22623523324728012,
      "num_tokens": 105392586.0,
      "step": 60765
    },
    {
      "entropy": 5.373070526123047,
      "epoch": 4.728107372106594,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002970869548885023,
      "loss": 4.7528,
      "mean_token_accuracy": 0.22661541104316713,
      "num_tokens": 105401092.0,
      "step": 60770
    },
    {
      "entropy": 5.358581781387329,
      "epoch": 4.728496401478312,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002970593729933591,
      "loss": 4.8154,
      "mean_token_accuracy": 0.21873392313718795,
      "num_tokens": 105410225.0,
      "step": 60775
    },
    {
      "entropy": 5.371318387985229,
      "epoch": 4.728885430850029,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029703179076350005,
      "loss": 4.7156,
      "mean_token_accuracy": 0.22888160943984986,
      "num_tokens": 105418214.0,
      "step": 60780
    },
    {
      "entropy": 5.443757772445679,
      "epoch": 4.729274460221747,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029700420819934363,
      "loss": 4.8575,
      "mean_token_accuracy": 0.22139487117528917,
      "num_tokens": 105427196.0,
      "step": 60785
    },
    {
      "entropy": 5.390106964111328,
      "epoch": 4.729663489593464,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029697662530130836,
      "loss": 4.7944,
      "mean_token_accuracy": 0.22351303100585937,
      "num_tokens": 105435496.0,
      "step": 60790
    },
    {
      "entropy": 5.390697336196899,
      "epoch": 4.730052518965182,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002969490420698129,
      "loss": 4.7258,
      "mean_token_accuracy": 0.22501311153173448,
      "num_tokens": 105443353.0,
      "step": 60795
    },
    {
      "entropy": 5.344409990310669,
      "epoch": 4.730441548336899,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029692145850527553,
      "loss": 4.7875,
      "mean_token_accuracy": 0.21797986924648285,
      "num_tokens": 105451969.0,
      "step": 60800
    },
    {
      "entropy": 5.397104358673095,
      "epoch": 4.730830577708617,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029689387460811494,
      "loss": 4.7027,
      "mean_token_accuracy": 0.22758535146713257,
      "num_tokens": 105460582.0,
      "step": 60805
    },
    {
      "entropy": 5.287304210662842,
      "epoch": 4.731219607080335,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002968662903787497,
      "loss": 4.6509,
      "mean_token_accuracy": 0.23784717917442322,
      "num_tokens": 105469045.0,
      "step": 60810
    },
    {
      "entropy": 5.4265531539917,
      "epoch": 4.731608636452052,
      "grad_norm": 1.25,
      "learning_rate": 0.0002968387058175983,
      "loss": 4.7923,
      "mean_token_accuracy": 0.21973309218883513,
      "num_tokens": 105477175.0,
      "step": 60815
    },
    {
      "entropy": 5.33932466506958,
      "epoch": 4.731997665823769,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002968111209250792,
      "loss": 4.7314,
      "mean_token_accuracy": 0.2341986194252968,
      "num_tokens": 105485569.0,
      "step": 60820
    },
    {
      "entropy": 5.47452301979065,
      "epoch": 4.732386695195487,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029678353570161117,
      "loss": 4.8646,
      "mean_token_accuracy": 0.2159319654107094,
      "num_tokens": 105494177.0,
      "step": 60825
    },
    {
      "entropy": 5.430749702453613,
      "epoch": 4.732775724567205,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002967559501476126,
      "loss": 4.8401,
      "mean_token_accuracy": 0.22461995333433152,
      "num_tokens": 105503678.0,
      "step": 60830
    },
    {
      "entropy": 5.4184314727783205,
      "epoch": 4.733164753938922,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029672836426350213,
      "loss": 4.8059,
      "mean_token_accuracy": 0.22504674643278122,
      "num_tokens": 105511988.0,
      "step": 60835
    },
    {
      "entropy": 5.489646673202515,
      "epoch": 4.73355378331064,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029670077804969835,
      "loss": 4.8226,
      "mean_token_accuracy": 0.22241744101047517,
      "num_tokens": 105520377.0,
      "step": 60840
    },
    {
      "entropy": 5.459394931793213,
      "epoch": 4.733942812682358,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029667319150661976,
      "loss": 4.7217,
      "mean_token_accuracy": 0.2245567485690117,
      "num_tokens": 105528863.0,
      "step": 60845
    },
    {
      "entropy": 5.296284008026123,
      "epoch": 4.734331842054075,
      "grad_norm": 1.203125,
      "learning_rate": 0.000296645604634685,
      "loss": 4.6307,
      "mean_token_accuracy": 0.23047300726175307,
      "num_tokens": 105537543.0,
      "step": 60850
    },
    {
      "entropy": 5.350412464141845,
      "epoch": 4.734720871425793,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002966180174343127,
      "loss": 4.7837,
      "mean_token_accuracy": 0.22362377643585205,
      "num_tokens": 105546895.0,
      "step": 60855
    },
    {
      "entropy": 5.445178842544555,
      "epoch": 4.73510990079751,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029659042990592125,
      "loss": 4.8893,
      "mean_token_accuracy": 0.20818587392568588,
      "num_tokens": 105555857.0,
      "step": 60860
    },
    {
      "entropy": 5.5068724155426025,
      "epoch": 4.735498930169228,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002965628420499295,
      "loss": 4.8321,
      "mean_token_accuracy": 0.2190171793103218,
      "num_tokens": 105563835.0,
      "step": 60865
    },
    {
      "entropy": 5.406539344787598,
      "epoch": 4.735887959540945,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002965352538667559,
      "loss": 4.7933,
      "mean_token_accuracy": 0.2189328595995903,
      "num_tokens": 105573001.0,
      "step": 60870
    },
    {
      "entropy": 5.335478782653809,
      "epoch": 4.736276988912663,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029650766535681917,
      "loss": 4.7314,
      "mean_token_accuracy": 0.23009627610445021,
      "num_tokens": 105581718.0,
      "step": 60875
    },
    {
      "entropy": 5.377077913284301,
      "epoch": 4.736666018284381,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029648007652053774,
      "loss": 4.7797,
      "mean_token_accuracy": 0.22028095722198487,
      "num_tokens": 105590932.0,
      "step": 60880
    },
    {
      "entropy": 5.445408153533935,
      "epoch": 4.737055047656098,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029645248735833045,
      "loss": 4.7512,
      "mean_token_accuracy": 0.23277106285095214,
      "num_tokens": 105599113.0,
      "step": 60885
    },
    {
      "entropy": 5.459812641143799,
      "epoch": 4.737444077027815,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029642489787061574,
      "loss": 4.7588,
      "mean_token_accuracy": 0.2190199926495552,
      "num_tokens": 105607596.0,
      "step": 60890
    },
    {
      "entropy": 5.399894905090332,
      "epoch": 4.737833106399533,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002963973080578123,
      "loss": 4.8338,
      "mean_token_accuracy": 0.22014791369438172,
      "num_tokens": 105615667.0,
      "step": 60895
    },
    {
      "entropy": 5.276633024215698,
      "epoch": 4.738222135771251,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029636971792033873,
      "loss": 4.6731,
      "mean_token_accuracy": 0.23273625522851943,
      "num_tokens": 105624231.0,
      "step": 60900
    },
    {
      "entropy": 5.3891528129577635,
      "epoch": 4.738611165142968,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002963421274586138,
      "loss": 4.7212,
      "mean_token_accuracy": 0.22100729793310164,
      "num_tokens": 105633415.0,
      "step": 60905
    },
    {
      "entropy": 5.4625410556793215,
      "epoch": 4.739000194514686,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029631453667305603,
      "loss": 4.8345,
      "mean_token_accuracy": 0.21750265955924988,
      "num_tokens": 105642203.0,
      "step": 60910
    },
    {
      "entropy": 5.499260425567627,
      "epoch": 4.739389223886404,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002962869455640841,
      "loss": 4.8637,
      "mean_token_accuracy": 0.2206459864974022,
      "num_tokens": 105650816.0,
      "step": 60915
    },
    {
      "entropy": 5.362947750091553,
      "epoch": 4.739778253258121,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002962593541321167,
      "loss": 4.7564,
      "mean_token_accuracy": 0.22616415470838547,
      "num_tokens": 105659770.0,
      "step": 60920
    },
    {
      "entropy": 5.442066192626953,
      "epoch": 4.740167282629838,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002962317623775723,
      "loss": 4.8324,
      "mean_token_accuracy": 0.21902460008859634,
      "num_tokens": 105668251.0,
      "step": 60925
    },
    {
      "entropy": 5.394265270233154,
      "epoch": 4.740556312001556,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029620417030086976,
      "loss": 4.7643,
      "mean_token_accuracy": 0.22255380749702453,
      "num_tokens": 105676945.0,
      "step": 60930
    },
    {
      "entropy": 5.491617107391358,
      "epoch": 4.740945341373274,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029617657790242767,
      "loss": 4.8873,
      "mean_token_accuracy": 0.21985943466424943,
      "num_tokens": 105686129.0,
      "step": 60935
    },
    {
      "entropy": 5.483660364151001,
      "epoch": 4.741334370744991,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029614898518266476,
      "loss": 4.8512,
      "mean_token_accuracy": 0.2255292519927025,
      "num_tokens": 105694837.0,
      "step": 60940
    },
    {
      "entropy": 5.393662357330323,
      "epoch": 4.741723400116709,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002961213921419997,
      "loss": 4.7576,
      "mean_token_accuracy": 0.2267066165804863,
      "num_tokens": 105702794.0,
      "step": 60945
    },
    {
      "entropy": 5.314022541046143,
      "epoch": 4.742112429488427,
      "grad_norm": 1.21875,
      "learning_rate": 0.000296093798780851,
      "loss": 4.6537,
      "mean_token_accuracy": 0.22447141706943513,
      "num_tokens": 105711049.0,
      "step": 60950
    },
    {
      "entropy": 5.385110902786255,
      "epoch": 4.7425014588601435,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002960662050996377,
      "loss": 4.7283,
      "mean_token_accuracy": 0.22648329436779022,
      "num_tokens": 105720071.0,
      "step": 60955
    },
    {
      "entropy": 5.34975962638855,
      "epoch": 4.742890488231861,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029603861109877807,
      "loss": 4.8014,
      "mean_token_accuracy": 0.22421453148126602,
      "num_tokens": 105728863.0,
      "step": 60960
    },
    {
      "entropy": 5.469671058654785,
      "epoch": 4.743279517603579,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029601101677869106,
      "loss": 4.8465,
      "mean_token_accuracy": 0.21905567049980162,
      "num_tokens": 105736727.0,
      "step": 60965
    },
    {
      "entropy": 5.441883325576782,
      "epoch": 4.743668546975297,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002959834221397952,
      "loss": 4.8332,
      "mean_token_accuracy": 0.219450481235981,
      "num_tokens": 105745111.0,
      "step": 60970
    },
    {
      "entropy": 5.32754373550415,
      "epoch": 4.744057576347014,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029595582718250954,
      "loss": 4.6389,
      "mean_token_accuracy": 0.2277310609817505,
      "num_tokens": 105753319.0,
      "step": 60975
    },
    {
      "entropy": 5.278686237335205,
      "epoch": 4.744446605718732,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002959282319072524,
      "loss": 4.6651,
      "mean_token_accuracy": 0.23588953018188477,
      "num_tokens": 105761973.0,
      "step": 60980
    },
    {
      "entropy": 5.37761435508728,
      "epoch": 4.74483563509045,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002959006363144427,
      "loss": 4.7499,
      "mean_token_accuracy": 0.22775333225727082,
      "num_tokens": 105770752.0,
      "step": 60985
    },
    {
      "entropy": 5.401414585113526,
      "epoch": 4.7452246644621665,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002958730404044991,
      "loss": 4.6861,
      "mean_token_accuracy": 0.23078614920377732,
      "num_tokens": 105779148.0,
      "step": 60990
    },
    {
      "entropy": 5.3770476341247555,
      "epoch": 4.745613693833884,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002958454441778403,
      "loss": 4.7756,
      "mean_token_accuracy": 0.2263579860329628,
      "num_tokens": 105787794.0,
      "step": 60995
    },
    {
      "entropy": 5.378298759460449,
      "epoch": 4.746002723205602,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029581784763488506,
      "loss": 4.8048,
      "mean_token_accuracy": 0.2203591674566269,
      "num_tokens": 105796827.0,
      "step": 61000
    },
    {
      "entropy": 5.479292201995849,
      "epoch": 4.74639175257732,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002957902507760522,
      "loss": 4.8009,
      "mean_token_accuracy": 0.2199700027704239,
      "num_tokens": 105805638.0,
      "step": 61005
    },
    {
      "entropy": 5.460744190216064,
      "epoch": 4.746780781949037,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002957626536017603,
      "loss": 4.7698,
      "mean_token_accuracy": 0.22634891271591187,
      "num_tokens": 105814374.0,
      "step": 61010
    },
    {
      "entropy": 5.35067458152771,
      "epoch": 4.747169811320755,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002957350561124282,
      "loss": 4.691,
      "mean_token_accuracy": 0.23094465136528014,
      "num_tokens": 105822938.0,
      "step": 61015
    },
    {
      "entropy": 5.442540836334229,
      "epoch": 4.747558840692472,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002957074583084747,
      "loss": 4.9308,
      "mean_token_accuracy": 0.2212826132774353,
      "num_tokens": 105831770.0,
      "step": 61020
    },
    {
      "entropy": 5.404641199111938,
      "epoch": 4.7479478700641895,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002956798601903184,
      "loss": 4.7397,
      "mean_token_accuracy": 0.23217836618423462,
      "num_tokens": 105840517.0,
      "step": 61025
    },
    {
      "entropy": 5.432840967178345,
      "epoch": 4.748336899435907,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029565226175837814,
      "loss": 4.6944,
      "mean_token_accuracy": 0.23149009943008422,
      "num_tokens": 105848299.0,
      "step": 61030
    },
    {
      "entropy": 5.339942216873169,
      "epoch": 4.748725928807625,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002956246630130727,
      "loss": 4.694,
      "mean_token_accuracy": 0.22761086672544478,
      "num_tokens": 105856430.0,
      "step": 61035
    },
    {
      "entropy": 5.325467538833618,
      "epoch": 4.749114958179343,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029559706395482075,
      "loss": 4.8155,
      "mean_token_accuracy": 0.22007054090499878,
      "num_tokens": 105865578.0,
      "step": 61040
    },
    {
      "entropy": 5.388486671447754,
      "epoch": 4.74950398755106,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029556946458404124,
      "loss": 4.8227,
      "mean_token_accuracy": 0.21596349030733109,
      "num_tokens": 105874380.0,
      "step": 61045
    },
    {
      "entropy": 5.479332637786865,
      "epoch": 4.749893016922778,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029554186490115284,
      "loss": 4.7569,
      "mean_token_accuracy": 0.22599659115076065,
      "num_tokens": 105882822.0,
      "step": 61050
    },
    {
      "entropy": 5.404797983169556,
      "epoch": 4.750282046294496,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002955142649065743,
      "loss": 4.7616,
      "mean_token_accuracy": 0.21996737718582154,
      "num_tokens": 105891904.0,
      "step": 61055
    },
    {
      "entropy": 5.385646867752075,
      "epoch": 4.7506710756662125,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002954866646007244,
      "loss": 4.7808,
      "mean_token_accuracy": 0.2255269855260849,
      "num_tokens": 105900103.0,
      "step": 61060
    },
    {
      "entropy": 5.359794664382934,
      "epoch": 4.75106010503793,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029545906398402205,
      "loss": 4.7479,
      "mean_token_accuracy": 0.23442851305007933,
      "num_tokens": 105908868.0,
      "step": 61065
    },
    {
      "entropy": 5.346950435638428,
      "epoch": 4.751449134409648,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029543146305688594,
      "loss": 4.6585,
      "mean_token_accuracy": 0.23449019193649293,
      "num_tokens": 105916846.0,
      "step": 61070
    },
    {
      "entropy": 5.292568397521973,
      "epoch": 4.751838163781366,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002954038618197349,
      "loss": 4.658,
      "mean_token_accuracy": 0.23252671211957932,
      "num_tokens": 105925497.0,
      "step": 61075
    },
    {
      "entropy": 5.440683078765869,
      "epoch": 4.752227193153083,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002953762602729877,
      "loss": 4.8037,
      "mean_token_accuracy": 0.2161116987466812,
      "num_tokens": 105934049.0,
      "step": 61080
    },
    {
      "entropy": 5.311450862884522,
      "epoch": 4.752616222524801,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029534865841706324,
      "loss": 4.6653,
      "mean_token_accuracy": 0.23404623866081237,
      "num_tokens": 105942170.0,
      "step": 61085
    },
    {
      "entropy": 5.355849409103394,
      "epoch": 4.753005251896518,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002953210562523803,
      "loss": 4.6929,
      "mean_token_accuracy": 0.23151392489671707,
      "num_tokens": 105950324.0,
      "step": 61090
    },
    {
      "entropy": 5.435976362228393,
      "epoch": 4.7533942812682355,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002952934537793576,
      "loss": 4.9333,
      "mean_token_accuracy": 0.21799732446670533,
      "num_tokens": 105959675.0,
      "step": 61095
    },
    {
      "entropy": 5.336900234222412,
      "epoch": 4.753783310639953,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029526585099841395,
      "loss": 4.6828,
      "mean_token_accuracy": 0.22844017148017884,
      "num_tokens": 105967856.0,
      "step": 61100
    },
    {
      "entropy": 5.396536064147949,
      "epoch": 4.754172340011671,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029523824790996845,
      "loss": 4.7912,
      "mean_token_accuracy": 0.22316533476114273,
      "num_tokens": 105976069.0,
      "step": 61105
    },
    {
      "entropy": 5.354799032211304,
      "epoch": 4.754561369383389,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029521064451443966,
      "loss": 4.6978,
      "mean_token_accuracy": 0.2298081785440445,
      "num_tokens": 105984249.0,
      "step": 61110
    },
    {
      "entropy": 5.384209728240966,
      "epoch": 4.754950398755106,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029518304081224656,
      "loss": 4.773,
      "mean_token_accuracy": 0.22394746989011766,
      "num_tokens": 105992551.0,
      "step": 61115
    },
    {
      "entropy": 5.4783533096313475,
      "epoch": 4.755339428126824,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029515543680380793,
      "loss": 4.9362,
      "mean_token_accuracy": 0.21320951581001282,
      "num_tokens": 106001687.0,
      "step": 61120
    },
    {
      "entropy": 5.423759126663208,
      "epoch": 4.755728457498541,
      "grad_norm": 1.25,
      "learning_rate": 0.00029512783248954253,
      "loss": 4.7815,
      "mean_token_accuracy": 0.22356802970170975,
      "num_tokens": 106009610.0,
      "step": 61125
    },
    {
      "entropy": 5.425961446762085,
      "epoch": 4.7561174868702585,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002951002278698695,
      "loss": 4.8413,
      "mean_token_accuracy": 0.2166310042142868,
      "num_tokens": 106017801.0,
      "step": 61130
    },
    {
      "entropy": 5.411542129516602,
      "epoch": 4.756506516241976,
      "grad_norm": 1.109375,
      "learning_rate": 0.00029507262294520726,
      "loss": 4.8092,
      "mean_token_accuracy": 0.2213703751564026,
      "num_tokens": 106026805.0,
      "step": 61135
    },
    {
      "entropy": 5.43509202003479,
      "epoch": 4.756895545613694,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002950450177159751,
      "loss": 4.7594,
      "mean_token_accuracy": 0.22402465492486953,
      "num_tokens": 106035685.0,
      "step": 61140
    },
    {
      "entropy": 5.384979295730591,
      "epoch": 4.757284574985412,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029501741218259167,
      "loss": 4.8505,
      "mean_token_accuracy": 0.21784700453281403,
      "num_tokens": 106044556.0,
      "step": 61145
    },
    {
      "entropy": 5.355625057220459,
      "epoch": 4.757673604357129,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029498980634547585,
      "loss": 4.7562,
      "mean_token_accuracy": 0.22328538894653321,
      "num_tokens": 106053706.0,
      "step": 61150
    },
    {
      "entropy": 5.336162424087524,
      "epoch": 4.758062633728846,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029496220020504656,
      "loss": 4.7532,
      "mean_token_accuracy": 0.2334196835756302,
      "num_tokens": 106062498.0,
      "step": 61155
    },
    {
      "entropy": 5.370337057113647,
      "epoch": 4.758451663100564,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029493459376172263,
      "loss": 4.6802,
      "mean_token_accuracy": 0.227053901553154,
      "num_tokens": 106071689.0,
      "step": 61160
    },
    {
      "entropy": 5.408340311050415,
      "epoch": 4.7588406924722815,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002949069870159229,
      "loss": 4.8178,
      "mean_token_accuracy": 0.22431112229824066,
      "num_tokens": 106080331.0,
      "step": 61165
    },
    {
      "entropy": 5.516005516052246,
      "epoch": 4.759229721843999,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029487937996806635,
      "loss": 4.896,
      "mean_token_accuracy": 0.20779773890972136,
      "num_tokens": 106089419.0,
      "step": 61170
    },
    {
      "entropy": 5.418132352828979,
      "epoch": 4.759618751215717,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029485177261857193,
      "loss": 4.7453,
      "mean_token_accuracy": 0.23355381786823273,
      "num_tokens": 106097588.0,
      "step": 61175
    },
    {
      "entropy": 5.371577167510987,
      "epoch": 4.760007780587435,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002948241649678584,
      "loss": 4.7629,
      "mean_token_accuracy": 0.22706762105226516,
      "num_tokens": 106107213.0,
      "step": 61180
    },
    {
      "entropy": 5.319773578643799,
      "epoch": 4.760396809959152,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002947965570163448,
      "loss": 4.6877,
      "mean_token_accuracy": 0.23044611066579818,
      "num_tokens": 106116230.0,
      "step": 61185
    },
    {
      "entropy": 5.427718019485473,
      "epoch": 4.76078583933087,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002947689487644499,
      "loss": 4.7767,
      "mean_token_accuracy": 0.2304047539830208,
      "num_tokens": 106124795.0,
      "step": 61190
    },
    {
      "entropy": 5.407531023025513,
      "epoch": 4.761174868702587,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002947413402125927,
      "loss": 4.8288,
      "mean_token_accuracy": 0.2258199483156204,
      "num_tokens": 106134562.0,
      "step": 61195
    },
    {
      "entropy": 5.4292937278747555,
      "epoch": 4.7615638980743045,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002947137313611919,
      "loss": 4.7817,
      "mean_token_accuracy": 0.22134263217449188,
      "num_tokens": 106142756.0,
      "step": 61200
    },
    {
      "entropy": 5.418309450149536,
      "epoch": 4.761952927446022,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029468612221066683,
      "loss": 4.8063,
      "mean_token_accuracy": 0.2242311105132103,
      "num_tokens": 106151576.0,
      "step": 61205
    },
    {
      "entropy": 5.362037324905396,
      "epoch": 4.76234195681774,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002946585127614361,
      "loss": 4.7431,
      "mean_token_accuracy": 0.22929263412952422,
      "num_tokens": 106160091.0,
      "step": 61210
    },
    {
      "entropy": 5.409893655776978,
      "epoch": 4.762730986189458,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002946309030139188,
      "loss": 4.8294,
      "mean_token_accuracy": 0.21888908445835115,
      "num_tokens": 106168314.0,
      "step": 61215
    },
    {
      "entropy": 5.351134443283081,
      "epoch": 4.763120015561174,
      "grad_norm": 1.25,
      "learning_rate": 0.0002946032929685337,
      "loss": 4.7285,
      "mean_token_accuracy": 0.23142897188663483,
      "num_tokens": 106176659.0,
      "step": 61220
    },
    {
      "entropy": 5.4166546821594235,
      "epoch": 4.763509044932892,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002945756826256999,
      "loss": 4.7906,
      "mean_token_accuracy": 0.22887581586837769,
      "num_tokens": 106185781.0,
      "step": 61225
    },
    {
      "entropy": 5.464412021636963,
      "epoch": 4.76389807430461,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002945480719858363,
      "loss": 4.7718,
      "mean_token_accuracy": 0.22435036301612854,
      "num_tokens": 106193851.0,
      "step": 61230
    },
    {
      "entropy": 5.310726118087769,
      "epoch": 4.7642871036763275,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002945204610493617,
      "loss": 4.7115,
      "mean_token_accuracy": 0.2256808266043663,
      "num_tokens": 106202148.0,
      "step": 61235
    },
    {
      "entropy": 5.300835371017456,
      "epoch": 4.764676133048045,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029449284981669524,
      "loss": 4.7886,
      "mean_token_accuracy": 0.22049281001091003,
      "num_tokens": 106210473.0,
      "step": 61240
    },
    {
      "entropy": 5.394700765609741,
      "epoch": 4.765065162419763,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029446523828825577,
      "loss": 4.8114,
      "mean_token_accuracy": 0.22652681469917296,
      "num_tokens": 106219307.0,
      "step": 61245
    },
    {
      "entropy": 5.401035499572754,
      "epoch": 4.7654541917914806,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002944376264644624,
      "loss": 4.7587,
      "mean_token_accuracy": 0.22903700321912765,
      "num_tokens": 106227618.0,
      "step": 61250
    },
    {
      "entropy": 5.448077869415283,
      "epoch": 4.765843221163198,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002944100143457339,
      "loss": 4.8556,
      "mean_token_accuracy": 0.21209297925233841,
      "num_tokens": 106235681.0,
      "step": 61255
    },
    {
      "entropy": 5.337272310256958,
      "epoch": 4.766232250534915,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002943824019324893,
      "loss": 4.6677,
      "mean_token_accuracy": 0.23288217335939407,
      "num_tokens": 106243786.0,
      "step": 61260
    },
    {
      "entropy": 5.361037826538086,
      "epoch": 4.766621279906633,
      "grad_norm": 1.25,
      "learning_rate": 0.0002943547892251476,
      "loss": 4.7331,
      "mean_token_accuracy": 0.22878089398145676,
      "num_tokens": 106252117.0,
      "step": 61265
    },
    {
      "entropy": 5.352073001861572,
      "epoch": 4.7670103092783505,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029432717622412784,
      "loss": 4.7609,
      "mean_token_accuracy": 0.2292614310979843,
      "num_tokens": 106260473.0,
      "step": 61270
    },
    {
      "entropy": 5.317076683044434,
      "epoch": 4.767399338650068,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029429956292984893,
      "loss": 4.6985,
      "mean_token_accuracy": 0.23215328603982927,
      "num_tokens": 106269713.0,
      "step": 61275
    },
    {
      "entropy": 5.398649787902832,
      "epoch": 4.767788368021786,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002942719493427299,
      "loss": 4.7944,
      "mean_token_accuracy": 0.2272867813706398,
      "num_tokens": 106278909.0,
      "step": 61280
    },
    {
      "entropy": 5.33880500793457,
      "epoch": 4.7681773973935035,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002942443354631897,
      "loss": 4.7311,
      "mean_token_accuracy": 0.2279257595539093,
      "num_tokens": 106287475.0,
      "step": 61285
    },
    {
      "entropy": 5.418565511703491,
      "epoch": 4.76856642676522,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029421672129164726,
      "loss": 4.8105,
      "mean_token_accuracy": 0.21864811778068544,
      "num_tokens": 106295605.0,
      "step": 61290
    },
    {
      "entropy": 5.42115216255188,
      "epoch": 4.768955456136938,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029418910682852176,
      "loss": 4.7803,
      "mean_token_accuracy": 0.22009591162204742,
      "num_tokens": 106303947.0,
      "step": 61295
    },
    {
      "entropy": 5.435758256912232,
      "epoch": 4.769344485508656,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029416149207423203,
      "loss": 4.8161,
      "mean_token_accuracy": 0.21683825701475143,
      "num_tokens": 106312423.0,
      "step": 61300
    },
    {
      "entropy": 5.327768564224243,
      "epoch": 4.7697335148803734,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029413387702919723,
      "loss": 4.6392,
      "mean_token_accuracy": 0.2345084249973297,
      "num_tokens": 106320904.0,
      "step": 61305
    },
    {
      "entropy": 5.335542058944702,
      "epoch": 4.770122544252091,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029410626169383623,
      "loss": 4.7497,
      "mean_token_accuracy": 0.2255782812833786,
      "num_tokens": 106330096.0,
      "step": 61310
    },
    {
      "entropy": 5.3593590259552,
      "epoch": 4.770511573623809,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029407864606856814,
      "loss": 4.8042,
      "mean_token_accuracy": 0.21647580116987228,
      "num_tokens": 106338980.0,
      "step": 61315
    },
    {
      "entropy": 5.422854900360107,
      "epoch": 4.7709006029955265,
      "grad_norm": 1.203125,
      "learning_rate": 0.000294051030153812,
      "loss": 4.799,
      "mean_token_accuracy": 0.21937923282384872,
      "num_tokens": 106347623.0,
      "step": 61320
    },
    {
      "entropy": 5.401127052307129,
      "epoch": 4.771289632367243,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002940234139499868,
      "loss": 4.7894,
      "mean_token_accuracy": 0.2201191484928131,
      "num_tokens": 106356051.0,
      "step": 61325
    },
    {
      "entropy": 5.441878032684326,
      "epoch": 4.771678661738961,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029399579745751156,
      "loss": 4.8022,
      "mean_token_accuracy": 0.22014029771089555,
      "num_tokens": 106364618.0,
      "step": 61330
    },
    {
      "entropy": 5.399563598632812,
      "epoch": 4.772067691110679,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029396818067680525,
      "loss": 4.7475,
      "mean_token_accuracy": 0.23158082813024522,
      "num_tokens": 106373455.0,
      "step": 61335
    },
    {
      "entropy": 5.413110208511353,
      "epoch": 4.772456720482396,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029394056360828707,
      "loss": 4.7769,
      "mean_token_accuracy": 0.22744117975234984,
      "num_tokens": 106383522.0,
      "step": 61340
    },
    {
      "entropy": 5.374820232391357,
      "epoch": 4.772845749854114,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029391294625237595,
      "loss": 4.7358,
      "mean_token_accuracy": 0.22248922884464264,
      "num_tokens": 106392277.0,
      "step": 61345
    },
    {
      "entropy": 5.405362319946289,
      "epoch": 4.773234779225832,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002938853286094911,
      "loss": 4.7773,
      "mean_token_accuracy": 0.2287943869829178,
      "num_tokens": 106401154.0,
      "step": 61350
    },
    {
      "entropy": 5.451045560836792,
      "epoch": 4.773623808597549,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029385771068005123,
      "loss": 4.7502,
      "mean_token_accuracy": 0.2267290860414505,
      "num_tokens": 106409452.0,
      "step": 61355
    },
    {
      "entropy": 5.411964368820191,
      "epoch": 4.774012837969266,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002938300924644757,
      "loss": 4.7711,
      "mean_token_accuracy": 0.21638796478509903,
      "num_tokens": 106417490.0,
      "step": 61360
    },
    {
      "entropy": 5.412882614135742,
      "epoch": 4.774401867340984,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029380247396318355,
      "loss": 4.7552,
      "mean_token_accuracy": 0.2285612240433693,
      "num_tokens": 106425913.0,
      "step": 61365
    },
    {
      "entropy": 5.476139163970947,
      "epoch": 4.774790896712702,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002937748551765937,
      "loss": 4.8423,
      "mean_token_accuracy": 0.22279808819293975,
      "num_tokens": 106434192.0,
      "step": 61370
    },
    {
      "entropy": 5.4088210105896,
      "epoch": 4.775179926084419,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002937472361051254,
      "loss": 4.7989,
      "mean_token_accuracy": 0.22200180739164352,
      "num_tokens": 106442761.0,
      "step": 61375
    },
    {
      "entropy": 5.423699474334716,
      "epoch": 4.775568955456137,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002937196167491976,
      "loss": 4.7804,
      "mean_token_accuracy": 0.22812281399965287,
      "num_tokens": 106451172.0,
      "step": 61380
    },
    {
      "entropy": 5.461946058273315,
      "epoch": 4.775957984827855,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029369199710922944,
      "loss": 4.8118,
      "mean_token_accuracy": 0.21687264740467072,
      "num_tokens": 106459482.0,
      "step": 61385
    },
    {
      "entropy": 5.313634300231934,
      "epoch": 4.7763470141995725,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029366437718563994,
      "loss": 4.6485,
      "mean_token_accuracy": 0.23619213551282883,
      "num_tokens": 106467959.0,
      "step": 61390
    },
    {
      "entropy": 5.384954643249512,
      "epoch": 4.776736043571289,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002936367569788482,
      "loss": 4.7418,
      "mean_token_accuracy": 0.22847955226898192,
      "num_tokens": 106475833.0,
      "step": 61395
    },
    {
      "entropy": 5.41038932800293,
      "epoch": 4.777125072943007,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002936091364892733,
      "loss": 4.8411,
      "mean_token_accuracy": 0.2217931792140007,
      "num_tokens": 106485045.0,
      "step": 61400
    },
    {
      "entropy": 5.383838033676147,
      "epoch": 4.777514102314725,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029358151571733446,
      "loss": 4.7673,
      "mean_token_accuracy": 0.22453367859125137,
      "num_tokens": 106494074.0,
      "step": 61405
    },
    {
      "entropy": 5.378242444992066,
      "epoch": 4.777903131686442,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029355389466345064,
      "loss": 4.7822,
      "mean_token_accuracy": 0.22574361860752107,
      "num_tokens": 106502706.0,
      "step": 61410
    },
    {
      "entropy": 5.36213173866272,
      "epoch": 4.77829216105816,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029352627332804104,
      "loss": 4.7506,
      "mean_token_accuracy": 0.22445164024829864,
      "num_tokens": 106511263.0,
      "step": 61415
    },
    {
      "entropy": 5.442114019393921,
      "epoch": 4.778681190429878,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029349865171152475,
      "loss": 4.804,
      "mean_token_accuracy": 0.23096578270196916,
      "num_tokens": 106520324.0,
      "step": 61420
    },
    {
      "entropy": 5.343964862823486,
      "epoch": 4.779070219801595,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029347102981432084,
      "loss": 4.6709,
      "mean_token_accuracy": 0.23022420853376388,
      "num_tokens": 106529175.0,
      "step": 61425
    },
    {
      "entropy": 5.336801195144654,
      "epoch": 4.779459249173312,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002934434076368484,
      "loss": 4.7508,
      "mean_token_accuracy": 0.22757611572742462,
      "num_tokens": 106538107.0,
      "step": 61430
    },
    {
      "entropy": 5.4528985023498535,
      "epoch": 4.77984827854503,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002934157851795266,
      "loss": 4.9114,
      "mean_token_accuracy": 0.2119425430893898,
      "num_tokens": 106546050.0,
      "step": 61435
    },
    {
      "entropy": 5.3592689514160154,
      "epoch": 4.780237307916748,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002933881624427746,
      "loss": 4.6905,
      "mean_token_accuracy": 0.22445545941591263,
      "num_tokens": 106554266.0,
      "step": 61440
    },
    {
      "entropy": 5.375132751464844,
      "epoch": 4.780626337288465,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002933605394270116,
      "loss": 4.7398,
      "mean_token_accuracy": 0.23191299736499787,
      "num_tokens": 106562897.0,
      "step": 61445
    },
    {
      "entropy": 5.291368913650513,
      "epoch": 4.781015366660183,
      "grad_norm": 1.125,
      "learning_rate": 0.0002933329161326565,
      "loss": 4.668,
      "mean_token_accuracy": 0.23306576311588287,
      "num_tokens": 106571815.0,
      "step": 61450
    },
    {
      "entropy": 5.350451803207397,
      "epoch": 4.781404396031901,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029330529256012863,
      "loss": 4.756,
      "mean_token_accuracy": 0.2275736689567566,
      "num_tokens": 106580074.0,
      "step": 61455
    },
    {
      "entropy": 5.442710065841675,
      "epoch": 4.781793425403618,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00029327766870984714,
      "loss": 4.8172,
      "mean_token_accuracy": 0.22009642869234086,
      "num_tokens": 106589385.0,
      "step": 61460
    },
    {
      "entropy": 5.423815155029297,
      "epoch": 4.782182454775335,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029325004458223097,
      "loss": 4.7553,
      "mean_token_accuracy": 0.22562260776758195,
      "num_tokens": 106598779.0,
      "step": 61465
    },
    {
      "entropy": 5.460942888259888,
      "epoch": 4.782571484147053,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029322242017769954,
      "loss": 4.8958,
      "mean_token_accuracy": 0.21253805011510848,
      "num_tokens": 106607412.0,
      "step": 61470
    },
    {
      "entropy": 5.350466012954712,
      "epoch": 4.782960513518771,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029319479549667183,
      "loss": 4.6233,
      "mean_token_accuracy": 0.23295226842164993,
      "num_tokens": 106615703.0,
      "step": 61475
    },
    {
      "entropy": 5.402147102355957,
      "epoch": 4.783349542890488,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002931671705395671,
      "loss": 4.7646,
      "mean_token_accuracy": 0.2222425401210785,
      "num_tokens": 106624541.0,
      "step": 61480
    },
    {
      "entropy": 5.37731704711914,
      "epoch": 4.783738572262206,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029313954530680447,
      "loss": 4.807,
      "mean_token_accuracy": 0.22690728306770325,
      "num_tokens": 106633307.0,
      "step": 61485
    },
    {
      "entropy": 5.484022188186645,
      "epoch": 4.784127601633923,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029311191979880313,
      "loss": 4.8812,
      "mean_token_accuracy": 0.21933328360319138,
      "num_tokens": 106642645.0,
      "step": 61490
    },
    {
      "entropy": 5.437813329696655,
      "epoch": 4.784516631005641,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002930842940159822,
      "loss": 4.7301,
      "mean_token_accuracy": 0.22371327877044678,
      "num_tokens": 106651860.0,
      "step": 61495
    },
    {
      "entropy": 5.36376166343689,
      "epoch": 4.784905660377358,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029305666795876093,
      "loss": 4.7141,
      "mean_token_accuracy": 0.22501543313264846,
      "num_tokens": 106660704.0,
      "step": 61500
    },
    {
      "entropy": 5.394883251190185,
      "epoch": 4.785294689749076,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029302904162755843,
      "loss": 4.7439,
      "mean_token_accuracy": 0.22623571306467055,
      "num_tokens": 106669190.0,
      "step": 61505
    },
    {
      "entropy": 5.394026041030884,
      "epoch": 4.785683719120794,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029300141502279395,
      "loss": 4.7466,
      "mean_token_accuracy": 0.221255424618721,
      "num_tokens": 106677018.0,
      "step": 61510
    },
    {
      "entropy": 5.351914548873902,
      "epoch": 4.786072748492511,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029297378814488657,
      "loss": 4.7035,
      "mean_token_accuracy": 0.2184475004673004,
      "num_tokens": 106685971.0,
      "step": 61515
    },
    {
      "entropy": 5.362134885787964,
      "epoch": 4.786461777864229,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00029294616099425556,
      "loss": 4.7014,
      "mean_token_accuracy": 0.22833310663700104,
      "num_tokens": 106694478.0,
      "step": 61520
    },
    {
      "entropy": 5.367424392700196,
      "epoch": 4.786850807235947,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002929185335713202,
      "loss": 4.7906,
      "mean_token_accuracy": 0.22128487527370452,
      "num_tokens": 106702527.0,
      "step": 61525
    },
    {
      "entropy": 5.394149923324585,
      "epoch": 4.787239836607664,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029289090587649965,
      "loss": 4.7777,
      "mean_token_accuracy": 0.22110269069671631,
      "num_tokens": 106710793.0,
      "step": 61530
    },
    {
      "entropy": 5.393852567672729,
      "epoch": 4.787628865979381,
      "grad_norm": 1.125,
      "learning_rate": 0.000292863277910213,
      "loss": 4.8084,
      "mean_token_accuracy": 0.22059060633182526,
      "num_tokens": 106718992.0,
      "step": 61535
    },
    {
      "entropy": 5.381119203567505,
      "epoch": 4.788017895351099,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002928356496728796,
      "loss": 4.8085,
      "mean_token_accuracy": 0.22847482413053513,
      "num_tokens": 106727921.0,
      "step": 61540
    },
    {
      "entropy": 5.404296636581421,
      "epoch": 4.788406924722817,
      "grad_norm": 1.125,
      "learning_rate": 0.00029280802116491854,
      "loss": 4.7632,
      "mean_token_accuracy": 0.21876602470874787,
      "num_tokens": 106736264.0,
      "step": 61545
    },
    {
      "entropy": 5.3030198097229,
      "epoch": 4.788795954094534,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029278039238674915,
      "loss": 4.7017,
      "mean_token_accuracy": 0.2334312990307808,
      "num_tokens": 106744531.0,
      "step": 61550
    },
    {
      "entropy": 5.368652629852295,
      "epoch": 4.789184983466251,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002927527633387906,
      "loss": 4.7854,
      "mean_token_accuracy": 0.2235810711979866,
      "num_tokens": 106752780.0,
      "step": 61555
    },
    {
      "entropy": 5.418041229248047,
      "epoch": 4.789574012837969,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029272513402146203,
      "loss": 4.7831,
      "mean_token_accuracy": 0.2270827278494835,
      "num_tokens": 106762276.0,
      "step": 61560
    },
    {
      "entropy": 5.393799734115601,
      "epoch": 4.789963042209687,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002926975044351828,
      "loss": 4.7752,
      "mean_token_accuracy": 0.22876435667276382,
      "num_tokens": 106771200.0,
      "step": 61565
    },
    {
      "entropy": 5.338265037536621,
      "epoch": 4.790352071581404,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002926698745803723,
      "loss": 4.706,
      "mean_token_accuracy": 0.22634213119745256,
      "num_tokens": 106779871.0,
      "step": 61570
    },
    {
      "entropy": 5.492870140075683,
      "epoch": 4.790741100953122,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029264224445744937,
      "loss": 4.9244,
      "mean_token_accuracy": 0.2143673062324524,
      "num_tokens": 106788980.0,
      "step": 61575
    },
    {
      "entropy": 5.502860403060913,
      "epoch": 4.79113013032484,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029261461406683353,
      "loss": 4.8785,
      "mean_token_accuracy": 0.21506026834249498,
      "num_tokens": 106798263.0,
      "step": 61580
    },
    {
      "entropy": 5.466037225723267,
      "epoch": 4.791519159696557,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029258698340894397,
      "loss": 4.8692,
      "mean_token_accuracy": 0.21764410138130189,
      "num_tokens": 106807593.0,
      "step": 61585
    },
    {
      "entropy": 5.472850513458252,
      "epoch": 4.791908189068275,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029255935248419983,
      "loss": 4.7926,
      "mean_token_accuracy": 0.2255539372563362,
      "num_tokens": 106815982.0,
      "step": 61590
    },
    {
      "entropy": 5.426155471801758,
      "epoch": 4.792297218439992,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002925317212930206,
      "loss": 4.7575,
      "mean_token_accuracy": 0.2253747284412384,
      "num_tokens": 106824264.0,
      "step": 61595
    },
    {
      "entropy": 5.363981103897094,
      "epoch": 4.79268624781171,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029250408983582527,
      "loss": 4.797,
      "mean_token_accuracy": 0.2229149594902992,
      "num_tokens": 106833572.0,
      "step": 61600
    },
    {
      "entropy": 5.377031326293945,
      "epoch": 4.793075277183427,
      "grad_norm": 1.25,
      "learning_rate": 0.0002924764581130333,
      "loss": 4.8131,
      "mean_token_accuracy": 0.21402206867933274,
      "num_tokens": 106841841.0,
      "step": 61605
    },
    {
      "entropy": 5.371603059768677,
      "epoch": 4.793464306555145,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000292448826125064,
      "loss": 4.776,
      "mean_token_accuracy": 0.22759679555892945,
      "num_tokens": 106850543.0,
      "step": 61610
    },
    {
      "entropy": 5.437587118148803,
      "epoch": 4.793853335926863,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002924211938723364,
      "loss": 4.7754,
      "mean_token_accuracy": 0.2211402729153633,
      "num_tokens": 106859078.0,
      "step": 61615
    },
    {
      "entropy": 5.391442966461182,
      "epoch": 4.79424236529858,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029239356135526994,
      "loss": 4.6844,
      "mean_token_accuracy": 0.23005143404006959,
      "num_tokens": 106867158.0,
      "step": 61620
    },
    {
      "entropy": 5.45987720489502,
      "epoch": 4.794631394670297,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000292365928574284,
      "loss": 4.7859,
      "mean_token_accuracy": 0.2250692993402481,
      "num_tokens": 106875072.0,
      "step": 61625
    },
    {
      "entropy": 5.40940227508545,
      "epoch": 4.795020424042015,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002923382955297975,
      "loss": 4.8063,
      "mean_token_accuracy": 0.22032802551984787,
      "num_tokens": 106883243.0,
      "step": 61630
    },
    {
      "entropy": 5.376550149917603,
      "epoch": 4.795409453413733,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002923106622222301,
      "loss": 4.7508,
      "mean_token_accuracy": 0.22749942392110825,
      "num_tokens": 106891572.0,
      "step": 61635
    },
    {
      "entropy": 5.373151540756226,
      "epoch": 4.79579848278545,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002922830286520009,
      "loss": 4.7769,
      "mean_token_accuracy": 0.22584119588136672,
      "num_tokens": 106899857.0,
      "step": 61640
    },
    {
      "entropy": 5.361700344085693,
      "epoch": 4.796187512157168,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00029225539481952934,
      "loss": 4.7327,
      "mean_token_accuracy": 0.22513916045427323,
      "num_tokens": 106909774.0,
      "step": 61645
    },
    {
      "entropy": 5.407968473434448,
      "epoch": 4.796576541528886,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002922277607252346,
      "loss": 4.7579,
      "mean_token_accuracy": 0.22356380969285966,
      "num_tokens": 106917952.0,
      "step": 61650
    },
    {
      "entropy": 5.424052000045776,
      "epoch": 4.796965570900603,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000292200126369536,
      "loss": 4.8569,
      "mean_token_accuracy": 0.2200043573975563,
      "num_tokens": 106926652.0,
      "step": 61655
    },
    {
      "entropy": 5.434649085998535,
      "epoch": 4.79735460027232,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029217249175285284,
      "loss": 4.8384,
      "mean_token_accuracy": 0.22024600058794022,
      "num_tokens": 106935134.0,
      "step": 61660
    },
    {
      "entropy": 5.38742241859436,
      "epoch": 4.797743629644038,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029214485687560437,
      "loss": 4.7964,
      "mean_token_accuracy": 0.22090229392051697,
      "num_tokens": 106943948.0,
      "step": 61665
    },
    {
      "entropy": 5.440089511871338,
      "epoch": 4.798132659015756,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002921172217382101,
      "loss": 4.8984,
      "mean_token_accuracy": 0.2167525440454483,
      "num_tokens": 106953226.0,
      "step": 61670
    },
    {
      "entropy": 5.4425108432769775,
      "epoch": 4.798521688387473,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029208958634108917,
      "loss": 4.8043,
      "mean_token_accuracy": 0.21372335106134416,
      "num_tokens": 106962153.0,
      "step": 61675
    },
    {
      "entropy": 5.402464103698731,
      "epoch": 4.798910717759191,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000292061950684661,
      "loss": 4.7613,
      "mean_token_accuracy": 0.2225642666220665,
      "num_tokens": 106970660.0,
      "step": 61680
    },
    {
      "entropy": 5.391481018066406,
      "epoch": 4.799299747130909,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029203431476934496,
      "loss": 4.7384,
      "mean_token_accuracy": 0.23120226562023163,
      "num_tokens": 106979676.0,
      "step": 61685
    },
    {
      "entropy": 5.428488731384277,
      "epoch": 4.7996887765026255,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029200667859556026,
      "loss": 4.8072,
      "mean_token_accuracy": 0.22111489474773408,
      "num_tokens": 106988819.0,
      "step": 61690
    },
    {
      "entropy": 5.375599384307861,
      "epoch": 4.800077805874343,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002919790421637262,
      "loss": 4.6894,
      "mean_token_accuracy": 0.22531514912843703,
      "num_tokens": 106998075.0,
      "step": 61695
    },
    {
      "entropy": 5.413203287124634,
      "epoch": 4.800466835246061,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002919514054742622,
      "loss": 4.7234,
      "mean_token_accuracy": 0.23364122956991196,
      "num_tokens": 107006598.0,
      "step": 61700
    },
    {
      "entropy": 5.408813953399658,
      "epoch": 4.800855864617779,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029192376852758774,
      "loss": 4.7747,
      "mean_token_accuracy": 0.22132117301225662,
      "num_tokens": 107015439.0,
      "step": 61705
    },
    {
      "entropy": 5.4127601146697994,
      "epoch": 4.801244893989496,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002918961313241219,
      "loss": 4.8091,
      "mean_token_accuracy": 0.2270084574818611,
      "num_tokens": 107023487.0,
      "step": 61710
    },
    {
      "entropy": 5.362775754928589,
      "epoch": 4.801633923361214,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002918684938642842,
      "loss": 4.7266,
      "mean_token_accuracy": 0.2315456598997116,
      "num_tokens": 107031905.0,
      "step": 61715
    },
    {
      "entropy": 5.352206230163574,
      "epoch": 4.802022952732932,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029184085614849403,
      "loss": 4.7331,
      "mean_token_accuracy": 0.23020475208759308,
      "num_tokens": 107040667.0,
      "step": 61720
    },
    {
      "entropy": 5.496997594833374,
      "epoch": 4.802411982104649,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002918132181771706,
      "loss": 4.9237,
      "mean_token_accuracy": 0.20938269942998886,
      "num_tokens": 107049032.0,
      "step": 61725
    },
    {
      "entropy": 5.415164756774902,
      "epoch": 4.802801011476366,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029178557995073335,
      "loss": 4.7844,
      "mean_token_accuracy": 0.2253414899110794,
      "num_tokens": 107057729.0,
      "step": 61730
    },
    {
      "entropy": 5.346363925933838,
      "epoch": 4.803190040848084,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002917579414696017,
      "loss": 4.67,
      "mean_token_accuracy": 0.23532794564962387,
      "num_tokens": 107066113.0,
      "step": 61735
    },
    {
      "entropy": 5.349463558197021,
      "epoch": 4.803579070219802,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029173030273419484,
      "loss": 4.6961,
      "mean_token_accuracy": 0.22976571172475815,
      "num_tokens": 107074873.0,
      "step": 61740
    },
    {
      "entropy": 5.263867664337158,
      "epoch": 4.803968099591519,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002917026637449323,
      "loss": 4.6254,
      "mean_token_accuracy": 0.23173534870147705,
      "num_tokens": 107083485.0,
      "step": 61745
    },
    {
      "entropy": 5.445455074310303,
      "epoch": 4.804357128963237,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002916750245022335,
      "loss": 4.8343,
      "mean_token_accuracy": 0.21726056039333344,
      "num_tokens": 107092064.0,
      "step": 61750
    },
    {
      "entropy": 5.451920747756958,
      "epoch": 4.804746158334955,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002916473850065177,
      "loss": 4.6678,
      "mean_token_accuracy": 0.23108033686876298,
      "num_tokens": 107100744.0,
      "step": 61755
    },
    {
      "entropy": 5.359065294265747,
      "epoch": 4.8051351877066715,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029161974525820434,
      "loss": 4.6868,
      "mean_token_accuracy": 0.2253183200955391,
      "num_tokens": 107109185.0,
      "step": 61760
    },
    {
      "entropy": 5.355554819107056,
      "epoch": 4.805524217078389,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002915921052577127,
      "loss": 4.7423,
      "mean_token_accuracy": 0.22153462916612626,
      "num_tokens": 107118628.0,
      "step": 61765
    },
    {
      "entropy": 5.363520765304566,
      "epoch": 4.805913246450107,
      "grad_norm": 1.25,
      "learning_rate": 0.00029156446500546243,
      "loss": 4.7569,
      "mean_token_accuracy": 0.22124791145324707,
      "num_tokens": 107128017.0,
      "step": 61770
    },
    {
      "entropy": 5.435674142837525,
      "epoch": 4.806302275821825,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029153682450187265,
      "loss": 4.7462,
      "mean_token_accuracy": 0.2220491275191307,
      "num_tokens": 107136393.0,
      "step": 61775
    },
    {
      "entropy": 5.412199878692627,
      "epoch": 4.806691305193542,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029150918374736294,
      "loss": 4.755,
      "mean_token_accuracy": 0.22044182121753692,
      "num_tokens": 107144936.0,
      "step": 61780
    },
    {
      "entropy": 5.330133628845215,
      "epoch": 4.80708033456526,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002914815427423526,
      "loss": 4.7177,
      "mean_token_accuracy": 0.23071602880954742,
      "num_tokens": 107153579.0,
      "step": 61785
    },
    {
      "entropy": 5.412717342376709,
      "epoch": 4.807469363936978,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002914539014872612,
      "loss": 4.811,
      "mean_token_accuracy": 0.22447938919067384,
      "num_tokens": 107162213.0,
      "step": 61790
    },
    {
      "entropy": 5.4631859302520756,
      "epoch": 4.8078583933086945,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029142625998250786,
      "loss": 4.9048,
      "mean_token_accuracy": 0.22075000852346421,
      "num_tokens": 107170493.0,
      "step": 61795
    },
    {
      "entropy": 5.41983733177185,
      "epoch": 4.808247422680412,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029139861822851233,
      "loss": 4.7613,
      "mean_token_accuracy": 0.22606029808521272,
      "num_tokens": 107179309.0,
      "step": 61800
    },
    {
      "entropy": 5.425087881088257,
      "epoch": 4.80863645205213,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029137097622569376,
      "loss": 4.755,
      "mean_token_accuracy": 0.22315050661563873,
      "num_tokens": 107187852.0,
      "step": 61805
    },
    {
      "entropy": 5.445657873153687,
      "epoch": 4.809025481423848,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002913433339744718,
      "loss": 4.8563,
      "mean_token_accuracy": 0.21579809486865997,
      "num_tokens": 107196198.0,
      "step": 61810
    },
    {
      "entropy": 5.493415832519531,
      "epoch": 4.809414510795565,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002913156914752656,
      "loss": 4.8607,
      "mean_token_accuracy": 0.21620033979415892,
      "num_tokens": 107205133.0,
      "step": 61815
    },
    {
      "entropy": 5.384306001663208,
      "epoch": 4.809803540167283,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002912880487284948,
      "loss": 4.7986,
      "mean_token_accuracy": 0.22225674986839294,
      "num_tokens": 107213962.0,
      "step": 61820
    },
    {
      "entropy": 5.445097303390503,
      "epoch": 4.810192569539,
      "grad_norm": 1.25,
      "learning_rate": 0.00029126040573457886,
      "loss": 4.7722,
      "mean_token_accuracy": 0.2181605726480484,
      "num_tokens": 107222692.0,
      "step": 61825
    },
    {
      "entropy": 5.3659978866577145,
      "epoch": 4.8105815989107175,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029123276249393707,
      "loss": 4.6539,
      "mean_token_accuracy": 0.23161963522434234,
      "num_tokens": 107231474.0,
      "step": 61830
    },
    {
      "entropy": 5.363829326629639,
      "epoch": 4.810970628282435,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029120511900698894,
      "loss": 4.781,
      "mean_token_accuracy": 0.2266690820455551,
      "num_tokens": 107239436.0,
      "step": 61835
    },
    {
      "entropy": 5.322675323486328,
      "epoch": 4.811359657654153,
      "grad_norm": 1.25,
      "learning_rate": 0.000291177475274154,
      "loss": 4.7081,
      "mean_token_accuracy": 0.23004014194011688,
      "num_tokens": 107247832.0,
      "step": 61840
    },
    {
      "entropy": 5.356645011901856,
      "epoch": 4.8117486870258706,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002911498312958516,
      "loss": 4.7586,
      "mean_token_accuracy": 0.22210778892040253,
      "num_tokens": 107256995.0,
      "step": 61845
    },
    {
      "entropy": 5.4442463397979735,
      "epoch": 4.812137716397588,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029112218707250115,
      "loss": 4.7302,
      "mean_token_accuracy": 0.23682008534669877,
      "num_tokens": 107265412.0,
      "step": 61850
    },
    {
      "entropy": 5.372488212585449,
      "epoch": 4.812526745769306,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029109454260452223,
      "loss": 4.7086,
      "mean_token_accuracy": 0.2288095012307167,
      "num_tokens": 107273882.0,
      "step": 61855
    },
    {
      "entropy": 5.400299882888794,
      "epoch": 4.812915775141024,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029106689789233425,
      "loss": 4.8299,
      "mean_token_accuracy": 0.21648874282836914,
      "num_tokens": 107283074.0,
      "step": 61860
    },
    {
      "entropy": 5.2955059051513675,
      "epoch": 4.8133048045127405,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002910392529363566,
      "loss": 4.6378,
      "mean_token_accuracy": 0.23276097923517228,
      "num_tokens": 107292072.0,
      "step": 61865
    },
    {
      "entropy": 5.454576826095581,
      "epoch": 4.813693833884458,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029101160773700887,
      "loss": 4.8266,
      "mean_token_accuracy": 0.22787625193595887,
      "num_tokens": 107300275.0,
      "step": 61870
    },
    {
      "entropy": 5.499374198913574,
      "epoch": 4.814082863256176,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002909839622947104,
      "loss": 4.8507,
      "mean_token_accuracy": 0.21894708275794983,
      "num_tokens": 107309157.0,
      "step": 61875
    },
    {
      "entropy": 5.451032400131226,
      "epoch": 4.8144718926278935,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002909563166098809,
      "loss": 4.8099,
      "mean_token_accuracy": 0.22242779433727264,
      "num_tokens": 107318408.0,
      "step": 61880
    },
    {
      "entropy": 5.44443826675415,
      "epoch": 4.814860921999611,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029092867068293966,
      "loss": 4.8147,
      "mean_token_accuracy": 0.22431010156869888,
      "num_tokens": 107327273.0,
      "step": 61885
    },
    {
      "entropy": 5.441891241073608,
      "epoch": 4.815249951371328,
      "grad_norm": 1.25,
      "learning_rate": 0.0002909010245143061,
      "loss": 4.8076,
      "mean_token_accuracy": 0.22640783488750457,
      "num_tokens": 107336047.0,
      "step": 61890
    },
    {
      "entropy": 5.380611801147461,
      "epoch": 4.815638980743046,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002908733781043998,
      "loss": 4.7614,
      "mean_token_accuracy": 0.22540556341409684,
      "num_tokens": 107344481.0,
      "step": 61895
    },
    {
      "entropy": 5.361806154251099,
      "epoch": 4.8160280101147634,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002908457314536403,
      "loss": 4.7118,
      "mean_token_accuracy": 0.2339036777615547,
      "num_tokens": 107353502.0,
      "step": 61900
    },
    {
      "entropy": 5.372123050689697,
      "epoch": 4.816417039486481,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029081808456244703,
      "loss": 4.7828,
      "mean_token_accuracy": 0.22410992681980133,
      "num_tokens": 107362067.0,
      "step": 61905
    },
    {
      "entropy": 5.404747676849365,
      "epoch": 4.816806068858199,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029079043743123953,
      "loss": 4.8182,
      "mean_token_accuracy": 0.22547973543405533,
      "num_tokens": 107371107.0,
      "step": 61910
    },
    {
      "entropy": 5.407158851623535,
      "epoch": 4.8171950982299165,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029076279006043723,
      "loss": 4.7563,
      "mean_token_accuracy": 0.22345586270093917,
      "num_tokens": 107379707.0,
      "step": 61915
    },
    {
      "entropy": 5.495927381515503,
      "epoch": 4.817584127601634,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002907351424504597,
      "loss": 4.8522,
      "mean_token_accuracy": 0.22566048800945282,
      "num_tokens": 107389481.0,
      "step": 61920
    },
    {
      "entropy": 5.435864639282227,
      "epoch": 4.817973156973352,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002907074946017265,
      "loss": 4.8059,
      "mean_token_accuracy": 0.22802118957042694,
      "num_tokens": 107397463.0,
      "step": 61925
    },
    {
      "entropy": 5.414382362365723,
      "epoch": 4.818362186345069,
      "grad_norm": 1.171875,
      "learning_rate": 0.000290679846514657,
      "loss": 4.7936,
      "mean_token_accuracy": 0.2212691605091095,
      "num_tokens": 107405563.0,
      "step": 61930
    },
    {
      "entropy": 5.343133974075317,
      "epoch": 4.818751215716786,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029065219818967075,
      "loss": 4.7205,
      "mean_token_accuracy": 0.23624372631311416,
      "num_tokens": 107414641.0,
      "step": 61935
    },
    {
      "entropy": 5.42290358543396,
      "epoch": 4.819140245088504,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029062454962718736,
      "loss": 4.751,
      "mean_token_accuracy": 0.22517256140708924,
      "num_tokens": 107423269.0,
      "step": 61940
    },
    {
      "entropy": 5.453370571136475,
      "epoch": 4.819529274460222,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002905969008276263,
      "loss": 4.7507,
      "mean_token_accuracy": 0.2292963147163391,
      "num_tokens": 107432025.0,
      "step": 61945
    },
    {
      "entropy": 5.360519886016846,
      "epoch": 4.8199183038319395,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002905692517914071,
      "loss": 4.6606,
      "mean_token_accuracy": 0.23068249970674515,
      "num_tokens": 107440454.0,
      "step": 61950
    },
    {
      "entropy": 5.30904598236084,
      "epoch": 4.820307333203657,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029054160251894924,
      "loss": 4.6882,
      "mean_token_accuracy": 0.2366346076130867,
      "num_tokens": 107448574.0,
      "step": 61955
    },
    {
      "entropy": 5.409383487701416,
      "epoch": 4.820696362575374,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002905139530106724,
      "loss": 4.8237,
      "mean_token_accuracy": 0.21846911013126374,
      "num_tokens": 107457102.0,
      "step": 61960
    },
    {
      "entropy": 5.458728694915772,
      "epoch": 4.821085391947092,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029048630326699594,
      "loss": 4.8606,
      "mean_token_accuracy": 0.21428277790546418,
      "num_tokens": 107465290.0,
      "step": 61965
    },
    {
      "entropy": 5.396546506881714,
      "epoch": 4.821474421318809,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029045865328833944,
      "loss": 4.7754,
      "mean_token_accuracy": 0.22858381867408753,
      "num_tokens": 107473724.0,
      "step": 61970
    },
    {
      "entropy": 5.398178291320801,
      "epoch": 4.821863450690527,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029043100307512247,
      "loss": 4.6844,
      "mean_token_accuracy": 0.23349391669034958,
      "num_tokens": 107482065.0,
      "step": 61975
    },
    {
      "entropy": 5.356892919540405,
      "epoch": 4.822252480062245,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029040335262776467,
      "loss": 4.7099,
      "mean_token_accuracy": 0.23352280557155608,
      "num_tokens": 107491630.0,
      "step": 61980
    },
    {
      "entropy": 5.427842283248902,
      "epoch": 4.8226415094339625,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002903757019466855,
      "loss": 4.836,
      "mean_token_accuracy": 0.21833476275205613,
      "num_tokens": 107500119.0,
      "step": 61985
    },
    {
      "entropy": 5.404071187973022,
      "epoch": 4.82303053880568,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029034805103230453,
      "loss": 4.752,
      "mean_token_accuracy": 0.2241002842783928,
      "num_tokens": 107508832.0,
      "step": 61990
    },
    {
      "entropy": 5.448879098892212,
      "epoch": 4.823419568177397,
      "grad_norm": 1.25,
      "learning_rate": 0.00029032039988504126,
      "loss": 4.7762,
      "mean_token_accuracy": 0.22887902706861496,
      "num_tokens": 107517432.0,
      "step": 61995
    },
    {
      "entropy": 5.3534400939941404,
      "epoch": 4.823808597549115,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029029274850531534,
      "loss": 4.6589,
      "mean_token_accuracy": 0.2358362630009651,
      "num_tokens": 107526700.0,
      "step": 62000
    },
    {
      "entropy": 5.362733173370361,
      "epoch": 4.824197626920832,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00029026509689354635,
      "loss": 4.785,
      "mean_token_accuracy": 0.21853710561990738,
      "num_tokens": 107536811.0,
      "step": 62005
    },
    {
      "entropy": 5.297612857818604,
      "epoch": 4.82458665629255,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029023744505015373,
      "loss": 4.6093,
      "mean_token_accuracy": 0.2395670160651207,
      "num_tokens": 107545919.0,
      "step": 62010
    },
    {
      "entropy": 5.31793098449707,
      "epoch": 4.824975685664268,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029020979297555717,
      "loss": 4.6837,
      "mean_token_accuracy": 0.23302997499704362,
      "num_tokens": 107554044.0,
      "step": 62015
    },
    {
      "entropy": 5.363417768478394,
      "epoch": 4.8253647150359855,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029018214067017624,
      "loss": 4.7431,
      "mean_token_accuracy": 0.22819169461727143,
      "num_tokens": 107563200.0,
      "step": 62020
    },
    {
      "entropy": 5.353820180892944,
      "epoch": 4.825753744407702,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002901544881344305,
      "loss": 4.7287,
      "mean_token_accuracy": 0.2338038444519043,
      "num_tokens": 107572219.0,
      "step": 62025
    },
    {
      "entropy": 5.47909951210022,
      "epoch": 4.82614277377942,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002901268353687394,
      "loss": 4.8699,
      "mean_token_accuracy": 0.2224046915769577,
      "num_tokens": 107580891.0,
      "step": 62030
    },
    {
      "entropy": 5.417860746383667,
      "epoch": 4.826531803151138,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029009918237352274,
      "loss": 4.8466,
      "mean_token_accuracy": 0.2248125985264778,
      "num_tokens": 107589342.0,
      "step": 62035
    },
    {
      "entropy": 5.459008836746216,
      "epoch": 4.826920832522855,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002900715291492,
      "loss": 4.8764,
      "mean_token_accuracy": 0.22367976903915404,
      "num_tokens": 107598748.0,
      "step": 62040
    },
    {
      "entropy": 5.4241491794586185,
      "epoch": 4.827309861894573,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002900438756961908,
      "loss": 4.7129,
      "mean_token_accuracy": 0.22881414443254472,
      "num_tokens": 107606733.0,
      "step": 62045
    },
    {
      "entropy": 5.419542360305786,
      "epoch": 4.827698891266291,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002900162220149148,
      "loss": 4.7938,
      "mean_token_accuracy": 0.2243785575032234,
      "num_tokens": 107615582.0,
      "step": 62050
    },
    {
      "entropy": 5.349026918411255,
      "epoch": 4.8280879206380085,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028998856810579144,
      "loss": 4.6771,
      "mean_token_accuracy": 0.2338649407029152,
      "num_tokens": 107624245.0,
      "step": 62055
    },
    {
      "entropy": 5.337031126022339,
      "epoch": 4.828476950009726,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002899609139692405,
      "loss": 4.7032,
      "mean_token_accuracy": 0.23185718208551406,
      "num_tokens": 107632398.0,
      "step": 62060
    },
    {
      "entropy": 5.301353454589844,
      "epoch": 4.828865979381443,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028993325960568133,
      "loss": 4.7901,
      "mean_token_accuracy": 0.22813630551099778,
      "num_tokens": 107641674.0,
      "step": 62065
    },
    {
      "entropy": 5.476647567749024,
      "epoch": 4.829255008753161,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002899056050155339,
      "loss": 4.804,
      "mean_token_accuracy": 0.22075988799333573,
      "num_tokens": 107650086.0,
      "step": 62070
    },
    {
      "entropy": 5.4539049625396725,
      "epoch": 4.829644038124878,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002898779501992176,
      "loss": 4.8115,
      "mean_token_accuracy": 0.22158151268959045,
      "num_tokens": 107658954.0,
      "step": 62075
    },
    {
      "entropy": 5.394329833984375,
      "epoch": 4.830033067496596,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000289850295157152,
      "loss": 4.7464,
      "mean_token_accuracy": 0.22531236857175826,
      "num_tokens": 107668130.0,
      "step": 62080
    },
    {
      "entropy": 5.503174734115601,
      "epoch": 4.830422096868314,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028982263988975687,
      "loss": 4.8771,
      "mean_token_accuracy": 0.21520103961229325,
      "num_tokens": 107676657.0,
      "step": 62085
    },
    {
      "entropy": 5.396610307693481,
      "epoch": 4.8308111262400315,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002897949843974518,
      "loss": 4.6795,
      "mean_token_accuracy": 0.2386665537953377,
      "num_tokens": 107685961.0,
      "step": 62090
    },
    {
      "entropy": 5.400670862197876,
      "epoch": 4.831200155611748,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002897673286806562,
      "loss": 4.7509,
      "mean_token_accuracy": 0.2308945521712303,
      "num_tokens": 107694781.0,
      "step": 62095
    },
    {
      "entropy": 5.395420074462891,
      "epoch": 4.831589184983466,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002897396727397901,
      "loss": 4.7726,
      "mean_token_accuracy": 0.22023675143718718,
      "num_tokens": 107703096.0,
      "step": 62100
    },
    {
      "entropy": 5.352065849304199,
      "epoch": 4.831978214355184,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002897120165752729,
      "loss": 4.7936,
      "mean_token_accuracy": 0.2225321963429451,
      "num_tokens": 107712390.0,
      "step": 62105
    },
    {
      "entropy": 5.394181537628174,
      "epoch": 4.832367243726901,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002896843601875243,
      "loss": 4.7314,
      "mean_token_accuracy": 0.23118022680282593,
      "num_tokens": 107721205.0,
      "step": 62110
    },
    {
      "entropy": 5.371083307266235,
      "epoch": 4.832756273098619,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028965670357696384,
      "loss": 4.7385,
      "mean_token_accuracy": 0.22614828497171402,
      "num_tokens": 107730217.0,
      "step": 62115
    },
    {
      "entropy": 5.377795028686523,
      "epoch": 4.833145302470337,
      "grad_norm": 1.171875,
      "learning_rate": 0.00028962904674401124,
      "loss": 4.7595,
      "mean_token_accuracy": 0.2262401208281517,
      "num_tokens": 107739153.0,
      "step": 62120
    },
    {
      "entropy": 5.309487247467041,
      "epoch": 4.8335343318420545,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002896013896890862,
      "loss": 4.6599,
      "mean_token_accuracy": 0.23767198771238326,
      "num_tokens": 107747551.0,
      "step": 62125
    },
    {
      "entropy": 5.401728010177612,
      "epoch": 4.833923361213771,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028957373241260816,
      "loss": 4.8308,
      "mean_token_accuracy": 0.21870179027318953,
      "num_tokens": 107755763.0,
      "step": 62130
    },
    {
      "entropy": 5.517660427093506,
      "epoch": 4.834312390585489,
      "grad_norm": 1.171875,
      "learning_rate": 0.000289546074914997,
      "loss": 4.8709,
      "mean_token_accuracy": 0.21933193355798722,
      "num_tokens": 107763431.0,
      "step": 62135
    },
    {
      "entropy": 5.483182001113891,
      "epoch": 4.834701419957207,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002895184171966724,
      "loss": 4.8738,
      "mean_token_accuracy": 0.21124197393655778,
      "num_tokens": 107772601.0,
      "step": 62140
    },
    {
      "entropy": 5.40276403427124,
      "epoch": 4.835090449328924,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002894907592580538,
      "loss": 4.847,
      "mean_token_accuracy": 0.22624280154705048,
      "num_tokens": 107780855.0,
      "step": 62145
    },
    {
      "entropy": 5.381205654144287,
      "epoch": 4.835479478700642,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028946310109956107,
      "loss": 4.7524,
      "mean_token_accuracy": 0.23215004801750183,
      "num_tokens": 107790428.0,
      "step": 62150
    },
    {
      "entropy": 5.518284463882447,
      "epoch": 4.83586850807236,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002894354427216138,
      "loss": 4.8762,
      "mean_token_accuracy": 0.2150247186422348,
      "num_tokens": 107798816.0,
      "step": 62155
    },
    {
      "entropy": 5.3940025806427006,
      "epoch": 4.836257537444077,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028940778412463164,
      "loss": 4.7636,
      "mean_token_accuracy": 0.22035638689994813,
      "num_tokens": 107807387.0,
      "step": 62160
    },
    {
      "entropy": 5.296510410308838,
      "epoch": 4.836646566815794,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028938012530903425,
      "loss": 4.6691,
      "mean_token_accuracy": 0.23735246062278748,
      "num_tokens": 107816447.0,
      "step": 62165
    },
    {
      "entropy": 5.318221616744995,
      "epoch": 4.837035596187512,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002893524662752414,
      "loss": 4.725,
      "mean_token_accuracy": 0.2266872227191925,
      "num_tokens": 107824852.0,
      "step": 62170
    },
    {
      "entropy": 5.297435569763183,
      "epoch": 4.83742462555923,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028932480702367264,
      "loss": 4.6459,
      "mean_token_accuracy": 0.23211088329553603,
      "num_tokens": 107833266.0,
      "step": 62175
    },
    {
      "entropy": 5.3077410697937015,
      "epoch": 4.837813654930947,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002892971475547478,
      "loss": 4.7377,
      "mean_token_accuracy": 0.23364885449409484,
      "num_tokens": 107841723.0,
      "step": 62180
    },
    {
      "entropy": 5.3862865447998045,
      "epoch": 4.838202684302665,
      "grad_norm": 1.125,
      "learning_rate": 0.00028926948786888655,
      "loss": 4.7705,
      "mean_token_accuracy": 0.2240111380815506,
      "num_tokens": 107850731.0,
      "step": 62185
    },
    {
      "entropy": 5.428891372680664,
      "epoch": 4.838591713674383,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028924182796650854,
      "loss": 4.7952,
      "mean_token_accuracy": 0.22418223172426224,
      "num_tokens": 107859251.0,
      "step": 62190
    },
    {
      "entropy": 5.484386396408081,
      "epoch": 4.8389807430461,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028921416784803336,
      "loss": 4.8003,
      "mean_token_accuracy": 0.22607420682907103,
      "num_tokens": 107867584.0,
      "step": 62195
    },
    {
      "entropy": 5.3803760528564455,
      "epoch": 4.839369772417817,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002891865075138809,
      "loss": 4.8142,
      "mean_token_accuracy": 0.22114376723766327,
      "num_tokens": 107876509.0,
      "step": 62200
    },
    {
      "entropy": 5.345442295074463,
      "epoch": 4.839758801789535,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028915884696447064,
      "loss": 4.6897,
      "mean_token_accuracy": 0.23121631294488906,
      "num_tokens": 107885591.0,
      "step": 62205
    },
    {
      "entropy": 5.384468841552734,
      "epoch": 4.840147831161253,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002891311862002225,
      "loss": 4.6574,
      "mean_token_accuracy": 0.23188535422086715,
      "num_tokens": 107893896.0,
      "step": 62210
    },
    {
      "entropy": 5.3528882503509525,
      "epoch": 4.84053686053297,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002891035252215561,
      "loss": 4.792,
      "mean_token_accuracy": 0.22083986848592757,
      "num_tokens": 107901769.0,
      "step": 62215
    },
    {
      "entropy": 5.363465690612793,
      "epoch": 4.840925889904688,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028907586402889115,
      "loss": 4.7788,
      "mean_token_accuracy": 0.22365833669900895,
      "num_tokens": 107909790.0,
      "step": 62220
    },
    {
      "entropy": 5.316484785079956,
      "epoch": 4.841314919276405,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002890482026226473,
      "loss": 4.7315,
      "mean_token_accuracy": 0.22493889778852463,
      "num_tokens": 107918239.0,
      "step": 62225
    },
    {
      "entropy": 5.364662170410156,
      "epoch": 4.841703948648123,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028902054100324445,
      "loss": 4.7756,
      "mean_token_accuracy": 0.22463614195585252,
      "num_tokens": 107926416.0,
      "step": 62230
    },
    {
      "entropy": 5.42995777130127,
      "epoch": 4.84209297801984,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028899287917110214,
      "loss": 4.7854,
      "mean_token_accuracy": 0.229339662194252,
      "num_tokens": 107935595.0,
      "step": 62235
    },
    {
      "entropy": 5.501901912689209,
      "epoch": 4.842482007391558,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002889652171266402,
      "loss": 4.8335,
      "mean_token_accuracy": 0.22634778320789337,
      "num_tokens": 107943214.0,
      "step": 62240
    },
    {
      "entropy": 5.304883623123169,
      "epoch": 4.842871036763276,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028893755487027827,
      "loss": 4.6607,
      "mean_token_accuracy": 0.23679470270872116,
      "num_tokens": 107951404.0,
      "step": 62245
    },
    {
      "entropy": 5.396559429168701,
      "epoch": 4.843260066134993,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028890989240243614,
      "loss": 4.8784,
      "mean_token_accuracy": 0.21596727520227432,
      "num_tokens": 107960611.0,
      "step": 62250
    },
    {
      "entropy": 5.354087591171265,
      "epoch": 4.843649095506711,
      "grad_norm": 1.171875,
      "learning_rate": 0.00028888222972353356,
      "loss": 4.7312,
      "mean_token_accuracy": 0.2322120800614357,
      "num_tokens": 107969217.0,
      "step": 62255
    },
    {
      "entropy": 5.4852887153625485,
      "epoch": 4.844038124878429,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028885456683399024,
      "loss": 4.8117,
      "mean_token_accuracy": 0.2228536769747734,
      "num_tokens": 107978448.0,
      "step": 62260
    },
    {
      "entropy": 5.429894351959229,
      "epoch": 4.844427154250146,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028882690373422586,
      "loss": 4.7439,
      "mean_token_accuracy": 0.2218604639172554,
      "num_tokens": 107987578.0,
      "step": 62265
    },
    {
      "entropy": 5.449800872802735,
      "epoch": 4.844816183621863,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002887992404246603,
      "loss": 4.832,
      "mean_token_accuracy": 0.222810734808445,
      "num_tokens": 107996868.0,
      "step": 62270
    },
    {
      "entropy": 5.478562498092652,
      "epoch": 4.845205212993581,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002887715769057132,
      "loss": 4.8057,
      "mean_token_accuracy": 0.22171455323696138,
      "num_tokens": 108005100.0,
      "step": 62275
    },
    {
      "entropy": 5.373879671096802,
      "epoch": 4.845594242365299,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028874391317780437,
      "loss": 4.7383,
      "mean_token_accuracy": 0.23460486382246018,
      "num_tokens": 108013712.0,
      "step": 62280
    },
    {
      "entropy": 5.444703149795532,
      "epoch": 4.845983271737016,
      "grad_norm": 1.25,
      "learning_rate": 0.0002887162492413534,
      "loss": 4.888,
      "mean_token_accuracy": 0.2198150336742401,
      "num_tokens": 108022423.0,
      "step": 62285
    },
    {
      "entropy": 5.38258810043335,
      "epoch": 4.846372301108734,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002886885850967803,
      "loss": 4.705,
      "mean_token_accuracy": 0.22793688774108886,
      "num_tokens": 108031531.0,
      "step": 62290
    },
    {
      "entropy": 5.391807556152344,
      "epoch": 4.846761330480451,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028866092074450467,
      "loss": 4.7515,
      "mean_token_accuracy": 0.21793063133955,
      "num_tokens": 108040753.0,
      "step": 62295
    },
    {
      "entropy": 5.422497320175171,
      "epoch": 4.847150359852169,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002886332561849464,
      "loss": 4.8703,
      "mean_token_accuracy": 0.22131213545799255,
      "num_tokens": 108050185.0,
      "step": 62300
    },
    {
      "entropy": 5.494942903518677,
      "epoch": 4.847539389223886,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00028860559141852504,
      "loss": 4.8565,
      "mean_token_accuracy": 0.2275303289294243,
      "num_tokens": 108058116.0,
      "step": 62305
    },
    {
      "entropy": 5.437143611907959,
      "epoch": 4.847928418595604,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028857792644566057,
      "loss": 4.7711,
      "mean_token_accuracy": 0.22440730333328246,
      "num_tokens": 108066150.0,
      "step": 62310
    },
    {
      "entropy": 5.330299615859985,
      "epoch": 4.848317447967322,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002885502612667727,
      "loss": 4.6802,
      "mean_token_accuracy": 0.23074383586645125,
      "num_tokens": 108075519.0,
      "step": 62315
    },
    {
      "entropy": 5.288139772415161,
      "epoch": 4.848706477339039,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028852259588228106,
      "loss": 4.6897,
      "mean_token_accuracy": 0.230577951669693,
      "num_tokens": 108084506.0,
      "step": 62320
    },
    {
      "entropy": 5.43466796875,
      "epoch": 4.849095506710757,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028849493029260573,
      "loss": 4.8113,
      "mean_token_accuracy": 0.23240932673215867,
      "num_tokens": 108093311.0,
      "step": 62325
    },
    {
      "entropy": 5.372525644302368,
      "epoch": 4.849484536082474,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002884672644981661,
      "loss": 4.7197,
      "mean_token_accuracy": 0.23194820433855057,
      "num_tokens": 108102110.0,
      "step": 62330
    },
    {
      "entropy": 5.4318877220153805,
      "epoch": 4.849873565454192,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002884395984993823,
      "loss": 4.7927,
      "mean_token_accuracy": 0.21765922605991364,
      "num_tokens": 108110805.0,
      "step": 62335
    },
    {
      "entropy": 5.481058120727539,
      "epoch": 4.850262594825909,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000288411932296674,
      "loss": 4.8604,
      "mean_token_accuracy": 0.2180154800415039,
      "num_tokens": 108119717.0,
      "step": 62340
    },
    {
      "entropy": 5.4526838779449465,
      "epoch": 4.850651624197627,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002883842658904609,
      "loss": 4.7512,
      "mean_token_accuracy": 0.23140333443880082,
      "num_tokens": 108128091.0,
      "step": 62345
    },
    {
      "entropy": 5.456583786010742,
      "epoch": 4.851040653569345,
      "grad_norm": 1.125,
      "learning_rate": 0.0002883565992811629,
      "loss": 4.7917,
      "mean_token_accuracy": 0.22166844457387924,
      "num_tokens": 108136766.0,
      "step": 62350
    },
    {
      "entropy": 5.403702592849731,
      "epoch": 4.851429682941062,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002883289324691998,
      "loss": 4.7881,
      "mean_token_accuracy": 0.22381279468536378,
      "num_tokens": 108145327.0,
      "step": 62355
    },
    {
      "entropy": 5.415755653381348,
      "epoch": 4.851818712312779,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002883012654549913,
      "loss": 4.742,
      "mean_token_accuracy": 0.22388166934251785,
      "num_tokens": 108153692.0,
      "step": 62360
    },
    {
      "entropy": 5.477291488647461,
      "epoch": 4.852207741684497,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028827359823895726,
      "loss": 4.8358,
      "mean_token_accuracy": 0.2089389219880104,
      "num_tokens": 108162406.0,
      "step": 62365
    },
    {
      "entropy": 5.474725532531738,
      "epoch": 4.852596771056215,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00028824593082151754,
      "loss": 4.9243,
      "mean_token_accuracy": 0.211528879404068,
      "num_tokens": 108170787.0,
      "step": 62370
    },
    {
      "entropy": 5.383820295333862,
      "epoch": 4.852985800427932,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002882182632030919,
      "loss": 4.7262,
      "mean_token_accuracy": 0.22840136289596558,
      "num_tokens": 108179719.0,
      "step": 62375
    },
    {
      "entropy": 5.434463548660278,
      "epoch": 4.85337482979965,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002881905953841002,
      "loss": 4.8469,
      "mean_token_accuracy": 0.22321785539388656,
      "num_tokens": 108188098.0,
      "step": 62380
    },
    {
      "entropy": 5.389695835113526,
      "epoch": 4.853763859171368,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002881629273649621,
      "loss": 4.8007,
      "mean_token_accuracy": 0.2230105072259903,
      "num_tokens": 108196912.0,
      "step": 62385
    },
    {
      "entropy": 5.3954854011535645,
      "epoch": 4.854152888543085,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028813525914609767,
      "loss": 4.6979,
      "mean_token_accuracy": 0.2234781041741371,
      "num_tokens": 108205574.0,
      "step": 62390
    },
    {
      "entropy": 5.508768224716187,
      "epoch": 4.854541917914803,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028810759072792644,
      "loss": 4.8489,
      "mean_token_accuracy": 0.2176006019115448,
      "num_tokens": 108213493.0,
      "step": 62395
    },
    {
      "entropy": 5.333204364776611,
      "epoch": 4.85493094728652,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002880799221108684,
      "loss": 4.6861,
      "mean_token_accuracy": 0.22731382995843888,
      "num_tokens": 108221852.0,
      "step": 62400
    },
    {
      "entropy": 5.405997467041016,
      "epoch": 4.855319976658238,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002880522532953435,
      "loss": 4.7073,
      "mean_token_accuracy": 0.22474607825279236,
      "num_tokens": 108229517.0,
      "step": 62405
    },
    {
      "entropy": 5.337581729888916,
      "epoch": 4.855709006029955,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002880245842817713,
      "loss": 4.789,
      "mean_token_accuracy": 0.22694811522960662,
      "num_tokens": 108238666.0,
      "step": 62410
    },
    {
      "entropy": 5.392370748519897,
      "epoch": 4.856098035401673,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002879969150705718,
      "loss": 4.8039,
      "mean_token_accuracy": 0.22382072508335113,
      "num_tokens": 108246971.0,
      "step": 62415
    },
    {
      "entropy": 5.480716180801392,
      "epoch": 4.856487064773391,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002879692456621649,
      "loss": 4.7948,
      "mean_token_accuracy": 0.2158706530928612,
      "num_tokens": 108255123.0,
      "step": 62420
    },
    {
      "entropy": 5.385247611999512,
      "epoch": 4.8568760941451075,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002879415760569702,
      "loss": 4.7671,
      "mean_token_accuracy": 0.22791349440813063,
      "num_tokens": 108263737.0,
      "step": 62425
    },
    {
      "entropy": 5.387014722824096,
      "epoch": 4.857265123516825,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002879139062554077,
      "loss": 4.7487,
      "mean_token_accuracy": 0.22671225816011428,
      "num_tokens": 108272356.0,
      "step": 62430
    },
    {
      "entropy": 5.5118879795074465,
      "epoch": 4.857654152888543,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028788623625789725,
      "loss": 4.9372,
      "mean_token_accuracy": 0.21839599758386613,
      "num_tokens": 108280431.0,
      "step": 62435
    },
    {
      "entropy": 5.399744319915771,
      "epoch": 4.8580431822602606,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002878585660648587,
      "loss": 4.6685,
      "mean_token_accuracy": 0.23119349777698517,
      "num_tokens": 108288966.0,
      "step": 62440
    },
    {
      "entropy": 5.308374118804932,
      "epoch": 4.858432211631978,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002878308956767118,
      "loss": 4.7086,
      "mean_token_accuracy": 0.2357080787420273,
      "num_tokens": 108297630.0,
      "step": 62445
    },
    {
      "entropy": 5.382984733581543,
      "epoch": 4.858821241003696,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002878032250938766,
      "loss": 4.7292,
      "mean_token_accuracy": 0.23048530220985414,
      "num_tokens": 108305828.0,
      "step": 62450
    },
    {
      "entropy": 5.541228485107422,
      "epoch": 4.859210270375414,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028777555431677276,
      "loss": 4.8799,
      "mean_token_accuracy": 0.21814603358507156,
      "num_tokens": 108315192.0,
      "step": 62455
    },
    {
      "entropy": 5.426833534240723,
      "epoch": 4.859599299747131,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028774788334582025,
      "loss": 4.7722,
      "mean_token_accuracy": 0.21955320090055466,
      "num_tokens": 108324934.0,
      "step": 62460
    },
    {
      "entropy": 5.372299337387085,
      "epoch": 4.859988329118848,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002877202121814389,
      "loss": 4.6612,
      "mean_token_accuracy": 0.23443253338336945,
      "num_tokens": 108333549.0,
      "step": 62465
    },
    {
      "entropy": 5.3668749809265135,
      "epoch": 4.860377358490566,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002876925408240485,
      "loss": 4.735,
      "mean_token_accuracy": 0.22402966022491455,
      "num_tokens": 108342782.0,
      "step": 62470
    },
    {
      "entropy": 5.46300892829895,
      "epoch": 4.8607663878622835,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002876648692740691,
      "loss": 4.8868,
      "mean_token_accuracy": 0.2169716700911522,
      "num_tokens": 108352272.0,
      "step": 62475
    },
    {
      "entropy": 5.391826725006103,
      "epoch": 4.861155417234001,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028763719753192044,
      "loss": 4.74,
      "mean_token_accuracy": 0.2264694795012474,
      "num_tokens": 108362015.0,
      "step": 62480
    },
    {
      "entropy": 5.361153173446655,
      "epoch": 4.861544446605719,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002876095255980224,
      "loss": 4.7118,
      "mean_token_accuracy": 0.23048152029514313,
      "num_tokens": 108370307.0,
      "step": 62485
    },
    {
      "entropy": 5.451849126815796,
      "epoch": 4.861933475977437,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028758185347279485,
      "loss": 4.7998,
      "mean_token_accuracy": 0.22439068108797072,
      "num_tokens": 108379803.0,
      "step": 62490
    },
    {
      "entropy": 5.46420316696167,
      "epoch": 4.8623225053491534,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002875541811566578,
      "loss": 4.7585,
      "mean_token_accuracy": 0.22957368642091752,
      "num_tokens": 108388499.0,
      "step": 62495
    },
    {
      "entropy": 5.3534770011901855,
      "epoch": 4.862711534720871,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002875265086500309,
      "loss": 4.6393,
      "mean_token_accuracy": 0.23172218054533006,
      "num_tokens": 108396437.0,
      "step": 62500
    },
    {
      "entropy": 5.336780118942261,
      "epoch": 4.863100564092589,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028749883595333424,
      "loss": 4.7119,
      "mean_token_accuracy": 0.23219569772481918,
      "num_tokens": 108405011.0,
      "step": 62505
    },
    {
      "entropy": 5.416940546035766,
      "epoch": 4.8634895934643065,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002874711630669876,
      "loss": 4.6986,
      "mean_token_accuracy": 0.23322720676660538,
      "num_tokens": 108413068.0,
      "step": 62510
    },
    {
      "entropy": 5.465639686584472,
      "epoch": 4.863878622836024,
      "grad_norm": 1.34375,
      "learning_rate": 0.000287443489991411,
      "loss": 4.7517,
      "mean_token_accuracy": 0.22750117778778076,
      "num_tokens": 108422088.0,
      "step": 62515
    },
    {
      "entropy": 5.459622192382812,
      "epoch": 4.864267652207742,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002874158167270242,
      "loss": 4.8817,
      "mean_token_accuracy": 0.21617176085710527,
      "num_tokens": 108430580.0,
      "step": 62520
    },
    {
      "entropy": 5.430044317245484,
      "epoch": 4.86465668157946,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028738814327424714,
      "loss": 4.7358,
      "mean_token_accuracy": 0.22372561544179917,
      "num_tokens": 108439187.0,
      "step": 62525
    },
    {
      "entropy": 5.46817626953125,
      "epoch": 4.865045710951176,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002873604696334997,
      "loss": 4.767,
      "mean_token_accuracy": 0.2224241927266121,
      "num_tokens": 108447547.0,
      "step": 62530
    },
    {
      "entropy": 5.462680435180664,
      "epoch": 4.865434740322894,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028733279580520185,
      "loss": 4.8172,
      "mean_token_accuracy": 0.22280648946762086,
      "num_tokens": 108456398.0,
      "step": 62535
    },
    {
      "entropy": 5.411099243164062,
      "epoch": 4.865823769694612,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028730512178977343,
      "loss": 4.7793,
      "mean_token_accuracy": 0.2245758056640625,
      "num_tokens": 108465156.0,
      "step": 62540
    },
    {
      "entropy": 5.399336194992065,
      "epoch": 4.8662127990663295,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028727744758763434,
      "loss": 4.7507,
      "mean_token_accuracy": 0.22699066549539565,
      "num_tokens": 108474133.0,
      "step": 62545
    },
    {
      "entropy": 5.449962902069092,
      "epoch": 4.866601828438047,
      "grad_norm": 1.25,
      "learning_rate": 0.00028724977319920453,
      "loss": 4.8236,
      "mean_token_accuracy": 0.22399901747703552,
      "num_tokens": 108482243.0,
      "step": 62550
    },
    {
      "entropy": 5.373311614990234,
      "epoch": 4.866990857809765,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028722209862490393,
      "loss": 4.7847,
      "mean_token_accuracy": 0.22300568521022796,
      "num_tokens": 108491892.0,
      "step": 62555
    },
    {
      "entropy": 5.450145530700683,
      "epoch": 4.867379887181482,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028719442386515245,
      "loss": 4.8743,
      "mean_token_accuracy": 0.21830102503299714,
      "num_tokens": 108500991.0,
      "step": 62560
    },
    {
      "entropy": 5.4401695251464846,
      "epoch": 4.867768916553199,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028716674892036994,
      "loss": 4.724,
      "mean_token_accuracy": 0.22529653012752532,
      "num_tokens": 108509182.0,
      "step": 62565
    },
    {
      "entropy": 5.429437446594238,
      "epoch": 4.868157945924917,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002871390737909765,
      "loss": 4.6777,
      "mean_token_accuracy": 0.23882465064525604,
      "num_tokens": 108517472.0,
      "step": 62570
    },
    {
      "entropy": 5.375689172744751,
      "epoch": 4.868546975296635,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028711139847739187,
      "loss": 4.7382,
      "mean_token_accuracy": 0.22955442517995833,
      "num_tokens": 108525874.0,
      "step": 62575
    },
    {
      "entropy": 5.421518564224243,
      "epoch": 4.8689360046683525,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028708372298003607,
      "loss": 4.8796,
      "mean_token_accuracy": 0.22388354539871216,
      "num_tokens": 108534964.0,
      "step": 62580
    },
    {
      "entropy": 5.505806541442871,
      "epoch": 4.86932503404007,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028705604729932895,
      "loss": 4.8472,
      "mean_token_accuracy": 0.2168989822268486,
      "num_tokens": 108543145.0,
      "step": 62585
    },
    {
      "entropy": 5.412068939208984,
      "epoch": 4.869714063411788,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028702837143569057,
      "loss": 4.8001,
      "mean_token_accuracy": 0.22655394971370696,
      "num_tokens": 108551434.0,
      "step": 62590
    },
    {
      "entropy": 5.3919289112091064,
      "epoch": 4.870103092783506,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028700069538954064,
      "loss": 4.7692,
      "mean_token_accuracy": 0.22733844667673112,
      "num_tokens": 108559968.0,
      "step": 62595
    },
    {
      "entropy": 5.388823127746582,
      "epoch": 4.870492122155222,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028697301916129944,
      "loss": 4.7882,
      "mean_token_accuracy": 0.22116948217153548,
      "num_tokens": 108569817.0,
      "step": 62600
    },
    {
      "entropy": 5.550229454040528,
      "epoch": 4.87088115152694,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002869453427513866,
      "loss": 4.8735,
      "mean_token_accuracy": 0.2204594776034355,
      "num_tokens": 108578487.0,
      "step": 62605
    },
    {
      "entropy": 5.337610244750977,
      "epoch": 4.871270180898658,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028691766616022234,
      "loss": 4.6799,
      "mean_token_accuracy": 0.2340213492512703,
      "num_tokens": 108587794.0,
      "step": 62610
    },
    {
      "entropy": 5.344323396682739,
      "epoch": 4.8716592102703755,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002868899893882264,
      "loss": 4.7506,
      "mean_token_accuracy": 0.23071829974651337,
      "num_tokens": 108595796.0,
      "step": 62615
    },
    {
      "entropy": 5.391664171218872,
      "epoch": 4.872048239642093,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028686231243581873,
      "loss": 4.712,
      "mean_token_accuracy": 0.2227457195520401,
      "num_tokens": 108604661.0,
      "step": 62620
    },
    {
      "entropy": 5.455424070358276,
      "epoch": 4.872437269013811,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028683463530341944,
      "loss": 4.7712,
      "mean_token_accuracy": 0.2272370532155037,
      "num_tokens": 108612569.0,
      "step": 62625
    },
    {
      "entropy": 5.346637535095215,
      "epoch": 4.872826298385528,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002868069579914483,
      "loss": 4.6788,
      "mean_token_accuracy": 0.22973189055919646,
      "num_tokens": 108620622.0,
      "step": 62630
    },
    {
      "entropy": 5.341037225723267,
      "epoch": 4.873215327757245,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028677928050032547,
      "loss": 4.6826,
      "mean_token_accuracy": 0.23535866886377335,
      "num_tokens": 108628775.0,
      "step": 62635
    },
    {
      "entropy": 5.379862594604492,
      "epoch": 4.873604357128963,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028675160283047073,
      "loss": 4.7358,
      "mean_token_accuracy": 0.22782426178455353,
      "num_tokens": 108637545.0,
      "step": 62640
    },
    {
      "entropy": 5.406108236312866,
      "epoch": 4.873993386500681,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028672392498230414,
      "loss": 4.7905,
      "mean_token_accuracy": 0.22281149476766587,
      "num_tokens": 108647074.0,
      "step": 62645
    },
    {
      "entropy": 5.3663191318511965,
      "epoch": 4.8743824158723985,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028669624695624573,
      "loss": 4.7257,
      "mean_token_accuracy": 0.22987242490053178,
      "num_tokens": 108654711.0,
      "step": 62650
    },
    {
      "entropy": 5.362139701843262,
      "epoch": 4.874771445244116,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028666856875271535,
      "loss": 4.8113,
      "mean_token_accuracy": 0.22371883541345597,
      "num_tokens": 108663641.0,
      "step": 62655
    },
    {
      "entropy": 5.294822120666504,
      "epoch": 4.875160474615834,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002866408903721329,
      "loss": 4.6864,
      "mean_token_accuracy": 0.23326313197612764,
      "num_tokens": 108672454.0,
      "step": 62660
    },
    {
      "entropy": 5.353326845169067,
      "epoch": 4.875549503987551,
      "grad_norm": 1.125,
      "learning_rate": 0.00028661321181491854,
      "loss": 4.8333,
      "mean_token_accuracy": 0.22589156180620193,
      "num_tokens": 108681356.0,
      "step": 62665
    },
    {
      "entropy": 5.530039262771607,
      "epoch": 4.875938533359268,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002865855330814922,
      "loss": 4.9034,
      "mean_token_accuracy": 0.21265966147184373,
      "num_tokens": 108690751.0,
      "step": 62670
    },
    {
      "entropy": 5.385809421539307,
      "epoch": 4.876327562730986,
      "grad_norm": 1.25,
      "learning_rate": 0.0002865578541722738,
      "loss": 4.7026,
      "mean_token_accuracy": 0.22740580439567565,
      "num_tokens": 108699747.0,
      "step": 62675
    },
    {
      "entropy": 5.393407249450684,
      "epoch": 4.876716592102704,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002865301750876834,
      "loss": 4.7501,
      "mean_token_accuracy": 0.2221317023038864,
      "num_tokens": 108708741.0,
      "step": 62680
    },
    {
      "entropy": 5.429852628707886,
      "epoch": 4.8771056214744215,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002865024958281409,
      "loss": 4.7515,
      "mean_token_accuracy": 0.2280532017350197,
      "num_tokens": 108717220.0,
      "step": 62685
    },
    {
      "entropy": 5.444710922241211,
      "epoch": 4.877494650846139,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028647481639406644,
      "loss": 4.8089,
      "mean_token_accuracy": 0.2285093456506729,
      "num_tokens": 108725939.0,
      "step": 62690
    },
    {
      "entropy": 5.441153717041016,
      "epoch": 4.877883680217856,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028644713678587983,
      "loss": 4.7856,
      "mean_token_accuracy": 0.23086303919553758,
      "num_tokens": 108734738.0,
      "step": 62695
    },
    {
      "entropy": 5.26664662361145,
      "epoch": 4.878272709589574,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028641945700400104,
      "loss": 4.6234,
      "mean_token_accuracy": 0.24106180518865586,
      "num_tokens": 108743755.0,
      "step": 62700
    },
    {
      "entropy": 5.406527662277222,
      "epoch": 4.878661738961291,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028639177704885033,
      "loss": 4.8695,
      "mean_token_accuracy": 0.21789133697748184,
      "num_tokens": 108752494.0,
      "step": 62705
    },
    {
      "entropy": 5.383164310455323,
      "epoch": 4.879050768333009,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028636409692084743,
      "loss": 4.7882,
      "mean_token_accuracy": 0.21786618530750274,
      "num_tokens": 108761669.0,
      "step": 62710
    },
    {
      "entropy": 5.369325351715088,
      "epoch": 4.879439797704727,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002863364166204125,
      "loss": 4.7069,
      "mean_token_accuracy": 0.2256480112671852,
      "num_tokens": 108769699.0,
      "step": 62715
    },
    {
      "entropy": 5.389443826675415,
      "epoch": 4.8798288270764445,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028630873614796543,
      "loss": 4.7486,
      "mean_token_accuracy": 0.21982354670763016,
      "num_tokens": 108777799.0,
      "step": 62720
    },
    {
      "entropy": 5.387789344787597,
      "epoch": 4.880217856448162,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028628105550392637,
      "loss": 4.8235,
      "mean_token_accuracy": 0.22200949639081954,
      "num_tokens": 108786547.0,
      "step": 62725
    },
    {
      "entropy": 5.463385343551636,
      "epoch": 4.88060688581988,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002862533746887152,
      "loss": 4.8269,
      "mean_token_accuracy": 0.22086931020021439,
      "num_tokens": 108795268.0,
      "step": 62730
    },
    {
      "entropy": 5.415080308914185,
      "epoch": 4.880995915191597,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028622569370275196,
      "loss": 4.724,
      "mean_token_accuracy": 0.2270495191216469,
      "num_tokens": 108804496.0,
      "step": 62735
    },
    {
      "entropy": 5.416050052642822,
      "epoch": 4.881384944563314,
      "grad_norm": 1.328125,
      "learning_rate": 0.00028619801254645675,
      "loss": 4.8093,
      "mean_token_accuracy": 0.22286012023687363,
      "num_tokens": 108813717.0,
      "step": 62740
    },
    {
      "entropy": 5.336078834533692,
      "epoch": 4.881773973935032,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002861703312202495,
      "loss": 4.7903,
      "mean_token_accuracy": 0.22591417878866196,
      "num_tokens": 108822265.0,
      "step": 62745
    },
    {
      "entropy": 5.435274648666382,
      "epoch": 4.88216300330675,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028614264972455023,
      "loss": 4.7968,
      "mean_token_accuracy": 0.22205123007297517,
      "num_tokens": 108830678.0,
      "step": 62750
    },
    {
      "entropy": 5.407142877578735,
      "epoch": 4.8825520326784675,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028611496805977906,
      "loss": 4.7069,
      "mean_token_accuracy": 0.2327314332127571,
      "num_tokens": 108839253.0,
      "step": 62755
    },
    {
      "entropy": 5.370106172561646,
      "epoch": 4.882941062050184,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028608728622635585,
      "loss": 4.7411,
      "mean_token_accuracy": 0.22633830606937408,
      "num_tokens": 108847175.0,
      "step": 62760
    },
    {
      "entropy": 5.460643243789673,
      "epoch": 4.883330091421902,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00028605960422470073,
      "loss": 4.8931,
      "mean_token_accuracy": 0.21166498810052872,
      "num_tokens": 108856640.0,
      "step": 62765
    },
    {
      "entropy": 5.459381484985352,
      "epoch": 4.88371912079362,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002860319220552338,
      "loss": 4.7997,
      "mean_token_accuracy": 0.22016754150390624,
      "num_tokens": 108865207.0,
      "step": 62770
    },
    {
      "entropy": 5.414637184143066,
      "epoch": 4.884108150165337,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028600423971837484,
      "loss": 4.7343,
      "mean_token_accuracy": 0.22254758328199387,
      "num_tokens": 108874841.0,
      "step": 62775
    },
    {
      "entropy": 5.448733901977539,
      "epoch": 4.884497179537055,
      "grad_norm": 1.421875,
      "learning_rate": 0.00028597655721454427,
      "loss": 4.8352,
      "mean_token_accuracy": 0.22089233100414277,
      "num_tokens": 108883974.0,
      "step": 62780
    },
    {
      "entropy": 5.376339864730835,
      "epoch": 4.884886208908773,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028594887454416176,
      "loss": 4.7555,
      "mean_token_accuracy": 0.23052518218755721,
      "num_tokens": 108892665.0,
      "step": 62785
    },
    {
      "entropy": 5.387379837036133,
      "epoch": 4.8852752382804905,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002859211917076476,
      "loss": 4.7214,
      "mean_token_accuracy": 0.24001228511333467,
      "num_tokens": 108901441.0,
      "step": 62790
    },
    {
      "entropy": 5.402598476409912,
      "epoch": 4.885664267652208,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028589350870542165,
      "loss": 4.7913,
      "mean_token_accuracy": 0.2221328377723694,
      "num_tokens": 108909958.0,
      "step": 62795
    },
    {
      "entropy": 5.478853178024292,
      "epoch": 4.886053297023925,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002858658255379041,
      "loss": 4.8495,
      "mean_token_accuracy": 0.22005851417779923,
      "num_tokens": 108918259.0,
      "step": 62800
    },
    {
      "entropy": 5.346837186813355,
      "epoch": 4.886442326395643,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028583814220551495,
      "loss": 4.6404,
      "mean_token_accuracy": 0.2347148761153221,
      "num_tokens": 108926170.0,
      "step": 62805
    },
    {
      "entropy": 5.371564435958862,
      "epoch": 4.88683135576736,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002858104587086742,
      "loss": 4.8006,
      "mean_token_accuracy": 0.22204282879829407,
      "num_tokens": 108934996.0,
      "step": 62810
    },
    {
      "entropy": 5.390116262435913,
      "epoch": 4.887220385139078,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000285782775047802,
      "loss": 4.7719,
      "mean_token_accuracy": 0.22440939992666245,
      "num_tokens": 108943402.0,
      "step": 62815
    },
    {
      "entropy": 5.404979562759399,
      "epoch": 4.887609414510796,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002857550912233183,
      "loss": 4.7263,
      "mean_token_accuracy": 0.22750645577907563,
      "num_tokens": 108951938.0,
      "step": 62820
    },
    {
      "entropy": 5.359935569763183,
      "epoch": 4.8879984438825135,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002857274072356432,
      "loss": 4.744,
      "mean_token_accuracy": 0.2267139360308647,
      "num_tokens": 108960656.0,
      "step": 62825
    },
    {
      "entropy": 5.40571231842041,
      "epoch": 4.88838747325423,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028569972308519674,
      "loss": 4.7535,
      "mean_token_accuracy": 0.22500782907009126,
      "num_tokens": 108969375.0,
      "step": 62830
    },
    {
      "entropy": 5.425194215774536,
      "epoch": 4.888776502625948,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028567203877239914,
      "loss": 4.8296,
      "mean_token_accuracy": 0.22712219953536988,
      "num_tokens": 108978804.0,
      "step": 62835
    },
    {
      "entropy": 5.442486572265625,
      "epoch": 4.889165531997666,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028564435429767035,
      "loss": 4.7175,
      "mean_token_accuracy": 0.2255354791879654,
      "num_tokens": 108986687.0,
      "step": 62840
    },
    {
      "entropy": 5.402191209793091,
      "epoch": 4.889554561369383,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002856166696614303,
      "loss": 4.7833,
      "mean_token_accuracy": 0.2247421771287918,
      "num_tokens": 108995664.0,
      "step": 62845
    },
    {
      "entropy": 5.370953893661499,
      "epoch": 4.889943590741101,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002855889848640992,
      "loss": 4.7292,
      "mean_token_accuracy": 0.2312014266848564,
      "num_tokens": 109004614.0,
      "step": 62850
    },
    {
      "entropy": 5.426415205001831,
      "epoch": 4.890332620112819,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002855612999060971,
      "loss": 4.7885,
      "mean_token_accuracy": 0.22780075669288635,
      "num_tokens": 109012982.0,
      "step": 62855
    },
    {
      "entropy": 5.392883062362671,
      "epoch": 4.8907216494845365,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002855336147878442,
      "loss": 4.7611,
      "mean_token_accuracy": 0.22393666058778763,
      "num_tokens": 109020913.0,
      "step": 62860
    },
    {
      "entropy": 5.386337852478027,
      "epoch": 4.891110678856253,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028550592950976036,
      "loss": 4.7644,
      "mean_token_accuracy": 0.22652415186166763,
      "num_tokens": 109029746.0,
      "step": 62865
    },
    {
      "entropy": 5.264884901046753,
      "epoch": 4.891499708227971,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028547824407226574,
      "loss": 4.6522,
      "mean_token_accuracy": 0.23311558812856675,
      "num_tokens": 109038980.0,
      "step": 62870
    },
    {
      "entropy": 5.361165857315063,
      "epoch": 4.891888737599689,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002854505584757805,
      "loss": 4.7285,
      "mean_token_accuracy": 0.22874753922224045,
      "num_tokens": 109047120.0,
      "step": 62875
    },
    {
      "entropy": 5.4155913352966305,
      "epoch": 4.892277766971406,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028542287272072475,
      "loss": 4.7419,
      "mean_token_accuracy": 0.22763350009918212,
      "num_tokens": 109056138.0,
      "step": 62880
    },
    {
      "entropy": 5.425210618972779,
      "epoch": 4.892666796343124,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002853951868075184,
      "loss": 4.7631,
      "mean_token_accuracy": 0.2293737143278122,
      "num_tokens": 109064977.0,
      "step": 62885
    },
    {
      "entropy": 5.444200897216797,
      "epoch": 4.893055825714842,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002853675007365816,
      "loss": 4.8157,
      "mean_token_accuracy": 0.22457858622074128,
      "num_tokens": 109073884.0,
      "step": 62890
    },
    {
      "entropy": 5.43964900970459,
      "epoch": 4.893444855086559,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002853398145083345,
      "loss": 4.8314,
      "mean_token_accuracy": 0.21981510818004607,
      "num_tokens": 109082804.0,
      "step": 62895
    },
    {
      "entropy": 5.37054328918457,
      "epoch": 4.893833884458276,
      "grad_norm": 1.125,
      "learning_rate": 0.00028531212812319724,
      "loss": 4.6965,
      "mean_token_accuracy": 0.23577572405338287,
      "num_tokens": 109092327.0,
      "step": 62900
    },
    {
      "entropy": 5.444842433929443,
      "epoch": 4.894222913829994,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002852844415815898,
      "loss": 4.8183,
      "mean_token_accuracy": 0.22608525454998016,
      "num_tokens": 109100366.0,
      "step": 62905
    },
    {
      "entropy": 5.411284923553467,
      "epoch": 4.894611943201712,
      "grad_norm": 1.125,
      "learning_rate": 0.00028525675488393234,
      "loss": 4.7711,
      "mean_token_accuracy": 0.2226783737540245,
      "num_tokens": 109109504.0,
      "step": 62910
    },
    {
      "entropy": 5.407434034347534,
      "epoch": 4.895000972573429,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028522906803064494,
      "loss": 4.7951,
      "mean_token_accuracy": 0.22046105712652206,
      "num_tokens": 109118538.0,
      "step": 62915
    },
    {
      "entropy": 5.42401614189148,
      "epoch": 4.895390001945147,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002852013810221477,
      "loss": 4.7672,
      "mean_token_accuracy": 0.22073305547237396,
      "num_tokens": 109127057.0,
      "step": 62920
    },
    {
      "entropy": 5.477816247940064,
      "epoch": 4.895779031316865,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002851736938588607,
      "loss": 4.8342,
      "mean_token_accuracy": 0.21699199676513672,
      "num_tokens": 109135836.0,
      "step": 62925
    },
    {
      "entropy": 5.4339141845703125,
      "epoch": 4.8961680606885825,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002851460065412042,
      "loss": 4.8137,
      "mean_token_accuracy": 0.21796138137578963,
      "num_tokens": 109144224.0,
      "step": 62930
    },
    {
      "entropy": 5.399669647216797,
      "epoch": 4.896557090060299,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002851183190695981,
      "loss": 4.7981,
      "mean_token_accuracy": 0.22110403776168824,
      "num_tokens": 109152899.0,
      "step": 62935
    },
    {
      "entropy": 5.369991302490234,
      "epoch": 4.896946119432017,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002850906314444627,
      "loss": 4.7663,
      "mean_token_accuracy": 0.22514845579862594,
      "num_tokens": 109161429.0,
      "step": 62940
    },
    {
      "entropy": 5.467168951034546,
      "epoch": 4.897335148803735,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028506294366621796,
      "loss": 4.8698,
      "mean_token_accuracy": 0.21920322328805925,
      "num_tokens": 109169954.0,
      "step": 62945
    },
    {
      "entropy": 5.393003654479981,
      "epoch": 4.897724178175452,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028503525573528413,
      "loss": 4.7436,
      "mean_token_accuracy": 0.22532564401626587,
      "num_tokens": 109178750.0,
      "step": 62950
    },
    {
      "entropy": 5.43347053527832,
      "epoch": 4.89811320754717,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002850075676520812,
      "loss": 4.7857,
      "mean_token_accuracy": 0.22208648025989533,
      "num_tokens": 109187728.0,
      "step": 62955
    },
    {
      "entropy": 5.39475417137146,
      "epoch": 4.898502236918888,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002849798794170294,
      "loss": 4.7695,
      "mean_token_accuracy": 0.2267370417714119,
      "num_tokens": 109196721.0,
      "step": 62960
    },
    {
      "entropy": 5.3381688594818115,
      "epoch": 4.898891266290605,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002849521910305488,
      "loss": 4.7356,
      "mean_token_accuracy": 0.22488946169614793,
      "num_tokens": 109205333.0,
      "step": 62965
    },
    {
      "entropy": 5.354612827301025,
      "epoch": 4.899280295662322,
      "grad_norm": 1.171875,
      "learning_rate": 0.00028492450249305957,
      "loss": 4.7185,
      "mean_token_accuracy": 0.23347699046134948,
      "num_tokens": 109213475.0,
      "step": 62970
    },
    {
      "entropy": 5.471864986419678,
      "epoch": 4.89966932503404,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00028489681380498177,
      "loss": 4.8205,
      "mean_token_accuracy": 0.22297414541244506,
      "num_tokens": 109220827.0,
      "step": 62975
    },
    {
      "entropy": 5.317643499374389,
      "epoch": 4.900058354405758,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028486912496673557,
      "loss": 4.6437,
      "mean_token_accuracy": 0.23712167888879776,
      "num_tokens": 109229188.0,
      "step": 62980
    },
    {
      "entropy": 5.3704406261444095,
      "epoch": 4.900447383777475,
      "grad_norm": 1.125,
      "learning_rate": 0.00028484143597874115,
      "loss": 4.789,
      "mean_token_accuracy": 0.2239796847105026,
      "num_tokens": 109237567.0,
      "step": 62985
    },
    {
      "entropy": 5.383840274810791,
      "epoch": 4.900836413149193,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002848137468414186,
      "loss": 4.7572,
      "mean_token_accuracy": 0.22657978385686875,
      "num_tokens": 109245907.0,
      "step": 62990
    },
    {
      "entropy": 5.451814651489258,
      "epoch": 4.901225442520911,
      "grad_norm": 1.234375,
      "learning_rate": 0.000284786057555188,
      "loss": 4.7499,
      "mean_token_accuracy": 0.22753514498472213,
      "num_tokens": 109254824.0,
      "step": 62995
    },
    {
      "entropy": 5.4983679294586185,
      "epoch": 4.9016144718926276,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002847583681204696,
      "loss": 4.8646,
      "mean_token_accuracy": 0.2200871616601944,
      "num_tokens": 109263385.0,
      "step": 63000
    },
    {
      "epoch": 4.9016144718926276,
      "eval_entropy": 5.17692167196742,
      "eval_loss": 4.915230751037598,
      "eval_mean_token_accuracy": 0.226935276271519,
      "eval_num_tokens": 109263385.0,
      "eval_runtime": 28.9113,
      "eval_samples_per_second": 1437.431,
      "eval_steps_per_second": 179.688,
      "step": 63000
    },
    {
      "entropy": 5.419268465042114,
      "epoch": 4.902003501264345,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028473067853768347,
      "loss": 4.7276,
      "mean_token_accuracy": 0.22985544949769973,
      "num_tokens": 109271567.0,
      "step": 63005
    },
    {
      "entropy": 5.394822359085083,
      "epoch": 4.902392530636063,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028470298880724985,
      "loss": 4.7959,
      "mean_token_accuracy": 0.2280418574810028,
      "num_tokens": 109281714.0,
      "step": 63010
    },
    {
      "entropy": 5.3772705554962155,
      "epoch": 4.902781560007781,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002846752989295888,
      "loss": 4.7206,
      "mean_token_accuracy": 0.23254650533199311,
      "num_tokens": 109289994.0,
      "step": 63015
    },
    {
      "entropy": 5.418342256546021,
      "epoch": 4.903170589379498,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002846476089051204,
      "loss": 4.7806,
      "mean_token_accuracy": 0.22455807477235795,
      "num_tokens": 109299088.0,
      "step": 63020
    },
    {
      "entropy": 5.41381869316101,
      "epoch": 4.903559618751216,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028461991873426494,
      "loss": 4.8353,
      "mean_token_accuracy": 0.22176656424999236,
      "num_tokens": 109308375.0,
      "step": 63025
    },
    {
      "entropy": 5.414546537399292,
      "epoch": 4.903948648122933,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002845922284174426,
      "loss": 4.747,
      "mean_token_accuracy": 0.22666281312704087,
      "num_tokens": 109317255.0,
      "step": 63030
    },
    {
      "entropy": 5.475347232818604,
      "epoch": 4.9043376774946505,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028456453795507335,
      "loss": 4.7772,
      "mean_token_accuracy": 0.22808314859867096,
      "num_tokens": 109325960.0,
      "step": 63035
    },
    {
      "entropy": 5.375293445587158,
      "epoch": 4.904726706866368,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002845368473475776,
      "loss": 4.6782,
      "mean_token_accuracy": 0.22865275144577027,
      "num_tokens": 109334629.0,
      "step": 63040
    },
    {
      "entropy": 5.486982822418213,
      "epoch": 4.905115736238086,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028450915659537524,
      "loss": 4.846,
      "mean_token_accuracy": 0.21578042060136796,
      "num_tokens": 109342899.0,
      "step": 63045
    },
    {
      "entropy": 5.283858728408814,
      "epoch": 4.905504765609804,
      "grad_norm": 1.125,
      "learning_rate": 0.0002844814656988866,
      "loss": 4.6175,
      "mean_token_accuracy": 0.24462829083204268,
      "num_tokens": 109351288.0,
      "step": 63050
    },
    {
      "entropy": 5.38515830039978,
      "epoch": 4.905893794981521,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028445377465853176,
      "loss": 4.8476,
      "mean_token_accuracy": 0.21698755770921707,
      "num_tokens": 109360131.0,
      "step": 63055
    },
    {
      "entropy": 5.445740222930908,
      "epoch": 4.906282824353239,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028442608347473093,
      "loss": 4.7859,
      "mean_token_accuracy": 0.2304176777601242,
      "num_tokens": 109368921.0,
      "step": 63060
    },
    {
      "entropy": 5.477333879470825,
      "epoch": 4.906671853724957,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028439839214790437,
      "loss": 4.7765,
      "mean_token_accuracy": 0.223940509557724,
      "num_tokens": 109377027.0,
      "step": 63065
    },
    {
      "entropy": 5.42976746559143,
      "epoch": 4.9070608830966735,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002843707006784722,
      "loss": 4.7029,
      "mean_token_accuracy": 0.23074968606233598,
      "num_tokens": 109385320.0,
      "step": 63070
    },
    {
      "entropy": 5.420126008987427,
      "epoch": 4.907449912468391,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002843430090668545,
      "loss": 4.8003,
      "mean_token_accuracy": 0.22134614288806914,
      "num_tokens": 109394365.0,
      "step": 63075
    },
    {
      "entropy": 5.411204528808594,
      "epoch": 4.907838941840109,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028431531731347155,
      "loss": 4.8139,
      "mean_token_accuracy": 0.2219046652317047,
      "num_tokens": 109402901.0,
      "step": 63080
    },
    {
      "entropy": 5.429065942764282,
      "epoch": 4.908227971211827,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002842876254187434,
      "loss": 4.8403,
      "mean_token_accuracy": 0.22110950499773024,
      "num_tokens": 109411490.0,
      "step": 63085
    },
    {
      "entropy": 5.405127429962159,
      "epoch": 4.908617000583544,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028425993338309036,
      "loss": 4.8539,
      "mean_token_accuracy": 0.21818794310092926,
      "num_tokens": 109421415.0,
      "step": 63090
    },
    {
      "entropy": 5.434456634521484,
      "epoch": 4.909006029955261,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002842322412069325,
      "loss": 4.8266,
      "mean_token_accuracy": 0.22471297681331634,
      "num_tokens": 109429845.0,
      "step": 63095
    },
    {
      "entropy": 5.458617067337036,
      "epoch": 4.909395059326979,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002842045488906902,
      "loss": 4.7846,
      "mean_token_accuracy": 0.22964616864919662,
      "num_tokens": 109438850.0,
      "step": 63100
    },
    {
      "entropy": 5.425916147232056,
      "epoch": 4.9097840886986965,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028417685643478347,
      "loss": 4.8063,
      "mean_token_accuracy": 0.22048405557870865,
      "num_tokens": 109447317.0,
      "step": 63105
    },
    {
      "entropy": 5.374573802947998,
      "epoch": 4.910173118070414,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028414916383963265,
      "loss": 4.7563,
      "mean_token_accuracy": 0.22972544133663178,
      "num_tokens": 109455404.0,
      "step": 63110
    },
    {
      "entropy": 5.308570671081543,
      "epoch": 4.910562147442132,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002841214711056577,
      "loss": 4.6981,
      "mean_token_accuracy": 0.22467648684978486,
      "num_tokens": 109463208.0,
      "step": 63115
    },
    {
      "entropy": 5.3775146484375,
      "epoch": 4.91095117681385,
      "grad_norm": 1.171875,
      "learning_rate": 0.000284093778233279,
      "loss": 4.7646,
      "mean_token_accuracy": 0.22987503856420516,
      "num_tokens": 109471981.0,
      "step": 63120
    },
    {
      "entropy": 5.436327791213989,
      "epoch": 4.911340206185567,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002840660852229166,
      "loss": 4.7736,
      "mean_token_accuracy": 0.2208330363035202,
      "num_tokens": 109480968.0,
      "step": 63125
    },
    {
      "entropy": 5.378834342956543,
      "epoch": 4.911729235557285,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002840383920749908,
      "loss": 4.871,
      "mean_token_accuracy": 0.20854863971471788,
      "num_tokens": 109489628.0,
      "step": 63130
    },
    {
      "entropy": 5.401764345169068,
      "epoch": 4.912118264929002,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002840106987899219,
      "loss": 4.7309,
      "mean_token_accuracy": 0.22283015102148057,
      "num_tokens": 109498293.0,
      "step": 63135
    },
    {
      "entropy": 5.357742023468018,
      "epoch": 4.9125072943007195,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002839830053681299,
      "loss": 4.6937,
      "mean_token_accuracy": 0.22523528933525086,
      "num_tokens": 109506589.0,
      "step": 63140
    },
    {
      "entropy": 5.398295211791992,
      "epoch": 4.912896323672437,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028395531181003516,
      "loss": 4.834,
      "mean_token_accuracy": 0.2291323721408844,
      "num_tokens": 109514838.0,
      "step": 63145
    },
    {
      "entropy": 5.374823474884034,
      "epoch": 4.913285353044155,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002839276181160577,
      "loss": 4.7791,
      "mean_token_accuracy": 0.22140555828809738,
      "num_tokens": 109523532.0,
      "step": 63150
    },
    {
      "entropy": 5.411004114151001,
      "epoch": 4.913674382415873,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000283899924286618,
      "loss": 4.8917,
      "mean_token_accuracy": 0.21481716632843018,
      "num_tokens": 109531851.0,
      "step": 63155
    },
    {
      "entropy": 5.428954362869263,
      "epoch": 4.91406341178759,
      "grad_norm": 1.203125,
      "learning_rate": 0.000283872230322136,
      "loss": 4.7585,
      "mean_token_accuracy": 0.23031252324581147,
      "num_tokens": 109540375.0,
      "step": 63160
    },
    {
      "entropy": 5.432012414932251,
      "epoch": 4.914452441159307,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028384453622303207,
      "loss": 4.7348,
      "mean_token_accuracy": 0.22383230477571486,
      "num_tokens": 109550174.0,
      "step": 63165
    },
    {
      "entropy": 5.3848670482635494,
      "epoch": 4.914841470531025,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028381684198972636,
      "loss": 4.7447,
      "mean_token_accuracy": 0.2323377773165703,
      "num_tokens": 109559261.0,
      "step": 63170
    },
    {
      "entropy": 5.419246673583984,
      "epoch": 4.9152304999027425,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028378914762263905,
      "loss": 4.9395,
      "mean_token_accuracy": 0.21558811366558076,
      "num_tokens": 109568314.0,
      "step": 63175
    },
    {
      "entropy": 5.3990076065063475,
      "epoch": 4.91561952927446,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028376145312219054,
      "loss": 4.8196,
      "mean_token_accuracy": 0.22069069147109985,
      "num_tokens": 109576795.0,
      "step": 63180
    },
    {
      "entropy": 5.4721112728118895,
      "epoch": 4.916008558646178,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002837337584888009,
      "loss": 4.7383,
      "mean_token_accuracy": 0.22581649124622344,
      "num_tokens": 109585292.0,
      "step": 63185
    },
    {
      "entropy": 5.390499877929687,
      "epoch": 4.916397588017896,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002837060637228903,
      "loss": 4.7338,
      "mean_token_accuracy": 0.2259834885597229,
      "num_tokens": 109594133.0,
      "step": 63190
    },
    {
      "entropy": 5.378627443313599,
      "epoch": 4.916786617389613,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028367836882487905,
      "loss": 4.8793,
      "mean_token_accuracy": 0.21766443848609923,
      "num_tokens": 109603638.0,
      "step": 63195
    },
    {
      "entropy": 5.318688535690308,
      "epoch": 4.91717564676133,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002836506737951875,
      "loss": 4.7023,
      "mean_token_accuracy": 0.22679270803928375,
      "num_tokens": 109612199.0,
      "step": 63200
    },
    {
      "entropy": 5.40428352355957,
      "epoch": 4.917564676133048,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002836229786342356,
      "loss": 4.7421,
      "mean_token_accuracy": 0.2264079213142395,
      "num_tokens": 109620655.0,
      "step": 63205
    },
    {
      "entropy": 5.466332817077637,
      "epoch": 4.9179537055047655,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002835952833424438,
      "loss": 4.8725,
      "mean_token_accuracy": 0.21746882945299148,
      "num_tokens": 109630150.0,
      "step": 63210
    },
    {
      "entropy": 5.447049951553344,
      "epoch": 4.918342734876483,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028356758792023217,
      "loss": 4.8085,
      "mean_token_accuracy": 0.22153561115264891,
      "num_tokens": 109638884.0,
      "step": 63215
    },
    {
      "entropy": 5.408453130722046,
      "epoch": 4.918731764248201,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028353989236802114,
      "loss": 4.733,
      "mean_token_accuracy": 0.22793289572000502,
      "num_tokens": 109647549.0,
      "step": 63220
    },
    {
      "entropy": 5.416606760025024,
      "epoch": 4.919120793619919,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028351219668623076,
      "loss": 4.7984,
      "mean_token_accuracy": 0.22233612686395646,
      "num_tokens": 109656302.0,
      "step": 63225
    },
    {
      "entropy": 5.3860808372497555,
      "epoch": 4.919509822991635,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002834845008752813,
      "loss": 4.8402,
      "mean_token_accuracy": 0.21991758942604064,
      "num_tokens": 109665219.0,
      "step": 63230
    },
    {
      "entropy": 5.330677938461304,
      "epoch": 4.919898852363353,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028345680493559313,
      "loss": 4.7249,
      "mean_token_accuracy": 0.22920001745224,
      "num_tokens": 109674001.0,
      "step": 63235
    },
    {
      "entropy": 5.373510408401489,
      "epoch": 4.920287881735071,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002834291088675864,
      "loss": 4.7253,
      "mean_token_accuracy": 0.22384131997823714,
      "num_tokens": 109682673.0,
      "step": 63240
    },
    {
      "entropy": 5.452417087554932,
      "epoch": 4.9206769111067885,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028340141267168135,
      "loss": 4.8224,
      "mean_token_accuracy": 0.22424065619707106,
      "num_tokens": 109692004.0,
      "step": 63245
    },
    {
      "entropy": 5.512565422058105,
      "epoch": 4.921065940478506,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028337371634829824,
      "loss": 4.8339,
      "mean_token_accuracy": 0.21913688778877258,
      "num_tokens": 109701018.0,
      "step": 63250
    },
    {
      "entropy": 5.400132465362549,
      "epoch": 4.921454969850224,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002833460198978573,
      "loss": 4.7341,
      "mean_token_accuracy": 0.22517583668231964,
      "num_tokens": 109708955.0,
      "step": 63255
    },
    {
      "entropy": 5.377708530426025,
      "epoch": 4.921843999221942,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002833183233207788,
      "loss": 4.7994,
      "mean_token_accuracy": 0.22168483287096025,
      "num_tokens": 109717343.0,
      "step": 63260
    },
    {
      "entropy": 5.35015721321106,
      "epoch": 4.922233028593659,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028329062661748295,
      "loss": 4.7214,
      "mean_token_accuracy": 0.22882865220308304,
      "num_tokens": 109725629.0,
      "step": 63265
    },
    {
      "entropy": 5.323612308502197,
      "epoch": 4.922622057965376,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028326292978839005,
      "loss": 4.736,
      "mean_token_accuracy": 0.2307807207107544,
      "num_tokens": 109734760.0,
      "step": 63270
    },
    {
      "entropy": 5.42139310836792,
      "epoch": 4.923011087337094,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028323523283392034,
      "loss": 4.746,
      "mean_token_accuracy": 0.22891620248556138,
      "num_tokens": 109743570.0,
      "step": 63275
    },
    {
      "entropy": 5.383662843704224,
      "epoch": 4.9234001167088115,
      "grad_norm": 1.25,
      "learning_rate": 0.00028320753575449415,
      "loss": 4.7597,
      "mean_token_accuracy": 0.22544941008090974,
      "num_tokens": 109752619.0,
      "step": 63280
    },
    {
      "entropy": 5.346390676498413,
      "epoch": 4.923789146080529,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002831798385505316,
      "loss": 4.7184,
      "mean_token_accuracy": 0.22730448693037034,
      "num_tokens": 109761137.0,
      "step": 63285
    },
    {
      "entropy": 5.361239814758301,
      "epoch": 4.924178175452247,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028315214122245295,
      "loss": 4.7325,
      "mean_token_accuracy": 0.2257252424955368,
      "num_tokens": 109769290.0,
      "step": 63290
    },
    {
      "entropy": 5.340511083602905,
      "epoch": 4.924567204823965,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002831244437706786,
      "loss": 4.7457,
      "mean_token_accuracy": 0.2321557268500328,
      "num_tokens": 109777906.0,
      "step": 63295
    },
    {
      "entropy": 5.303054475784302,
      "epoch": 4.924956234195681,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002830967461956288,
      "loss": 4.6952,
      "mean_token_accuracy": 0.2333715409040451,
      "num_tokens": 109786649.0,
      "step": 63300
    },
    {
      "entropy": 5.364556455612183,
      "epoch": 4.925345263567399,
      "grad_norm": 1.25,
      "learning_rate": 0.00028306904849772374,
      "loss": 4.6962,
      "mean_token_accuracy": 0.23071868866682052,
      "num_tokens": 109795183.0,
      "step": 63305
    },
    {
      "entropy": 5.304232883453369,
      "epoch": 4.925734292939117,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002830413506773837,
      "loss": 4.6737,
      "mean_token_accuracy": 0.22667007893323898,
      "num_tokens": 109804062.0,
      "step": 63310
    },
    {
      "entropy": 5.381059074401856,
      "epoch": 4.9261233223108345,
      "grad_norm": 1.1875,
      "learning_rate": 0.000283013652735029,
      "loss": 4.8327,
      "mean_token_accuracy": 0.21904452592134477,
      "num_tokens": 109813258.0,
      "step": 63315
    },
    {
      "entropy": 5.389257574081421,
      "epoch": 4.926512351682552,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028298595467107985,
      "loss": 4.74,
      "mean_token_accuracy": 0.2296806275844574,
      "num_tokens": 109821772.0,
      "step": 63320
    },
    {
      "entropy": 5.524547147750854,
      "epoch": 4.92690138105427,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002829582564859565,
      "loss": 4.9333,
      "mean_token_accuracy": 0.21344909220933914,
      "num_tokens": 109830204.0,
      "step": 63325
    },
    {
      "entropy": 5.356841373443603,
      "epoch": 4.927290410425988,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002829305581800793,
      "loss": 4.7627,
      "mean_token_accuracy": 0.2246478393673897,
      "num_tokens": 109838128.0,
      "step": 63330
    },
    {
      "entropy": 5.35327115058899,
      "epoch": 4.927679439797704,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002829028597538685,
      "loss": 4.8031,
      "mean_token_accuracy": 0.22021936923265456,
      "num_tokens": 109846887.0,
      "step": 63335
    },
    {
      "entropy": 5.404262256622315,
      "epoch": 4.928068469169422,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028287516120774445,
      "loss": 4.7697,
      "mean_token_accuracy": 0.23114107996225358,
      "num_tokens": 109855980.0,
      "step": 63340
    },
    {
      "entropy": 5.391396856307983,
      "epoch": 4.92845749854114,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002828474625421273,
      "loss": 4.6779,
      "mean_token_accuracy": 0.2259627491235733,
      "num_tokens": 109864568.0,
      "step": 63345
    },
    {
      "entropy": 5.436896657943725,
      "epoch": 4.9288465279128575,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002828197637574374,
      "loss": 4.8261,
      "mean_token_accuracy": 0.22185804396867753,
      "num_tokens": 109873351.0,
      "step": 63350
    },
    {
      "entropy": 5.42801570892334,
      "epoch": 4.929235557284575,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028279206485409505,
      "loss": 4.7474,
      "mean_token_accuracy": 0.2223198816180229,
      "num_tokens": 109882221.0,
      "step": 63355
    },
    {
      "entropy": 5.454281806945801,
      "epoch": 4.929624586656293,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002827643658325205,
      "loss": 4.8162,
      "mean_token_accuracy": 0.2198670208454132,
      "num_tokens": 109890701.0,
      "step": 63360
    },
    {
      "entropy": 5.461422252655029,
      "epoch": 4.93001361602801,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000282736666693134,
      "loss": 4.7902,
      "mean_token_accuracy": 0.22446550875902177,
      "num_tokens": 109899201.0,
      "step": 63365
    },
    {
      "entropy": 5.387375736236573,
      "epoch": 4.930402645399727,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002827089674363559,
      "loss": 4.6903,
      "mean_token_accuracy": 0.23478329330682754,
      "num_tokens": 109907134.0,
      "step": 63370
    },
    {
      "entropy": 5.383121061325073,
      "epoch": 4.930791674771445,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002826812680626066,
      "loss": 4.6772,
      "mean_token_accuracy": 0.23114899843931197,
      "num_tokens": 109915317.0,
      "step": 63375
    },
    {
      "entropy": 5.437287521362305,
      "epoch": 4.931180704143163,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002826535685723062,
      "loss": 4.8874,
      "mean_token_accuracy": 0.21267596632242203,
      "num_tokens": 109924408.0,
      "step": 63380
    },
    {
      "entropy": 5.421556806564331,
      "epoch": 4.9315697335148805,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028262586896587517,
      "loss": 4.8011,
      "mean_token_accuracy": 0.23014312833547593,
      "num_tokens": 109933138.0,
      "step": 63385
    },
    {
      "entropy": 5.487618780136108,
      "epoch": 4.931958762886598,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002825981692437336,
      "loss": 4.8723,
      "mean_token_accuracy": 0.2169387087225914,
      "num_tokens": 109941857.0,
      "step": 63390
    },
    {
      "entropy": 5.34007158279419,
      "epoch": 4.932347792258316,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028257046940630197,
      "loss": 4.7738,
      "mean_token_accuracy": 0.22501509189605712,
      "num_tokens": 109950676.0,
      "step": 63395
    },
    {
      "entropy": 5.35067343711853,
      "epoch": 4.932736821630033,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028254276945400056,
      "loss": 4.6749,
      "mean_token_accuracy": 0.2400803431868553,
      "num_tokens": 109959326.0,
      "step": 63400
    },
    {
      "entropy": 5.411648082733154,
      "epoch": 4.93312585100175,
      "grad_norm": 1.328125,
      "learning_rate": 0.00028251506938724953,
      "loss": 4.7474,
      "mean_token_accuracy": 0.23206363320350648,
      "num_tokens": 109967369.0,
      "step": 63405
    },
    {
      "entropy": 5.410942077636719,
      "epoch": 4.933514880373468,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028248736920646927,
      "loss": 4.7733,
      "mean_token_accuracy": 0.2258870631456375,
      "num_tokens": 109975849.0,
      "step": 63410
    },
    {
      "entropy": 5.458699655532837,
      "epoch": 4.933903909745186,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028245966891208017,
      "loss": 4.7707,
      "mean_token_accuracy": 0.22098077237606048,
      "num_tokens": 109985042.0,
      "step": 63415
    },
    {
      "entropy": 5.419143438339233,
      "epoch": 4.9342929391169035,
      "grad_norm": 1.25,
      "learning_rate": 0.0002824319685045025,
      "loss": 4.7666,
      "mean_token_accuracy": 0.21788403689861296,
      "num_tokens": 109994330.0,
      "step": 63420
    },
    {
      "entropy": 5.385769701004028,
      "epoch": 4.934681968488621,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028240426798415645,
      "loss": 4.7273,
      "mean_token_accuracy": 0.22928309440612793,
      "num_tokens": 110003565.0,
      "step": 63425
    },
    {
      "entropy": 5.284177970886231,
      "epoch": 4.935070997860338,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002823765673514625,
      "loss": 4.6803,
      "mean_token_accuracy": 0.23430462181568146,
      "num_tokens": 110012501.0,
      "step": 63430
    },
    {
      "entropy": 5.358765506744385,
      "epoch": 4.935460027232056,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028234886660684084,
      "loss": 4.7839,
      "mean_token_accuracy": 0.22104683220386506,
      "num_tokens": 110020831.0,
      "step": 63435
    },
    {
      "entropy": 5.44643726348877,
      "epoch": 4.935849056603773,
      "grad_norm": 1.125,
      "learning_rate": 0.0002823211657507118,
      "loss": 4.8673,
      "mean_token_accuracy": 0.21490197479724885,
      "num_tokens": 110030838.0,
      "step": 63440
    },
    {
      "entropy": 5.378885412216187,
      "epoch": 4.936238085975491,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002822934647834958,
      "loss": 4.7103,
      "mean_token_accuracy": 0.23360484689474106,
      "num_tokens": 110040178.0,
      "step": 63445
    },
    {
      "entropy": 5.391190004348755,
      "epoch": 4.936627115347209,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028226576370561304,
      "loss": 4.7704,
      "mean_token_accuracy": 0.22496458292007446,
      "num_tokens": 110048838.0,
      "step": 63450
    },
    {
      "entropy": 5.379943323135376,
      "epoch": 4.9370161447189265,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028223806251748384,
      "loss": 4.7434,
      "mean_token_accuracy": 0.23784064501523972,
      "num_tokens": 110057140.0,
      "step": 63455
    },
    {
      "entropy": 5.432900905609131,
      "epoch": 4.937405174090644,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002822103612195286,
      "loss": 4.7125,
      "mean_token_accuracy": 0.23619811981916428,
      "num_tokens": 110065294.0,
      "step": 63460
    },
    {
      "entropy": 5.4168439388275145,
      "epoch": 4.937794203462362,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002821826598121676,
      "loss": 4.8214,
      "mean_token_accuracy": 0.22423195838928223,
      "num_tokens": 110073810.0,
      "step": 63465
    },
    {
      "entropy": 5.415780925750733,
      "epoch": 4.938183232834079,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002821549582958212,
      "loss": 4.8079,
      "mean_token_accuracy": 0.21808325201272966,
      "num_tokens": 110082446.0,
      "step": 63470
    },
    {
      "entropy": 5.405220174789429,
      "epoch": 4.938572262205796,
      "grad_norm": 1.125,
      "learning_rate": 0.0002821272566709096,
      "loss": 4.7684,
      "mean_token_accuracy": 0.23130155950784684,
      "num_tokens": 110090815.0,
      "step": 63475
    },
    {
      "entropy": 5.387920665740967,
      "epoch": 4.938961291577514,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028209955493785333,
      "loss": 4.7436,
      "mean_token_accuracy": 0.21858742386102675,
      "num_tokens": 110100295.0,
      "step": 63480
    },
    {
      "entropy": 5.433146810531616,
      "epoch": 4.939350320949232,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002820718530970726,
      "loss": 4.8315,
      "mean_token_accuracy": 0.21839747726917266,
      "num_tokens": 110109194.0,
      "step": 63485
    },
    {
      "entropy": 5.403715562820435,
      "epoch": 4.9397393503209495,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028204415114898765,
      "loss": 4.7963,
      "mean_token_accuracy": 0.21995267868041993,
      "num_tokens": 110117554.0,
      "step": 63490
    },
    {
      "entropy": 5.413893556594848,
      "epoch": 4.940128379692667,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028201644909401903,
      "loss": 4.8388,
      "mean_token_accuracy": 0.22244620472192764,
      "num_tokens": 110126166.0,
      "step": 63495
    },
    {
      "entropy": 5.428742980957031,
      "epoch": 4.940517409064384,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028198874693258696,
      "loss": 4.7995,
      "mean_token_accuracy": 0.22940473556518554,
      "num_tokens": 110134517.0,
      "step": 63500
    },
    {
      "entropy": 5.3981846332550045,
      "epoch": 4.940906438436102,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002819610446651117,
      "loss": 4.7358,
      "mean_token_accuracy": 0.22687412798404694,
      "num_tokens": 110143172.0,
      "step": 63505
    },
    {
      "entropy": 5.408468341827392,
      "epoch": 4.941295467807819,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028193334229201374,
      "loss": 4.8155,
      "mean_token_accuracy": 0.21854631304740907,
      "num_tokens": 110152871.0,
      "step": 63510
    },
    {
      "entropy": 5.419909286499023,
      "epoch": 4.941684497179537,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028190563981371334,
      "loss": 4.7163,
      "mean_token_accuracy": 0.22186665683984758,
      "num_tokens": 110160993.0,
      "step": 63515
    },
    {
      "entropy": 5.44317364692688,
      "epoch": 4.942073526551255,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002818779372306308,
      "loss": 4.7379,
      "mean_token_accuracy": 0.22576926052570342,
      "num_tokens": 110169701.0,
      "step": 63520
    },
    {
      "entropy": 5.428468179702759,
      "epoch": 4.9424625559229725,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028185023454318646,
      "loss": 4.8236,
      "mean_token_accuracy": 0.21897818446159362,
      "num_tokens": 110178354.0,
      "step": 63525
    },
    {
      "entropy": 5.3757377624511715,
      "epoch": 4.94285158529469,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028182253175180084,
      "loss": 4.758,
      "mean_token_accuracy": 0.2247410848736763,
      "num_tokens": 110186865.0,
      "step": 63530
    },
    {
      "entropy": 5.425506734848023,
      "epoch": 4.943240614666407,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002817948288568942,
      "loss": 4.7999,
      "mean_token_accuracy": 0.22424442172050477,
      "num_tokens": 110195333.0,
      "step": 63535
    },
    {
      "entropy": 5.282755708694458,
      "epoch": 4.943629644038125,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002817671258588868,
      "loss": 4.5886,
      "mean_token_accuracy": 0.23865402936935426,
      "num_tokens": 110204375.0,
      "step": 63540
    },
    {
      "entropy": 5.323008966445923,
      "epoch": 4.944018673409842,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002817394227581989,
      "loss": 4.7516,
      "mean_token_accuracy": 0.23320356011390686,
      "num_tokens": 110212833.0,
      "step": 63545
    },
    {
      "entropy": 5.3404326915740965,
      "epoch": 4.94440770278156,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002817117195552511,
      "loss": 4.6994,
      "mean_token_accuracy": 0.22850479781627656,
      "num_tokens": 110220667.0,
      "step": 63550
    },
    {
      "entropy": 5.325035810470581,
      "epoch": 4.944796732153278,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028168401625046363,
      "loss": 4.5693,
      "mean_token_accuracy": 0.2424585700035095,
      "num_tokens": 110228981.0,
      "step": 63555
    },
    {
      "entropy": 5.337361240386963,
      "epoch": 4.9451857615249954,
      "grad_norm": 1.125,
      "learning_rate": 0.0002816563128442568,
      "loss": 4.7297,
      "mean_token_accuracy": 0.2248943939805031,
      "num_tokens": 110237449.0,
      "step": 63560
    },
    {
      "entropy": 5.425177335739136,
      "epoch": 4.945574790896712,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028162860933705105,
      "loss": 4.7893,
      "mean_token_accuracy": 0.2260083809494972,
      "num_tokens": 110246175.0,
      "step": 63565
    },
    {
      "entropy": 5.375520515441894,
      "epoch": 4.94596382026843,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002816009057292667,
      "loss": 4.724,
      "mean_token_accuracy": 0.23735830783843995,
      "num_tokens": 110255381.0,
      "step": 63570
    },
    {
      "entropy": 5.460589981079101,
      "epoch": 4.946352849640148,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028157320202132417,
      "loss": 4.798,
      "mean_token_accuracy": 0.22056447267532348,
      "num_tokens": 110264508.0,
      "step": 63575
    },
    {
      "entropy": 5.4892627716064455,
      "epoch": 4.946741879011865,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028154549821364375,
      "loss": 4.92,
      "mean_token_accuracy": 0.2124575361609459,
      "num_tokens": 110273542.0,
      "step": 63580
    },
    {
      "entropy": 5.3830235481262205,
      "epoch": 4.947130908383583,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028151779430664575,
      "loss": 4.7575,
      "mean_token_accuracy": 0.22668575644493102,
      "num_tokens": 110282310.0,
      "step": 63585
    },
    {
      "entropy": 5.337215709686279,
      "epoch": 4.947519937755301,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028149009030075056,
      "loss": 4.6841,
      "mean_token_accuracy": 0.23601720184087754,
      "num_tokens": 110290596.0,
      "step": 63590
    },
    {
      "entropy": 5.356709146499634,
      "epoch": 4.947908967127018,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028146238619637865,
      "loss": 4.6437,
      "mean_token_accuracy": 0.23606915920972824,
      "num_tokens": 110298520.0,
      "step": 63595
    },
    {
      "entropy": 5.385938024520874,
      "epoch": 4.948297996498736,
      "grad_norm": 1.25,
      "learning_rate": 0.00028143468199395036,
      "loss": 4.8028,
      "mean_token_accuracy": 0.22657362520694732,
      "num_tokens": 110307615.0,
      "step": 63600
    },
    {
      "entropy": 5.378319168090821,
      "epoch": 4.948687025870453,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028140697769388597,
      "loss": 4.7325,
      "mean_token_accuracy": 0.23165423274040223,
      "num_tokens": 110316166.0,
      "step": 63605
    },
    {
      "entropy": 5.394776153564453,
      "epoch": 4.949076055242171,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002813792732966059,
      "loss": 4.6605,
      "mean_token_accuracy": 0.23094343394041061,
      "num_tokens": 110324562.0,
      "step": 63610
    },
    {
      "entropy": 5.39465446472168,
      "epoch": 4.949465084613888,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002813515688025305,
      "loss": 4.8263,
      "mean_token_accuracy": 0.21294835358858108,
      "num_tokens": 110332195.0,
      "step": 63615
    },
    {
      "entropy": 5.311608791351318,
      "epoch": 4.949854113985606,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002813238642120802,
      "loss": 4.681,
      "mean_token_accuracy": 0.23552799075841904,
      "num_tokens": 110341360.0,
      "step": 63620
    },
    {
      "entropy": 5.409696006774903,
      "epoch": 4.950243143357324,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028129615952567517,
      "loss": 4.7787,
      "mean_token_accuracy": 0.22397578954696656,
      "num_tokens": 110349609.0,
      "step": 63625
    },
    {
      "entropy": 5.386757469177246,
      "epoch": 4.9506321727290405,
      "grad_norm": 1.25,
      "learning_rate": 0.0002812684547437361,
      "loss": 4.7645,
      "mean_token_accuracy": 0.22112909257411956,
      "num_tokens": 110357703.0,
      "step": 63630
    },
    {
      "entropy": 5.429359197616577,
      "epoch": 4.951021202100758,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002812407498666831,
      "loss": 4.7438,
      "mean_token_accuracy": 0.22850308567285538,
      "num_tokens": 110365822.0,
      "step": 63635
    },
    {
      "entropy": 5.337197113037109,
      "epoch": 4.951410231472476,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002812130448949367,
      "loss": 4.7528,
      "mean_token_accuracy": 0.22177010774612427,
      "num_tokens": 110375352.0,
      "step": 63640
    },
    {
      "entropy": 5.358320236206055,
      "epoch": 4.951799260844194,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028118533982891727,
      "loss": 4.6219,
      "mean_token_accuracy": 0.22965431213378906,
      "num_tokens": 110383616.0,
      "step": 63645
    },
    {
      "entropy": 5.44491491317749,
      "epoch": 4.952188290215911,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002811576346690451,
      "loss": 4.8532,
      "mean_token_accuracy": 0.21897365897893906,
      "num_tokens": 110392425.0,
      "step": 63650
    },
    {
      "entropy": 5.3238396644592285,
      "epoch": 4.952577319587629,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028112992941574074,
      "loss": 4.6472,
      "mean_token_accuracy": 0.2351665362715721,
      "num_tokens": 110400621.0,
      "step": 63655
    },
    {
      "entropy": 5.402384185791016,
      "epoch": 4.952966348959347,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028110222406942424,
      "loss": 4.8391,
      "mean_token_accuracy": 0.21989277452230455,
      "num_tokens": 110410123.0,
      "step": 63660
    },
    {
      "entropy": 5.413518714904785,
      "epoch": 4.953355378331064,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002810745186305164,
      "loss": 4.7862,
      "mean_token_accuracy": 0.21957554668188095,
      "num_tokens": 110418775.0,
      "step": 63665
    },
    {
      "entropy": 5.4003688335418705,
      "epoch": 4.953744407702781,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002810468130994373,
      "loss": 4.7881,
      "mean_token_accuracy": 0.21991329193115233,
      "num_tokens": 110428202.0,
      "step": 63670
    },
    {
      "entropy": 5.4698686599731445,
      "epoch": 4.954133437074499,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028101910747660745,
      "loss": 4.8182,
      "mean_token_accuracy": 0.21917980164289474,
      "num_tokens": 110437398.0,
      "step": 63675
    },
    {
      "entropy": 5.466535425186157,
      "epoch": 4.954522466446217,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002809914017624472,
      "loss": 4.8484,
      "mean_token_accuracy": 0.22227211445569992,
      "num_tokens": 110446366.0,
      "step": 63680
    },
    {
      "entropy": 5.522869729995728,
      "epoch": 4.954911495817934,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002809636959573771,
      "loss": 4.8576,
      "mean_token_accuracy": 0.2190161943435669,
      "num_tokens": 110454951.0,
      "step": 63685
    },
    {
      "entropy": 5.42654128074646,
      "epoch": 4.955300525189652,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002809359900618173,
      "loss": 4.7635,
      "mean_token_accuracy": 0.23206537812948227,
      "num_tokens": 110463331.0,
      "step": 63690
    },
    {
      "entropy": 5.493383550643921,
      "epoch": 4.95568955456137,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028090828407618837,
      "loss": 4.8741,
      "mean_token_accuracy": 0.21942332088947297,
      "num_tokens": 110472999.0,
      "step": 63695
    },
    {
      "entropy": 5.389278841018677,
      "epoch": 4.9560785839330865,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028088057800091054,
      "loss": 4.7786,
      "mean_token_accuracy": 0.22670643776655197,
      "num_tokens": 110482132.0,
      "step": 63700
    },
    {
      "entropy": 5.458418178558349,
      "epoch": 4.956467613304804,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002808528718364043,
      "loss": 4.8122,
      "mean_token_accuracy": 0.21769326478242873,
      "num_tokens": 110491402.0,
      "step": 63705
    },
    {
      "entropy": 5.4723974704742435,
      "epoch": 4.956856642676522,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028082516558309005,
      "loss": 4.7719,
      "mean_token_accuracy": 0.225288924574852,
      "num_tokens": 110500148.0,
      "step": 63710
    },
    {
      "entropy": 5.345379114151001,
      "epoch": 4.95724567204824,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002807974592413883,
      "loss": 4.713,
      "mean_token_accuracy": 0.23639550656080247,
      "num_tokens": 110508297.0,
      "step": 63715
    },
    {
      "entropy": 5.323996305465698,
      "epoch": 4.957634701419957,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028076975281171926,
      "loss": 4.7126,
      "mean_token_accuracy": 0.23683362156152726,
      "num_tokens": 110516428.0,
      "step": 63720
    },
    {
      "entropy": 5.350346612930298,
      "epoch": 4.958023730791675,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028074204629450334,
      "loss": 4.7615,
      "mean_token_accuracy": 0.22429319471120834,
      "num_tokens": 110525488.0,
      "step": 63725
    },
    {
      "entropy": 5.3793299198150635,
      "epoch": 4.958412760163393,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002807143396901611,
      "loss": 4.6642,
      "mean_token_accuracy": 0.2296088606119156,
      "num_tokens": 110533665.0,
      "step": 63730
    },
    {
      "entropy": 5.466081428527832,
      "epoch": 4.9588017895351095,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028068663299911286,
      "loss": 4.7907,
      "mean_token_accuracy": 0.22824931889772415,
      "num_tokens": 110541812.0,
      "step": 63735
    },
    {
      "entropy": 5.375364160537719,
      "epoch": 4.959190818906827,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002806589262217789,
      "loss": 4.688,
      "mean_token_accuracy": 0.23833245784044266,
      "num_tokens": 110550231.0,
      "step": 63740
    },
    {
      "entropy": 5.392983293533325,
      "epoch": 4.959579848278545,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028063121935857985,
      "loss": 4.7773,
      "mean_token_accuracy": 0.22600287497043609,
      "num_tokens": 110559192.0,
      "step": 63745
    },
    {
      "entropy": 5.362807989120483,
      "epoch": 4.959968877650263,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002806035124099359,
      "loss": 4.766,
      "mean_token_accuracy": 0.22804066985845567,
      "num_tokens": 110567998.0,
      "step": 63750
    },
    {
      "entropy": 5.344998073577881,
      "epoch": 4.96035790702198,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028057580537626766,
      "loss": 4.6868,
      "mean_token_accuracy": 0.22626933306455613,
      "num_tokens": 110576910.0,
      "step": 63755
    },
    {
      "entropy": 5.348239707946777,
      "epoch": 4.960746936393698,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028054809825799537,
      "loss": 4.6894,
      "mean_token_accuracy": 0.23627488166093827,
      "num_tokens": 110585574.0,
      "step": 63760
    },
    {
      "entropy": 5.428102254867554,
      "epoch": 4.961135965765415,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028052039105553965,
      "loss": 4.7711,
      "mean_token_accuracy": 0.22135983854532243,
      "num_tokens": 110593612.0,
      "step": 63765
    },
    {
      "entropy": 5.369566488265991,
      "epoch": 4.9615249951371325,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002804926837693207,
      "loss": 4.8402,
      "mean_token_accuracy": 0.22553509920835496,
      "num_tokens": 110601796.0,
      "step": 63770
    },
    {
      "entropy": 5.376843738555908,
      "epoch": 4.96191402450885,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000280464976399759,
      "loss": 4.7534,
      "mean_token_accuracy": 0.23063101172447203,
      "num_tokens": 110610104.0,
      "step": 63775
    },
    {
      "entropy": 5.406345319747925,
      "epoch": 4.962303053880568,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000280437268947275,
      "loss": 4.7028,
      "mean_token_accuracy": 0.23415306806564332,
      "num_tokens": 110618980.0,
      "step": 63780
    },
    {
      "entropy": 5.427701139450074,
      "epoch": 4.962692083252286,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002804095614122891,
      "loss": 4.8143,
      "mean_token_accuracy": 0.22397271394729615,
      "num_tokens": 110627804.0,
      "step": 63785
    },
    {
      "entropy": 5.402082347869873,
      "epoch": 4.963081112624003,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028038185379522176,
      "loss": 4.8311,
      "mean_token_accuracy": 0.22169987559318544,
      "num_tokens": 110636149.0,
      "step": 63790
    },
    {
      "entropy": 5.342545700073242,
      "epoch": 4.963470141995721,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028035414609649327,
      "loss": 4.695,
      "mean_token_accuracy": 0.23131568729877472,
      "num_tokens": 110644430.0,
      "step": 63795
    },
    {
      "entropy": 5.434930467605591,
      "epoch": 4.963859171367439,
      "grad_norm": 1.125,
      "learning_rate": 0.00028032643831652415,
      "loss": 4.8285,
      "mean_token_accuracy": 0.22235576212406158,
      "num_tokens": 110654451.0,
      "step": 63800
    },
    {
      "entropy": 5.488545560836792,
      "epoch": 4.9642482007391555,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002802987304557349,
      "loss": 4.7944,
      "mean_token_accuracy": 0.2286959633231163,
      "num_tokens": 110662519.0,
      "step": 63805
    },
    {
      "entropy": 5.457192277908325,
      "epoch": 4.964637230110873,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002802710225145458,
      "loss": 4.8624,
      "mean_token_accuracy": 0.22379558384418488,
      "num_tokens": 110670954.0,
      "step": 63810
    },
    {
      "entropy": 5.433698415756226,
      "epoch": 4.965026259482591,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028024331449337726,
      "loss": 4.8576,
      "mean_token_accuracy": 0.21783299148082733,
      "num_tokens": 110679628.0,
      "step": 63815
    },
    {
      "entropy": 5.408075189590454,
      "epoch": 4.965415288854309,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028021560639264985,
      "loss": 4.6989,
      "mean_token_accuracy": 0.23267098516225815,
      "num_tokens": 110688372.0,
      "step": 63820
    },
    {
      "entropy": 5.474450206756591,
      "epoch": 4.965804318226026,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002801878982127838,
      "loss": 4.7977,
      "mean_token_accuracy": 0.22783046811819077,
      "num_tokens": 110696567.0,
      "step": 63825
    },
    {
      "entropy": 5.383289051055908,
      "epoch": 4.966193347597744,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002801601899541998,
      "loss": 4.7597,
      "mean_token_accuracy": 0.22773262858390808,
      "num_tokens": 110705650.0,
      "step": 63830
    },
    {
      "entropy": 5.320143795013427,
      "epoch": 4.966582376969461,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000280132481617318,
      "loss": 4.7217,
      "mean_token_accuracy": 0.22731700539588928,
      "num_tokens": 110714793.0,
      "step": 63835
    },
    {
      "entropy": 5.315478277206421,
      "epoch": 4.9669714063411785,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028010477320255904,
      "loss": 4.6885,
      "mean_token_accuracy": 0.23174546658992767,
      "num_tokens": 110722914.0,
      "step": 63840
    },
    {
      "entropy": 5.431572914123535,
      "epoch": 4.967360435712896,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002800770647103433,
      "loss": 4.7829,
      "mean_token_accuracy": 0.23041789531707763,
      "num_tokens": 110731702.0,
      "step": 63845
    },
    {
      "entropy": 5.445021867752075,
      "epoch": 4.967749465084614,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002800493561410911,
      "loss": 4.7724,
      "mean_token_accuracy": 0.23257410973310472,
      "num_tokens": 110740252.0,
      "step": 63850
    },
    {
      "entropy": 5.458088731765747,
      "epoch": 4.968138494456332,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000280021647495223,
      "loss": 4.9148,
      "mean_token_accuracy": 0.21065183877944946,
      "num_tokens": 110750369.0,
      "step": 63855
    },
    {
      "entropy": 5.544083547592163,
      "epoch": 4.968527523828049,
      "grad_norm": 1.203125,
      "learning_rate": 0.00027999393877315944,
      "loss": 4.9339,
      "mean_token_accuracy": 0.20953608602285384,
      "num_tokens": 110759149.0,
      "step": 63860
    },
    {
      "entropy": 5.509131526947021,
      "epoch": 4.968916553199767,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002799662299753207,
      "loss": 4.8083,
      "mean_token_accuracy": 0.22567250579595566,
      "num_tokens": 110767438.0,
      "step": 63865
    },
    {
      "entropy": 5.344177961349487,
      "epoch": 4.969305582571484,
      "grad_norm": 1.125,
      "learning_rate": 0.00027993852110212735,
      "loss": 4.7007,
      "mean_token_accuracy": 0.2332465097308159,
      "num_tokens": 110775061.0,
      "step": 63870
    },
    {
      "entropy": 5.462042331695557,
      "epoch": 4.9696946119432015,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002799108121539999,
      "loss": 4.9758,
      "mean_token_accuracy": 0.2123553514480591,
      "num_tokens": 110783596.0,
      "step": 63875
    },
    {
      "entropy": 5.412834358215332,
      "epoch": 4.970083641314919,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00027988310313135857,
      "loss": 4.7892,
      "mean_token_accuracy": 0.2268381506204605,
      "num_tokens": 110792897.0,
      "step": 63880
    },
    {
      "entropy": 5.4380754947662355,
      "epoch": 4.970472670686637,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00027985539403462407,
      "loss": 4.8223,
      "mean_token_accuracy": 0.22155838906764985,
      "num_tokens": 110801769.0,
      "step": 63885
    },
    {
      "entropy": 5.289813470840454,
      "epoch": 4.970861700058355,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027982768486421657,
      "loss": 4.6498,
      "mean_token_accuracy": 0.23127272576093674,
      "num_tokens": 110811170.0,
      "step": 63890
    },
    {
      "entropy": 5.434076118469238,
      "epoch": 4.971250729430072,
      "grad_norm": 1.125,
      "learning_rate": 0.00027979997562055663,
      "loss": 4.8518,
      "mean_token_accuracy": 0.21941933929920196,
      "num_tokens": 110819681.0,
      "step": 63895
    },
    {
      "entropy": 5.37553973197937,
      "epoch": 4.971639758801789,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00027977226630406476,
      "loss": 4.72,
      "mean_token_accuracy": 0.2285078153014183,
      "num_tokens": 110828362.0,
      "step": 63900
    },
    {
      "entropy": 5.4117316722869875,
      "epoch": 4.972028788173507,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002797445569151613,
      "loss": 4.796,
      "mean_token_accuracy": 0.22335676550865174,
      "num_tokens": 110836768.0,
      "step": 63905
    },
    {
      "entropy": 5.427364015579224,
      "epoch": 4.9724178175452245,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002797168474542668,
      "loss": 4.8445,
      "mean_token_accuracy": 0.20969960689544678,
      "num_tokens": 110846238.0,
      "step": 63910
    },
    {
      "entropy": 5.3649420738220215,
      "epoch": 4.972806846916942,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00027968913792180154,
      "loss": 4.7401,
      "mean_token_accuracy": 0.23378137946128846,
      "num_tokens": 110854845.0,
      "step": 63915
    },
    {
      "entropy": 5.3783773422241214,
      "epoch": 4.97319587628866,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002796614283181862,
      "loss": 4.749,
      "mean_token_accuracy": 0.22303928434848785,
      "num_tokens": 110863393.0,
      "step": 63920
    },
    {
      "entropy": 5.448866271972657,
      "epoch": 4.973584905660378,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000279633718643841,
      "loss": 4.8192,
      "mean_token_accuracy": 0.22450248748064042,
      "num_tokens": 110871963.0,
      "step": 63925
    },
    {
      "entropy": 5.376678848266602,
      "epoch": 4.973973935032095,
      "grad_norm": 1.109375,
      "learning_rate": 0.00027960600889918657,
      "loss": 4.6893,
      "mean_token_accuracy": 0.22881503254175187,
      "num_tokens": 110881080.0,
      "step": 63930
    },
    {
      "entropy": 5.460013008117675,
      "epoch": 4.974362964403813,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002795782990846433,
      "loss": 4.8095,
      "mean_token_accuracy": 0.22481611222028733,
      "num_tokens": 110890449.0,
      "step": 63935
    },
    {
      "entropy": 5.457606363296509,
      "epoch": 4.97475199377553,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002795505892006316,
      "loss": 4.8531,
      "mean_token_accuracy": 0.2138216257095337,
      "num_tokens": 110899559.0,
      "step": 63940
    },
    {
      "entropy": 5.377870750427246,
      "epoch": 4.9751410231472475,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002795228792475719,
      "loss": 4.7381,
      "mean_token_accuracy": 0.23463571816682816,
      "num_tokens": 110908499.0,
      "step": 63945
    },
    {
      "entropy": 5.403241443634033,
      "epoch": 4.975530052518965,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002794951692258848,
      "loss": 4.7575,
      "mean_token_accuracy": 0.22593780755996704,
      "num_tokens": 110917279.0,
      "step": 63950
    },
    {
      "entropy": 5.41656904220581,
      "epoch": 4.975919081890683,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002794674591359906,
      "loss": 4.7866,
      "mean_token_accuracy": 0.22414452135562896,
      "num_tokens": 110927126.0,
      "step": 63955
    },
    {
      "entropy": 5.39356164932251,
      "epoch": 4.976308111262401,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002794397489783098,
      "loss": 4.758,
      "mean_token_accuracy": 0.22097909599542617,
      "num_tokens": 110935850.0,
      "step": 63960
    },
    {
      "entropy": 5.390287113189697,
      "epoch": 4.976697140634117,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002794120387532629,
      "loss": 4.7317,
      "mean_token_accuracy": 0.2292109414935112,
      "num_tokens": 110944279.0,
      "step": 63965
    },
    {
      "entropy": 5.371587800979614,
      "epoch": 4.977086170005835,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002793843284612703,
      "loss": 4.7566,
      "mean_token_accuracy": 0.22940073758363724,
      "num_tokens": 110952354.0,
      "step": 63970
    },
    {
      "entropy": 5.351665830612182,
      "epoch": 4.977475199377553,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002793566181027526,
      "loss": 4.7677,
      "mean_token_accuracy": 0.2244884788990021,
      "num_tokens": 110961216.0,
      "step": 63975
    },
    {
      "entropy": 5.445906639099121,
      "epoch": 4.9778642287492705,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027932890767813005,
      "loss": 4.7774,
      "mean_token_accuracy": 0.22335783541202545,
      "num_tokens": 110969162.0,
      "step": 63980
    },
    {
      "entropy": 5.415161037445069,
      "epoch": 4.978253258120988,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002793011971878233,
      "loss": 4.7289,
      "mean_token_accuracy": 0.2312667429447174,
      "num_tokens": 110977746.0,
      "step": 63985
    },
    {
      "entropy": 5.355083465576172,
      "epoch": 4.978642287492706,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002792734866322526,
      "loss": 4.738,
      "mean_token_accuracy": 0.2325270652770996,
      "num_tokens": 110986387.0,
      "step": 63990
    },
    {
      "entropy": 5.412277460098267,
      "epoch": 4.979031316864424,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002792457760118387,
      "loss": 4.8192,
      "mean_token_accuracy": 0.22542303651571274,
      "num_tokens": 110995709.0,
      "step": 63995
    },
    {
      "entropy": 5.435282278060913,
      "epoch": 4.979420346236141,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027921806532700186,
      "loss": 4.8013,
      "mean_token_accuracy": 0.2206418216228485,
      "num_tokens": 111003825.0,
      "step": 64000
    },
    {
      "entropy": 5.418638563156128,
      "epoch": 4.979809375607858,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002791903545781625,
      "loss": 4.7931,
      "mean_token_accuracy": 0.22615060955286026,
      "num_tokens": 111013106.0,
      "step": 64005
    },
    {
      "entropy": 5.425638103485108,
      "epoch": 4.980198404979576,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027916264376574124,
      "loss": 4.8247,
      "mean_token_accuracy": 0.2186003014445305,
      "num_tokens": 111021755.0,
      "step": 64010
    },
    {
      "entropy": 5.394620418548584,
      "epoch": 4.9805874343512935,
      "grad_norm": 1.078125,
      "learning_rate": 0.00027913493289015857,
      "loss": 4.7222,
      "mean_token_accuracy": 0.2289113849401474,
      "num_tokens": 111030815.0,
      "step": 64015
    },
    {
      "entropy": 5.360451698303223,
      "epoch": 4.980976463723011,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00027910722195183474,
      "loss": 4.7518,
      "mean_token_accuracy": 0.23672803044319152,
      "num_tokens": 111039398.0,
      "step": 64020
    },
    {
      "entropy": 5.391369867324829,
      "epoch": 4.981365493094729,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002790795109511904,
      "loss": 4.7599,
      "mean_token_accuracy": 0.2260148212313652,
      "num_tokens": 111048075.0,
      "step": 64025
    },
    {
      "entropy": 5.439908981323242,
      "epoch": 4.981754522466447,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027905179988864593,
      "loss": 4.7714,
      "mean_token_accuracy": 0.23065318465232848,
      "num_tokens": 111056270.0,
      "step": 64030
    },
    {
      "entropy": 5.414389562606812,
      "epoch": 4.982143551838163,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002790240887646219,
      "loss": 4.7733,
      "mean_token_accuracy": 0.2269987791776657,
      "num_tokens": 111065416.0,
      "step": 64035
    },
    {
      "entropy": 5.423352384567261,
      "epoch": 4.982532581209881,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027899637757953876,
      "loss": 4.8406,
      "mean_token_accuracy": 0.2221213921904564,
      "num_tokens": 111073715.0,
      "step": 64040
    },
    {
      "entropy": 5.482796907424927,
      "epoch": 4.982921610581599,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027896866633381694,
      "loss": 4.885,
      "mean_token_accuracy": 0.21826411485671998,
      "num_tokens": 111082801.0,
      "step": 64045
    },
    {
      "entropy": 5.474845838546753,
      "epoch": 4.9833106399533165,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002789409550278769,
      "loss": 4.8251,
      "mean_token_accuracy": 0.21935138255357742,
      "num_tokens": 111091583.0,
      "step": 64050
    },
    {
      "entropy": 5.354418039321899,
      "epoch": 4.983699669325034,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002789132436621391,
      "loss": 4.6305,
      "mean_token_accuracy": 0.24013043195009232,
      "num_tokens": 111100008.0,
      "step": 64055
    },
    {
      "entropy": 5.389859437942505,
      "epoch": 4.984088698696752,
      "grad_norm": 1.125,
      "learning_rate": 0.0002788855322370241,
      "loss": 4.7594,
      "mean_token_accuracy": 0.22591807395219804,
      "num_tokens": 111108536.0,
      "step": 64060
    },
    {
      "entropy": 5.515747690200806,
      "epoch": 4.98447772806847,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027885782075295234,
      "loss": 4.8969,
      "mean_token_accuracy": 0.21636863946914672,
      "num_tokens": 111116866.0,
      "step": 64065
    },
    {
      "entropy": 5.418009614944458,
      "epoch": 4.984866757440186,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002788301092103443,
      "loss": 4.75,
      "mean_token_accuracy": 0.22806775718927383,
      "num_tokens": 111125385.0,
      "step": 64070
    },
    {
      "entropy": 5.411872625350952,
      "epoch": 4.985255786811904,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00027880239760962044,
      "loss": 4.8021,
      "mean_token_accuracy": 0.2322111964225769,
      "num_tokens": 111134548.0,
      "step": 64075
    },
    {
      "entropy": 5.351829576492309,
      "epoch": 4.985644816183622,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00027877468595120125,
      "loss": 4.7152,
      "mean_token_accuracy": 0.22233431935310363,
      "num_tokens": 111143082.0,
      "step": 64080
    },
    {
      "entropy": 5.4896769523620605,
      "epoch": 4.9860338455553395,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002787469742355072,
      "loss": 4.8748,
      "mean_token_accuracy": 0.22104825973510742,
      "num_tokens": 111152722.0,
      "step": 64085
    },
    {
      "entropy": 5.507826042175293,
      "epoch": 4.986422874927057,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002787192624629587,
      "loss": 4.9148,
      "mean_token_accuracy": 0.22559223920106888,
      "num_tokens": 111162416.0,
      "step": 64090
    },
    {
      "entropy": 5.4191066265106205,
      "epoch": 4.986811904298775,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002786915506339764,
      "loss": 4.7845,
      "mean_token_accuracy": 0.21721385568380355,
      "num_tokens": 111171501.0,
      "step": 64095
    },
    {
      "entropy": 5.452735757827758,
      "epoch": 4.987200933670492,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002786638387489807,
      "loss": 4.8333,
      "mean_token_accuracy": 0.2147175043821335,
      "num_tokens": 111180765.0,
      "step": 64100
    },
    {
      "entropy": 5.475985193252564,
      "epoch": 4.987589963042209,
      "grad_norm": 1.140625,
      "learning_rate": 0.00027863612680839204,
      "loss": 4.865,
      "mean_token_accuracy": 0.2151520788669586,
      "num_tokens": 111189477.0,
      "step": 64105
    },
    {
      "entropy": 5.442052268981934,
      "epoch": 4.987978992413927,
      "grad_norm": 1.125,
      "learning_rate": 0.000278608414812631,
      "loss": 4.7809,
      "mean_token_accuracy": 0.21815894693136215,
      "num_tokens": 111198239.0,
      "step": 64110
    },
    {
      "entropy": 5.365978479385376,
      "epoch": 4.988368021785645,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027858070276211804,
      "loss": 4.7841,
      "mean_token_accuracy": 0.22990026772022248,
      "num_tokens": 111207330.0,
      "step": 64115
    },
    {
      "entropy": 5.3681844711303714,
      "epoch": 4.9887570511573625,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002785529906572735,
      "loss": 4.7392,
      "mean_token_accuracy": 0.2290705293416977,
      "num_tokens": 111216038.0,
      "step": 64120
    },
    {
      "entropy": 5.465754842758178,
      "epoch": 4.98914608052908,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000278525278498518,
      "loss": 4.8131,
      "mean_token_accuracy": 0.21883846670389176,
      "num_tokens": 111224980.0,
      "step": 64125
    },
    {
      "entropy": 5.487655448913574,
      "epoch": 4.989535109900798,
      "grad_norm": 1.125,
      "learning_rate": 0.00027849756628627205,
      "loss": 4.8133,
      "mean_token_accuracy": 0.22330766320228576,
      "num_tokens": 111234394.0,
      "step": 64130
    },
    {
      "entropy": 5.377928781509399,
      "epoch": 4.9899241392725155,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002784698540209561,
      "loss": 4.7178,
      "mean_token_accuracy": 0.2260890558362007,
      "num_tokens": 111242260.0,
      "step": 64135
    },
    {
      "entropy": 5.387384605407715,
      "epoch": 4.990313168644232,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027844214170299065,
      "loss": 4.6443,
      "mean_token_accuracy": 0.24159384220838548,
      "num_tokens": 111250706.0,
      "step": 64140
    },
    {
      "entropy": 5.372367334365845,
      "epoch": 4.99070219801595,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00027841442933279614,
      "loss": 4.8146,
      "mean_token_accuracy": 0.21741727739572525,
      "num_tokens": 111259965.0,
      "step": 64145
    },
    {
      "entropy": 5.3532792091369625,
      "epoch": 4.991091227387668,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027838671691079314,
      "loss": 4.6869,
      "mean_token_accuracy": 0.22441308498382567,
      "num_tokens": 111269422.0,
      "step": 64150
    },
    {
      "entropy": 5.437767076492309,
      "epoch": 4.9914802567593854,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002783590044374021,
      "loss": 4.7344,
      "mean_token_accuracy": 0.23423292338848115,
      "num_tokens": 111278505.0,
      "step": 64155
    },
    {
      "entropy": 5.4433564186096195,
      "epoch": 4.991869286131103,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002783312919130435,
      "loss": 4.7472,
      "mean_token_accuracy": 0.22648418992757796,
      "num_tokens": 111288194.0,
      "step": 64160
    },
    {
      "entropy": 5.359408617019653,
      "epoch": 4.992258315502821,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002783035793381378,
      "loss": 4.7842,
      "mean_token_accuracy": 0.22051894813776016,
      "num_tokens": 111297425.0,
      "step": 64165
    },
    {
      "entropy": 5.413225793838501,
      "epoch": 4.992647344874538,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027827586671310556,
      "loss": 4.8335,
      "mean_token_accuracy": 0.22535801380872728,
      "num_tokens": 111307008.0,
      "step": 64170
    },
    {
      "entropy": 5.4240659236907955,
      "epoch": 4.993036374246255,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002782481540383673,
      "loss": 4.7318,
      "mean_token_accuracy": 0.23277881890535354,
      "num_tokens": 111315311.0,
      "step": 64175
    },
    {
      "entropy": 5.461198186874389,
      "epoch": 4.993425403617973,
      "grad_norm": 1.15625,
      "learning_rate": 0.00027822044131434344,
      "loss": 4.7862,
      "mean_token_accuracy": 0.22127892524003984,
      "num_tokens": 111323734.0,
      "step": 64180
    },
    {
      "entropy": 5.394426155090332,
      "epoch": 4.993814432989691,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002781927285414545,
      "loss": 4.6455,
      "mean_token_accuracy": 0.23338836431503296,
      "num_tokens": 111331696.0,
      "step": 64185
    },
    {
      "entropy": 5.3852911472320555,
      "epoch": 4.994203462361408,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027816501572012104,
      "loss": 4.8584,
      "mean_token_accuracy": 0.21642314344644548,
      "num_tokens": 111341266.0,
      "step": 64190
    },
    {
      "entropy": 5.443745422363281,
      "epoch": 4.994592491733126,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002781373028507634,
      "loss": 4.8619,
      "mean_token_accuracy": 0.2204653352499008,
      "num_tokens": 111350509.0,
      "step": 64195
    },
    {
      "entropy": 5.448437023162842,
      "epoch": 4.994981521104844,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002781095899338023,
      "loss": 4.8222,
      "mean_token_accuracy": 0.2221800908446312,
      "num_tokens": 111359408.0,
      "step": 64200
    },
    {
      "entropy": 5.422222042083741,
      "epoch": 4.995370550476561,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002780818769696581,
      "loss": 4.7367,
      "mean_token_accuracy": 0.22704196870326995,
      "num_tokens": 111367523.0,
      "step": 64205
    },
    {
      "entropy": 5.487494802474975,
      "epoch": 4.995759579848278,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027805416395875127,
      "loss": 4.9156,
      "mean_token_accuracy": 0.21983126997947694,
      "num_tokens": 111376157.0,
      "step": 64210
    },
    {
      "entropy": 5.340842342376709,
      "epoch": 4.996148609219996,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002780264509015024,
      "loss": 4.6969,
      "mean_token_accuracy": 0.22905329763889312,
      "num_tokens": 111384300.0,
      "step": 64215
    },
    {
      "entropy": 5.3841142654418945,
      "epoch": 4.996537638591714,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027799873779833193,
      "loss": 4.7079,
      "mean_token_accuracy": 0.23045638650655748,
      "num_tokens": 111393207.0,
      "step": 64220
    },
    {
      "entropy": 5.443645286560058,
      "epoch": 4.996926667963431,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002779710246496604,
      "loss": 4.7432,
      "mean_token_accuracy": 0.22990184128284455,
      "num_tokens": 111401543.0,
      "step": 64225
    },
    {
      "entropy": 5.4928313255310055,
      "epoch": 4.997315697335149,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002779433114559083,
      "loss": 4.9022,
      "mean_token_accuracy": 0.21773283034563065,
      "num_tokens": 111410055.0,
      "step": 64230
    },
    {
      "entropy": 5.425173377990722,
      "epoch": 4.997704726706866,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027791559821749615,
      "loss": 4.8164,
      "mean_token_accuracy": 0.21817171722650527,
      "num_tokens": 111419244.0,
      "step": 64235
    },
    {
      "entropy": 5.422532844543457,
      "epoch": 4.998093756078584,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00027788788493484447,
      "loss": 4.6954,
      "mean_token_accuracy": 0.2294187068939209,
      "num_tokens": 111428144.0,
      "step": 64240
    },
    {
      "entropy": 5.393340396881103,
      "epoch": 4.998482785450301,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002778601716083736,
      "loss": 4.7409,
      "mean_token_accuracy": 0.22522320300340654,
      "num_tokens": 111437459.0,
      "step": 64245
    },
    {
      "entropy": 5.494318962097168,
      "epoch": 4.998871814822019,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002778324582385043,
      "loss": 4.9344,
      "mean_token_accuracy": 0.21052779853343964,
      "num_tokens": 111446771.0,
      "step": 64250
    },
    {
      "entropy": 5.3928529739379885,
      "epoch": 4.999260844193737,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00027780474482565687,
      "loss": 4.7686,
      "mean_token_accuracy": 0.22472599744796753,
      "num_tokens": 111455725.0,
      "step": 64255
    },
    {
      "entropy": 5.491280794143677,
      "epoch": 4.999649873565454,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00027777703137025193,
      "loss": 4.8446,
      "mean_token_accuracy": 0.2185389444231987,
      "num_tokens": 111464343.0,
      "step": 64260
    },
    {
      "entropy": 5.467421054840088,
      "epoch": 5.0,
      "grad_norm": 2.765625,
      "learning_rate": 0.00027774931787271,
      "loss": 4.7557,
      "mean_token_accuracy": 0.22578904032707214,
      "num_tokens": 111471835.0,
      "step": 64265
    },
    {
      "entropy": 5.393160772323609,
      "epoch": 5.000389029371718,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002777216043334516,
      "loss": 4.6693,
      "mean_token_accuracy": 0.22942399978637695,
      "num_tokens": 111481644.0,
      "step": 64270
    },
    {
      "entropy": 5.359787368774414,
      "epoch": 5.000778058743435,
      "grad_norm": 1.078125,
      "learning_rate": 0.00027769389075289714,
      "loss": 4.7041,
      "mean_token_accuracy": 0.23059123158454894,
      "num_tokens": 111490873.0,
      "step": 64275
    },
    {
      "entropy": 5.350889062881469,
      "epoch": 5.001167088115153,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002776661771314671,
      "loss": 4.6615,
      "mean_token_accuracy": 0.23821038901805877,
      "num_tokens": 111499059.0,
      "step": 64280
    },
    {
      "entropy": 5.390213966369629,
      "epoch": 5.00155611748687,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002776384634695821,
      "loss": 4.6287,
      "mean_token_accuracy": 0.24267079383134843,
      "num_tokens": 111507406.0,
      "step": 64285
    },
    {
      "entropy": 5.384152269363403,
      "epoch": 5.001945146858588,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002776107497676626,
      "loss": 4.6374,
      "mean_token_accuracy": 0.23686433881521224,
      "num_tokens": 111516998.0,
      "step": 64290
    },
    {
      "entropy": 5.326612710952759,
      "epoch": 5.002334176230305,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002775830360261291,
      "loss": 4.7304,
      "mean_token_accuracy": 0.23271764516830445,
      "num_tokens": 111525636.0,
      "step": 64295
    },
    {
      "entropy": 5.372388887405395,
      "epoch": 5.002723205602023,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027755532224540226,
      "loss": 4.6879,
      "mean_token_accuracy": 0.2348104938864708,
      "num_tokens": 111533057.0,
      "step": 64300
    },
    {
      "entropy": 5.346791553497314,
      "epoch": 5.003112234973741,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027752760842590245,
      "loss": 4.6558,
      "mean_token_accuracy": 0.23345253318548204,
      "num_tokens": 111541852.0,
      "step": 64305
    },
    {
      "entropy": 5.452053833007812,
      "epoch": 5.003501264345458,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002774998945680502,
      "loss": 4.7031,
      "mean_token_accuracy": 0.23394353538751603,
      "num_tokens": 111551195.0,
      "step": 64310
    },
    {
      "entropy": 5.407734823226929,
      "epoch": 5.003890293717176,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000277472180672266,
      "loss": 4.7192,
      "mean_token_accuracy": 0.23333755880594254,
      "num_tokens": 111559483.0,
      "step": 64315
    },
    {
      "entropy": 5.4497997760772705,
      "epoch": 5.004279323088893,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027744446673897034,
      "loss": 4.7625,
      "mean_token_accuracy": 0.222392338514328,
      "num_tokens": 111568410.0,
      "step": 64320
    },
    {
      "entropy": 5.417685174942017,
      "epoch": 5.004668352460611,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027741675276858374,
      "loss": 4.7283,
      "mean_token_accuracy": 0.2289823114871979,
      "num_tokens": 111577095.0,
      "step": 64325
    },
    {
      "entropy": 5.346709585189819,
      "epoch": 5.005057381832328,
      "grad_norm": 1.140625,
      "learning_rate": 0.00027738903876152693,
      "loss": 4.5628,
      "mean_token_accuracy": 0.24600160270929336,
      "num_tokens": 111585592.0,
      "step": 64330
    },
    {
      "entropy": 5.392195463180542,
      "epoch": 5.005446411204046,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002773613247182202,
      "loss": 4.6627,
      "mean_token_accuracy": 0.2382856711745262,
      "num_tokens": 111594012.0,
      "step": 64335
    },
    {
      "entropy": 5.352856159210205,
      "epoch": 5.005835440575764,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027733361063908405,
      "loss": 4.6551,
      "mean_token_accuracy": 0.24531556218862532,
      "num_tokens": 111602756.0,
      "step": 64340
    },
    {
      "entropy": 5.435971736907959,
      "epoch": 5.006224469947481,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002773058965245393,
      "loss": 4.7776,
      "mean_token_accuracy": 0.23149231374263762,
      "num_tokens": 111611411.0,
      "step": 64345
    },
    {
      "entropy": 5.4332164287567135,
      "epoch": 5.006613499319198,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000277278182375006,
      "loss": 4.7224,
      "mean_token_accuracy": 0.22553129196166993,
      "num_tokens": 111621190.0,
      "step": 64350
    },
    {
      "entropy": 5.503264713287353,
      "epoch": 5.007002528690916,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002772504681909049,
      "loss": 4.812,
      "mean_token_accuracy": 0.22305875569581984,
      "num_tokens": 111629834.0,
      "step": 64355
    },
    {
      "entropy": 5.3610021591186525,
      "epoch": 5.007391558062634,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002772227539726566,
      "loss": 4.6114,
      "mean_token_accuracy": 0.23325600177049638,
      "num_tokens": 111638260.0,
      "step": 64360
    },
    {
      "entropy": 5.362676095962525,
      "epoch": 5.007780587434351,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002771950397206816,
      "loss": 4.672,
      "mean_token_accuracy": 0.23826683461666107,
      "num_tokens": 111646643.0,
      "step": 64365
    },
    {
      "entropy": 5.44324951171875,
      "epoch": 5.008169616806069,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002771673254354003,
      "loss": 4.7551,
      "mean_token_accuracy": 0.22555161416530609,
      "num_tokens": 111654725.0,
      "step": 64370
    },
    {
      "entropy": 5.302894878387451,
      "epoch": 5.008558646177787,
      "grad_norm": 1.15625,
      "learning_rate": 0.00027713961111723335,
      "loss": 4.614,
      "mean_token_accuracy": 0.23836666196584702,
      "num_tokens": 111663638.0,
      "step": 64375
    },
    {
      "entropy": 5.3154296875,
      "epoch": 5.008947675549504,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00027711189676660116,
      "loss": 4.5682,
      "mean_token_accuracy": 0.2404208540916443,
      "num_tokens": 111672855.0,
      "step": 64380
    },
    {
      "entropy": 5.399192476272583,
      "epoch": 5.009336704921221,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00027708418238392435,
      "loss": 4.6691,
      "mean_token_accuracy": 0.23871610015630723,
      "num_tokens": 111682328.0,
      "step": 64385
    },
    {
      "entropy": 5.41549596786499,
      "epoch": 5.009725734292939,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002770564679696233,
      "loss": 4.6888,
      "mean_token_accuracy": 0.23409728556871415,
      "num_tokens": 111690713.0,
      "step": 64390
    },
    {
      "entropy": 5.362019491195679,
      "epoch": 5.010114763664657,
      "grad_norm": 1.25,
      "learning_rate": 0.00027702875352411866,
      "loss": 4.6916,
      "mean_token_accuracy": 0.23450776189565659,
      "num_tokens": 111699466.0,
      "step": 64395
    },
    {
      "entropy": 5.403151226043701,
      "epoch": 5.010503793036374,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002770010390478309,
      "loss": 4.7285,
      "mean_token_accuracy": 0.22748700976371766,
      "num_tokens": 111708331.0,
      "step": 64400
    },
    {
      "entropy": 5.39559006690979,
      "epoch": 5.010892822408092,
      "grad_norm": 1.140625,
      "learning_rate": 0.00027697332454118057,
      "loss": 4.7008,
      "mean_token_accuracy": 0.2342975452542305,
      "num_tokens": 111716564.0,
      "step": 64405
    },
    {
      "entropy": 5.425977182388306,
      "epoch": 5.01128185177981,
      "grad_norm": 1.140625,
      "learning_rate": 0.00027694561000458825,
      "loss": 4.678,
      "mean_token_accuracy": 0.23416243344545365,
      "num_tokens": 111725241.0,
      "step": 64410
    },
    {
      "entropy": 5.457909631729126,
      "epoch": 5.011670881151527,
      "grad_norm": 1.25,
      "learning_rate": 0.0002769178954384743,
      "loss": 4.7774,
      "mean_token_accuracy": 0.22220711261034012,
      "num_tokens": 111733569.0,
      "step": 64415
    },
    {
      "entropy": 5.39597339630127,
      "epoch": 5.012059910523244,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002768901808432594,
      "loss": 4.7009,
      "mean_token_accuracy": 0.24136213958263397,
      "num_tokens": 111742599.0,
      "step": 64420
    },
    {
      "entropy": 5.424000406265259,
      "epoch": 5.012448939894962,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000276862466219364,
      "loss": 4.6896,
      "mean_token_accuracy": 0.2359243080019951,
      "num_tokens": 111751562.0,
      "step": 64425
    },
    {
      "entropy": 5.375463199615479,
      "epoch": 5.01283796926668,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002768347515672086,
      "loss": 4.6018,
      "mean_token_accuracy": 0.23400613218545913,
      "num_tokens": 111760649.0,
      "step": 64430
    },
    {
      "entropy": 5.432074165344238,
      "epoch": 5.013226998638397,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027680703688721383,
      "loss": 4.7361,
      "mean_token_accuracy": 0.2317538008093834,
      "num_tokens": 111768850.0,
      "step": 64435
    },
    {
      "entropy": 5.378588008880615,
      "epoch": 5.013616028010115,
      "grad_norm": 1.15625,
      "learning_rate": 0.00027677932217980004,
      "loss": 4.684,
      "mean_token_accuracy": 0.2459005519747734,
      "num_tokens": 111777791.0,
      "step": 64440
    },
    {
      "entropy": 5.402051067352295,
      "epoch": 5.014005057381833,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000276751607445388,
      "loss": 4.7213,
      "mean_token_accuracy": 0.2279883250594139,
      "num_tokens": 111787248.0,
      "step": 64445
    },
    {
      "entropy": 5.33976035118103,
      "epoch": 5.0143940867535495,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000276723892684398,
      "loss": 4.6474,
      "mean_token_accuracy": 0.2304422900080681,
      "num_tokens": 111795218.0,
      "step": 64450
    },
    {
      "entropy": 5.4128748893737795,
      "epoch": 5.014783116125267,
      "grad_norm": 1.140625,
      "learning_rate": 0.00027669617789725074,
      "loss": 4.7052,
      "mean_token_accuracy": 0.23688170909881592,
      "num_tokens": 111803769.0,
      "step": 64455
    },
    {
      "entropy": 5.403263854980469,
      "epoch": 5.015172145496985,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002766684630843667,
      "loss": 4.6115,
      "mean_token_accuracy": 0.2430353507399559,
      "num_tokens": 111811841.0,
      "step": 64460
    },
    {
      "entropy": 5.392855405807495,
      "epoch": 5.015561174868703,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002766407482461663,
      "loss": 4.697,
      "mean_token_accuracy": 0.2374156594276428,
      "num_tokens": 111820836.0,
      "step": 64465
    },
    {
      "entropy": 5.3530598163604735,
      "epoch": 5.01595020424042,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002766130333830702,
      "loss": 4.6653,
      "mean_token_accuracy": 0.23851159065961838,
      "num_tokens": 111829917.0,
      "step": 64470
    },
    {
      "entropy": 5.363078546524048,
      "epoch": 5.016339233612138,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027658531849549894,
      "loss": 4.6537,
      "mean_token_accuracy": 0.2352581024169922,
      "num_tokens": 111839336.0,
      "step": 64475
    },
    {
      "entropy": 5.442423343658447,
      "epoch": 5.016728262983856,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002765576035838729,
      "loss": 4.7161,
      "mean_token_accuracy": 0.23403845727443695,
      "num_tokens": 111847495.0,
      "step": 64480
    },
    {
      "entropy": 5.318526029586792,
      "epoch": 5.0171172923555725,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002765298886486128,
      "loss": 4.6567,
      "mean_token_accuracy": 0.23740802109241485,
      "num_tokens": 111855903.0,
      "step": 64485
    },
    {
      "entropy": 5.3797626972198485,
      "epoch": 5.01750632172729,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002765021736901389,
      "loss": 4.7385,
      "mean_token_accuracy": 0.23324352502822876,
      "num_tokens": 111864910.0,
      "step": 64490
    },
    {
      "entropy": 5.407120323181152,
      "epoch": 5.017895351099008,
      "grad_norm": 1.25,
      "learning_rate": 0.0002764744587088721,
      "loss": 4.7157,
      "mean_token_accuracy": 0.2326302334666252,
      "num_tokens": 111873201.0,
      "step": 64495
    },
    {
      "entropy": 5.350489091873169,
      "epoch": 5.018284380470726,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002764467437052327,
      "loss": 4.6321,
      "mean_token_accuracy": 0.23700974136590958,
      "num_tokens": 111882002.0,
      "step": 64500
    },
    {
      "entropy": 5.387865352630615,
      "epoch": 5.018673409842443,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00027641902867964126,
      "loss": 4.7366,
      "mean_token_accuracy": 0.22644797414541246,
      "num_tokens": 111891330.0,
      "step": 64505
    },
    {
      "entropy": 5.409853124618531,
      "epoch": 5.019062439214161,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027639131363251825,
      "loss": 4.7406,
      "mean_token_accuracy": 0.23686837255954743,
      "num_tokens": 111899743.0,
      "step": 64510
    },
    {
      "entropy": 5.42524151802063,
      "epoch": 5.019451468585879,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027636359856428436,
      "loss": 4.7075,
      "mean_token_accuracy": 0.24052108079195023,
      "num_tokens": 111907844.0,
      "step": 64515
    },
    {
      "entropy": 5.401770973205567,
      "epoch": 5.0198404979575955,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002763358834753599,
      "loss": 4.7308,
      "mean_token_accuracy": 0.23089895248413086,
      "num_tokens": 111916151.0,
      "step": 64520
    },
    {
      "entropy": 5.403537559509277,
      "epoch": 5.020229527329313,
      "grad_norm": 1.1875,
      "learning_rate": 0.00027630816836616564,
      "loss": 4.6592,
      "mean_token_accuracy": 0.23611665815114974,
      "num_tokens": 111924576.0,
      "step": 64525
    },
    {
      "entropy": 5.414581775665283,
      "epoch": 5.020618556701031,
      "grad_norm": 1.328125,
      "learning_rate": 0.00027628045323712194,
      "loss": 4.7652,
      "mean_token_accuracy": 0.23140973150730132,
      "num_tokens": 111932909.0,
      "step": 64530
    },
    {
      "entropy": 5.4493695259094235,
      "epoch": 5.0210075860727486,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027625273808864955,
      "loss": 4.7276,
      "mean_token_accuracy": 0.23673717975616454,
      "num_tokens": 111941539.0,
      "step": 64535
    },
    {
      "entropy": 5.416542291641235,
      "epoch": 5.021396615444466,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002762250229211687,
      "loss": 4.759,
      "mean_token_accuracy": 0.22537968456745147,
      "num_tokens": 111950186.0,
      "step": 64540
    },
    {
      "entropy": 5.418787908554077,
      "epoch": 5.021785644816184,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027619730773510015,
      "loss": 4.8008,
      "mean_token_accuracy": 0.22744613587856294,
      "num_tokens": 111959264.0,
      "step": 64545
    },
    {
      "entropy": 5.3736530303955075,
      "epoch": 5.022174674187901,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027616959253086437,
      "loss": 4.6842,
      "mean_token_accuracy": 0.23682726174592972,
      "num_tokens": 111968553.0,
      "step": 64550
    },
    {
      "entropy": 5.474446868896484,
      "epoch": 5.0225637035596185,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002761418773088818,
      "loss": 4.7216,
      "mean_token_accuracy": 0.23695174008607864,
      "num_tokens": 111977380.0,
      "step": 64555
    },
    {
      "entropy": 5.4269242763519285,
      "epoch": 5.022952732931336,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002761141620695731,
      "loss": 4.7051,
      "mean_token_accuracy": 0.23576796650886536,
      "num_tokens": 111986008.0,
      "step": 64560
    },
    {
      "entropy": 5.252368068695068,
      "epoch": 5.023341762303054,
      "grad_norm": 1.15625,
      "learning_rate": 0.00027608644681335876,
      "loss": 4.5546,
      "mean_token_accuracy": 0.24408508092164993,
      "num_tokens": 111994863.0,
      "step": 64565
    },
    {
      "entropy": 5.48540768623352,
      "epoch": 5.0237307916747715,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027605873154065935,
      "loss": 4.8269,
      "mean_token_accuracy": 0.22523840963840486,
      "num_tokens": 112003612.0,
      "step": 64570
    },
    {
      "entropy": 5.342392873764038,
      "epoch": 5.024119821046489,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027603101625189543,
      "loss": 4.5647,
      "mean_token_accuracy": 0.2354205936193466,
      "num_tokens": 112012034.0,
      "step": 64575
    },
    {
      "entropy": 5.478159093856812,
      "epoch": 5.024508850418207,
      "grad_norm": 1.25,
      "learning_rate": 0.0002760033009474874,
      "loss": 4.8373,
      "mean_token_accuracy": 0.2164052426815033,
      "num_tokens": 112021207.0,
      "step": 64580
    },
    {
      "entropy": 5.329247331619262,
      "epoch": 5.024897879789924,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002759755856278558,
      "loss": 4.7051,
      "mean_token_accuracy": 0.22684061527252197,
      "num_tokens": 112029677.0,
      "step": 64585
    },
    {
      "entropy": 5.306162738800049,
      "epoch": 5.0252869091616414,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002759478702934213,
      "loss": 4.5952,
      "mean_token_accuracy": 0.24477420896291732,
      "num_tokens": 112038024.0,
      "step": 64590
    },
    {
      "entropy": 5.462326431274414,
      "epoch": 5.025675938533359,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027592015494460443,
      "loss": 4.7151,
      "mean_token_accuracy": 0.22498405426740647,
      "num_tokens": 112046320.0,
      "step": 64595
    },
    {
      "entropy": 5.471684360504151,
      "epoch": 5.026064967905077,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027589243958182566,
      "loss": 4.8028,
      "mean_token_accuracy": 0.22004290372133256,
      "num_tokens": 112055539.0,
      "step": 64600
    },
    {
      "entropy": 5.40683798789978,
      "epoch": 5.0264539972767945,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027586472420550553,
      "loss": 4.6509,
      "mean_token_accuracy": 0.22683300077915192,
      "num_tokens": 112064282.0,
      "step": 64605
    },
    {
      "entropy": 5.414402341842651,
      "epoch": 5.026843026648512,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002758370088160646,
      "loss": 4.6514,
      "mean_token_accuracy": 0.23657518923282622,
      "num_tokens": 112072641.0,
      "step": 64610
    },
    {
      "entropy": 5.2838493347167965,
      "epoch": 5.02723205602023,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002758092934139234,
      "loss": 4.587,
      "mean_token_accuracy": 0.2451772376894951,
      "num_tokens": 112081026.0,
      "step": 64615
    },
    {
      "entropy": 5.342365550994873,
      "epoch": 5.027621085391947,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002757815779995023,
      "loss": 4.6906,
      "mean_token_accuracy": 0.23365252763032912,
      "num_tokens": 112089810.0,
      "step": 64620
    },
    {
      "entropy": 5.492216348648071,
      "epoch": 5.028010114763664,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002757538625732222,
      "loss": 4.8311,
      "mean_token_accuracy": 0.2206197127699852,
      "num_tokens": 112098279.0,
      "step": 64625
    },
    {
      "entropy": 5.436621856689453,
      "epoch": 5.028399144135382,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002757261471355034,
      "loss": 4.7505,
      "mean_token_accuracy": 0.23636455535888673,
      "num_tokens": 112107456.0,
      "step": 64630
    },
    {
      "entropy": 5.374758863449097,
      "epoch": 5.0287881735071,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00027569843168676647,
      "loss": 4.6771,
      "mean_token_accuracy": 0.22461806386709213,
      "num_tokens": 112116595.0,
      "step": 64635
    },
    {
      "entropy": 5.347405624389649,
      "epoch": 5.0291772028788175,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002756707162274319,
      "loss": 4.6954,
      "mean_token_accuracy": 0.22683172523975373,
      "num_tokens": 112126468.0,
      "step": 64640
    },
    {
      "entropy": 5.335866498947143,
      "epoch": 5.029566232250535,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00027564300075792026,
      "loss": 4.6587,
      "mean_token_accuracy": 0.2302749916911125,
      "num_tokens": 112135178.0,
      "step": 64645
    },
    {
      "entropy": 5.377702140808106,
      "epoch": 5.029955261622252,
      "grad_norm": 1.25,
      "learning_rate": 0.00027561528527865215,
      "loss": 4.6622,
      "mean_token_accuracy": 0.24113797545433044,
      "num_tokens": 112143792.0,
      "step": 64650
    },
    {
      "entropy": 5.344615173339844,
      "epoch": 5.03034429099397,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002755875697900481,
      "loss": 4.6074,
      "mean_token_accuracy": 0.23290932476520537,
      "num_tokens": 112152160.0,
      "step": 64655
    },
    {
      "entropy": 5.395511770248413,
      "epoch": 5.030733320365687,
      "grad_norm": 1.125,
      "learning_rate": 0.00027555985429252857,
      "loss": 4.7694,
      "mean_token_accuracy": 0.23126326501369476,
      "num_tokens": 112161223.0,
      "step": 64660
    },
    {
      "entropy": 5.353583240509034,
      "epoch": 5.031122349737405,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002755321387865141,
      "loss": 4.6049,
      "mean_token_accuracy": 0.24054396897554398,
      "num_tokens": 112169031.0,
      "step": 64665
    },
    {
      "entropy": 5.3820137023925785,
      "epoch": 5.031511379109123,
      "grad_norm": 1.25,
      "learning_rate": 0.0002755044232724253,
      "loss": 4.6492,
      "mean_token_accuracy": 0.2416839987039566,
      "num_tokens": 112177481.0,
      "step": 64670
    },
    {
      "entropy": 5.321760463714599,
      "epoch": 5.0319004084808405,
      "grad_norm": 1.203125,
      "learning_rate": 0.00027547670775068274,
      "loss": 4.6423,
      "mean_token_accuracy": 0.23692483752965926,
      "num_tokens": 112185494.0,
      "step": 64675
    },
    {
      "entropy": 5.310522651672363,
      "epoch": 5.032289437852558,
      "grad_norm": 1.203125,
      "learning_rate": 0.00027544899222170684,
      "loss": 4.681,
      "mean_token_accuracy": 0.23392193466424943,
      "num_tokens": 112194110.0,
      "step": 64680
    },
    {
      "entropy": 5.31445198059082,
      "epoch": 5.032678467224275,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027542127668591824,
      "loss": 4.6469,
      "mean_token_accuracy": 0.23335846215486528,
      "num_tokens": 112204135.0,
      "step": 64685
    },
    {
      "entropy": 5.364391708374024,
      "epoch": 5.033067496595993,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002753935611437374,
      "loss": 4.6408,
      "mean_token_accuracy": 0.22939473539590835,
      "num_tokens": 112212610.0,
      "step": 64690
    },
    {
      "entropy": 5.372741413116455,
      "epoch": 5.03345652596771,
      "grad_norm": 1.125,
      "learning_rate": 0.00027536584559558494,
      "loss": 4.733,
      "mean_token_accuracy": 0.22856444865465164,
      "num_tokens": 112222095.0,
      "step": 64695
    },
    {
      "entropy": 5.3948705196380615,
      "epoch": 5.033845555339428,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002753381300418813,
      "loss": 4.6953,
      "mean_token_accuracy": 0.23481495529413224,
      "num_tokens": 112230699.0,
      "step": 64700
    },
    {
      "entropy": 5.31956262588501,
      "epoch": 5.034234584711146,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002753104144830471,
      "loss": 4.6664,
      "mean_token_accuracy": 0.23793037831783295,
      "num_tokens": 112240462.0,
      "step": 64705
    },
    {
      "entropy": 5.333866310119629,
      "epoch": 5.0346236140828635,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027528269891950285,
      "loss": 4.6705,
      "mean_token_accuracy": 0.22964534759521485,
      "num_tokens": 112249237.0,
      "step": 64710
    },
    {
      "entropy": 5.405124282836914,
      "epoch": 5.035012643454581,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002752549833516691,
      "loss": 4.7483,
      "mean_token_accuracy": 0.22826498597860337,
      "num_tokens": 112258498.0,
      "step": 64715
    },
    {
      "entropy": 5.457015609741211,
      "epoch": 5.035401672826298,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002752272677799664,
      "loss": 4.6826,
      "mean_token_accuracy": 0.2325483113527298,
      "num_tokens": 112266749.0,
      "step": 64720
    },
    {
      "entropy": 5.365266418457031,
      "epoch": 5.035790702198016,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027519955220481525,
      "loss": 4.6682,
      "mean_token_accuracy": 0.23372434377670287,
      "num_tokens": 112275871.0,
      "step": 64725
    },
    {
      "entropy": 5.3376555919647215,
      "epoch": 5.036179731569733,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002751718366266362,
      "loss": 4.7062,
      "mean_token_accuracy": 0.2284228429198265,
      "num_tokens": 112284371.0,
      "step": 64730
    },
    {
      "entropy": 5.420383071899414,
      "epoch": 5.036568760941451,
      "grad_norm": 1.25,
      "learning_rate": 0.0002751441210458498,
      "loss": 4.7812,
      "mean_token_accuracy": 0.22827626913785934,
      "num_tokens": 112293481.0,
      "step": 64735
    },
    {
      "entropy": 5.485768175125122,
      "epoch": 5.036957790313169,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027511640546287665,
      "loss": 4.7898,
      "mean_token_accuracy": 0.21717797815799714,
      "num_tokens": 112301482.0,
      "step": 64740
    },
    {
      "entropy": 5.411526775360107,
      "epoch": 5.0373468196848865,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002750886898781372,
      "loss": 4.6733,
      "mean_token_accuracy": 0.23777484893798828,
      "num_tokens": 112309687.0,
      "step": 64745
    },
    {
      "entropy": 5.335828590393066,
      "epoch": 5.037735849056604,
      "grad_norm": 1.1875,
      "learning_rate": 0.000275060974292052,
      "loss": 4.6842,
      "mean_token_accuracy": 0.23297895938158036,
      "num_tokens": 112318351.0,
      "step": 64750
    },
    {
      "entropy": 5.378765439987182,
      "epoch": 5.038124878428321,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002750332587050416,
      "loss": 4.7215,
      "mean_token_accuracy": 0.2273263469338417,
      "num_tokens": 112327937.0,
      "step": 64755
    },
    {
      "entropy": 5.484892416000366,
      "epoch": 5.038513907800039,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002750055431175266,
      "loss": 4.836,
      "mean_token_accuracy": 0.22341853827238084,
      "num_tokens": 112337336.0,
      "step": 64760
    },
    {
      "entropy": 5.449134349822998,
      "epoch": 5.038902937171756,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00027497782752992743,
      "loss": 4.6902,
      "mean_token_accuracy": 0.23522359728813172,
      "num_tokens": 112345486.0,
      "step": 64765
    },
    {
      "entropy": 5.424623155593872,
      "epoch": 5.039291966543474,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002749501119426647,
      "loss": 4.668,
      "mean_token_accuracy": 0.23256232887506484,
      "num_tokens": 112354642.0,
      "step": 64770
    },
    {
      "entropy": 5.3984095573425295,
      "epoch": 5.039680995915192,
      "grad_norm": 1.25,
      "learning_rate": 0.00027492239635615903,
      "loss": 4.7035,
      "mean_token_accuracy": 0.2291734039783478,
      "num_tokens": 112362979.0,
      "step": 64775
    },
    {
      "entropy": 5.329232931137085,
      "epoch": 5.0400700252869095,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002748946807708308,
      "loss": 4.5869,
      "mean_token_accuracy": 0.24202586561441422,
      "num_tokens": 112371710.0,
      "step": 64780
    },
    {
      "entropy": 5.290570163726807,
      "epoch": 5.040459054658626,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002748669651871006,
      "loss": 4.5967,
      "mean_token_accuracy": 0.24343384355306624,
      "num_tokens": 112380174.0,
      "step": 64785
    },
    {
      "entropy": 5.45467848777771,
      "epoch": 5.040848084030344,
      "grad_norm": 1.1875,
      "learning_rate": 0.00027483924960538907,
      "loss": 4.8011,
      "mean_token_accuracy": 0.22138925790786743,
      "num_tokens": 112388531.0,
      "step": 64790
    },
    {
      "entropy": 5.4193531513214115,
      "epoch": 5.041237113402062,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027481153402611664,
      "loss": 4.6948,
      "mean_token_accuracy": 0.2288118839263916,
      "num_tokens": 112397586.0,
      "step": 64795
    },
    {
      "entropy": 5.419966173171997,
      "epoch": 5.041626142773779,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027478381844970386,
      "loss": 4.7385,
      "mean_token_accuracy": 0.23690458089113237,
      "num_tokens": 112406459.0,
      "step": 64800
    },
    {
      "entropy": 5.423429107666015,
      "epoch": 5.042015172145497,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002747561028765713,
      "loss": 4.7269,
      "mean_token_accuracy": 0.22894801795482636,
      "num_tokens": 112415010.0,
      "step": 64805
    },
    {
      "entropy": 5.505279207229615,
      "epoch": 5.042404201517215,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027472838730713956,
      "loss": 4.7615,
      "mean_token_accuracy": 0.2295281931757927,
      "num_tokens": 112424209.0,
      "step": 64810
    },
    {
      "entropy": 5.381301164627075,
      "epoch": 5.0427932308889325,
      "grad_norm": 1.265625,
      "learning_rate": 0.00027470067174182905,
      "loss": 4.6349,
      "mean_token_accuracy": 0.23473345935344697,
      "num_tokens": 112432449.0,
      "step": 64815
    },
    {
      "entropy": 5.3743537902832035,
      "epoch": 5.043182260260649,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002746729561810604,
      "loss": 4.7099,
      "mean_token_accuracy": 0.23365440219640732,
      "num_tokens": 112440919.0,
      "step": 64820
    },
    {
      "entropy": 5.331893825531006,
      "epoch": 5.043571289632367,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027464524062525405,
      "loss": 4.6665,
      "mean_token_accuracy": 0.23441773205995559,
      "num_tokens": 112449843.0,
      "step": 64825
    },
    {
      "entropy": 5.529669141769409,
      "epoch": 5.043960319004085,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002746175250748306,
      "loss": 4.8027,
      "mean_token_accuracy": 0.2144353449344635,
      "num_tokens": 112459361.0,
      "step": 64830
    },
    {
      "entropy": 5.446399259567261,
      "epoch": 5.044349348375802,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002745898095302106,
      "loss": 4.7635,
      "mean_token_accuracy": 0.22380660474300385,
      "num_tokens": 112469496.0,
      "step": 64835
    },
    {
      "entropy": 5.403471231460571,
      "epoch": 5.04473837774752,
      "grad_norm": 1.328125,
      "learning_rate": 0.00027456209399181466,
      "loss": 4.6529,
      "mean_token_accuracy": 0.23784971237182617,
      "num_tokens": 112477886.0,
      "step": 64840
    },
    {
      "entropy": 5.345551586151123,
      "epoch": 5.045127407119238,
      "grad_norm": 1.15625,
      "learning_rate": 0.00027453437846006325,
      "loss": 4.7229,
      "mean_token_accuracy": 0.22638411521911622,
      "num_tokens": 112486993.0,
      "step": 64845
    },
    {
      "entropy": 5.504258251190185,
      "epoch": 5.0455164364909555,
      "grad_norm": 1.1875,
      "learning_rate": 0.00027450666293537683,
      "loss": 4.7581,
      "mean_token_accuracy": 0.2275029182434082,
      "num_tokens": 112495560.0,
      "step": 64850
    },
    {
      "entropy": 5.343051719665527,
      "epoch": 5.045905465862672,
      "grad_norm": 1.296875,
      "learning_rate": 0.000274478947418176,
      "loss": 4.6575,
      "mean_token_accuracy": 0.2371336653828621,
      "num_tokens": 112504104.0,
      "step": 64855
    },
    {
      "entropy": 5.413698434829712,
      "epoch": 5.04629449523439,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002744512319088814,
      "loss": 4.814,
      "mean_token_accuracy": 0.2279406428337097,
      "num_tokens": 112512453.0,
      "step": 64860
    },
    {
      "entropy": 5.41165828704834,
      "epoch": 5.046683524606108,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002744235164079135,
      "loss": 4.7538,
      "mean_token_accuracy": 0.2235019952058792,
      "num_tokens": 112520734.0,
      "step": 64865
    },
    {
      "entropy": 5.38701696395874,
      "epoch": 5.047072553977825,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002743958009156928,
      "loss": 4.7143,
      "mean_token_accuracy": 0.2301914468407631,
      "num_tokens": 112529235.0,
      "step": 64870
    },
    {
      "entropy": 5.322866678237915,
      "epoch": 5.047461583349543,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002743680854326398,
      "loss": 4.636,
      "mean_token_accuracy": 0.23839350640773774,
      "num_tokens": 112538244.0,
      "step": 64875
    },
    {
      "entropy": 5.352987384796142,
      "epoch": 5.047850612721261,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002743403699591751,
      "loss": 4.6793,
      "mean_token_accuracy": 0.2254127651453018,
      "num_tokens": 112546341.0,
      "step": 64880
    },
    {
      "entropy": 5.454830265045166,
      "epoch": 5.048239642092978,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00027431265449571937,
      "loss": 4.7723,
      "mean_token_accuracy": 0.2201244443655014,
      "num_tokens": 112554820.0,
      "step": 64885
    },
    {
      "entropy": 5.421988534927368,
      "epoch": 5.048628671464695,
      "grad_norm": 1.1875,
      "learning_rate": 0.00027428493904269285,
      "loss": 4.713,
      "mean_token_accuracy": 0.22906819730997086,
      "num_tokens": 112563071.0,
      "step": 64890
    },
    {
      "entropy": 5.397017478942871,
      "epoch": 5.049017700836413,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002742572236005164,
      "loss": 4.6691,
      "mean_token_accuracy": 0.2386128529906273,
      "num_tokens": 112572121.0,
      "step": 64895
    },
    {
      "entropy": 5.374479722976685,
      "epoch": 5.049406730208131,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027422950816961025,
      "loss": 4.6976,
      "mean_token_accuracy": 0.23231341391801835,
      "num_tokens": 112580812.0,
      "step": 64900
    },
    {
      "entropy": 5.417215061187744,
      "epoch": 5.049795759579848,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000274201792750395,
      "loss": 4.7232,
      "mean_token_accuracy": 0.22661048024892808,
      "num_tokens": 112589196.0,
      "step": 64905
    },
    {
      "entropy": 5.360439968109131,
      "epoch": 5.050184788951566,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002741740773432914,
      "loss": 4.6688,
      "mean_token_accuracy": 0.2244074523448944,
      "num_tokens": 112597741.0,
      "step": 64910
    },
    {
      "entropy": 5.319122266769409,
      "epoch": 5.050573818323284,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002741463619487199,
      "loss": 4.5822,
      "mean_token_accuracy": 0.25064396262168886,
      "num_tokens": 112605755.0,
      "step": 64915
    },
    {
      "entropy": 5.383066082000733,
      "epoch": 5.050962847695001,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002741186465671011,
      "loss": 4.7315,
      "mean_token_accuracy": 0.23073919862508774,
      "num_tokens": 112615068.0,
      "step": 64920
    },
    {
      "entropy": 5.438505744934082,
      "epoch": 5.051351877066718,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002740909311988553,
      "loss": 4.819,
      "mean_token_accuracy": 0.22049048393964768,
      "num_tokens": 112624507.0,
      "step": 64925
    },
    {
      "entropy": 5.39831314086914,
      "epoch": 5.051740906438436,
      "grad_norm": 1.25,
      "learning_rate": 0.0002740632158444031,
      "loss": 4.6433,
      "mean_token_accuracy": 0.24046834260225297,
      "num_tokens": 112632276.0,
      "step": 64930
    },
    {
      "entropy": 5.400159549713135,
      "epoch": 5.052129935810154,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002740355005041652,
      "loss": 4.7268,
      "mean_token_accuracy": 0.23603378236293793,
      "num_tokens": 112641333.0,
      "step": 64935
    },
    {
      "entropy": 5.357839870452881,
      "epoch": 5.052518965181871,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002740077851785621,
      "loss": 4.5922,
      "mean_token_accuracy": 0.23776448667049407,
      "num_tokens": 112649807.0,
      "step": 64940
    },
    {
      "entropy": 5.367248630523681,
      "epoch": 5.052907994553589,
      "grad_norm": 1.1875,
      "learning_rate": 0.00027398006986801415,
      "loss": 4.7497,
      "mean_token_accuracy": 0.23069133162498473,
      "num_tokens": 112658794.0,
      "step": 64945
    },
    {
      "entropy": 5.328583240509033,
      "epoch": 5.053297023925307,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002739523545729421,
      "loss": 4.5711,
      "mean_token_accuracy": 0.24034045040607452,
      "num_tokens": 112667748.0,
      "step": 64950
    },
    {
      "entropy": 5.442044496536255,
      "epoch": 5.053686053297024,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002739246392937664,
      "loss": 4.7749,
      "mean_token_accuracy": 0.22431700527668,
      "num_tokens": 112675750.0,
      "step": 64955
    },
    {
      "entropy": 5.436061525344849,
      "epoch": 5.054075082668741,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027389692403090754,
      "loss": 4.7306,
      "mean_token_accuracy": 0.23467322438955307,
      "num_tokens": 112683997.0,
      "step": 64960
    },
    {
      "entropy": 5.26345911026001,
      "epoch": 5.054464112040459,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027386920878478616,
      "loss": 4.5149,
      "mean_token_accuracy": 0.2488034784793854,
      "num_tokens": 112692664.0,
      "step": 64965
    },
    {
      "entropy": 5.322584772109986,
      "epoch": 5.054853141412177,
      "grad_norm": 1.265625,
      "learning_rate": 0.00027384149355582266,
      "loss": 4.6581,
      "mean_token_accuracy": 0.2341347649693489,
      "num_tokens": 112701265.0,
      "step": 64970
    },
    {
      "entropy": 5.436297416687012,
      "epoch": 5.055242170783894,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002738137783444377,
      "loss": 4.7608,
      "mean_token_accuracy": 0.23288995921611785,
      "num_tokens": 112709657.0,
      "step": 64975
    },
    {
      "entropy": 5.34898738861084,
      "epoch": 5.055631200155612,
      "grad_norm": 1.125,
      "learning_rate": 0.00027378606315105175,
      "loss": 4.6874,
      "mean_token_accuracy": 0.23395895212888718,
      "num_tokens": 112718782.0,
      "step": 64980
    },
    {
      "entropy": 5.392288494110107,
      "epoch": 5.056020229527329,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002737583479760854,
      "loss": 4.809,
      "mean_token_accuracy": 0.2315194472670555,
      "num_tokens": 112727564.0,
      "step": 64985
    },
    {
      "entropy": 5.466008520126342,
      "epoch": 5.056409258899047,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002737306328199591,
      "loss": 4.7463,
      "mean_token_accuracy": 0.22870464622974396,
      "num_tokens": 112735916.0,
      "step": 64990
    },
    {
      "entropy": 5.410093593597412,
      "epoch": 5.056798288270764,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027370291768309354,
      "loss": 4.6586,
      "mean_token_accuracy": 0.23945223838090895,
      "num_tokens": 112744995.0,
      "step": 64995
    },
    {
      "entropy": 5.35861029624939,
      "epoch": 5.057187317642482,
      "grad_norm": 1.21875,
      "learning_rate": 0.000273675202565909,
      "loss": 4.7128,
      "mean_token_accuracy": 0.2328049585223198,
      "num_tokens": 112753342.0,
      "step": 65000
    },
    {
      "entropy": 5.372970867156982,
      "epoch": 5.0575763470142,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027364748746882623,
      "loss": 4.6495,
      "mean_token_accuracy": 0.2309429958462715,
      "num_tokens": 112761341.0,
      "step": 65005
    },
    {
      "entropy": 5.42521333694458,
      "epoch": 5.057965376385917,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027361977239226566,
      "loss": 4.7348,
      "mean_token_accuracy": 0.23603052496910096,
      "num_tokens": 112770373.0,
      "step": 65010
    },
    {
      "entropy": 5.380546712875367,
      "epoch": 5.058354405757635,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002735920573366478,
      "loss": 4.709,
      "mean_token_accuracy": 0.2269685983657837,
      "num_tokens": 112779362.0,
      "step": 65015
    },
    {
      "entropy": 5.356908655166626,
      "epoch": 5.058743435129352,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002735643423023933,
      "loss": 4.6269,
      "mean_token_accuracy": 0.23911878019571303,
      "num_tokens": 112787402.0,
      "step": 65020
    },
    {
      "entropy": 5.519896221160889,
      "epoch": 5.05913246450107,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002735366272899226,
      "loss": 4.7903,
      "mean_token_accuracy": 0.2192848339676857,
      "num_tokens": 112796505.0,
      "step": 65025
    },
    {
      "entropy": 5.463559675216675,
      "epoch": 5.059521493872787,
      "grad_norm": 1.1875,
      "learning_rate": 0.00027350891229965623,
      "loss": 4.7898,
      "mean_token_accuracy": 0.2208764836192131,
      "num_tokens": 112805333.0,
      "step": 65030
    },
    {
      "entropy": 5.3769806861877445,
      "epoch": 5.059910523244505,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002734811973320147,
      "loss": 4.6343,
      "mean_token_accuracy": 0.24255779832601548,
      "num_tokens": 112813661.0,
      "step": 65035
    },
    {
      "entropy": 5.279697418212891,
      "epoch": 5.060299552616223,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027345348238741864,
      "loss": 4.5951,
      "mean_token_accuracy": 0.24546021968126297,
      "num_tokens": 112822053.0,
      "step": 65040
    },
    {
      "entropy": 5.304648065567017,
      "epoch": 5.06068858198794,
      "grad_norm": 1.15625,
      "learning_rate": 0.00027342576746628854,
      "loss": 4.6234,
      "mean_token_accuracy": 0.2434864327311516,
      "num_tokens": 112831009.0,
      "step": 65045
    },
    {
      "entropy": 5.364735317230225,
      "epoch": 5.061077611359658,
      "grad_norm": 1.25,
      "learning_rate": 0.0002733980525690449,
      "loss": 4.6682,
      "mean_token_accuracy": 0.23628709465265274,
      "num_tokens": 112839648.0,
      "step": 65050
    },
    {
      "entropy": 5.45759015083313,
      "epoch": 5.061466640731375,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027337033769610827,
      "loss": 4.7636,
      "mean_token_accuracy": 0.22799090296030045,
      "num_tokens": 112848645.0,
      "step": 65055
    },
    {
      "entropy": 5.39937539100647,
      "epoch": 5.061855670103093,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002733426228478991,
      "loss": 4.6222,
      "mean_token_accuracy": 0.2395276427268982,
      "num_tokens": 112856610.0,
      "step": 65060
    },
    {
      "entropy": 5.31654920578003,
      "epoch": 5.06224469947481,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002733149080248381,
      "loss": 4.6494,
      "mean_token_accuracy": 0.23598543852567672,
      "num_tokens": 112865604.0,
      "step": 65065
    },
    {
      "entropy": 5.351459407806397,
      "epoch": 5.062633728846528,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002732871932273456,
      "loss": 4.6582,
      "mean_token_accuracy": 0.23824625462293625,
      "num_tokens": 112874135.0,
      "step": 65070
    },
    {
      "entropy": 5.466893672943115,
      "epoch": 5.063022758218246,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002732594784558423,
      "loss": 4.7951,
      "mean_token_accuracy": 0.22274016886949538,
      "num_tokens": 112882914.0,
      "step": 65075
    },
    {
      "entropy": 5.40595965385437,
      "epoch": 5.063411787589963,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002732317637107486,
      "loss": 4.7132,
      "mean_token_accuracy": 0.2303855001926422,
      "num_tokens": 112892385.0,
      "step": 65080
    },
    {
      "entropy": 5.481871557235718,
      "epoch": 5.063800816961681,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027320404899248507,
      "loss": 4.7562,
      "mean_token_accuracy": 0.23073528707027435,
      "num_tokens": 112901279.0,
      "step": 65085
    },
    {
      "entropy": 5.504454040527344,
      "epoch": 5.064189846333398,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002731763343014722,
      "loss": 4.9156,
      "mean_token_accuracy": 0.21177886724472045,
      "num_tokens": 112911068.0,
      "step": 65090
    },
    {
      "entropy": 5.389797019958496,
      "epoch": 5.064578875705116,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00027314861963813065,
      "loss": 4.7077,
      "mean_token_accuracy": 0.23947293907403946,
      "num_tokens": 112920301.0,
      "step": 65095
    },
    {
      "entropy": 5.4264328956604,
      "epoch": 5.064967905076833,
      "grad_norm": 1.25,
      "learning_rate": 0.00027312090500288065,
      "loss": 4.6787,
      "mean_token_accuracy": 0.23058042526245118,
      "num_tokens": 112929139.0,
      "step": 65100
    },
    {
      "entropy": 5.384863758087159,
      "epoch": 5.065356934448551,
      "grad_norm": 1.125,
      "learning_rate": 0.0002730931903961431,
      "loss": 4.7021,
      "mean_token_accuracy": 0.22919297367334365,
      "num_tokens": 112938609.0,
      "step": 65105
    },
    {
      "entropy": 5.383109092712402,
      "epoch": 5.065745963820269,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027306547581833836,
      "loss": 4.7303,
      "mean_token_accuracy": 0.23381359577178956,
      "num_tokens": 112946881.0,
      "step": 65110
    },
    {
      "entropy": 5.427600622177124,
      "epoch": 5.066134993191986,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027303776126988687,
      "loss": 4.7089,
      "mean_token_accuracy": 0.22714405953884126,
      "num_tokens": 112955286.0,
      "step": 65115
    },
    {
      "entropy": 5.396959447860718,
      "epoch": 5.066524022563703,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027301004675120926,
      "loss": 4.6582,
      "mean_token_accuracy": 0.2346645712852478,
      "num_tokens": 112963778.0,
      "step": 65120
    },
    {
      "entropy": 5.3928210735321045,
      "epoch": 5.066913051935421,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002729823322627261,
      "loss": 4.7172,
      "mean_token_accuracy": 0.23274625539779664,
      "num_tokens": 112972434.0,
      "step": 65125
    },
    {
      "entropy": 5.357746934890747,
      "epoch": 5.0673020813071386,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002729546178048577,
      "loss": 4.6821,
      "mean_token_accuracy": 0.23654775470495223,
      "num_tokens": 112980342.0,
      "step": 65130
    },
    {
      "entropy": 5.521769857406616,
      "epoch": 5.067691110678856,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002729269033780248,
      "loss": 4.8674,
      "mean_token_accuracy": 0.22088621109724044,
      "num_tokens": 112990387.0,
      "step": 65135
    },
    {
      "entropy": 5.366834115982056,
      "epoch": 5.068080140050574,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027289918898264784,
      "loss": 4.6231,
      "mean_token_accuracy": 0.23591775000095366,
      "num_tokens": 112999166.0,
      "step": 65140
    },
    {
      "entropy": 5.44498839378357,
      "epoch": 5.068469169422292,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002728714746191473,
      "loss": 4.759,
      "mean_token_accuracy": 0.2263491556048393,
      "num_tokens": 113007604.0,
      "step": 65145
    },
    {
      "entropy": 5.435696029663086,
      "epoch": 5.068858198794009,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027284376028794376,
      "loss": 4.725,
      "mean_token_accuracy": 0.23516846299171448,
      "num_tokens": 113015265.0,
      "step": 65150
    },
    {
      "entropy": 5.4280181407928465,
      "epoch": 5.069247228165726,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027281604598945774,
      "loss": 4.7268,
      "mean_token_accuracy": 0.2252932757139206,
      "num_tokens": 113023541.0,
      "step": 65155
    },
    {
      "entropy": 5.355860853195191,
      "epoch": 5.069636257537444,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002727883317241098,
      "loss": 4.6716,
      "mean_token_accuracy": 0.24173108041286467,
      "num_tokens": 113032726.0,
      "step": 65160
    },
    {
      "entropy": 5.357446765899658,
      "epoch": 5.0700252869091615,
      "grad_norm": 1.15625,
      "learning_rate": 0.00027276061749232034,
      "loss": 4.6673,
      "mean_token_accuracy": 0.23370827734470367,
      "num_tokens": 113041418.0,
      "step": 65165
    },
    {
      "entropy": 5.371289491653442,
      "epoch": 5.070414316280879,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002727329032945099,
      "loss": 4.6632,
      "mean_token_accuracy": 0.2274712309241295,
      "num_tokens": 113050097.0,
      "step": 65170
    },
    {
      "entropy": 5.4087423324584964,
      "epoch": 5.070803345652597,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002727051891310991,
      "loss": 4.712,
      "mean_token_accuracy": 0.23062759190797805,
      "num_tokens": 113058010.0,
      "step": 65175
    },
    {
      "entropy": 5.38252592086792,
      "epoch": 5.071192375024315,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002726774750025084,
      "loss": 4.7098,
      "mean_token_accuracy": 0.2291048213839531,
      "num_tokens": 113066307.0,
      "step": 65180
    },
    {
      "entropy": 5.402387189865112,
      "epoch": 5.071581404396032,
      "grad_norm": 1.15625,
      "learning_rate": 0.00027264976090915835,
      "loss": 4.7185,
      "mean_token_accuracy": 0.22869402319192886,
      "num_tokens": 113074653.0,
      "step": 65185
    },
    {
      "entropy": 5.326604843139648,
      "epoch": 5.071970433767749,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002726220468514694,
      "loss": 4.6287,
      "mean_token_accuracy": 0.2319682404398918,
      "num_tokens": 113082505.0,
      "step": 65190
    },
    {
      "entropy": 5.336994791030884,
      "epoch": 5.072359463139467,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002725943328298621,
      "loss": 4.6523,
      "mean_token_accuracy": 0.24229722768068312,
      "num_tokens": 113091262.0,
      "step": 65195
    },
    {
      "entropy": 5.43653359413147,
      "epoch": 5.0727484925111845,
      "grad_norm": 1.296875,
      "learning_rate": 0.000272566618844757,
      "loss": 4.7382,
      "mean_token_accuracy": 0.23399693369865418,
      "num_tokens": 113100122.0,
      "step": 65200
    },
    {
      "entropy": 5.354515600204468,
      "epoch": 5.073137521882902,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027253890489657457,
      "loss": 4.6275,
      "mean_token_accuracy": 0.2403170570731163,
      "num_tokens": 113108502.0,
      "step": 65205
    },
    {
      "entropy": 5.3438811779022215,
      "epoch": 5.07352655125462,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00027251119098573533,
      "loss": 4.6775,
      "mean_token_accuracy": 0.23231572657823563,
      "num_tokens": 113117791.0,
      "step": 65210
    },
    {
      "entropy": 5.3999104499816895,
      "epoch": 5.073915580626338,
      "grad_norm": 1.25,
      "learning_rate": 0.00027248347711265983,
      "loss": 4.7369,
      "mean_token_accuracy": 0.22982853055000305,
      "num_tokens": 113126387.0,
      "step": 65215
    },
    {
      "entropy": 5.368460369110108,
      "epoch": 5.074304609998054,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002724557632777686,
      "loss": 4.7231,
      "mean_token_accuracy": 0.2350810706615448,
      "num_tokens": 113134856.0,
      "step": 65220
    },
    {
      "entropy": 5.325463533401489,
      "epoch": 5.074693639369772,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00027242804948148197,
      "loss": 4.593,
      "mean_token_accuracy": 0.24041689038276673,
      "num_tokens": 113143431.0,
      "step": 65225
    },
    {
      "entropy": 5.397089242935181,
      "epoch": 5.07508266874149,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002724003357242207,
      "loss": 4.6766,
      "mean_token_accuracy": 0.23118810951709748,
      "num_tokens": 113151471.0,
      "step": 65230
    },
    {
      "entropy": 5.38531002998352,
      "epoch": 5.0754716981132075,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002723726220064051,
      "loss": 4.7356,
      "mean_token_accuracy": 0.22910382747650146,
      "num_tokens": 113160486.0,
      "step": 65235
    },
    {
      "entropy": 5.378416967391968,
      "epoch": 5.075860727484925,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002723449083284559,
      "loss": 4.604,
      "mean_token_accuracy": 0.2464513823390007,
      "num_tokens": 113168141.0,
      "step": 65240
    },
    {
      "entropy": 5.363914918899536,
      "epoch": 5.076249756856643,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027231719469079344,
      "loss": 4.7157,
      "mean_token_accuracy": 0.2285948872566223,
      "num_tokens": 113177185.0,
      "step": 65245
    },
    {
      "entropy": 5.381011819839477,
      "epoch": 5.076638786228361,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027228948109383826,
      "loss": 4.6883,
      "mean_token_accuracy": 0.23764241188764573,
      "num_tokens": 113185557.0,
      "step": 65250
    },
    {
      "entropy": 5.366472244262695,
      "epoch": 5.077027815600077,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027226176753801095,
      "loss": 4.6643,
      "mean_token_accuracy": 0.2350686401128769,
      "num_tokens": 113194235.0,
      "step": 65255
    },
    {
      "entropy": 5.368444204330444,
      "epoch": 5.077416844971795,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00027223405402373194,
      "loss": 4.7636,
      "mean_token_accuracy": 0.22911095768213272,
      "num_tokens": 113203843.0,
      "step": 65260
    },
    {
      "entropy": 5.346606731414795,
      "epoch": 5.077805874343513,
      "grad_norm": 1.328125,
      "learning_rate": 0.00027220634055142174,
      "loss": 4.682,
      "mean_token_accuracy": 0.23661319613456727,
      "num_tokens": 113212229.0,
      "step": 65265
    },
    {
      "entropy": 5.415178871154785,
      "epoch": 5.0781949037152305,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002721786271215009,
      "loss": 4.8116,
      "mean_token_accuracy": 0.21662216782569885,
      "num_tokens": 113220577.0,
      "step": 65270
    },
    {
      "entropy": 5.301629161834716,
      "epoch": 5.078583933086948,
      "grad_norm": 1.375,
      "learning_rate": 0.0002721509137343899,
      "loss": 4.5159,
      "mean_token_accuracy": 0.24785857349634172,
      "num_tokens": 113228445.0,
      "step": 65275
    },
    {
      "entropy": 5.333738899230957,
      "epoch": 5.078972962458666,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027212320039050924,
      "loss": 4.599,
      "mean_token_accuracy": 0.2419826328754425,
      "num_tokens": 113236988.0,
      "step": 65280
    },
    {
      "entropy": 5.380702638626099,
      "epoch": 5.079361991830384,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027209548709027944,
      "loss": 4.7219,
      "mean_token_accuracy": 0.22917523980140686,
      "num_tokens": 113245655.0,
      "step": 65285
    },
    {
      "entropy": 5.335486555099488,
      "epoch": 5.0797510212021,
      "grad_norm": 1.171875,
      "learning_rate": 0.000272067773834121,
      "loss": 4.695,
      "mean_token_accuracy": 0.22945841401815414,
      "num_tokens": 113254900.0,
      "step": 65290
    },
    {
      "entropy": 5.298296976089477,
      "epoch": 5.080140050573818,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002720400606224544,
      "loss": 4.5718,
      "mean_token_accuracy": 0.23297424912452697,
      "num_tokens": 113262983.0,
      "step": 65295
    },
    {
      "entropy": 5.412521028518677,
      "epoch": 5.080529079945536,
      "grad_norm": 1.203125,
      "learning_rate": 0.00027201234745570014,
      "loss": 4.786,
      "mean_token_accuracy": 0.21966168880462647,
      "num_tokens": 113271704.0,
      "step": 65300
    },
    {
      "entropy": 5.358068656921387,
      "epoch": 5.0809181093172535,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002719846343342788,
      "loss": 4.6154,
      "mean_token_accuracy": 0.23228641003370284,
      "num_tokens": 113280564.0,
      "step": 65305
    },
    {
      "entropy": 5.348234415054321,
      "epoch": 5.081307138688971,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002719569212586108,
      "loss": 4.5945,
      "mean_token_accuracy": 0.24256988763809204,
      "num_tokens": 113288462.0,
      "step": 65310
    },
    {
      "entropy": 5.337446546554565,
      "epoch": 5.081696168060689,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002719292082291167,
      "loss": 4.7182,
      "mean_token_accuracy": 0.23657677173614503,
      "num_tokens": 113296554.0,
      "step": 65315
    },
    {
      "entropy": 5.418079853057861,
      "epoch": 5.082085197432406,
      "grad_norm": 1.1875,
      "learning_rate": 0.000271901495246217,
      "loss": 4.7363,
      "mean_token_accuracy": 0.2304416850209236,
      "num_tokens": 113305359.0,
      "step": 65320
    },
    {
      "entropy": 5.430291318893433,
      "epoch": 5.082474226804123,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002718737823103322,
      "loss": 4.7093,
      "mean_token_accuracy": 0.2349811926484108,
      "num_tokens": 113313902.0,
      "step": 65325
    },
    {
      "entropy": 5.367583560943603,
      "epoch": 5.082863256175841,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027184606942188275,
      "loss": 4.6627,
      "mean_token_accuracy": 0.23261507153511046,
      "num_tokens": 113322150.0,
      "step": 65330
    },
    {
      "entropy": 5.370856618881225,
      "epoch": 5.083252285547559,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00027181835658128913,
      "loss": 4.6895,
      "mean_token_accuracy": 0.24248858243227006,
      "num_tokens": 113331066.0,
      "step": 65335
    },
    {
      "entropy": 5.417171812057495,
      "epoch": 5.0836413149192765,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002717906437889719,
      "loss": 4.8017,
      "mean_token_accuracy": 0.23062849491834642,
      "num_tokens": 113339795.0,
      "step": 65340
    },
    {
      "entropy": 5.418705415725708,
      "epoch": 5.084030344290994,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027176293104535154,
      "loss": 4.698,
      "mean_token_accuracy": 0.23209146708250045,
      "num_tokens": 113348153.0,
      "step": 65345
    },
    {
      "entropy": 5.344709396362305,
      "epoch": 5.084419373662712,
      "grad_norm": 1.203125,
      "learning_rate": 0.00027173521835084863,
      "loss": 4.6283,
      "mean_token_accuracy": 0.24109508246183395,
      "num_tokens": 113356887.0,
      "step": 65350
    },
    {
      "entropy": 5.311906099319458,
      "epoch": 5.084808403034429,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002717075057058835,
      "loss": 4.5629,
      "mean_token_accuracy": 0.2501395508646965,
      "num_tokens": 113364883.0,
      "step": 65355
    },
    {
      "entropy": 5.401915597915649,
      "epoch": 5.085197432406146,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002716797931108768,
      "loss": 4.7574,
      "mean_token_accuracy": 0.2282385140657425,
      "num_tokens": 113374157.0,
      "step": 65360
    },
    {
      "entropy": 5.328366088867187,
      "epoch": 5.085586461777864,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002716520805662489,
      "loss": 4.5946,
      "mean_token_accuracy": 0.24097496122121811,
      "num_tokens": 113382463.0,
      "step": 65365
    },
    {
      "entropy": 5.347125434875489,
      "epoch": 5.085975491149582,
      "grad_norm": 1.203125,
      "learning_rate": 0.00027162436807242024,
      "loss": 4.6249,
      "mean_token_accuracy": 0.22953520715236664,
      "num_tokens": 113391960.0,
      "step": 65370
    },
    {
      "entropy": 5.399878263473511,
      "epoch": 5.0863645205212995,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00027159665562981154,
      "loss": 4.7765,
      "mean_token_accuracy": 0.2258735254406929,
      "num_tokens": 113401119.0,
      "step": 65375
    },
    {
      "entropy": 5.429216384887695,
      "epoch": 5.086753549893017,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002715689432388433,
      "loss": 4.7464,
      "mean_token_accuracy": 0.2259046271443367,
      "num_tokens": 113409048.0,
      "step": 65380
    },
    {
      "entropy": 5.396190357208252,
      "epoch": 5.087142579264735,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002715412308999357,
      "loss": 4.7887,
      "mean_token_accuracy": 0.22731771171092988,
      "num_tokens": 113417572.0,
      "step": 65385
    },
    {
      "entropy": 5.407434415817261,
      "epoch": 5.087531608636452,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027151351861350953,
      "loss": 4.6586,
      "mean_token_accuracy": 0.23978892117738723,
      "num_tokens": 113426043.0,
      "step": 65390
    },
    {
      "entropy": 5.431021404266358,
      "epoch": 5.087920638008169,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002714858063799851,
      "loss": 4.6963,
      "mean_token_accuracy": 0.23497094660997392,
      "num_tokens": 113435268.0,
      "step": 65395
    },
    {
      "entropy": 5.3180399417877195,
      "epoch": 5.088309667379887,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027145809419978293,
      "loss": 4.6335,
      "mean_token_accuracy": 0.2320336729288101,
      "num_tokens": 113443495.0,
      "step": 65400
    },
    {
      "entropy": 5.387393283843994,
      "epoch": 5.088698696751605,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027143038207332363,
      "loss": 4.6805,
      "mean_token_accuracy": 0.23501917570829392,
      "num_tokens": 113451615.0,
      "step": 65405
    },
    {
      "entropy": 5.356422996520996,
      "epoch": 5.0890877261233225,
      "grad_norm": 1.203125,
      "learning_rate": 0.00027140267000102755,
      "loss": 4.6329,
      "mean_token_accuracy": 0.230201655626297,
      "num_tokens": 113460134.0,
      "step": 65410
    },
    {
      "entropy": 5.404710054397583,
      "epoch": 5.08947675549504,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027137495798331524,
      "loss": 4.6868,
      "mean_token_accuracy": 0.23152797371149064,
      "num_tokens": 113468388.0,
      "step": 65415
    },
    {
      "entropy": 5.401102113723755,
      "epoch": 5.089865784866757,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027134724602060717,
      "loss": 4.7242,
      "mean_token_accuracy": 0.23266421854496003,
      "num_tokens": 113477562.0,
      "step": 65420
    },
    {
      "entropy": 5.400466060638427,
      "epoch": 5.090254814238475,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027131953411332384,
      "loss": 4.7524,
      "mean_token_accuracy": 0.23158220052719117,
      "num_tokens": 113485798.0,
      "step": 65425
    },
    {
      "entropy": 5.515975522994995,
      "epoch": 5.090643843610192,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027129182226188575,
      "loss": 4.8185,
      "mean_token_accuracy": 0.21520659923553467,
      "num_tokens": 113494396.0,
      "step": 65430
    },
    {
      "entropy": 5.372452163696289,
      "epoch": 5.09103287298191,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002712641104667132,
      "loss": 4.6472,
      "mean_token_accuracy": 0.23959426432847977,
      "num_tokens": 113503017.0,
      "step": 65435
    },
    {
      "entropy": 5.408333730697632,
      "epoch": 5.091421902353628,
      "grad_norm": 1.296875,
      "learning_rate": 0.000271236398728227,
      "loss": 4.695,
      "mean_token_accuracy": 0.2335971936583519,
      "num_tokens": 113512059.0,
      "step": 65440
    },
    {
      "entropy": 5.366698408126831,
      "epoch": 5.0918109317253455,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027120868704684736,
      "loss": 4.6929,
      "mean_token_accuracy": 0.23536107540130616,
      "num_tokens": 113520368.0,
      "step": 65445
    },
    {
      "entropy": 5.3476447582244875,
      "epoch": 5.092199961097063,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002711809754229949,
      "loss": 4.7475,
      "mean_token_accuracy": 0.22523743361234666,
      "num_tokens": 113528486.0,
      "step": 65450
    },
    {
      "entropy": 5.452898168563843,
      "epoch": 5.09258899046878,
      "grad_norm": 1.1875,
      "learning_rate": 0.00027115326385709004,
      "loss": 4.8047,
      "mean_token_accuracy": 0.22994998544454576,
      "num_tokens": 113537390.0,
      "step": 65455
    },
    {
      "entropy": 5.30726957321167,
      "epoch": 5.092978019840498,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002711255523495534,
      "loss": 4.6275,
      "mean_token_accuracy": 0.242209230363369,
      "num_tokens": 113545388.0,
      "step": 65460
    },
    {
      "entropy": 5.369461107254028,
      "epoch": 5.093367049212215,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002710978409008052,
      "loss": 4.6726,
      "mean_token_accuracy": 0.2417624920606613,
      "num_tokens": 113553341.0,
      "step": 65465
    },
    {
      "entropy": 5.307066631317139,
      "epoch": 5.093756078583933,
      "grad_norm": 1.203125,
      "learning_rate": 0.00027107012951126607,
      "loss": 4.6792,
      "mean_token_accuracy": 0.23374118208885192,
      "num_tokens": 113561970.0,
      "step": 65470
    },
    {
      "entropy": 5.3758939743042,
      "epoch": 5.094145107955651,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00027104241818135655,
      "loss": 4.7168,
      "mean_token_accuracy": 0.22601673156023025,
      "num_tokens": 113570088.0,
      "step": 65475
    },
    {
      "entropy": 5.509513330459595,
      "epoch": 5.0945341373273685,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027101470691149693,
      "loss": 4.8637,
      "mean_token_accuracy": 0.21794642508029938,
      "num_tokens": 113579161.0,
      "step": 65480
    },
    {
      "entropy": 5.440929889678955,
      "epoch": 5.094923166699086,
      "grad_norm": 1.265625,
      "learning_rate": 0.00027098699570210787,
      "loss": 4.7084,
      "mean_token_accuracy": 0.22382608950138091,
      "num_tokens": 113588195.0,
      "step": 65485
    },
    {
      "entropy": 5.36814661026001,
      "epoch": 5.095312196070803,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027095928455360977,
      "loss": 4.6277,
      "mean_token_accuracy": 0.24170744717121123,
      "num_tokens": 113596730.0,
      "step": 65490
    },
    {
      "entropy": 5.403841590881347,
      "epoch": 5.095701225442521,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000270931573466423,
      "loss": 4.685,
      "mean_token_accuracy": 0.23899894654750825,
      "num_tokens": 113604889.0,
      "step": 65495
    },
    {
      "entropy": 5.318054246902466,
      "epoch": 5.096090254814238,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002709038624409681,
      "loss": 4.6618,
      "mean_token_accuracy": 0.2315066635608673,
      "num_tokens": 113614387.0,
      "step": 65500
    },
    {
      "entropy": 5.405959224700927,
      "epoch": 5.096479284185956,
      "grad_norm": 1.375,
      "learning_rate": 0.00027087615147766567,
      "loss": 4.6254,
      "mean_token_accuracy": 0.23597293049097062,
      "num_tokens": 113623083.0,
      "step": 65505
    },
    {
      "entropy": 5.296598625183106,
      "epoch": 5.096868313557674,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000270848440576936,
      "loss": 4.6202,
      "mean_token_accuracy": 0.23955265879631044,
      "num_tokens": 113631435.0,
      "step": 65510
    },
    {
      "entropy": 5.342639589309693,
      "epoch": 5.0972573429293915,
      "grad_norm": 1.265625,
      "learning_rate": 0.00027082072973919974,
      "loss": 4.7256,
      "mean_token_accuracy": 0.23370078653097154,
      "num_tokens": 113640562.0,
      "step": 65515
    },
    {
      "entropy": 5.352981519699097,
      "epoch": 5.097646372301108,
      "grad_norm": 1.140625,
      "learning_rate": 0.00027079301896487716,
      "loss": 4.6539,
      "mean_token_accuracy": 0.2374591961503029,
      "num_tokens": 113649388.0,
      "step": 65520
    },
    {
      "entropy": 5.470557737350464,
      "epoch": 5.098035401672826,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002707653082543889,
      "loss": 4.8449,
      "mean_token_accuracy": 0.2149989828467369,
      "num_tokens": 113658212.0,
      "step": 65525
    },
    {
      "entropy": 5.412865018844604,
      "epoch": 5.098424431044544,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002707375976081553,
      "loss": 4.7714,
      "mean_token_accuracy": 0.22632641047239305,
      "num_tokens": 113666773.0,
      "step": 65530
    },
    {
      "entropy": 5.326788997650146,
      "epoch": 5.098813460416261,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027070988702659687,
      "loss": 4.6479,
      "mean_token_accuracy": 0.23933408856391908,
      "num_tokens": 113675429.0,
      "step": 65535
    },
    {
      "entropy": 5.367677450180054,
      "epoch": 5.099202489787979,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027068217651013415,
      "loss": 4.703,
      "mean_token_accuracy": 0.22788191437721253,
      "num_tokens": 113684085.0,
      "step": 65540
    },
    {
      "entropy": 5.40313491821289,
      "epoch": 5.099591519159697,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002706544660591875,
      "loss": 4.6913,
      "mean_token_accuracy": 0.22958648651838304,
      "num_tokens": 113692185.0,
      "step": 65545
    },
    {
      "entropy": 5.346821403503418,
      "epoch": 5.0999805485314145,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00027062675567417736,
      "loss": 4.6662,
      "mean_token_accuracy": 0.2285730227828026,
      "num_tokens": 113700537.0,
      "step": 65550
    },
    {
      "entropy": 5.368384170532226,
      "epoch": 5.100369577903131,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027059904535552424,
      "loss": 4.7745,
      "mean_token_accuracy": 0.23148514032363893,
      "num_tokens": 113709052.0,
      "step": 65555
    },
    {
      "entropy": 5.401798248291016,
      "epoch": 5.100758607274849,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027057133510364866,
      "loss": 4.7635,
      "mean_token_accuracy": 0.23300737738609315,
      "num_tokens": 113717350.0,
      "step": 65560
    },
    {
      "entropy": 5.4771534442901615,
      "epoch": 5.101147636646567,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002705436249189709,
      "loss": 4.77,
      "mean_token_accuracy": 0.22975452542304992,
      "num_tokens": 113726090.0,
      "step": 65565
    },
    {
      "entropy": 5.424429130554199,
      "epoch": 5.101536666018284,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027051591480191167,
      "loss": 4.8082,
      "mean_token_accuracy": 0.22800733149051666,
      "num_tokens": 113734445.0,
      "step": 65570
    },
    {
      "entropy": 5.45571460723877,
      "epoch": 5.101925695390002,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002704882047528912,
      "loss": 4.832,
      "mean_token_accuracy": 0.2218543455004692,
      "num_tokens": 113742743.0,
      "step": 65575
    },
    {
      "entropy": 5.377057123184204,
      "epoch": 5.10231472476172,
      "grad_norm": 1.25,
      "learning_rate": 0.00027046049477233013,
      "loss": 4.6393,
      "mean_token_accuracy": 0.2363483041524887,
      "num_tokens": 113751339.0,
      "step": 65580
    },
    {
      "entropy": 5.462052917480468,
      "epoch": 5.1027037541334375,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002704327848606488,
      "loss": 4.7444,
      "mean_token_accuracy": 0.23309819400310516,
      "num_tokens": 113759372.0,
      "step": 65585
    },
    {
      "entropy": 5.395617198944092,
      "epoch": 5.103092783505154,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002704050750182677,
      "loss": 4.7836,
      "mean_token_accuracy": 0.22500190734863282,
      "num_tokens": 113768121.0,
      "step": 65590
    },
    {
      "entropy": 5.432155609130859,
      "epoch": 5.103481812876872,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027037736524560717,
      "loss": 4.7744,
      "mean_token_accuracy": 0.22170761227607727,
      "num_tokens": 113776583.0,
      "step": 65595
    },
    {
      "entropy": 5.3910950183868405,
      "epoch": 5.10387084224859,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002703496555430879,
      "loss": 4.6661,
      "mean_token_accuracy": 0.2371536076068878,
      "num_tokens": 113784364.0,
      "step": 65600
    },
    {
      "entropy": 5.42143235206604,
      "epoch": 5.104259871620307,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027032194591113015,
      "loss": 4.7656,
      "mean_token_accuracy": 0.22701077312231063,
      "num_tokens": 113792714.0,
      "step": 65605
    },
    {
      "entropy": 5.3945355892181395,
      "epoch": 5.104648900992025,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002702942363501544,
      "loss": 4.7606,
      "mean_token_accuracy": 0.23320869654417037,
      "num_tokens": 113801401.0,
      "step": 65610
    },
    {
      "entropy": 5.428937339782715,
      "epoch": 5.105037930363743,
      "grad_norm": 1.25,
      "learning_rate": 0.00027026652686058117,
      "loss": 4.7788,
      "mean_token_accuracy": 0.2287493571639061,
      "num_tokens": 113810110.0,
      "step": 65615
    },
    {
      "entropy": 5.3436315059661865,
      "epoch": 5.1054269597354605,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00027023881744283083,
      "loss": 4.582,
      "mean_token_accuracy": 0.24200268536806108,
      "num_tokens": 113818795.0,
      "step": 65620
    },
    {
      "entropy": 5.358218717575073,
      "epoch": 5.105815989107177,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002702111080973239,
      "loss": 4.6583,
      "mean_token_accuracy": 0.23327534794807434,
      "num_tokens": 113827323.0,
      "step": 65625
    },
    {
      "entropy": 5.340360164642334,
      "epoch": 5.106205018478895,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002701833988244808,
      "loss": 4.708,
      "mean_token_accuracy": 0.23821092545986175,
      "num_tokens": 113835837.0,
      "step": 65630
    },
    {
      "entropy": 5.423778963088989,
      "epoch": 5.106594047850613,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00027015568962472175,
      "loss": 4.7425,
      "mean_token_accuracy": 0.2352370023727417,
      "num_tokens": 113844730.0,
      "step": 65635
    },
    {
      "entropy": 5.402863264083862,
      "epoch": 5.10698307722233,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027012798049846757,
      "loss": 4.6995,
      "mean_token_accuracy": 0.2298349305987358,
      "num_tokens": 113853546.0,
      "step": 65640
    },
    {
      "entropy": 5.437662982940674,
      "epoch": 5.107372106594048,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002701002714461385,
      "loss": 4.7095,
      "mean_token_accuracy": 0.22861822247505187,
      "num_tokens": 113862361.0,
      "step": 65645
    },
    {
      "entropy": 5.375363492965699,
      "epoch": 5.107761135965766,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002700725624681551,
      "loss": 4.6,
      "mean_token_accuracy": 0.23485542833805084,
      "num_tokens": 113871000.0,
      "step": 65650
    },
    {
      "entropy": 5.4757880687713625,
      "epoch": 5.108150165337483,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00027004485356493763,
      "loss": 4.7978,
      "mean_token_accuracy": 0.2202759027481079,
      "num_tokens": 113879285.0,
      "step": 65655
    },
    {
      "entropy": 5.403604364395141,
      "epoch": 5.1085391947092,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027001714473690665,
      "loss": 4.722,
      "mean_token_accuracy": 0.22698740363121034,
      "num_tokens": 113888453.0,
      "step": 65660
    },
    {
      "entropy": 5.306671524047852,
      "epoch": 5.108928224080918,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026998943598448256,
      "loss": 4.6321,
      "mean_token_accuracy": 0.2407928854227066,
      "num_tokens": 113896652.0,
      "step": 65665
    },
    {
      "entropy": 5.407342100143433,
      "epoch": 5.109317253452636,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002699617273080858,
      "loss": 4.6927,
      "mean_token_accuracy": 0.22324788719415664,
      "num_tokens": 113904697.0,
      "step": 65670
    },
    {
      "entropy": 5.438044118881225,
      "epoch": 5.109706282824353,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002699340187081368,
      "loss": 4.8255,
      "mean_token_accuracy": 0.2273674115538597,
      "num_tokens": 113913775.0,
      "step": 65675
    },
    {
      "entropy": 5.447125053405761,
      "epoch": 5.110095312196071,
      "grad_norm": 1.265625,
      "learning_rate": 0.000269906310185056,
      "loss": 4.7095,
      "mean_token_accuracy": 0.22697632610797883,
      "num_tokens": 113922265.0,
      "step": 65680
    },
    {
      "entropy": 5.417784309387207,
      "epoch": 5.110484341567789,
      "grad_norm": 1.125,
      "learning_rate": 0.00026987860173926387,
      "loss": 4.707,
      "mean_token_accuracy": 0.2313396766781807,
      "num_tokens": 113931204.0,
      "step": 65685
    },
    {
      "entropy": 5.449784898757935,
      "epoch": 5.1108733709395056,
      "grad_norm": 1.4375,
      "learning_rate": 0.00026985089337118077,
      "loss": 4.7865,
      "mean_token_accuracy": 0.2242250993847847,
      "num_tokens": 113939764.0,
      "step": 65690
    },
    {
      "entropy": 5.374001741409302,
      "epoch": 5.111262400311223,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00026982318508122724,
      "loss": 4.6958,
      "mean_token_accuracy": 0.2310997411608696,
      "num_tokens": 113949574.0,
      "step": 65695
    },
    {
      "entropy": 5.4805694103240965,
      "epoch": 5.111651429682941,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002697954768698235,
      "loss": 4.7899,
      "mean_token_accuracy": 0.22505026459693908,
      "num_tokens": 113958104.0,
      "step": 65700
    },
    {
      "entropy": 5.4374746799469,
      "epoch": 5.112040459054659,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026976776873739017,
      "loss": 4.6728,
      "mean_token_accuracy": 0.23078602999448777,
      "num_tokens": 113967492.0,
      "step": 65705
    },
    {
      "entropy": 5.427159309387207,
      "epoch": 5.112429488426376,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002697400606843477,
      "loss": 4.7818,
      "mean_token_accuracy": 0.2219594955444336,
      "num_tokens": 113975663.0,
      "step": 65710
    },
    {
      "entropy": 5.414811325073242,
      "epoch": 5.112818517798094,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026971235271111636,
      "loss": 4.8132,
      "mean_token_accuracy": 0.22500250935554506,
      "num_tokens": 113984977.0,
      "step": 65715
    },
    {
      "entropy": 5.481187534332276,
      "epoch": 5.113207547169812,
      "grad_norm": 1.109375,
      "learning_rate": 0.00026968464481811667,
      "loss": 4.7458,
      "mean_token_accuracy": 0.2319274663925171,
      "num_tokens": 113994016.0,
      "step": 65720
    },
    {
      "entropy": 5.492456102371216,
      "epoch": 5.1135965765415285,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026965693700576904,
      "loss": 4.8247,
      "mean_token_accuracy": 0.22845372259616853,
      "num_tokens": 114002129.0,
      "step": 65725
    },
    {
      "entropy": 5.46475510597229,
      "epoch": 5.113985605913246,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026962922927449394,
      "loss": 4.7254,
      "mean_token_accuracy": 0.22603096812963486,
      "num_tokens": 114011250.0,
      "step": 65730
    },
    {
      "entropy": 5.4055990219116214,
      "epoch": 5.114374635284964,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002696015216247117,
      "loss": 4.6904,
      "mean_token_accuracy": 0.22688232511281967,
      "num_tokens": 114020909.0,
      "step": 65735
    },
    {
      "entropy": 5.5107887268066404,
      "epoch": 5.114763664656682,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002695738140568427,
      "loss": 4.8162,
      "mean_token_accuracy": 0.22227773815393448,
      "num_tokens": 114029972.0,
      "step": 65740
    },
    {
      "entropy": 5.3652996063232425,
      "epoch": 5.115152694028399,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002695461065713075,
      "loss": 4.6303,
      "mean_token_accuracy": 0.24354414492845536,
      "num_tokens": 114038290.0,
      "step": 65745
    },
    {
      "entropy": 5.392593622207642,
      "epoch": 5.115541723400117,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026951839916852647,
      "loss": 4.6824,
      "mean_token_accuracy": 0.23378140181303025,
      "num_tokens": 114046959.0,
      "step": 65750
    },
    {
      "entropy": 5.472382307052612,
      "epoch": 5.115930752771834,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026949069184892,
      "loss": 4.7363,
      "mean_token_accuracy": 0.2257254406809807,
      "num_tokens": 114055463.0,
      "step": 65755
    },
    {
      "entropy": 5.448762607574463,
      "epoch": 5.1163197821435515,
      "grad_norm": 1.25,
      "learning_rate": 0.00026946298461290845,
      "loss": 4.7524,
      "mean_token_accuracy": 0.23017695546150208,
      "num_tokens": 114064477.0,
      "step": 65760
    },
    {
      "entropy": 5.388959980010986,
      "epoch": 5.116708811515269,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002694352774609123,
      "loss": 4.7226,
      "mean_token_accuracy": 0.22926197201013565,
      "num_tokens": 114073531.0,
      "step": 65765
    },
    {
      "entropy": 5.366010665893555,
      "epoch": 5.117097840886987,
      "grad_norm": 1.28125,
      "learning_rate": 0.000269407570393352,
      "loss": 4.6486,
      "mean_token_accuracy": 0.23572139739990233,
      "num_tokens": 114081624.0,
      "step": 65770
    },
    {
      "entropy": 5.388136386871338,
      "epoch": 5.117486870258705,
      "grad_norm": 1.21875,
      "learning_rate": 0.000269379863410648,
      "loss": 4.7061,
      "mean_token_accuracy": 0.237593112885952,
      "num_tokens": 114090026.0,
      "step": 65775
    },
    {
      "entropy": 5.3389817237854,
      "epoch": 5.117875899630422,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002693521565132205,
      "loss": 4.6523,
      "mean_token_accuracy": 0.23590691536664962,
      "num_tokens": 114099687.0,
      "step": 65780
    },
    {
      "entropy": 5.447558450698852,
      "epoch": 5.11826492900214,
      "grad_norm": 1.15625,
      "learning_rate": 0.00026932444970149006,
      "loss": 4.6958,
      "mean_token_accuracy": 0.23552740961313248,
      "num_tokens": 114108150.0,
      "step": 65785
    },
    {
      "entropy": 5.406824254989624,
      "epoch": 5.118653958373857,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002692967429758771,
      "loss": 4.6336,
      "mean_token_accuracy": 0.24625136107206344,
      "num_tokens": 114117387.0,
      "step": 65790
    },
    {
      "entropy": 5.425927400588989,
      "epoch": 5.1190429877455745,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026926903633680193,
      "loss": 4.7813,
      "mean_token_accuracy": 0.2240218460559845,
      "num_tokens": 114126295.0,
      "step": 65795
    },
    {
      "entropy": 5.3068506717681885,
      "epoch": 5.119432017117292,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026924132978468505,
      "loss": 4.6641,
      "mean_token_accuracy": 0.23176830857992173,
      "num_tokens": 114134953.0,
      "step": 65800
    },
    {
      "entropy": 5.48882269859314,
      "epoch": 5.11982104648901,
      "grad_norm": 1.140625,
      "learning_rate": 0.00026921362331994684,
      "loss": 4.7152,
      "mean_token_accuracy": 0.23103980273008345,
      "num_tokens": 114143780.0,
      "step": 65805
    },
    {
      "entropy": 5.473871421813965,
      "epoch": 5.120210075860728,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026918591694300776,
      "loss": 4.8075,
      "mean_token_accuracy": 0.22674141228199005,
      "num_tokens": 114151605.0,
      "step": 65810
    },
    {
      "entropy": 5.3538306713104244,
      "epoch": 5.120599105232445,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026915821065428804,
      "loss": 4.6416,
      "mean_token_accuracy": 0.24022527188062667,
      "num_tokens": 114159833.0,
      "step": 65815
    },
    {
      "entropy": 5.430729770660401,
      "epoch": 5.120988134604163,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002691305044542082,
      "loss": 4.7281,
      "mean_token_accuracy": 0.23793275207281112,
      "num_tokens": 114168199.0,
      "step": 65820
    },
    {
      "entropy": 5.390360879898071,
      "epoch": 5.12137716397588,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026910279834318855,
      "loss": 4.696,
      "mean_token_accuracy": 0.22909157425165178,
      "num_tokens": 114176652.0,
      "step": 65825
    },
    {
      "entropy": 5.355378150939941,
      "epoch": 5.1217661933475975,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026907509232164964,
      "loss": 4.6321,
      "mean_token_accuracy": 0.23824428468942643,
      "num_tokens": 114185323.0,
      "step": 65830
    },
    {
      "entropy": 5.365657949447632,
      "epoch": 5.122155222719315,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026904738639001163,
      "loss": 4.5944,
      "mean_token_accuracy": 0.23856932371854783,
      "num_tokens": 114193747.0,
      "step": 65835
    },
    {
      "entropy": 5.3128900051116945,
      "epoch": 5.122544252091033,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002690196805486952,
      "loss": 4.5883,
      "mean_token_accuracy": 0.24285732209682465,
      "num_tokens": 114202116.0,
      "step": 65840
    },
    {
      "entropy": 5.269998836517334,
      "epoch": 5.122933281462751,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00026899197479812053,
      "loss": 4.5588,
      "mean_token_accuracy": 0.2432888925075531,
      "num_tokens": 114210269.0,
      "step": 65845
    },
    {
      "entropy": 5.386642932891846,
      "epoch": 5.123322310834468,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002689642691387081,
      "loss": 4.7266,
      "mean_token_accuracy": 0.22190095782279967,
      "num_tokens": 114219206.0,
      "step": 65850
    },
    {
      "entropy": 5.456344747543335,
      "epoch": 5.123711340206185,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026893656357087833,
      "loss": 4.7169,
      "mean_token_accuracy": 0.2329949989914894,
      "num_tokens": 114228597.0,
      "step": 65855
    },
    {
      "entropy": 5.505204582214356,
      "epoch": 5.124100369577903,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002689088580950515,
      "loss": 4.7598,
      "mean_token_accuracy": 0.2297585353255272,
      "num_tokens": 114236768.0,
      "step": 65860
    },
    {
      "entropy": 5.4713225841522215,
      "epoch": 5.1244893989496205,
      "grad_norm": 1.203125,
      "learning_rate": 0.000268881152711648,
      "loss": 4.8354,
      "mean_token_accuracy": 0.2224276155233383,
      "num_tokens": 114245297.0,
      "step": 65865
    },
    {
      "entropy": 5.429941272735595,
      "epoch": 5.124878428321338,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002688534474210884,
      "loss": 4.721,
      "mean_token_accuracy": 0.2285181149840355,
      "num_tokens": 114253955.0,
      "step": 65870
    },
    {
      "entropy": 5.322048139572144,
      "epoch": 5.125267457693056,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00026882574222379283,
      "loss": 4.6408,
      "mean_token_accuracy": 0.24228358268737793,
      "num_tokens": 114262513.0,
      "step": 65875
    },
    {
      "entropy": 5.36185131072998,
      "epoch": 5.125656487064774,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026879803712018187,
      "loss": 4.6845,
      "mean_token_accuracy": 0.23195052295923232,
      "num_tokens": 114271172.0,
      "step": 65880
    },
    {
      "entropy": 5.3886323928833,
      "epoch": 5.126045516436491,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002687703321106758,
      "loss": 4.716,
      "mean_token_accuracy": 0.22585140615701677,
      "num_tokens": 114278877.0,
      "step": 65885
    },
    {
      "entropy": 5.4015813827514645,
      "epoch": 5.126434545808208,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002687426271956951,
      "loss": 4.6576,
      "mean_token_accuracy": 0.23162325769662856,
      "num_tokens": 114287163.0,
      "step": 65890
    },
    {
      "entropy": 5.400324916839599,
      "epoch": 5.126823575179926,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00026871492237566,
      "loss": 4.7194,
      "mean_token_accuracy": 0.2343520149588585,
      "num_tokens": 114296187.0,
      "step": 65895
    },
    {
      "entropy": 5.314103889465332,
      "epoch": 5.1272126045516435,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026868721765099086,
      "loss": 4.5324,
      "mean_token_accuracy": 0.24911223948001862,
      "num_tokens": 114304548.0,
      "step": 65900
    },
    {
      "entropy": 5.495078992843628,
      "epoch": 5.127601633923361,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002686595130221082,
      "loss": 4.8412,
      "mean_token_accuracy": 0.22773391455411912,
      "num_tokens": 114313163.0,
      "step": 65905
    },
    {
      "entropy": 5.356654357910156,
      "epoch": 5.127990663295079,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026863180848943244,
      "loss": 4.6622,
      "mean_token_accuracy": 0.24243556708097458,
      "num_tokens": 114321350.0,
      "step": 65910
    },
    {
      "entropy": 5.39908995628357,
      "epoch": 5.128379692666797,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002686041040533838,
      "loss": 4.6888,
      "mean_token_accuracy": 0.2335049331188202,
      "num_tokens": 114330155.0,
      "step": 65915
    },
    {
      "entropy": 5.362762022018432,
      "epoch": 5.128768722038514,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002685763997143827,
      "loss": 4.6502,
      "mean_token_accuracy": 0.233978570997715,
      "num_tokens": 114338932.0,
      "step": 65920
    },
    {
      "entropy": 5.450249576568604,
      "epoch": 5.129157751410231,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00026854869547284947,
      "loss": 4.7581,
      "mean_token_accuracy": 0.23521080911159514,
      "num_tokens": 114347239.0,
      "step": 65925
    },
    {
      "entropy": 5.304029130935669,
      "epoch": 5.129546780781949,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026852099132920455,
      "loss": 4.6467,
      "mean_token_accuracy": 0.23812558650970458,
      "num_tokens": 114356421.0,
      "step": 65930
    },
    {
      "entropy": 5.402892255783081,
      "epoch": 5.1299358101536665,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002684932872838683,
      "loss": 4.6567,
      "mean_token_accuracy": 0.23382177650928498,
      "num_tokens": 114365524.0,
      "step": 65935
    },
    {
      "entropy": 5.406023597717285,
      "epoch": 5.130324839525384,
      "grad_norm": 1.296875,
      "learning_rate": 0.000268465583337261,
      "loss": 4.7535,
      "mean_token_accuracy": 0.2272989943623543,
      "num_tokens": 114374731.0,
      "step": 65940
    },
    {
      "entropy": 5.475456047058105,
      "epoch": 5.130713868897102,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002684378794898031,
      "loss": 4.8441,
      "mean_token_accuracy": 0.22046840339899063,
      "num_tokens": 114383051.0,
      "step": 65945
    },
    {
      "entropy": 5.450339984893799,
      "epoch": 5.13110289826882,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026841017574191496,
      "loss": 4.7089,
      "mean_token_accuracy": 0.23366149365901948,
      "num_tokens": 114391512.0,
      "step": 65950
    },
    {
      "entropy": 5.4329948902130125,
      "epoch": 5.131491927640537,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002683824720940168,
      "loss": 4.7347,
      "mean_token_accuracy": 0.23419105410575866,
      "num_tokens": 114399808.0,
      "step": 65955
    },
    {
      "entropy": 5.403705644607544,
      "epoch": 5.131880957012254,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026835476854652926,
      "loss": 4.6794,
      "mean_token_accuracy": 0.23160973638296128,
      "num_tokens": 114408593.0,
      "step": 65960
    },
    {
      "entropy": 5.3698242664337155,
      "epoch": 5.132269986383972,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002683270650998724,
      "loss": 4.751,
      "mean_token_accuracy": 0.2253437116742134,
      "num_tokens": 114416705.0,
      "step": 65965
    },
    {
      "entropy": 5.364916133880615,
      "epoch": 5.1326590157556895,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002682993617544667,
      "loss": 4.6134,
      "mean_token_accuracy": 0.24272384792566298,
      "num_tokens": 114425255.0,
      "step": 65970
    },
    {
      "entropy": 5.380217695236206,
      "epoch": 5.133048045127407,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002682716585107326,
      "loss": 4.6489,
      "mean_token_accuracy": 0.2349974974989891,
      "num_tokens": 114432832.0,
      "step": 65975
    },
    {
      "entropy": 5.449956846237183,
      "epoch": 5.133437074499125,
      "grad_norm": 1.46875,
      "learning_rate": 0.00026824395536909036,
      "loss": 4.8199,
      "mean_token_accuracy": 0.22321844547986985,
      "num_tokens": 114442060.0,
      "step": 65980
    },
    {
      "entropy": 5.497336149215698,
      "epoch": 5.133826103870843,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002682162523299603,
      "loss": 4.8619,
      "mean_token_accuracy": 0.22482421100139618,
      "num_tokens": 114450586.0,
      "step": 65985
    },
    {
      "entropy": 5.427227735519409,
      "epoch": 5.134215133242559,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00026818854939376284,
      "loss": 4.6956,
      "mean_token_accuracy": 0.23442237377166747,
      "num_tokens": 114458636.0,
      "step": 65990
    },
    {
      "entropy": 5.400297927856445,
      "epoch": 5.134604162614277,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026816084656091835,
      "loss": 4.7037,
      "mean_token_accuracy": 0.23371173739433287,
      "num_tokens": 114467920.0,
      "step": 65995
    },
    {
      "entropy": 5.39310884475708,
      "epoch": 5.134993191985995,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026813314383184714,
      "loss": 4.7067,
      "mean_token_accuracy": 0.240615876019001,
      "num_tokens": 114476131.0,
      "step": 66000
    },
    {
      "epoch": 5.134993191985995,
      "eval_entropy": 5.139688380712265,
      "eval_loss": 4.9139790534973145,
      "eval_mean_token_accuracy": 0.22861279457434652,
      "eval_num_tokens": 114476131.0,
      "eval_runtime": 28.7239,
      "eval_samples_per_second": 1446.808,
      "eval_steps_per_second": 180.86,
      "step": 66000
    },
    {
      "entropy": 5.417403602600098,
      "epoch": 5.1353822213577125,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026810544120696943,
      "loss": 4.7298,
      "mean_token_accuracy": 0.2260174885392189,
      "num_tokens": 114484709.0,
      "step": 66005
    },
    {
      "entropy": 5.424469900131226,
      "epoch": 5.13577125072943,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026807773868670575,
      "loss": 4.7065,
      "mean_token_accuracy": 0.231908118724823,
      "num_tokens": 114493045.0,
      "step": 66010
    },
    {
      "entropy": 5.375827789306641,
      "epoch": 5.136160280101148,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002680500362714764,
      "loss": 4.68,
      "mean_token_accuracy": 0.22803279459476472,
      "num_tokens": 114501506.0,
      "step": 66015
    },
    {
      "entropy": 5.491504335403443,
      "epoch": 5.136549309472866,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026802233396170164,
      "loss": 4.775,
      "mean_token_accuracy": 0.22315797358751296,
      "num_tokens": 114510353.0,
      "step": 66020
    },
    {
      "entropy": 5.4479516506195065,
      "epoch": 5.136938338844582,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002679946317578018,
      "loss": 4.7926,
      "mean_token_accuracy": 0.22691000998020172,
      "num_tokens": 114519363.0,
      "step": 66025
    },
    {
      "entropy": 5.430318593978882,
      "epoch": 5.1373273682163,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002679669296601973,
      "loss": 4.7204,
      "mean_token_accuracy": 0.23339899480342866,
      "num_tokens": 114528002.0,
      "step": 66030
    },
    {
      "entropy": 5.467751789093017,
      "epoch": 5.137716397588018,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026793922766930845,
      "loss": 4.8392,
      "mean_token_accuracy": 0.22155335992574693,
      "num_tokens": 114537683.0,
      "step": 66035
    },
    {
      "entropy": 5.372335815429688,
      "epoch": 5.1381054269597355,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026791152578555563,
      "loss": 4.6484,
      "mean_token_accuracy": 0.23907072395086287,
      "num_tokens": 114546946.0,
      "step": 66040
    },
    {
      "entropy": 5.406434679031372,
      "epoch": 5.138494456331453,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026788382400935913,
      "loss": 4.658,
      "mean_token_accuracy": 0.24150171726942063,
      "num_tokens": 114555906.0,
      "step": 66045
    },
    {
      "entropy": 5.354548025131225,
      "epoch": 5.138883485703171,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002678561223411392,
      "loss": 4.6268,
      "mean_token_accuracy": 0.22765659391880036,
      "num_tokens": 114564658.0,
      "step": 66050
    },
    {
      "entropy": 5.385910558700561,
      "epoch": 5.139272515074889,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002678284207813163,
      "loss": 4.7293,
      "mean_token_accuracy": 0.23187315464019775,
      "num_tokens": 114573316.0,
      "step": 66055
    },
    {
      "entropy": 5.332181835174561,
      "epoch": 5.139661544446605,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002678007193303107,
      "loss": 4.6617,
      "mean_token_accuracy": 0.2324550211429596,
      "num_tokens": 114581058.0,
      "step": 66060
    },
    {
      "entropy": 5.4508984088897705,
      "epoch": 5.140050573818323,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002677730179885427,
      "loss": 4.8537,
      "mean_token_accuracy": 0.2183619111776352,
      "num_tokens": 114590991.0,
      "step": 66065
    },
    {
      "entropy": 5.376085042953491,
      "epoch": 5.140439603190041,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026774531675643267,
      "loss": 4.6951,
      "mean_token_accuracy": 0.2333232879638672,
      "num_tokens": 114599656.0,
      "step": 66070
    },
    {
      "entropy": 5.453762054443359,
      "epoch": 5.1408286325617585,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002677176156344009,
      "loss": 4.7911,
      "mean_token_accuracy": 0.22020739316940308,
      "num_tokens": 114608176.0,
      "step": 66075
    },
    {
      "entropy": 5.406126832962036,
      "epoch": 5.141217661933476,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002676899146228678,
      "loss": 4.6754,
      "mean_token_accuracy": 0.23196159303188324,
      "num_tokens": 114617445.0,
      "step": 66080
    },
    {
      "entropy": 5.323542165756225,
      "epoch": 5.141606691305194,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026766221372225354,
      "loss": 4.6911,
      "mean_token_accuracy": 0.2348223090171814,
      "num_tokens": 114626433.0,
      "step": 66085
    },
    {
      "entropy": 5.447685241699219,
      "epoch": 5.141995720676911,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002676345129329785,
      "loss": 4.7943,
      "mean_token_accuracy": 0.22686877697706223,
      "num_tokens": 114635008.0,
      "step": 66090
    },
    {
      "entropy": 5.373812437057495,
      "epoch": 5.142384750048628,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026760681225546296,
      "loss": 4.7182,
      "mean_token_accuracy": 0.23642105758190154,
      "num_tokens": 114643595.0,
      "step": 66095
    },
    {
      "entropy": 5.338349294662476,
      "epoch": 5.142773779420346,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002675791116901273,
      "loss": 4.6492,
      "mean_token_accuracy": 0.23315968066453935,
      "num_tokens": 114652595.0,
      "step": 66100
    },
    {
      "entropy": 5.405515575408936,
      "epoch": 5.143162808792064,
      "grad_norm": 1.25,
      "learning_rate": 0.0002675514112373918,
      "loss": 4.6724,
      "mean_token_accuracy": 0.24144977033138276,
      "num_tokens": 114660765.0,
      "step": 66105
    },
    {
      "entropy": 5.385930585861206,
      "epoch": 5.1435518381637815,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026752371089767686,
      "loss": 4.62,
      "mean_token_accuracy": 0.23673540055751802,
      "num_tokens": 114668894.0,
      "step": 66110
    },
    {
      "entropy": 5.527282285690307,
      "epoch": 5.143940867535499,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002674960106714026,
      "loss": 4.8314,
      "mean_token_accuracy": 0.2227616935968399,
      "num_tokens": 114676896.0,
      "step": 66115
    },
    {
      "entropy": 5.429725551605225,
      "epoch": 5.144329896907217,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026746831055898953,
      "loss": 4.7674,
      "mean_token_accuracy": 0.22142336070537566,
      "num_tokens": 114685945.0,
      "step": 66120
    },
    {
      "entropy": 5.416826295852661,
      "epoch": 5.144718926278934,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002674406105608578,
      "loss": 4.7137,
      "mean_token_accuracy": 0.24404677301645278,
      "num_tokens": 114695039.0,
      "step": 66125
    },
    {
      "entropy": 5.343321990966797,
      "epoch": 5.145107955650651,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002674129106774278,
      "loss": 4.7136,
      "mean_token_accuracy": 0.229978746175766,
      "num_tokens": 114704357.0,
      "step": 66130
    },
    {
      "entropy": 5.48281626701355,
      "epoch": 5.145496985022369,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026738521090911984,
      "loss": 4.8355,
      "mean_token_accuracy": 0.2230332687497139,
      "num_tokens": 114713766.0,
      "step": 66135
    },
    {
      "entropy": 5.5071120262146,
      "epoch": 5.145886014394087,
      "grad_norm": 1.125,
      "learning_rate": 0.0002673575112563541,
      "loss": 4.748,
      "mean_token_accuracy": 0.23243640661239623,
      "num_tokens": 114723840.0,
      "step": 66140
    },
    {
      "entropy": 5.404505443572998,
      "epoch": 5.1462750437658045,
      "grad_norm": 1.234375,
      "learning_rate": 0.000267329811719551,
      "loss": 4.7124,
      "mean_token_accuracy": 0.23389067649841308,
      "num_tokens": 114733083.0,
      "step": 66145
    },
    {
      "entropy": 5.374289703369141,
      "epoch": 5.146664073137522,
      "grad_norm": 1.25,
      "learning_rate": 0.0002673021122991308,
      "loss": 4.6446,
      "mean_token_accuracy": 0.23164769858121873,
      "num_tokens": 114741444.0,
      "step": 66150
    },
    {
      "entropy": 5.3094391345977785,
      "epoch": 5.14705310250924,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002672744129955138,
      "loss": 4.6335,
      "mean_token_accuracy": 0.22914784401655197,
      "num_tokens": 114750095.0,
      "step": 66155
    },
    {
      "entropy": 5.374258804321289,
      "epoch": 5.147442131880957,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002672467138091204,
      "loss": 4.7474,
      "mean_token_accuracy": 0.22606673538684846,
      "num_tokens": 114759075.0,
      "step": 66160
    },
    {
      "entropy": 5.388504266738892,
      "epoch": 5.147831161252674,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002672190147403705,
      "loss": 4.6435,
      "mean_token_accuracy": 0.2358704164624214,
      "num_tokens": 114767909.0,
      "step": 66165
    },
    {
      "entropy": 5.507593059539795,
      "epoch": 5.148220190624392,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002671913157896849,
      "loss": 4.8484,
      "mean_token_accuracy": 0.219556924700737,
      "num_tokens": 114777329.0,
      "step": 66170
    },
    {
      "entropy": 5.430430555343628,
      "epoch": 5.14860921999611,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00026716361695748353,
      "loss": 4.7953,
      "mean_token_accuracy": 0.2295788899064064,
      "num_tokens": 114785971.0,
      "step": 66175
    },
    {
      "entropy": 5.4729015827178955,
      "epoch": 5.1489982493678275,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002671359182441868,
      "loss": 4.8283,
      "mean_token_accuracy": 0.22070123702287675,
      "num_tokens": 114794682.0,
      "step": 66180
    },
    {
      "entropy": 5.341976881027222,
      "epoch": 5.149387278739545,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026710821965021507,
      "loss": 4.6541,
      "mean_token_accuracy": 0.23373156934976577,
      "num_tokens": 114803989.0,
      "step": 66185
    },
    {
      "entropy": 5.469167900085449,
      "epoch": 5.149776308111262,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002670805211759886,
      "loss": 4.8043,
      "mean_token_accuracy": 0.22123969793319703,
      "num_tokens": 114812176.0,
      "step": 66190
    },
    {
      "entropy": 5.375462675094605,
      "epoch": 5.15016533748298,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002670528228219275,
      "loss": 4.6541,
      "mean_token_accuracy": 0.23139166831970215,
      "num_tokens": 114820697.0,
      "step": 66195
    },
    {
      "entropy": 5.462036609649658,
      "epoch": 5.150554366854697,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026702512458845224,
      "loss": 4.7687,
      "mean_token_accuracy": 0.22422267645597457,
      "num_tokens": 114829907.0,
      "step": 66200
    },
    {
      "entropy": 5.476194286346436,
      "epoch": 5.150943396226415,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026699742647598294,
      "loss": 4.7265,
      "mean_token_accuracy": 0.22197712510824202,
      "num_tokens": 114839254.0,
      "step": 66205
    },
    {
      "entropy": 5.432609701156617,
      "epoch": 5.151332425598133,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026696972848494005,
      "loss": 4.6426,
      "mean_token_accuracy": 0.22954562306404114,
      "num_tokens": 114847599.0,
      "step": 66210
    },
    {
      "entropy": 5.4027856349945065,
      "epoch": 5.1517214549698505,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026694203061574366,
      "loss": 4.7726,
      "mean_token_accuracy": 0.22356075197458267,
      "num_tokens": 114857020.0,
      "step": 66215
    },
    {
      "entropy": 5.3387861251831055,
      "epoch": 5.152110484341568,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026691433286881415,
      "loss": 4.6544,
      "mean_token_accuracy": 0.23844123184680938,
      "num_tokens": 114865861.0,
      "step": 66220
    },
    {
      "entropy": 5.3935342788696286,
      "epoch": 5.152499513713285,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002668866352445718,
      "loss": 4.6296,
      "mean_token_accuracy": 0.2350833833217621,
      "num_tokens": 114874128.0,
      "step": 66225
    },
    {
      "entropy": 5.331000328063965,
      "epoch": 5.152888543085003,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002668589377434368,
      "loss": 4.5404,
      "mean_token_accuracy": 0.24898185282945634,
      "num_tokens": 114882685.0,
      "step": 66230
    },
    {
      "entropy": 5.271511220932007,
      "epoch": 5.15327757245672,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002668312403658295,
      "loss": 4.568,
      "mean_token_accuracy": 0.23852672725915908,
      "num_tokens": 114891624.0,
      "step": 66235
    },
    {
      "entropy": 5.307132053375244,
      "epoch": 5.153666601828438,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00026680354311217006,
      "loss": 4.6324,
      "mean_token_accuracy": 0.23841010183095931,
      "num_tokens": 114900438.0,
      "step": 66240
    },
    {
      "entropy": 5.299925899505615,
      "epoch": 5.154055631200156,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026677584598287893,
      "loss": 4.6376,
      "mean_token_accuracy": 0.23247755765914918,
      "num_tokens": 114909625.0,
      "step": 66245
    },
    {
      "entropy": 5.411219072341919,
      "epoch": 5.1544446605718734,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026674814897837616,
      "loss": 4.7552,
      "mean_token_accuracy": 0.23449126332998277,
      "num_tokens": 114918807.0,
      "step": 66250
    },
    {
      "entropy": 5.360559749603271,
      "epoch": 5.154833689943591,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026672045209908215,
      "loss": 4.6051,
      "mean_token_accuracy": 0.2413678452372551,
      "num_tokens": 114927267.0,
      "step": 66255
    },
    {
      "entropy": 5.377671957015991,
      "epoch": 5.155222719315308,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026669275534541703,
      "loss": 4.7546,
      "mean_token_accuracy": 0.2295221731066704,
      "num_tokens": 114935877.0,
      "step": 66260
    },
    {
      "entropy": 5.4117724895477295,
      "epoch": 5.155611748687026,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026666505871780115,
      "loss": 4.675,
      "mean_token_accuracy": 0.2331179141998291,
      "num_tokens": 114943998.0,
      "step": 66265
    },
    {
      "entropy": 5.41001558303833,
      "epoch": 5.156000778058743,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002666373622166548,
      "loss": 4.7276,
      "mean_token_accuracy": 0.2355026423931122,
      "num_tokens": 114952802.0,
      "step": 66270
    },
    {
      "entropy": 5.377454710006714,
      "epoch": 5.156389807430461,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026660966584239815,
      "loss": 4.6809,
      "mean_token_accuracy": 0.23454331010580062,
      "num_tokens": 114961680.0,
      "step": 66275
    },
    {
      "entropy": 5.387762403488159,
      "epoch": 5.156778836802179,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026658196959545145,
      "loss": 4.7556,
      "mean_token_accuracy": 0.22974208295345305,
      "num_tokens": 114970395.0,
      "step": 66280
    },
    {
      "entropy": 5.428138065338135,
      "epoch": 5.157167866173896,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026655427347623494,
      "loss": 4.7717,
      "mean_token_accuracy": 0.2305280700325966,
      "num_tokens": 114979245.0,
      "step": 66285
    },
    {
      "entropy": 5.349743509292603,
      "epoch": 5.157556895545614,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002665265774851689,
      "loss": 4.6679,
      "mean_token_accuracy": 0.23415318876504898,
      "num_tokens": 114988826.0,
      "step": 66290
    },
    {
      "entropy": 5.334540176391601,
      "epoch": 5.157945924917331,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026649888162267357,
      "loss": 4.5987,
      "mean_token_accuracy": 0.2385523021221161,
      "num_tokens": 114997214.0,
      "step": 66295
    },
    {
      "entropy": 5.477506685256958,
      "epoch": 5.158334954289049,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002664711858891691,
      "loss": 4.775,
      "mean_token_accuracy": 0.22298480272293092,
      "num_tokens": 115005600.0,
      "step": 66300
    },
    {
      "entropy": 5.496365880966186,
      "epoch": 5.158723983660766,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002664434902850759,
      "loss": 4.7822,
      "mean_token_accuracy": 0.22870756089687347,
      "num_tokens": 115014605.0,
      "step": 66305
    },
    {
      "entropy": 5.541705894470215,
      "epoch": 5.159113013032484,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026641579481081405,
      "loss": 4.8555,
      "mean_token_accuracy": 0.22103175520896912,
      "num_tokens": 115023810.0,
      "step": 66310
    },
    {
      "entropy": 5.452320337295532,
      "epoch": 5.159502042404202,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026638809946680393,
      "loss": 4.6593,
      "mean_token_accuracy": 0.23280279338359833,
      "num_tokens": 115033129.0,
      "step": 66315
    },
    {
      "entropy": 5.497965288162232,
      "epoch": 5.159891071775919,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026636040425346565,
      "loss": 4.8625,
      "mean_token_accuracy": 0.22063197642564775,
      "num_tokens": 115042204.0,
      "step": 66320
    },
    {
      "entropy": 5.513740348815918,
      "epoch": 5.160280101147636,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002663327091712195,
      "loss": 4.8601,
      "mean_token_accuracy": 0.2238878384232521,
      "num_tokens": 115052103.0,
      "step": 66325
    },
    {
      "entropy": 5.372989463806152,
      "epoch": 5.160669130519354,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002663050142204856,
      "loss": 4.5859,
      "mean_token_accuracy": 0.24200529754161834,
      "num_tokens": 115060212.0,
      "step": 66330
    },
    {
      "entropy": 5.3416666984558105,
      "epoch": 5.161058159891072,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026627731940168445,
      "loss": 4.5768,
      "mean_token_accuracy": 0.24907450377941132,
      "num_tokens": 115069203.0,
      "step": 66335
    },
    {
      "entropy": 5.399899101257324,
      "epoch": 5.161447189262789,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026624962471523595,
      "loss": 4.6831,
      "mean_token_accuracy": 0.23013395667076111,
      "num_tokens": 115077847.0,
      "step": 66340
    },
    {
      "entropy": 5.400329065322876,
      "epoch": 5.161836218634507,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026622193016156054,
      "loss": 4.7399,
      "mean_token_accuracy": 0.226437309384346,
      "num_tokens": 115086300.0,
      "step": 66345
    },
    {
      "entropy": 5.424882936477661,
      "epoch": 5.162225248006225,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002661942357410784,
      "loss": 4.6344,
      "mean_token_accuracy": 0.23460484594106673,
      "num_tokens": 115095041.0,
      "step": 66350
    },
    {
      "entropy": 5.354291343688965,
      "epoch": 5.162614277377942,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002661665414542097,
      "loss": 4.6027,
      "mean_token_accuracy": 0.23665731996297837,
      "num_tokens": 115103131.0,
      "step": 66355
    },
    {
      "entropy": 5.464250564575195,
      "epoch": 5.163003306749659,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026613884730137454,
      "loss": 4.7487,
      "mean_token_accuracy": 0.23160839974880218,
      "num_tokens": 115112336.0,
      "step": 66360
    },
    {
      "entropy": 5.344486999511719,
      "epoch": 5.163392336121377,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002661111532829933,
      "loss": 4.6744,
      "mean_token_accuracy": 0.23536170423030853,
      "num_tokens": 115121488.0,
      "step": 66365
    },
    {
      "entropy": 5.343558597564697,
      "epoch": 5.163781365493095,
      "grad_norm": 1.15625,
      "learning_rate": 0.00026608345939948625,
      "loss": 4.6278,
      "mean_token_accuracy": 0.24175651222467423,
      "num_tokens": 115129536.0,
      "step": 66370
    },
    {
      "entropy": 5.3458113193511965,
      "epoch": 5.164170394864812,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002660557656512734,
      "loss": 4.6711,
      "mean_token_accuracy": 0.23355360329151154,
      "num_tokens": 115137793.0,
      "step": 66375
    },
    {
      "entropy": 5.3754056930542,
      "epoch": 5.16455942423653,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026602807203877525,
      "loss": 4.6943,
      "mean_token_accuracy": 0.23153086155653,
      "num_tokens": 115146628.0,
      "step": 66380
    },
    {
      "entropy": 5.3893506050109865,
      "epoch": 5.164948453608248,
      "grad_norm": 1.15625,
      "learning_rate": 0.00026600037856241174,
      "loss": 4.7137,
      "mean_token_accuracy": 0.22357339709997176,
      "num_tokens": 115155423.0,
      "step": 66385
    },
    {
      "entropy": 5.36907992362976,
      "epoch": 5.1653374829799645,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026597268522260317,
      "loss": 4.7184,
      "mean_token_accuracy": 0.23372651040554046,
      "num_tokens": 115163632.0,
      "step": 66390
    },
    {
      "entropy": 5.419540786743164,
      "epoch": 5.165726512351682,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026594499201976966,
      "loss": 4.7339,
      "mean_token_accuracy": 0.22615587413311006,
      "num_tokens": 115172738.0,
      "step": 66395
    },
    {
      "entropy": 5.523875761032104,
      "epoch": 5.1661155417234,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026591729895433157,
      "loss": 4.8184,
      "mean_token_accuracy": 0.2119523987174034,
      "num_tokens": 115181587.0,
      "step": 66400
    },
    {
      "entropy": 5.348265314102173,
      "epoch": 5.166504571095118,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026588960602670896,
      "loss": 4.5959,
      "mean_token_accuracy": 0.24473153799772263,
      "num_tokens": 115190243.0,
      "step": 66405
    },
    {
      "entropy": 5.359602737426758,
      "epoch": 5.166893600466835,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026586191323732213,
      "loss": 4.689,
      "mean_token_accuracy": 0.24254054725170135,
      "num_tokens": 115198725.0,
      "step": 66410
    },
    {
      "entropy": 5.411100435256958,
      "epoch": 5.167282629838553,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002658342205865912,
      "loss": 4.7322,
      "mean_token_accuracy": 0.22872708290815352,
      "num_tokens": 115207344.0,
      "step": 66415
    },
    {
      "entropy": 5.454765367507934,
      "epoch": 5.167671659210271,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002658065280749363,
      "loss": 4.7395,
      "mean_token_accuracy": 0.2290185958147049,
      "num_tokens": 115216744.0,
      "step": 66420
    },
    {
      "entropy": 5.335331201553345,
      "epoch": 5.1680606885819875,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026577883570277786,
      "loss": 4.7042,
      "mean_token_accuracy": 0.2287403479218483,
      "num_tokens": 115226243.0,
      "step": 66425
    },
    {
      "entropy": 5.423353958129883,
      "epoch": 5.168449717953705,
      "grad_norm": 1.25,
      "learning_rate": 0.00026575114347053574,
      "loss": 4.7179,
      "mean_token_accuracy": 0.2296246334910393,
      "num_tokens": 115234687.0,
      "step": 66430
    },
    {
      "entropy": 5.429052209854126,
      "epoch": 5.168838747325423,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002657234513786304,
      "loss": 4.7533,
      "mean_token_accuracy": 0.23035000115633011,
      "num_tokens": 115243799.0,
      "step": 66435
    },
    {
      "entropy": 5.414677143096924,
      "epoch": 5.169227776697141,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002656957594274819,
      "loss": 4.774,
      "mean_token_accuracy": 0.231318461894989,
      "num_tokens": 115252637.0,
      "step": 66440
    },
    {
      "entropy": 5.404609060287475,
      "epoch": 5.169616806068858,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026566806761751043,
      "loss": 4.7208,
      "mean_token_accuracy": 0.23430444151163102,
      "num_tokens": 115261093.0,
      "step": 66445
    },
    {
      "entropy": 5.304970026016235,
      "epoch": 5.170005835440576,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002656403759491362,
      "loss": 4.5581,
      "mean_token_accuracy": 0.24215611666440964,
      "num_tokens": 115270148.0,
      "step": 66450
    },
    {
      "entropy": 5.422722005844117,
      "epoch": 5.170394864812294,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002656126844227794,
      "loss": 4.7347,
      "mean_token_accuracy": 0.2373960554599762,
      "num_tokens": 115278491.0,
      "step": 66455
    },
    {
      "entropy": 5.39317626953125,
      "epoch": 5.1707838941840105,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026558499303886015,
      "loss": 4.636,
      "mean_token_accuracy": 0.23231945484876632,
      "num_tokens": 115287756.0,
      "step": 66460
    },
    {
      "entropy": 5.414976596832275,
      "epoch": 5.171172923555728,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026555730179779864,
      "loss": 4.7326,
      "mean_token_accuracy": 0.2338121324777603,
      "num_tokens": 115296817.0,
      "step": 66465
    },
    {
      "entropy": 5.419347667694092,
      "epoch": 5.171561952927446,
      "grad_norm": 1.21875,
      "learning_rate": 0.000265529610700015,
      "loss": 4.6616,
      "mean_token_accuracy": 0.23219528198242187,
      "num_tokens": 115305145.0,
      "step": 66470
    },
    {
      "entropy": 5.364060878753662,
      "epoch": 5.171950982299164,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002655019197459295,
      "loss": 4.6635,
      "mean_token_accuracy": 0.2307695910334587,
      "num_tokens": 115314040.0,
      "step": 66475
    },
    {
      "entropy": 5.3408125877380375,
      "epoch": 5.172340011670881,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026547422893596223,
      "loss": 4.6804,
      "mean_token_accuracy": 0.2298760488629341,
      "num_tokens": 115321936.0,
      "step": 66480
    },
    {
      "entropy": 5.344593906402588,
      "epoch": 5.172729041042599,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026544653827053336,
      "loss": 4.6997,
      "mean_token_accuracy": 0.22792448252439498,
      "num_tokens": 115331148.0,
      "step": 66485
    },
    {
      "entropy": 5.4350574016571045,
      "epoch": 5.173118070414317,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002654188477500631,
      "loss": 4.7181,
      "mean_token_accuracy": 0.22526077181100845,
      "num_tokens": 115339599.0,
      "step": 66490
    },
    {
      "entropy": 5.406511545181274,
      "epoch": 5.1735070997860335,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002653911573749715,
      "loss": 4.6486,
      "mean_token_accuracy": 0.24284175932407379,
      "num_tokens": 115347850.0,
      "step": 66495
    },
    {
      "entropy": 5.361261892318725,
      "epoch": 5.173896129157751,
      "grad_norm": 1.25,
      "learning_rate": 0.00026536346714567887,
      "loss": 4.6237,
      "mean_token_accuracy": 0.23030744791030883,
      "num_tokens": 115356686.0,
      "step": 66500
    },
    {
      "entropy": 5.364866447448731,
      "epoch": 5.174285158529469,
      "grad_norm": 1.40625,
      "learning_rate": 0.00026533577706260524,
      "loss": 4.7521,
      "mean_token_accuracy": 0.2323892369866371,
      "num_tokens": 115366003.0,
      "step": 66505
    },
    {
      "entropy": 5.355849647521973,
      "epoch": 5.174674187901187,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002653080871261709,
      "loss": 4.6653,
      "mean_token_accuracy": 0.2354801744222641,
      "num_tokens": 115373876.0,
      "step": 66510
    },
    {
      "entropy": 5.458825588226318,
      "epoch": 5.175063217272904,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002652803973367958,
      "loss": 4.783,
      "mean_token_accuracy": 0.22823418974876403,
      "num_tokens": 115383175.0,
      "step": 66515
    },
    {
      "entropy": 5.411269998550415,
      "epoch": 5.175452246644622,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00026525270769490024,
      "loss": 4.6751,
      "mean_token_accuracy": 0.24501416236162185,
      "num_tokens": 115391885.0,
      "step": 66520
    },
    {
      "entropy": 5.35424542427063,
      "epoch": 5.175841276016339,
      "grad_norm": 1.15625,
      "learning_rate": 0.00026522501820090433,
      "loss": 4.6143,
      "mean_token_accuracy": 0.23891518861055375,
      "num_tokens": 115400948.0,
      "step": 66525
    },
    {
      "entropy": 5.367395305633545,
      "epoch": 5.1762303053880565,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026519732885522826,
      "loss": 4.6984,
      "mean_token_accuracy": 0.23093351870775222,
      "num_tokens": 115409702.0,
      "step": 66530
    },
    {
      "entropy": 5.3480242729187015,
      "epoch": 5.176619334759774,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026516963965829204,
      "loss": 4.69,
      "mean_token_accuracy": 0.23169441819190978,
      "num_tokens": 115418213.0,
      "step": 66535
    },
    {
      "entropy": 5.407758331298828,
      "epoch": 5.177008364131492,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000265141950610516,
      "loss": 4.7319,
      "mean_token_accuracy": 0.232973350584507,
      "num_tokens": 115426918.0,
      "step": 66540
    },
    {
      "entropy": 5.437516117095948,
      "epoch": 5.17739739350321,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002651142617123201,
      "loss": 4.7501,
      "mean_token_accuracy": 0.23080938011407853,
      "num_tokens": 115435624.0,
      "step": 66545
    },
    {
      "entropy": 5.407209634780884,
      "epoch": 5.177786422874927,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002650865729641246,
      "loss": 4.6489,
      "mean_token_accuracy": 0.2442161485552788,
      "num_tokens": 115443921.0,
      "step": 66550
    },
    {
      "entropy": 5.38073239326477,
      "epoch": 5.178175452246645,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002650588843663495,
      "loss": 4.7763,
      "mean_token_accuracy": 0.22102256417274474,
      "num_tokens": 115453110.0,
      "step": 66555
    },
    {
      "entropy": 5.353304004669189,
      "epoch": 5.178564481618362,
      "grad_norm": 1.25,
      "learning_rate": 0.000265031195919415,
      "loss": 4.6581,
      "mean_token_accuracy": 0.23622562885284423,
      "num_tokens": 115461374.0,
      "step": 66560
    },
    {
      "entropy": 5.458454895019531,
      "epoch": 5.1789535109900795,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026500350762374115,
      "loss": 4.8031,
      "mean_token_accuracy": 0.2235411912202835,
      "num_tokens": 115470483.0,
      "step": 66565
    },
    {
      "entropy": 5.350240468978882,
      "epoch": 5.179342540361797,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026497581947974836,
      "loss": 4.5439,
      "mean_token_accuracy": 0.24542067795991898,
      "num_tokens": 115479644.0,
      "step": 66570
    },
    {
      "entropy": 5.407918453216553,
      "epoch": 5.179731569733515,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026494813148785645,
      "loss": 4.7614,
      "mean_token_accuracy": 0.22682550698518752,
      "num_tokens": 115488353.0,
      "step": 66575
    },
    {
      "entropy": 5.357215642929077,
      "epoch": 5.180120599105233,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026492044364848566,
      "loss": 4.6195,
      "mean_token_accuracy": 0.23776522427797317,
      "num_tokens": 115497137.0,
      "step": 66580
    },
    {
      "entropy": 5.2926239490509035,
      "epoch": 5.18050962847695,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002648927559620561,
      "loss": 4.5748,
      "mean_token_accuracy": 0.24359971284866333,
      "num_tokens": 115506289.0,
      "step": 66585
    },
    {
      "entropy": 5.433597755432129,
      "epoch": 5.180898657848668,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002648650684289879,
      "loss": 4.7014,
      "mean_token_accuracy": 0.22783830761909485,
      "num_tokens": 115514704.0,
      "step": 66590
    },
    {
      "entropy": 5.312577915191651,
      "epoch": 5.181287687220385,
      "grad_norm": 1.25,
      "learning_rate": 0.0002648373810497011,
      "loss": 4.6575,
      "mean_token_accuracy": 0.2387799620628357,
      "num_tokens": 115523381.0,
      "step": 66595
    },
    {
      "entropy": 5.405880546569824,
      "epoch": 5.1816767165921025,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002648096938246159,
      "loss": 4.7305,
      "mean_token_accuracy": 0.23448034673929213,
      "num_tokens": 115532128.0,
      "step": 66600
    },
    {
      "entropy": 5.462495756149292,
      "epoch": 5.18206574596382,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002647820067541524,
      "loss": 4.7757,
      "mean_token_accuracy": 0.23255939781665802,
      "num_tokens": 115541018.0,
      "step": 66605
    },
    {
      "entropy": 5.4276063442230225,
      "epoch": 5.182454775335538,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026475431983873066,
      "loss": 4.7561,
      "mean_token_accuracy": 0.22834536731243132,
      "num_tokens": 115550273.0,
      "step": 66610
    },
    {
      "entropy": 5.474047756195068,
      "epoch": 5.182843804707256,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002647266330787708,
      "loss": 4.7131,
      "mean_token_accuracy": 0.23367710262537003,
      "num_tokens": 115558735.0,
      "step": 66615
    },
    {
      "entropy": 5.479162406921387,
      "epoch": 5.183232834078973,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026469894647469295,
      "loss": 4.8239,
      "mean_token_accuracy": 0.22698970437049865,
      "num_tokens": 115567359.0,
      "step": 66620
    },
    {
      "entropy": 5.355253219604492,
      "epoch": 5.183621863450691,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002646712600269172,
      "loss": 4.6481,
      "mean_token_accuracy": 0.2414157435297966,
      "num_tokens": 115575119.0,
      "step": 66625
    },
    {
      "entropy": 5.4224800109863285,
      "epoch": 5.184010892822408,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002646435737358635,
      "loss": 4.7406,
      "mean_token_accuracy": 0.23100375831127168,
      "num_tokens": 115583691.0,
      "step": 66630
    },
    {
      "entropy": 5.4344069480896,
      "epoch": 5.1843999221941255,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002646158876019522,
      "loss": 4.7263,
      "mean_token_accuracy": 0.23520067632198333,
      "num_tokens": 115592425.0,
      "step": 66635
    },
    {
      "entropy": 5.417125701904297,
      "epoch": 5.184788951565843,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002645882016256033,
      "loss": 4.685,
      "mean_token_accuracy": 0.2304583102464676,
      "num_tokens": 115602317.0,
      "step": 66640
    },
    {
      "entropy": 5.391387271881103,
      "epoch": 5.185177980937561,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002645605158072368,
      "loss": 4.7441,
      "mean_token_accuracy": 0.23463061451911926,
      "num_tokens": 115610596.0,
      "step": 66645
    },
    {
      "entropy": 5.371836709976196,
      "epoch": 5.185567010309279,
      "grad_norm": 1.3125,
      "learning_rate": 0.000264532830147273,
      "loss": 4.6892,
      "mean_token_accuracy": 0.22938281297683716,
      "num_tokens": 115619737.0,
      "step": 66650
    },
    {
      "entropy": 5.425459289550782,
      "epoch": 5.185956039680996,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026450514464613177,
      "loss": 4.7041,
      "mean_token_accuracy": 0.22937609106302262,
      "num_tokens": 115628160.0,
      "step": 66655
    },
    {
      "entropy": 5.3927388191223145,
      "epoch": 5.186345069052713,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002644774593042332,
      "loss": 4.7353,
      "mean_token_accuracy": 0.23127149492502214,
      "num_tokens": 115636267.0,
      "step": 66660
    },
    {
      "entropy": 5.396825504302979,
      "epoch": 5.186734098424431,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002644497741219975,
      "loss": 4.6858,
      "mean_token_accuracy": 0.23586197346448898,
      "num_tokens": 115645043.0,
      "step": 66665
    },
    {
      "entropy": 5.3639326095581055,
      "epoch": 5.1871231277961485,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002644220890998447,
      "loss": 4.5971,
      "mean_token_accuracy": 0.24752693325281144,
      "num_tokens": 115653073.0,
      "step": 66670
    },
    {
      "entropy": 5.354261684417724,
      "epoch": 5.187512157167866,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026439440423819483,
      "loss": 4.7086,
      "mean_token_accuracy": 0.23109793215990065,
      "num_tokens": 115661979.0,
      "step": 66675
    },
    {
      "entropy": 5.378603410720825,
      "epoch": 5.187901186539584,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000264366719537468,
      "loss": 4.6686,
      "mean_token_accuracy": 0.23499181121587753,
      "num_tokens": 115670623.0,
      "step": 66680
    },
    {
      "entropy": 5.409667587280273,
      "epoch": 5.188290215911302,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002643390349980843,
      "loss": 4.7159,
      "mean_token_accuracy": 0.22925227284431457,
      "num_tokens": 115679127.0,
      "step": 66685
    },
    {
      "entropy": 5.349870204925537,
      "epoch": 5.188679245283019,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002643113506204637,
      "loss": 4.6897,
      "mean_token_accuracy": 0.23326584398746492,
      "num_tokens": 115687918.0,
      "step": 66690
    },
    {
      "entropy": 5.386233949661255,
      "epoch": 5.189068274654736,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002642836664050263,
      "loss": 4.638,
      "mean_token_accuracy": 0.24373760521411897,
      "num_tokens": 115696002.0,
      "step": 66695
    },
    {
      "entropy": 5.3983948707580565,
      "epoch": 5.189457304026454,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026425598235219235,
      "loss": 4.7082,
      "mean_token_accuracy": 0.2280642047524452,
      "num_tokens": 115705158.0,
      "step": 66700
    },
    {
      "entropy": 5.398994398117066,
      "epoch": 5.1898463333981715,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026422829846238174,
      "loss": 4.6625,
      "mean_token_accuracy": 0.23604964315891266,
      "num_tokens": 115713674.0,
      "step": 66705
    },
    {
      "entropy": 5.471413421630859,
      "epoch": 5.190235362769889,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026420061473601445,
      "loss": 4.7565,
      "mean_token_accuracy": 0.22825206071138382,
      "num_tokens": 115722535.0,
      "step": 66710
    },
    {
      "entropy": 5.4131622314453125,
      "epoch": 5.190624392141607,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026417293117351077,
      "loss": 4.694,
      "mean_token_accuracy": 0.23339079171419144,
      "num_tokens": 115730997.0,
      "step": 66715
    },
    {
      "entropy": 5.424811506271363,
      "epoch": 5.191013421513325,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026414524777529055,
      "loss": 4.7293,
      "mean_token_accuracy": 0.22487339228391648,
      "num_tokens": 115739653.0,
      "step": 66720
    },
    {
      "entropy": 5.4809552192687985,
      "epoch": 5.191402450885041,
      "grad_norm": 1.375,
      "learning_rate": 0.0002641175645417739,
      "loss": 4.7618,
      "mean_token_accuracy": 0.22200549393892288,
      "num_tokens": 115747778.0,
      "step": 66725
    },
    {
      "entropy": 5.416953086853027,
      "epoch": 5.191791480256759,
      "grad_norm": 1.171875,
      "learning_rate": 0.000264089881473381,
      "loss": 4.7557,
      "mean_token_accuracy": 0.22009914070367814,
      "num_tokens": 115757451.0,
      "step": 66730
    },
    {
      "entropy": 5.49046573638916,
      "epoch": 5.192180509628477,
      "grad_norm": 1.25,
      "learning_rate": 0.0002640621985705318,
      "loss": 4.7453,
      "mean_token_accuracy": 0.22851253747940065,
      "num_tokens": 115766091.0,
      "step": 66735
    },
    {
      "entropy": 5.461507320404053,
      "epoch": 5.1925695390001945,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026403451583364616,
      "loss": 4.7116,
      "mean_token_accuracy": 0.23616015613079072,
      "num_tokens": 115774329.0,
      "step": 66740
    },
    {
      "entropy": 5.366433906555176,
      "epoch": 5.192958568371912,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002640068332631444,
      "loss": 4.6621,
      "mean_token_accuracy": 0.23427434414625167,
      "num_tokens": 115782601.0,
      "step": 66745
    },
    {
      "entropy": 5.393208360671997,
      "epoch": 5.19334759774363,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002639791508594464,
      "loss": 4.704,
      "mean_token_accuracy": 0.23692605644464493,
      "num_tokens": 115791261.0,
      "step": 66750
    },
    {
      "entropy": 5.3969056606292725,
      "epoch": 5.193736627115348,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026395146862297225,
      "loss": 4.7063,
      "mean_token_accuracy": 0.2316366031765938,
      "num_tokens": 115800268.0,
      "step": 66755
    },
    {
      "entropy": 5.394565582275391,
      "epoch": 5.194125656487064,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000263923786554142,
      "loss": 4.6543,
      "mean_token_accuracy": 0.23227832168340684,
      "num_tokens": 115809098.0,
      "step": 66760
    },
    {
      "entropy": 5.4379448890686035,
      "epoch": 5.194514685858782,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002638961046533756,
      "loss": 4.7823,
      "mean_token_accuracy": 0.22195144593715668,
      "num_tokens": 115817694.0,
      "step": 66765
    },
    {
      "entropy": 5.40190019607544,
      "epoch": 5.1949037152305,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002638684229210932,
      "loss": 4.7034,
      "mean_token_accuracy": 0.23486540615558624,
      "num_tokens": 115826684.0,
      "step": 66770
    },
    {
      "entropy": 5.426560115814209,
      "epoch": 5.1952927446022175,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026384074135771474,
      "loss": 4.7482,
      "mean_token_accuracy": 0.22626017034053802,
      "num_tokens": 115835229.0,
      "step": 66775
    },
    {
      "entropy": 5.4902793884277346,
      "epoch": 5.195681773973935,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002638130599636603,
      "loss": 4.7247,
      "mean_token_accuracy": 0.23835522681474686,
      "num_tokens": 115843442.0,
      "step": 66780
    },
    {
      "entropy": 5.466753244400024,
      "epoch": 5.196070803345653,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026378537873934983,
      "loss": 4.7745,
      "mean_token_accuracy": 0.2245756655931473,
      "num_tokens": 115852427.0,
      "step": 66785
    },
    {
      "entropy": 5.420783042907715,
      "epoch": 5.1964598327173706,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026375769768520344,
      "loss": 4.734,
      "mean_token_accuracy": 0.22221490144729614,
      "num_tokens": 115860651.0,
      "step": 66790
    },
    {
      "entropy": 5.426042366027832,
      "epoch": 5.196848862089087,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000263730016801641,
      "loss": 4.7933,
      "mean_token_accuracy": 0.22274277359247208,
      "num_tokens": 115868344.0,
      "step": 66795
    },
    {
      "entropy": 5.418788146972656,
      "epoch": 5.197237891460805,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002637023360890827,
      "loss": 4.7246,
      "mean_token_accuracy": 0.22883989065885543,
      "num_tokens": 115876652.0,
      "step": 66800
    },
    {
      "entropy": 5.293851757049561,
      "epoch": 5.197626920832523,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026367465554794843,
      "loss": 4.6189,
      "mean_token_accuracy": 0.23460931032896043,
      "num_tokens": 115885478.0,
      "step": 66805
    },
    {
      "entropy": 5.3742919921875,
      "epoch": 5.1980159502042405,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026364697517865834,
      "loss": 4.7583,
      "mean_token_accuracy": 0.2306802451610565,
      "num_tokens": 115894196.0,
      "step": 66810
    },
    {
      "entropy": 5.429883623123169,
      "epoch": 5.198404979575958,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002636192949816322,
      "loss": 4.7055,
      "mean_token_accuracy": 0.23416150063276292,
      "num_tokens": 115902559.0,
      "step": 66815
    },
    {
      "entropy": 5.519118928909302,
      "epoch": 5.198794008947676,
      "grad_norm": 1.375,
      "learning_rate": 0.0002635916149572902,
      "loss": 4.8137,
      "mean_token_accuracy": 0.22269026190042496,
      "num_tokens": 115910710.0,
      "step": 66820
    },
    {
      "entropy": 5.446938800811767,
      "epoch": 5.1991830383193935,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026356393510605224,
      "loss": 4.7921,
      "mean_token_accuracy": 0.22143012285232544,
      "num_tokens": 115919306.0,
      "step": 66825
    },
    {
      "entropy": 5.419488382339478,
      "epoch": 5.19957206769111,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002635362554283383,
      "loss": 4.7385,
      "mean_token_accuracy": 0.23185072392225264,
      "num_tokens": 115928446.0,
      "step": 66830
    },
    {
      "entropy": 5.4105418682098385,
      "epoch": 5.199961097062828,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002635085759245686,
      "loss": 4.7009,
      "mean_token_accuracy": 0.23526550084352493,
      "num_tokens": 115937487.0,
      "step": 66835
    },
    {
      "entropy": 5.422021770477295,
      "epoch": 5.200350126434546,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002634808965951629,
      "loss": 4.7143,
      "mean_token_accuracy": 0.23393051624298095,
      "num_tokens": 115946056.0,
      "step": 66840
    },
    {
      "entropy": 5.394704055786133,
      "epoch": 5.2007391558062634,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002634532174405413,
      "loss": 4.8149,
      "mean_token_accuracy": 0.2219460964202881,
      "num_tokens": 115954773.0,
      "step": 66845
    },
    {
      "entropy": 5.358224010467529,
      "epoch": 5.201128185177981,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026342553846112364,
      "loss": 4.6886,
      "mean_token_accuracy": 0.22903091013431548,
      "num_tokens": 115964305.0,
      "step": 66850
    },
    {
      "entropy": 5.377748250961304,
      "epoch": 5.201517214549699,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002633978596573301,
      "loss": 4.6304,
      "mean_token_accuracy": 0.23878708779811858,
      "num_tokens": 115972276.0,
      "step": 66855
    },
    {
      "entropy": 5.37067666053772,
      "epoch": 5.201906243921416,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026337018102958056,
      "loss": 4.6633,
      "mean_token_accuracy": 0.23262891322374343,
      "num_tokens": 115980284.0,
      "step": 66860
    },
    {
      "entropy": 5.391755771636963,
      "epoch": 5.202295273293133,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000263342502578295,
      "loss": 4.6712,
      "mean_token_accuracy": 0.2420029178261757,
      "num_tokens": 115988793.0,
      "step": 66865
    },
    {
      "entropy": 5.443074893951416,
      "epoch": 5.202684302664851,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026331482430389343,
      "loss": 4.7163,
      "mean_token_accuracy": 0.23009722232818602,
      "num_tokens": 115997141.0,
      "step": 66870
    },
    {
      "entropy": 5.453098154067993,
      "epoch": 5.203073332036569,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026328714620679576,
      "loss": 4.7571,
      "mean_token_accuracy": 0.2267243191599846,
      "num_tokens": 116005096.0,
      "step": 66875
    },
    {
      "entropy": 5.284905672073364,
      "epoch": 5.203462361408286,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026325946828742203,
      "loss": 4.6652,
      "mean_token_accuracy": 0.23756748288869858,
      "num_tokens": 116013294.0,
      "step": 66880
    },
    {
      "entropy": 5.303067398071289,
      "epoch": 5.203851390780004,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026323179054619215,
      "loss": 4.5792,
      "mean_token_accuracy": 0.2421253055334091,
      "num_tokens": 116021607.0,
      "step": 66885
    },
    {
      "entropy": 5.350574779510498,
      "epoch": 5.204240420151722,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002632041129835262,
      "loss": 4.6495,
      "mean_token_accuracy": 0.23439411967992782,
      "num_tokens": 116030082.0,
      "step": 66890
    },
    {
      "entropy": 5.409378337860107,
      "epoch": 5.204629449523439,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026317643559984396,
      "loss": 4.7582,
      "mean_token_accuracy": 0.2289731428027153,
      "num_tokens": 116038469.0,
      "step": 66895
    },
    {
      "entropy": 5.37337703704834,
      "epoch": 5.205018478895156,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002631487583955655,
      "loss": 4.6208,
      "mean_token_accuracy": 0.23720172345638274,
      "num_tokens": 116046703.0,
      "step": 66900
    },
    {
      "entropy": 5.35559983253479,
      "epoch": 5.205407508266874,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026312108137111086,
      "loss": 4.6157,
      "mean_token_accuracy": 0.2374376654624939,
      "num_tokens": 116055140.0,
      "step": 66905
    },
    {
      "entropy": 5.414861679077148,
      "epoch": 5.205796537638592,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002630934045268998,
      "loss": 4.7378,
      "mean_token_accuracy": 0.23329027444124223,
      "num_tokens": 116064279.0,
      "step": 66910
    },
    {
      "entropy": 5.375687837600708,
      "epoch": 5.206185567010309,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002630657278633525,
      "loss": 4.6508,
      "mean_token_accuracy": 0.23344203531742097,
      "num_tokens": 116072847.0,
      "step": 66915
    },
    {
      "entropy": 5.411289358139038,
      "epoch": 5.206574596382027,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002630380513808887,
      "loss": 4.7431,
      "mean_token_accuracy": 0.2310434341430664,
      "num_tokens": 116081536.0,
      "step": 66920
    },
    {
      "entropy": 5.443298625946045,
      "epoch": 5.206963625753745,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002630103750799284,
      "loss": 4.6894,
      "mean_token_accuracy": 0.2339365914463997,
      "num_tokens": 116089936.0,
      "step": 66925
    },
    {
      "entropy": 5.404892826080323,
      "epoch": 5.207352655125462,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026298269896089164,
      "loss": 4.6914,
      "mean_token_accuracy": 0.2374357610940933,
      "num_tokens": 116098202.0,
      "step": 66930
    },
    {
      "entropy": 5.314939165115357,
      "epoch": 5.207741684497179,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026295502302419826,
      "loss": 4.6423,
      "mean_token_accuracy": 0.2358327627182007,
      "num_tokens": 116107429.0,
      "step": 66935
    },
    {
      "entropy": 5.273540449142456,
      "epoch": 5.208130713868897,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002629273472702683,
      "loss": 4.568,
      "mean_token_accuracy": 0.2419932261109352,
      "num_tokens": 116116294.0,
      "step": 66940
    },
    {
      "entropy": 5.397609519958496,
      "epoch": 5.208519743240615,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002628996716995215,
      "loss": 4.7699,
      "mean_token_accuracy": 0.2224861592054367,
      "num_tokens": 116125102.0,
      "step": 66945
    },
    {
      "entropy": 5.394065713882446,
      "epoch": 5.208908772612332,
      "grad_norm": 1.3125,
      "learning_rate": 0.000262871996312378,
      "loss": 4.6644,
      "mean_token_accuracy": 0.23367744237184523,
      "num_tokens": 116133467.0,
      "step": 66950
    },
    {
      "entropy": 5.414559459686279,
      "epoch": 5.20929780198405,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026284432110925765,
      "loss": 4.6755,
      "mean_token_accuracy": 0.2348521575331688,
      "num_tokens": 116142374.0,
      "step": 66955
    },
    {
      "entropy": 5.523514366149902,
      "epoch": 5.209686831355767,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026281664609058037,
      "loss": 4.8171,
      "mean_token_accuracy": 0.2244883418083191,
      "num_tokens": 116151222.0,
      "step": 66960
    },
    {
      "entropy": 5.414645004272461,
      "epoch": 5.210075860727485,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002627889712567661,
      "loss": 4.7078,
      "mean_token_accuracy": 0.23553937077522277,
      "num_tokens": 116160621.0,
      "step": 66965
    },
    {
      "entropy": 5.408946990966797,
      "epoch": 5.210464890099202,
      "grad_norm": 1.140625,
      "learning_rate": 0.00026276129660823475,
      "loss": 4.7879,
      "mean_token_accuracy": 0.23229963928461075,
      "num_tokens": 116170419.0,
      "step": 66970
    },
    {
      "entropy": 5.421690893173218,
      "epoch": 5.21085391947092,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026273362214540623,
      "loss": 4.7362,
      "mean_token_accuracy": 0.2221705883741379,
      "num_tokens": 116179357.0,
      "step": 66975
    },
    {
      "entropy": 5.329740762710571,
      "epoch": 5.211242948842638,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002627059478687005,
      "loss": 4.5871,
      "mean_token_accuracy": 0.2394294708967209,
      "num_tokens": 116188306.0,
      "step": 66980
    },
    {
      "entropy": 5.442807054519653,
      "epoch": 5.211631978214355,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002626782737785374,
      "loss": 4.7631,
      "mean_token_accuracy": 0.2313667953014374,
      "num_tokens": 116197100.0,
      "step": 66985
    },
    {
      "entropy": 5.348980045318603,
      "epoch": 5.212021007586073,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026265059987533697,
      "loss": 4.7781,
      "mean_token_accuracy": 0.23036125302314758,
      "num_tokens": 116205924.0,
      "step": 66990
    },
    {
      "entropy": 5.35806188583374,
      "epoch": 5.21241003695779,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026262292615951893,
      "loss": 4.6892,
      "mean_token_accuracy": 0.23715207427740098,
      "num_tokens": 116215255.0,
      "step": 66995
    },
    {
      "entropy": 5.432737493515015,
      "epoch": 5.212799066329508,
      "grad_norm": 1.734375,
      "learning_rate": 0.0002625952526315034,
      "loss": 4.7348,
      "mean_token_accuracy": 0.2308162823319435,
      "num_tokens": 116224723.0,
      "step": 67000
    },
    {
      "entropy": 5.419402456283569,
      "epoch": 5.213188095701225,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002625675792917101,
      "loss": 4.6977,
      "mean_token_accuracy": 0.23954665362834932,
      "num_tokens": 116234251.0,
      "step": 67005
    },
    {
      "entropy": 5.424380588531494,
      "epoch": 5.213577125072943,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000262539906140559,
      "loss": 4.7674,
      "mean_token_accuracy": 0.22445859014987946,
      "num_tokens": 116243402.0,
      "step": 67010
    },
    {
      "entropy": 5.346519184112549,
      "epoch": 5.213966154444661,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026251223317847003,
      "loss": 4.6723,
      "mean_token_accuracy": 0.23416320681571962,
      "num_tokens": 116251883.0,
      "step": 67015
    },
    {
      "entropy": 5.480128765106201,
      "epoch": 5.214355183816378,
      "grad_norm": 1.15625,
      "learning_rate": 0.000262484560405863,
      "loss": 4.8746,
      "mean_token_accuracy": 0.21486998796463014,
      "num_tokens": 116260721.0,
      "step": 67020
    },
    {
      "entropy": 5.449476337432861,
      "epoch": 5.214744213188096,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026245688782315787,
      "loss": 4.6959,
      "mean_token_accuracy": 0.23218224048614503,
      "num_tokens": 116270008.0,
      "step": 67025
    },
    {
      "entropy": 5.464039897918701,
      "epoch": 5.215133242559813,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026242921543077435,
      "loss": 4.8087,
      "mean_token_accuracy": 0.22978552728891372,
      "num_tokens": 116278920.0,
      "step": 67030
    },
    {
      "entropy": 5.393578147888183,
      "epoch": 5.215522271931531,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002624015432291326,
      "loss": 4.7506,
      "mean_token_accuracy": 0.2244909703731537,
      "num_tokens": 116287896.0,
      "step": 67035
    },
    {
      "entropy": 5.337471342086792,
      "epoch": 5.215911301303248,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002623738712186525,
      "loss": 4.6171,
      "mean_token_accuracy": 0.24403032809495925,
      "num_tokens": 116297117.0,
      "step": 67040
    },
    {
      "entropy": 5.37034797668457,
      "epoch": 5.216300330674966,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026234619939975365,
      "loss": 4.6886,
      "mean_token_accuracy": 0.2300478160381317,
      "num_tokens": 116305729.0,
      "step": 67045
    },
    {
      "entropy": 5.3906717777252195,
      "epoch": 5.216689360046684,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026231852777285606,
      "loss": 4.683,
      "mean_token_accuracy": 0.23300980031490326,
      "num_tokens": 116314542.0,
      "step": 67050
    },
    {
      "entropy": 5.41131625175476,
      "epoch": 5.217078389418401,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026229085633837975,
      "loss": 4.7058,
      "mean_token_accuracy": 0.22886818945407866,
      "num_tokens": 116324315.0,
      "step": 67055
    },
    {
      "entropy": 5.433721494674683,
      "epoch": 5.217467418790118,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002622631850967444,
      "loss": 4.7022,
      "mean_token_accuracy": 0.23028027266263962,
      "num_tokens": 116332837.0,
      "step": 67060
    },
    {
      "entropy": 5.395877552032471,
      "epoch": 5.217856448161836,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00026223551404837,
      "loss": 4.6219,
      "mean_token_accuracy": 0.24128084182739257,
      "num_tokens": 116341219.0,
      "step": 67065
    },
    {
      "entropy": 5.310512924194336,
      "epoch": 5.218245477533554,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026220784319367626,
      "loss": 4.6805,
      "mean_token_accuracy": 0.2342594474554062,
      "num_tokens": 116349233.0,
      "step": 67070
    },
    {
      "entropy": 5.4932088375091555,
      "epoch": 5.218634506905271,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002621801725330832,
      "loss": 4.8079,
      "mean_token_accuracy": 0.2190869376063347,
      "num_tokens": 116358449.0,
      "step": 67075
    },
    {
      "entropy": 5.42773699760437,
      "epoch": 5.219023536276989,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002621525020670106,
      "loss": 4.7383,
      "mean_token_accuracy": 0.22296229004859924,
      "num_tokens": 116366608.0,
      "step": 67080
    },
    {
      "entropy": 5.517432928085327,
      "epoch": 5.219412565648707,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002621248317958784,
      "loss": 4.8073,
      "mean_token_accuracy": 0.2235269770026207,
      "num_tokens": 116374906.0,
      "step": 67085
    },
    {
      "entropy": 5.472274494171143,
      "epoch": 5.219801595020424,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026209716172010636,
      "loss": 4.7647,
      "mean_token_accuracy": 0.22768107205629348,
      "num_tokens": 116384630.0,
      "step": 67090
    },
    {
      "entropy": 5.39162311553955,
      "epoch": 5.220190624392141,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002620694918401142,
      "loss": 4.6125,
      "mean_token_accuracy": 0.23564833998680115,
      "num_tokens": 116393473.0,
      "step": 67095
    },
    {
      "entropy": 5.327333450317383,
      "epoch": 5.220579653763859,
      "grad_norm": 1.3125,
      "learning_rate": 0.000262041822156322,
      "loss": 4.5598,
      "mean_token_accuracy": 0.2506711736321449,
      "num_tokens": 116401144.0,
      "step": 67100
    },
    {
      "entropy": 5.332838439941407,
      "epoch": 5.220968683135577,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002620141526691496,
      "loss": 4.6817,
      "mean_token_accuracy": 0.23285489082336425,
      "num_tokens": 116409994.0,
      "step": 67105
    },
    {
      "entropy": 5.343691492080689,
      "epoch": 5.221357712507294,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026198648337901674,
      "loss": 4.6639,
      "mean_token_accuracy": 0.24353963881731033,
      "num_tokens": 116418399.0,
      "step": 67110
    },
    {
      "entropy": 5.38607120513916,
      "epoch": 5.221746741879012,
      "grad_norm": 1.125,
      "learning_rate": 0.00026195881428634334,
      "loss": 4.6536,
      "mean_token_accuracy": 0.23533177226781846,
      "num_tokens": 116427683.0,
      "step": 67115
    },
    {
      "entropy": 5.348475408554077,
      "epoch": 5.22213577125073,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00026193114539154903,
      "loss": 4.661,
      "mean_token_accuracy": 0.23248441368341446,
      "num_tokens": 116436333.0,
      "step": 67120
    },
    {
      "entropy": 5.402314376831055,
      "epoch": 5.222524800622447,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002619034766950539,
      "loss": 4.714,
      "mean_token_accuracy": 0.23208766132593156,
      "num_tokens": 116444823.0,
      "step": 67125
    },
    {
      "entropy": 5.390159177780151,
      "epoch": 5.222913829994164,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002618758081972776,
      "loss": 4.6974,
      "mean_token_accuracy": 0.23342971056699752,
      "num_tokens": 116453245.0,
      "step": 67130
    },
    {
      "entropy": 5.428356504440307,
      "epoch": 5.223302859365882,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002618481398986401,
      "loss": 4.8284,
      "mean_token_accuracy": 0.22293556928634645,
      "num_tokens": 116461921.0,
      "step": 67135
    },
    {
      "entropy": 5.405839490890503,
      "epoch": 5.2236918887376,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00026182047179956107,
      "loss": 4.7624,
      "mean_token_accuracy": 0.23294555246829987,
      "num_tokens": 116470037.0,
      "step": 67140
    },
    {
      "entropy": 5.316297435760498,
      "epoch": 5.224080918109317,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026179280390046047,
      "loss": 4.668,
      "mean_token_accuracy": 0.24213544875383378,
      "num_tokens": 116478881.0,
      "step": 67145
    },
    {
      "entropy": 5.34333176612854,
      "epoch": 5.224469947481035,
      "grad_norm": 1.25,
      "learning_rate": 0.00026176513620175795,
      "loss": 4.6799,
      "mean_token_accuracy": 0.2334248125553131,
      "num_tokens": 116487305.0,
      "step": 67150
    },
    {
      "entropy": 5.371088123321533,
      "epoch": 5.224858976852753,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002617374687038735,
      "loss": 4.6598,
      "mean_token_accuracy": 0.23291896432638168,
      "num_tokens": 116496646.0,
      "step": 67155
    },
    {
      "entropy": 5.412616014480591,
      "epoch": 5.22524800622447,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026170980140722677,
      "loss": 4.6666,
      "mean_token_accuracy": 0.23399246037006377,
      "num_tokens": 116505356.0,
      "step": 67160
    },
    {
      "entropy": 5.385659837722779,
      "epoch": 5.225637035596187,
      "grad_norm": 1.25,
      "learning_rate": 0.00026168213431223777,
      "loss": 4.7726,
      "mean_token_accuracy": 0.22493771314620972,
      "num_tokens": 116514008.0,
      "step": 67165
    },
    {
      "entropy": 5.385939455032348,
      "epoch": 5.226026064967905,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026165446741932606,
      "loss": 4.5951,
      "mean_token_accuracy": 0.24019682705402373,
      "num_tokens": 116522513.0,
      "step": 67170
    },
    {
      "entropy": 5.331109285354614,
      "epoch": 5.226415094339623,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026162680072891166,
      "loss": 4.6217,
      "mean_token_accuracy": 0.23934606313705445,
      "num_tokens": 116530706.0,
      "step": 67175
    },
    {
      "entropy": 5.32786455154419,
      "epoch": 5.22680412371134,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026159913424141426,
      "loss": 4.6696,
      "mean_token_accuracy": 0.23382638841867448,
      "num_tokens": 116539277.0,
      "step": 67180
    },
    {
      "entropy": 5.389493083953857,
      "epoch": 5.227193153083058,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002615714679572536,
      "loss": 4.6894,
      "mean_token_accuracy": 0.23388569951057434,
      "num_tokens": 116548586.0,
      "step": 67185
    },
    {
      "entropy": 5.300423336029053,
      "epoch": 5.227582182454776,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026154380187684963,
      "loss": 4.5889,
      "mean_token_accuracy": 0.23865025490522385,
      "num_tokens": 116557121.0,
      "step": 67190
    },
    {
      "entropy": 5.37852110862732,
      "epoch": 5.2279712118264925,
      "grad_norm": 1.265625,
      "learning_rate": 0.000261516136000622,
      "loss": 4.6776,
      "mean_token_accuracy": 0.22680604457855225,
      "num_tokens": 116565698.0,
      "step": 67195
    },
    {
      "entropy": 5.383540678024292,
      "epoch": 5.22836024119821,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002614884703289906,
      "loss": 4.7322,
      "mean_token_accuracy": 0.23250954300165178,
      "num_tokens": 116573751.0,
      "step": 67200
    },
    {
      "entropy": 5.31975531578064,
      "epoch": 5.228749270569928,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026146080486237513,
      "loss": 4.6481,
      "mean_token_accuracy": 0.242208856344223,
      "num_tokens": 116581901.0,
      "step": 67205
    },
    {
      "entropy": 5.407828235626221,
      "epoch": 5.229138299941646,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026143313960119534,
      "loss": 4.7359,
      "mean_token_accuracy": 0.2288748502731323,
      "num_tokens": 116590827.0,
      "step": 67210
    },
    {
      "entropy": 5.426082563400269,
      "epoch": 5.229527329313363,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002614054745458711,
      "loss": 4.7335,
      "mean_token_accuracy": 0.23252734690904617,
      "num_tokens": 116599319.0,
      "step": 67215
    },
    {
      "entropy": 5.333526134490967,
      "epoch": 5.229916358685081,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026137780969682215,
      "loss": 4.664,
      "mean_token_accuracy": 0.2276516705751419,
      "num_tokens": 116607573.0,
      "step": 67220
    },
    {
      "entropy": 5.275151062011719,
      "epoch": 5.230305388056799,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002613501450544683,
      "loss": 4.6454,
      "mean_token_accuracy": 0.23069819808006287,
      "num_tokens": 116615998.0,
      "step": 67225
    },
    {
      "entropy": 5.397999095916748,
      "epoch": 5.2306944174285155,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002613224806192291,
      "loss": 4.6758,
      "mean_token_accuracy": 0.2346273899078369,
      "num_tokens": 116624557.0,
      "step": 67230
    },
    {
      "entropy": 5.395716428756714,
      "epoch": 5.231083446800233,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002612948163915247,
      "loss": 4.7019,
      "mean_token_accuracy": 0.22342079281806945,
      "num_tokens": 116633443.0,
      "step": 67235
    },
    {
      "entropy": 5.437121343612671,
      "epoch": 5.231472476171951,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002612671523717745,
      "loss": 4.8846,
      "mean_token_accuracy": 0.21760348826646805,
      "num_tokens": 116641988.0,
      "step": 67240
    },
    {
      "entropy": 5.438926267623901,
      "epoch": 5.231861505543669,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002612394885603985,
      "loss": 4.7298,
      "mean_token_accuracy": 0.2255731701850891,
      "num_tokens": 116649831.0,
      "step": 67245
    },
    {
      "entropy": 5.42825493812561,
      "epoch": 5.232250534915386,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002612118249578163,
      "loss": 4.6929,
      "mean_token_accuracy": 0.23583969175815583,
      "num_tokens": 116658856.0,
      "step": 67250
    },
    {
      "entropy": 5.404086637496948,
      "epoch": 5.232639564287104,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002611841615644477,
      "loss": 4.6937,
      "mean_token_accuracy": 0.2371744155883789,
      "num_tokens": 116666731.0,
      "step": 67255
    },
    {
      "entropy": 5.398430395126343,
      "epoch": 5.233028593658822,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026115649838071253,
      "loss": 4.6949,
      "mean_token_accuracy": 0.23271084427833558,
      "num_tokens": 116674953.0,
      "step": 67260
    },
    {
      "entropy": 5.374135446548462,
      "epoch": 5.2334176230305385,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002611288354070304,
      "loss": 4.6578,
      "mean_token_accuracy": 0.24367496222257615,
      "num_tokens": 116683825.0,
      "step": 67265
    },
    {
      "entropy": 5.48495922088623,
      "epoch": 5.233806652402256,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002611011726438212,
      "loss": 4.9284,
      "mean_token_accuracy": 0.2215125471353531,
      "num_tokens": 116693128.0,
      "step": 67270
    },
    {
      "entropy": 5.399789476394654,
      "epoch": 5.234195681773974,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00026107351009150453,
      "loss": 4.7169,
      "mean_token_accuracy": 0.2307500198483467,
      "num_tokens": 116701789.0,
      "step": 67275
    },
    {
      "entropy": 5.401145935058594,
      "epoch": 5.234584711145692,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002610458477505002,
      "loss": 4.6972,
      "mean_token_accuracy": 0.23181137144565583,
      "num_tokens": 116710187.0,
      "step": 67280
    },
    {
      "entropy": 5.442689895629883,
      "epoch": 5.234973740517409,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002610181856212279,
      "loss": 4.7713,
      "mean_token_accuracy": 0.22508362978696822,
      "num_tokens": 116719245.0,
      "step": 67285
    },
    {
      "entropy": 5.376486968994141,
      "epoch": 5.235362769889127,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002609905237041074,
      "loss": 4.6593,
      "mean_token_accuracy": 0.23711671233177184,
      "num_tokens": 116726977.0,
      "step": 67290
    },
    {
      "entropy": 5.418247127532959,
      "epoch": 5.235751799260844,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002609628619995582,
      "loss": 4.68,
      "mean_token_accuracy": 0.23080794215202333,
      "num_tokens": 116735078.0,
      "step": 67295
    },
    {
      "entropy": 5.425756454467773,
      "epoch": 5.2361408286325615,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026093520050800046,
      "loss": 4.7432,
      "mean_token_accuracy": 0.2296899974346161,
      "num_tokens": 116744015.0,
      "step": 67300
    },
    {
      "entropy": 5.441672277450562,
      "epoch": 5.236529858004279,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002609075392298536,
      "loss": 4.8129,
      "mean_token_accuracy": 0.22739205956459047,
      "num_tokens": 116752152.0,
      "step": 67305
    },
    {
      "entropy": 5.380702829360962,
      "epoch": 5.236918887375997,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002608798781655374,
      "loss": 4.6371,
      "mean_token_accuracy": 0.23781922310590745,
      "num_tokens": 116761598.0,
      "step": 67310
    },
    {
      "entropy": 5.358177280426025,
      "epoch": 5.237307916747715,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002608522173154716,
      "loss": 4.5965,
      "mean_token_accuracy": 0.24128767251968383,
      "num_tokens": 116770673.0,
      "step": 67315
    },
    {
      "entropy": 5.379529762268066,
      "epoch": 5.237696946119432,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002608245566800759,
      "loss": 4.7258,
      "mean_token_accuracy": 0.22561502754688262,
      "num_tokens": 116779750.0,
      "step": 67320
    },
    {
      "entropy": 5.484083604812622,
      "epoch": 5.23808597549115,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026079689625976995,
      "loss": 4.7156,
      "mean_token_accuracy": 0.22590613663196563,
      "num_tokens": 116788044.0,
      "step": 67325
    },
    {
      "entropy": 5.433552026748657,
      "epoch": 5.238475004862867,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026076923605497347,
      "loss": 4.7516,
      "mean_token_accuracy": 0.22788045108318328,
      "num_tokens": 116795981.0,
      "step": 67330
    },
    {
      "entropy": 5.445416641235352,
      "epoch": 5.2388640342345845,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026074157606610626,
      "loss": 4.7017,
      "mean_token_accuracy": 0.2383514702320099,
      "num_tokens": 116804610.0,
      "step": 67335
    },
    {
      "entropy": 5.479588651657105,
      "epoch": 5.239253063606302,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026071391629358786,
      "loss": 4.825,
      "mean_token_accuracy": 0.22355996519327165,
      "num_tokens": 116813499.0,
      "step": 67340
    },
    {
      "entropy": 5.453675270080566,
      "epoch": 5.23964209297802,
      "grad_norm": 1.25,
      "learning_rate": 0.00026068625673783807,
      "loss": 4.7813,
      "mean_token_accuracy": 0.22704491913318633,
      "num_tokens": 116823208.0,
      "step": 67345
    },
    {
      "entropy": 5.4371918678283695,
      "epoch": 5.240031122349738,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026065859739927657,
      "loss": 4.7373,
      "mean_token_accuracy": 0.23173634111881256,
      "num_tokens": 116831335.0,
      "step": 67350
    },
    {
      "entropy": 5.508289813995361,
      "epoch": 5.240420151721455,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000260630938278323,
      "loss": 4.8155,
      "mean_token_accuracy": 0.22185618281364441,
      "num_tokens": 116839897.0,
      "step": 67355
    },
    {
      "entropy": 5.414023447036743,
      "epoch": 5.240809181093173,
      "grad_norm": 1.25,
      "learning_rate": 0.00026060327937539704,
      "loss": 4.6748,
      "mean_token_accuracy": 0.23026626259088517,
      "num_tokens": 116847919.0,
      "step": 67360
    },
    {
      "entropy": 5.376636409759522,
      "epoch": 5.24119821046489,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026057562069091845,
      "loss": 4.6979,
      "mean_token_accuracy": 0.23264960944652557,
      "num_tokens": 116856139.0,
      "step": 67365
    },
    {
      "entropy": 5.349032783508301,
      "epoch": 5.2415872398366075,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026054796222530684,
      "loss": 4.6513,
      "mean_token_accuracy": 0.24196869879961014,
      "num_tokens": 116864211.0,
      "step": 67370
    },
    {
      "entropy": 5.428495407104492,
      "epoch": 5.241976269208325,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002605203039789819,
      "loss": 4.7557,
      "mean_token_accuracy": 0.22287316173315047,
      "num_tokens": 116873032.0,
      "step": 67375
    },
    {
      "entropy": 5.462880229949951,
      "epoch": 5.242365298580043,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002604926459523633,
      "loss": 4.7643,
      "mean_token_accuracy": 0.2276371479034424,
      "num_tokens": 116880941.0,
      "step": 67380
    },
    {
      "entropy": 5.400443649291992,
      "epoch": 5.2427543279517606,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026046498814587076,
      "loss": 4.7012,
      "mean_token_accuracy": 0.2283831387758255,
      "num_tokens": 116888833.0,
      "step": 67385
    },
    {
      "entropy": 5.387718534469604,
      "epoch": 5.243143357323478,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026043733055992386,
      "loss": 4.6753,
      "mean_token_accuracy": 0.23279907703399658,
      "num_tokens": 116897450.0,
      "step": 67390
    },
    {
      "entropy": 5.312741088867187,
      "epoch": 5.243532386695195,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026040967319494225,
      "loss": 4.6465,
      "mean_token_accuracy": 0.23524246513843536,
      "num_tokens": 116906615.0,
      "step": 67395
    },
    {
      "entropy": 5.385815095901489,
      "epoch": 5.243921416066913,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026038201605134567,
      "loss": 4.6932,
      "mean_token_accuracy": 0.23378511071205138,
      "num_tokens": 116914972.0,
      "step": 67400
    },
    {
      "entropy": 5.345600748062134,
      "epoch": 5.2443104454386305,
      "grad_norm": 1.25,
      "learning_rate": 0.00026035435912955375,
      "loss": 4.6834,
      "mean_token_accuracy": 0.2387162059545517,
      "num_tokens": 116923316.0,
      "step": 67405
    },
    {
      "entropy": 5.354609155654908,
      "epoch": 5.244699474810348,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002603267024299861,
      "loss": 4.7167,
      "mean_token_accuracy": 0.23606501668691635,
      "num_tokens": 116932180.0,
      "step": 67410
    },
    {
      "entropy": 5.3945677280426025,
      "epoch": 5.245088504182066,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002602990459530623,
      "loss": 4.7133,
      "mean_token_accuracy": 0.23605251759290696,
      "num_tokens": 116941102.0,
      "step": 67415
    },
    {
      "entropy": 5.449324369430542,
      "epoch": 5.2454775335537835,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002602713896992022,
      "loss": 4.6862,
      "mean_token_accuracy": 0.2336167261004448,
      "num_tokens": 116948888.0,
      "step": 67420
    },
    {
      "entropy": 5.447108888626099,
      "epoch": 5.245866562925501,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002602437336688252,
      "loss": 4.7862,
      "mean_token_accuracy": 0.22331737726926804,
      "num_tokens": 116958386.0,
      "step": 67425
    },
    {
      "entropy": 5.33681287765503,
      "epoch": 5.246255592297218,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002602160778623511,
      "loss": 4.6224,
      "mean_token_accuracy": 0.24246330559253693,
      "num_tokens": 116967378.0,
      "step": 67430
    },
    {
      "entropy": 5.490798616409302,
      "epoch": 5.246644621668936,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026018842228019953,
      "loss": 4.8148,
      "mean_token_accuracy": 0.22208707630634308,
      "num_tokens": 116976537.0,
      "step": 67435
    },
    {
      "entropy": 5.374784803390503,
      "epoch": 5.2470336510406534,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026016076692279003,
      "loss": 4.6598,
      "mean_token_accuracy": 0.23650529235601425,
      "num_tokens": 116984911.0,
      "step": 67440
    },
    {
      "entropy": 5.3797601699829105,
      "epoch": 5.247422680412371,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002601331117905423,
      "loss": 4.6939,
      "mean_token_accuracy": 0.23245563209056855,
      "num_tokens": 116993847.0,
      "step": 67445
    },
    {
      "entropy": 5.355009078979492,
      "epoch": 5.247811709784089,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026010545688387595,
      "loss": 4.6476,
      "mean_token_accuracy": 0.23593928962945937,
      "num_tokens": 117002114.0,
      "step": 67450
    },
    {
      "entropy": 5.369286060333252,
      "epoch": 5.2482007391558065,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00026007780220321054,
      "loss": 4.7077,
      "mean_token_accuracy": 0.23278801888227463,
      "num_tokens": 117010617.0,
      "step": 67455
    },
    {
      "entropy": 5.366733980178833,
      "epoch": 5.248589768527524,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026005014774896574,
      "loss": 4.6698,
      "mean_token_accuracy": 0.23100989013910295,
      "num_tokens": 117019235.0,
      "step": 67460
    },
    {
      "entropy": 5.406934213638306,
      "epoch": 5.248978797899241,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026002249352156124,
      "loss": 4.7963,
      "mean_token_accuracy": 0.22444510757923125,
      "num_tokens": 117027911.0,
      "step": 67465
    },
    {
      "entropy": 5.4959064483642575,
      "epoch": 5.249367827270959,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0002599948395214165,
      "loss": 4.7685,
      "mean_token_accuracy": 0.22609095573425292,
      "num_tokens": 117036331.0,
      "step": 67470
    },
    {
      "entropy": 5.459526109695434,
      "epoch": 5.249756856642676,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00025996718574895116,
      "loss": 4.769,
      "mean_token_accuracy": 0.22066522240638733,
      "num_tokens": 117043707.0,
      "step": 67475
    },
    {
      "entropy": 5.475381326675415,
      "epoch": 5.250145886014394,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025993953220458487,
      "loss": 4.7812,
      "mean_token_accuracy": 0.23207593262195586,
      "num_tokens": 117052050.0,
      "step": 67480
    },
    {
      "entropy": 5.437155437469483,
      "epoch": 5.250534915386112,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002599118788887372,
      "loss": 4.7063,
      "mean_token_accuracy": 0.23764725774526596,
      "num_tokens": 117060793.0,
      "step": 67485
    },
    {
      "entropy": 5.381167268753051,
      "epoch": 5.2509239447578295,
      "grad_norm": 1.09375,
      "learning_rate": 0.00025988422580182775,
      "loss": 4.6712,
      "mean_token_accuracy": 0.22808973342180253,
      "num_tokens": 117070221.0,
      "step": 67490
    },
    {
      "entropy": 5.394053363800049,
      "epoch": 5.251312974129547,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002598565729442761,
      "loss": 4.7542,
      "mean_token_accuracy": 0.22572365701198577,
      "num_tokens": 117079792.0,
      "step": 67495
    },
    {
      "entropy": 5.412895202636719,
      "epoch": 5.251702003501264,
      "grad_norm": 1.203125,
      "learning_rate": 0.00025982892031650186,
      "loss": 4.6812,
      "mean_token_accuracy": 0.22866483330726622,
      "num_tokens": 117088455.0,
      "step": 67500
    },
    {
      "entropy": 5.3610734939575195,
      "epoch": 5.252091032872982,
      "grad_norm": 1.25,
      "learning_rate": 0.0002598012679189246,
      "loss": 4.7354,
      "mean_token_accuracy": 0.23251037001609803,
      "num_tokens": 117096902.0,
      "step": 67505
    },
    {
      "entropy": 5.3349268436431885,
      "epoch": 5.252480062244699,
      "grad_norm": 1.34375,
      "learning_rate": 0.00025977361575196394,
      "loss": 4.6687,
      "mean_token_accuracy": 0.23329002261161805,
      "num_tokens": 117105252.0,
      "step": 67510
    },
    {
      "entropy": 5.308884572982788,
      "epoch": 5.252869091616417,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00025974596381603943,
      "loss": 4.5642,
      "mean_token_accuracy": 0.2355449005961418,
      "num_tokens": 117113288.0,
      "step": 67515
    },
    {
      "entropy": 5.37100396156311,
      "epoch": 5.253258120988135,
      "grad_norm": 1.25,
      "learning_rate": 0.0002597183121115707,
      "loss": 4.6524,
      "mean_token_accuracy": 0.24688079506158828,
      "num_tokens": 117121942.0,
      "step": 67520
    },
    {
      "entropy": 5.419775581359863,
      "epoch": 5.2536471503598525,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002596906606389771,
      "loss": 4.686,
      "mean_token_accuracy": 0.23545729219913483,
      "num_tokens": 117130133.0,
      "step": 67525
    },
    {
      "entropy": 5.404507112503052,
      "epoch": 5.254036179731569,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00025966300939867843,
      "loss": 4.6881,
      "mean_token_accuracy": 0.23277096152305604,
      "num_tokens": 117138979.0,
      "step": 67530
    },
    {
      "entropy": 5.356218576431274,
      "epoch": 5.254425209103287,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002596353583910942,
      "loss": 4.7375,
      "mean_token_accuracy": 0.22256511002779006,
      "num_tokens": 117148348.0,
      "step": 67535
    },
    {
      "entropy": 5.343806552886963,
      "epoch": 5.254814238475005,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025960770761664396,
      "loss": 4.601,
      "mean_token_accuracy": 0.2461948126554489,
      "num_tokens": 117156440.0,
      "step": 67540
    },
    {
      "entropy": 5.320831966400147,
      "epoch": 5.255203267846722,
      "grad_norm": 1.140625,
      "learning_rate": 0.00025958005707574726,
      "loss": 4.6852,
      "mean_token_accuracy": 0.2345167353749275,
      "num_tokens": 117165858.0,
      "step": 67545
    },
    {
      "entropy": 5.390328502655029,
      "epoch": 5.25559229721844,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002595524067688237,
      "loss": 4.7412,
      "mean_token_accuracy": 0.22469160556793213,
      "num_tokens": 117175984.0,
      "step": 67550
    },
    {
      "entropy": 5.510164499282837,
      "epoch": 5.255981326590158,
      "grad_norm": 1.15625,
      "learning_rate": 0.00025952475669629267,
      "loss": 4.7593,
      "mean_token_accuracy": 0.2326524406671524,
      "num_tokens": 117185166.0,
      "step": 67555
    },
    {
      "entropy": 5.4184250831604,
      "epoch": 5.2563703559618755,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025949710685857383,
      "loss": 4.7112,
      "mean_token_accuracy": 0.22882448732852936,
      "num_tokens": 117193701.0,
      "step": 67560
    },
    {
      "entropy": 5.366162443161011,
      "epoch": 5.256759385333592,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025946945725608675,
      "loss": 4.7371,
      "mean_token_accuracy": 0.2323720335960388,
      "num_tokens": 117202683.0,
      "step": 67565
    },
    {
      "entropy": 5.367905616760254,
      "epoch": 5.25714841470531,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002594418078892509,
      "loss": 4.6549,
      "mean_token_accuracy": 0.23404959738254547,
      "num_tokens": 117211164.0,
      "step": 67570
    },
    {
      "entropy": 5.435426712036133,
      "epoch": 5.257537444077028,
      "grad_norm": 1.28125,
      "learning_rate": 0.000259414158758486,
      "loss": 4.7261,
      "mean_token_accuracy": 0.2247810110449791,
      "num_tokens": 117219033.0,
      "step": 67575
    },
    {
      "entropy": 5.452024650573731,
      "epoch": 5.257926473448745,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025938650986421125,
      "loss": 4.7605,
      "mean_token_accuracy": 0.22904880195856095,
      "num_tokens": 117227948.0,
      "step": 67580
    },
    {
      "entropy": 5.441284894943237,
      "epoch": 5.258315502820463,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002593588612068465,
      "loss": 4.7213,
      "mean_token_accuracy": 0.24183682799339296,
      "num_tokens": 117236169.0,
      "step": 67585
    },
    {
      "entropy": 5.34754433631897,
      "epoch": 5.258704532192181,
      "grad_norm": 1.25,
      "learning_rate": 0.000259331212786811,
      "loss": 4.783,
      "mean_token_accuracy": 0.22686605304479598,
      "num_tokens": 117245371.0,
      "step": 67590
    },
    {
      "entropy": 5.3409229755401615,
      "epoch": 5.259093561563898,
      "grad_norm": 1.15625,
      "learning_rate": 0.00025930356460452454,
      "loss": 4.6949,
      "mean_token_accuracy": 0.2237337589263916,
      "num_tokens": 117254357.0,
      "step": 67595
    },
    {
      "entropy": 5.432412433624267,
      "epoch": 5.259482590935615,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002592759166604064,
      "loss": 4.7034,
      "mean_token_accuracy": 0.2325456351041794,
      "num_tokens": 117263188.0,
      "step": 67600
    },
    {
      "entropy": 5.438701725006103,
      "epoch": 5.259871620307333,
      "grad_norm": 1.265625,
      "learning_rate": 0.00025924826895487623,
      "loss": 4.7823,
      "mean_token_accuracy": 0.23559650033712387,
      "num_tokens": 117272103.0,
      "step": 67605
    },
    {
      "entropy": 5.3339701175689695,
      "epoch": 5.260260649679051,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025922062148835357,
      "loss": 4.715,
      "mean_token_accuracy": 0.2287560820579529,
      "num_tokens": 117280356.0,
      "step": 67610
    },
    {
      "entropy": 5.4158868312835695,
      "epoch": 5.260649679050768,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002591929742612578,
      "loss": 4.7132,
      "mean_token_accuracy": 0.22876811772584915,
      "num_tokens": 117288173.0,
      "step": 67615
    },
    {
      "entropy": 5.414004898071289,
      "epoch": 5.261038708422486,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002591653272740085,
      "loss": 4.7235,
      "mean_token_accuracy": 0.22858828753232957,
      "num_tokens": 117295795.0,
      "step": 67620
    },
    {
      "entropy": 5.434411430358887,
      "epoch": 5.261427737794204,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00025913768052702514,
      "loss": 4.734,
      "mean_token_accuracy": 0.2232011839747429,
      "num_tokens": 117304509.0,
      "step": 67625
    },
    {
      "entropy": 5.44519305229187,
      "epoch": 5.261816767165921,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002591100340207272,
      "loss": 4.7315,
      "mean_token_accuracy": 0.23461957573890685,
      "num_tokens": 117312151.0,
      "step": 67630
    },
    {
      "entropy": 5.305275964736938,
      "epoch": 5.262205796537638,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002590823877555343,
      "loss": 4.6711,
      "mean_token_accuracy": 0.23494940400123596,
      "num_tokens": 117320743.0,
      "step": 67635
    },
    {
      "entropy": 5.323049068450928,
      "epoch": 5.262594825909356,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002590547417318657,
      "loss": 4.6972,
      "mean_token_accuracy": 0.2328971281647682,
      "num_tokens": 117330169.0,
      "step": 67640
    },
    {
      "entropy": 5.3841142654418945,
      "epoch": 5.262983855281074,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002590270959501411,
      "loss": 4.6729,
      "mean_token_accuracy": 0.23634839355945586,
      "num_tokens": 117338081.0,
      "step": 67645
    },
    {
      "entropy": 5.398157453536987,
      "epoch": 5.263372884652791,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025899945041077995,
      "loss": 4.6693,
      "mean_token_accuracy": 0.23173351883888244,
      "num_tokens": 117346770.0,
      "step": 67650
    },
    {
      "entropy": 5.403488445281982,
      "epoch": 5.263761914024509,
      "grad_norm": 1.265625,
      "learning_rate": 0.00025897180511420164,
      "loss": 4.6893,
      "mean_token_accuracy": 0.23138634413480758,
      "num_tokens": 117355646.0,
      "step": 67655
    },
    {
      "entropy": 5.3779195785522464,
      "epoch": 5.264150943396227,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025894416006082564,
      "loss": 4.7019,
      "mean_token_accuracy": 0.2267584979534149,
      "num_tokens": 117364592.0,
      "step": 67660
    },
    {
      "entropy": 5.419485998153687,
      "epoch": 5.264539972767944,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00025891651525107145,
      "loss": 4.7146,
      "mean_token_accuracy": 0.2303535148501396,
      "num_tokens": 117373302.0,
      "step": 67665
    },
    {
      "entropy": 5.39093861579895,
      "epoch": 5.264929002139661,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025888887068535853,
      "loss": 4.7086,
      "mean_token_accuracy": 0.23065055459737777,
      "num_tokens": 117381310.0,
      "step": 67670
    },
    {
      "entropy": 5.4331550121307375,
      "epoch": 5.265318031511379,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025886122636410643,
      "loss": 4.7539,
      "mean_token_accuracy": 0.22767216414213182,
      "num_tokens": 117389680.0,
      "step": 67675
    },
    {
      "entropy": 5.518896675109863,
      "epoch": 5.265707060883097,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025883358228773445,
      "loss": 4.806,
      "mean_token_accuracy": 0.2279020980000496,
      "num_tokens": 117398253.0,
      "step": 67680
    },
    {
      "entropy": 5.487711095809937,
      "epoch": 5.266096090254814,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002588059384566622,
      "loss": 4.773,
      "mean_token_accuracy": 0.22800550609827042,
      "num_tokens": 117406633.0,
      "step": 67685
    },
    {
      "entropy": 5.406185674667358,
      "epoch": 5.266485119626532,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00025877829487130894,
      "loss": 4.7308,
      "mean_token_accuracy": 0.23067347705364227,
      "num_tokens": 117414755.0,
      "step": 67690
    },
    {
      "entropy": 5.457627964019776,
      "epoch": 5.26687414899825,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002587506515320943,
      "loss": 4.7708,
      "mean_token_accuracy": 0.23172007501125336,
      "num_tokens": 117423310.0,
      "step": 67695
    },
    {
      "entropy": 5.413216638565063,
      "epoch": 5.267263178369967,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002587230084394377,
      "loss": 4.6736,
      "mean_token_accuracy": 0.22678678035736083,
      "num_tokens": 117431931.0,
      "step": 67700
    },
    {
      "entropy": 5.463682794570923,
      "epoch": 5.267652207741684,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002586953655937585,
      "loss": 4.7249,
      "mean_token_accuracy": 0.2321800947189331,
      "num_tokens": 117440233.0,
      "step": 67705
    },
    {
      "entropy": 5.419193410873413,
      "epoch": 5.268041237113402,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002586677229954761,
      "loss": 4.7588,
      "mean_token_accuracy": 0.22933555394411087,
      "num_tokens": 117449679.0,
      "step": 67710
    },
    {
      "entropy": 5.390128946304321,
      "epoch": 5.26843026648512,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002586400806450102,
      "loss": 4.7052,
      "mean_token_accuracy": 0.23210811614990234,
      "num_tokens": 117458000.0,
      "step": 67715
    },
    {
      "entropy": 5.475764513015747,
      "epoch": 5.268819295856837,
      "grad_norm": 1.15625,
      "learning_rate": 0.00025861243854277986,
      "loss": 4.8332,
      "mean_token_accuracy": 0.2188028946518898,
      "num_tokens": 117467856.0,
      "step": 67720
    },
    {
      "entropy": 5.455661106109619,
      "epoch": 5.269208325228555,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002585847966892048,
      "loss": 4.7305,
      "mean_token_accuracy": 0.22966277003288268,
      "num_tokens": 117475723.0,
      "step": 67725
    },
    {
      "entropy": 5.44646201133728,
      "epoch": 5.269597354600272,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002585571550847042,
      "loss": 4.734,
      "mean_token_accuracy": 0.22961940318346025,
      "num_tokens": 117484784.0,
      "step": 67730
    },
    {
      "entropy": 5.358654975891113,
      "epoch": 5.26998638397199,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002585295137296978,
      "loss": 4.6778,
      "mean_token_accuracy": 0.23610725849866868,
      "num_tokens": 117493490.0,
      "step": 67735
    },
    {
      "entropy": 5.398290395736694,
      "epoch": 5.270375413343707,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002585018726246047,
      "loss": 4.779,
      "mean_token_accuracy": 0.2247351661324501,
      "num_tokens": 117502582.0,
      "step": 67740
    },
    {
      "entropy": 5.341506052017212,
      "epoch": 5.270764442715425,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002584742317698444,
      "loss": 4.6499,
      "mean_token_accuracy": 0.23633422553539277,
      "num_tokens": 117511381.0,
      "step": 67745
    },
    {
      "entropy": 5.442134237289428,
      "epoch": 5.271153472087143,
      "grad_norm": 1.375,
      "learning_rate": 0.00025844659116583635,
      "loss": 4.8114,
      "mean_token_accuracy": 0.2309001460671425,
      "num_tokens": 117520177.0,
      "step": 67750
    },
    {
      "entropy": 5.466410827636719,
      "epoch": 5.27154250145886,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002584189508129999,
      "loss": 4.7937,
      "mean_token_accuracy": 0.22736624479293824,
      "num_tokens": 117529763.0,
      "step": 67755
    },
    {
      "entropy": 5.47055344581604,
      "epoch": 5.271931530830578,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002583913107117544,
      "loss": 4.7429,
      "mean_token_accuracy": 0.23290761262178422,
      "num_tokens": 117538309.0,
      "step": 67760
    },
    {
      "entropy": 5.38823971748352,
      "epoch": 5.272320560202295,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002583636708625195,
      "loss": 4.8049,
      "mean_token_accuracy": 0.22759881019592285,
      "num_tokens": 117546608.0,
      "step": 67765
    },
    {
      "entropy": 5.428650712966919,
      "epoch": 5.272709589574013,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002583360312657143,
      "loss": 4.7177,
      "mean_token_accuracy": 0.23121925294399262,
      "num_tokens": 117555136.0,
      "step": 67770
    },
    {
      "entropy": 5.425867080688477,
      "epoch": 5.27309861894573,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025830839192175835,
      "loss": 4.731,
      "mean_token_accuracy": 0.234566330909729,
      "num_tokens": 117563304.0,
      "step": 67775
    },
    {
      "entropy": 5.465944814682007,
      "epoch": 5.273487648317448,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000258280752831071,
      "loss": 4.8256,
      "mean_token_accuracy": 0.22526848912239075,
      "num_tokens": 117572707.0,
      "step": 67780
    },
    {
      "entropy": 5.508855772018433,
      "epoch": 5.273876677689166,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025825311399407157,
      "loss": 4.8184,
      "mean_token_accuracy": 0.2193262681365013,
      "num_tokens": 117581169.0,
      "step": 67785
    },
    {
      "entropy": 5.457820177078247,
      "epoch": 5.274265707060883,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002582254754111795,
      "loss": 4.6832,
      "mean_token_accuracy": 0.23896281868219377,
      "num_tokens": 117589508.0,
      "step": 67790
    },
    {
      "entropy": 5.416562128067016,
      "epoch": 5.274654736432601,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002581978370828142,
      "loss": 4.7467,
      "mean_token_accuracy": 0.225620137155056,
      "num_tokens": 117598211.0,
      "step": 67795
    },
    {
      "entropy": 5.39215931892395,
      "epoch": 5.275043765804318,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002581701990093949,
      "loss": 4.6348,
      "mean_token_accuracy": 0.23864491879940034,
      "num_tokens": 117606425.0,
      "step": 67800
    },
    {
      "entropy": 5.40304970741272,
      "epoch": 5.275432795176036,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002581425611913411,
      "loss": 4.6841,
      "mean_token_accuracy": 0.2346534788608551,
      "num_tokens": 117615420.0,
      "step": 67805
    },
    {
      "entropy": 5.384714078903198,
      "epoch": 5.275821824547753,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002581149236290721,
      "loss": 4.6623,
      "mean_token_accuracy": 0.24034376740455626,
      "num_tokens": 117624427.0,
      "step": 67810
    },
    {
      "entropy": 5.4182086944580075,
      "epoch": 5.276210853919471,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002580872863230072,
      "loss": 4.7328,
      "mean_token_accuracy": 0.2357745960354805,
      "num_tokens": 117633097.0,
      "step": 67815
    },
    {
      "entropy": 5.387362623214722,
      "epoch": 5.276599883291189,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025805964927356584,
      "loss": 4.7471,
      "mean_token_accuracy": 0.23142085820436478,
      "num_tokens": 117642493.0,
      "step": 67820
    },
    {
      "entropy": 5.417189264297486,
      "epoch": 5.276988912662906,
      "grad_norm": 1.25,
      "learning_rate": 0.00025803201248116725,
      "loss": 4.7572,
      "mean_token_accuracy": 0.22613672465085982,
      "num_tokens": 117651064.0,
      "step": 67825
    },
    {
      "entropy": 5.475321435928345,
      "epoch": 5.277377942034624,
      "grad_norm": 1.15625,
      "learning_rate": 0.00025800437594623095,
      "loss": 4.8464,
      "mean_token_accuracy": 0.21559946089982987,
      "num_tokens": 117660461.0,
      "step": 67830
    },
    {
      "entropy": 5.472397327423096,
      "epoch": 5.277766971406341,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002579767396691762,
      "loss": 4.753,
      "mean_token_accuracy": 0.22670711129903792,
      "num_tokens": 117669169.0,
      "step": 67835
    },
    {
      "entropy": 5.391211748123169,
      "epoch": 5.278156000778059,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025794910365042233,
      "loss": 4.6124,
      "mean_token_accuracy": 0.2359376445412636,
      "num_tokens": 117677284.0,
      "step": 67840
    },
    {
      "entropy": 5.430466890335083,
      "epoch": 5.278545030149776,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025792146789038875,
      "loss": 4.7589,
      "mean_token_accuracy": 0.22917403131723404,
      "num_tokens": 117685708.0,
      "step": 67845
    },
    {
      "entropy": 5.344880771636963,
      "epoch": 5.278934059521494,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002578938323894946,
      "loss": 4.7167,
      "mean_token_accuracy": 0.23063499331474305,
      "num_tokens": 117693994.0,
      "step": 67850
    },
    {
      "entropy": 5.432592248916626,
      "epoch": 5.279323088893212,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025786619714815933,
      "loss": 4.7211,
      "mean_token_accuracy": 0.2311262920498848,
      "num_tokens": 117702384.0,
      "step": 67855
    },
    {
      "entropy": 5.441815614700317,
      "epoch": 5.279712118264929,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002578385621668023,
      "loss": 4.6635,
      "mean_token_accuracy": 0.23401136249303817,
      "num_tokens": 117710707.0,
      "step": 67860
    },
    {
      "entropy": 5.43959641456604,
      "epoch": 5.280101147636646,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002578109274458427,
      "loss": 4.7296,
      "mean_token_accuracy": 0.23360033035278321,
      "num_tokens": 117719321.0,
      "step": 67865
    },
    {
      "entropy": 5.473988056182861,
      "epoch": 5.280490177008364,
      "grad_norm": 1.1875,
      "learning_rate": 0.00025778329298569997,
      "loss": 4.8444,
      "mean_token_accuracy": 0.2215265616774559,
      "num_tokens": 117729867.0,
      "step": 67870
    },
    {
      "entropy": 5.572522401809692,
      "epoch": 5.280879206380082,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002577556587867934,
      "loss": 4.8544,
      "mean_token_accuracy": 0.2229031503200531,
      "num_tokens": 117738921.0,
      "step": 67875
    },
    {
      "entropy": 5.421004343032837,
      "epoch": 5.281268235751799,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025772802484954217,
      "loss": 4.7026,
      "mean_token_accuracy": 0.2268058553338051,
      "num_tokens": 117747182.0,
      "step": 67880
    },
    {
      "entropy": 5.373906660079956,
      "epoch": 5.281657265123517,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00025770039117436566,
      "loss": 4.6493,
      "mean_token_accuracy": 0.23034662306308745,
      "num_tokens": 117755454.0,
      "step": 67885
    },
    {
      "entropy": 5.362284851074219,
      "epoch": 5.282046294495235,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002576727577616832,
      "loss": 4.7456,
      "mean_token_accuracy": 0.23015681952238082,
      "num_tokens": 117764940.0,
      "step": 67890
    },
    {
      "entropy": 5.475304698944091,
      "epoch": 5.282435323866952,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00025764512461191413,
      "loss": 4.7215,
      "mean_token_accuracy": 0.2319142132997513,
      "num_tokens": 117772699.0,
      "step": 67895
    },
    {
      "entropy": 5.399413299560547,
      "epoch": 5.282824353238669,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002576174917254776,
      "loss": 4.6185,
      "mean_token_accuracy": 0.2434467077255249,
      "num_tokens": 117781793.0,
      "step": 67900
    },
    {
      "entropy": 5.451322460174561,
      "epoch": 5.283213382610387,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025758985910279304,
      "loss": 4.7603,
      "mean_token_accuracy": 0.22349606901407243,
      "num_tokens": 117791675.0,
      "step": 67905
    },
    {
      "entropy": 5.331875371932983,
      "epoch": 5.283602411982105,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002575622267442796,
      "loss": 4.6529,
      "mean_token_accuracy": 0.22651276737451553,
      "num_tokens": 117801328.0,
      "step": 67910
    },
    {
      "entropy": 5.399914693832398,
      "epoch": 5.283991441353822,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002575345946503566,
      "loss": 4.752,
      "mean_token_accuracy": 0.2273305058479309,
      "num_tokens": 117810320.0,
      "step": 67915
    },
    {
      "entropy": 5.423849248886109,
      "epoch": 5.28438047072554,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025750696282144335,
      "loss": 4.6999,
      "mean_token_accuracy": 0.23206643760204315,
      "num_tokens": 117818411.0,
      "step": 67920
    },
    {
      "entropy": 5.4297792434692385,
      "epoch": 5.284769500097258,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002574793312579592,
      "loss": 4.7654,
      "mean_token_accuracy": 0.2329715833067894,
      "num_tokens": 117827437.0,
      "step": 67925
    },
    {
      "entropy": 5.388472080230713,
      "epoch": 5.2851585294689745,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025745169996032317,
      "loss": 4.6584,
      "mean_token_accuracy": 0.2284606173634529,
      "num_tokens": 117835061.0,
      "step": 67930
    },
    {
      "entropy": 5.365268087387085,
      "epoch": 5.285547558840692,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002574240689289547,
      "loss": 4.7698,
      "mean_token_accuracy": 0.22489264905452727,
      "num_tokens": 117843205.0,
      "step": 67935
    },
    {
      "entropy": 5.34759488105774,
      "epoch": 5.28593658821241,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00025739643816427305,
      "loss": 4.6317,
      "mean_token_accuracy": 0.2360741138458252,
      "num_tokens": 117851797.0,
      "step": 67940
    },
    {
      "entropy": 5.4044536590576175,
      "epoch": 5.286325617584128,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002573688076666974,
      "loss": 4.7246,
      "mean_token_accuracy": 0.2280576005578041,
      "num_tokens": 117859915.0,
      "step": 67945
    },
    {
      "entropy": 5.379363965988159,
      "epoch": 5.286714646955845,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025734117743664704,
      "loss": 4.7473,
      "mean_token_accuracy": 0.22872845381498336,
      "num_tokens": 117868877.0,
      "step": 67950
    },
    {
      "entropy": 5.359675025939941,
      "epoch": 5.287103676327563,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002573135474745412,
      "loss": 4.6431,
      "mean_token_accuracy": 0.2399894878268242,
      "num_tokens": 117877930.0,
      "step": 67955
    },
    {
      "entropy": 5.3367678165435795,
      "epoch": 5.287492705699281,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00025728591778079906,
      "loss": 4.6348,
      "mean_token_accuracy": 0.2404622033238411,
      "num_tokens": 117886083.0,
      "step": 67960
    },
    {
      "entropy": 5.377387523651123,
      "epoch": 5.2878817350709975,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025725828835584,
      "loss": 4.7019,
      "mean_token_accuracy": 0.22656186521053315,
      "num_tokens": 117894047.0,
      "step": 67965
    },
    {
      "entropy": 5.451760959625244,
      "epoch": 5.288270764442715,
      "grad_norm": 1.28125,
      "learning_rate": 0.00025723065920008314,
      "loss": 4.7284,
      "mean_token_accuracy": 0.2294162020087242,
      "num_tokens": 117902875.0,
      "step": 67970
    },
    {
      "entropy": 5.445099592208862,
      "epoch": 5.288659793814433,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002572030303139477,
      "loss": 4.7645,
      "mean_token_accuracy": 0.22684448808431626,
      "num_tokens": 117911090.0,
      "step": 67975
    },
    {
      "entropy": 5.465509128570557,
      "epoch": 5.2890488231861505,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025717540169785304,
      "loss": 4.7575,
      "mean_token_accuracy": 0.23597677648067475,
      "num_tokens": 117920388.0,
      "step": 67980
    },
    {
      "entropy": 5.473395252227784,
      "epoch": 5.289437852557868,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002571477733522183,
      "loss": 4.78,
      "mean_token_accuracy": 0.22574895173311232,
      "num_tokens": 117929458.0,
      "step": 67985
    },
    {
      "entropy": 5.39021258354187,
      "epoch": 5.289826881929586,
      "grad_norm": 1.171875,
      "learning_rate": 0.00025712014527746264,
      "loss": 4.7066,
      "mean_token_accuracy": 0.2375793218612671,
      "num_tokens": 117939239.0,
      "step": 67990
    },
    {
      "entropy": 5.42416410446167,
      "epoch": 5.290215911301304,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002570925174740053,
      "loss": 4.7968,
      "mean_token_accuracy": 0.2296088680624962,
      "num_tokens": 117949010.0,
      "step": 67995
    },
    {
      "entropy": 5.372650957107544,
      "epoch": 5.2906049406730205,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002570648899422655,
      "loss": 4.6425,
      "mean_token_accuracy": 0.23721289485692978,
      "num_tokens": 117957431.0,
      "step": 68000
    },
    {
      "entropy": 5.448591518402099,
      "epoch": 5.290993970044738,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002570372626826625,
      "loss": 4.7441,
      "mean_token_accuracy": 0.22351339608430862,
      "num_tokens": 117965679.0,
      "step": 68005
    },
    {
      "entropy": 5.4630186557769775,
      "epoch": 5.291382999416456,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002570096356956155,
      "loss": 4.6965,
      "mean_token_accuracy": 0.2300425574183464,
      "num_tokens": 117974326.0,
      "step": 68010
    },
    {
      "entropy": 5.439398288726807,
      "epoch": 5.2917720287881735,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025698200898154354,
      "loss": 4.725,
      "mean_token_accuracy": 0.23586793392896652,
      "num_tokens": 117984038.0,
      "step": 68015
    },
    {
      "entropy": 5.339920282363892,
      "epoch": 5.292161058159891,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025695438254086595,
      "loss": 4.6318,
      "mean_token_accuracy": 0.24165019094944,
      "num_tokens": 117992402.0,
      "step": 68020
    },
    {
      "entropy": 5.34862961769104,
      "epoch": 5.292550087531609,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002569267563740018,
      "loss": 4.7125,
      "mean_token_accuracy": 0.23222738802433013,
      "num_tokens": 118001762.0,
      "step": 68025
    },
    {
      "entropy": 5.344294309616089,
      "epoch": 5.292939116903327,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002568991304813704,
      "loss": 4.6577,
      "mean_token_accuracy": 0.23492414504289627,
      "num_tokens": 118010198.0,
      "step": 68030
    },
    {
      "entropy": 5.40525369644165,
      "epoch": 5.2933281462750434,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025687150486339094,
      "loss": 4.6835,
      "mean_token_accuracy": 0.22730460166931152,
      "num_tokens": 118019027.0,
      "step": 68035
    },
    {
      "entropy": 5.432329320907593,
      "epoch": 5.293717175646761,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00025684387952048247,
      "loss": 4.785,
      "mean_token_accuracy": 0.23295140415430068,
      "num_tokens": 118027507.0,
      "step": 68040
    },
    {
      "entropy": 5.4012213230133055,
      "epoch": 5.294106205018479,
      "grad_norm": 1.125,
      "learning_rate": 0.00025681625445306435,
      "loss": 4.7557,
      "mean_token_accuracy": 0.23130976408720016,
      "num_tokens": 118036269.0,
      "step": 68045
    },
    {
      "entropy": 5.417764329910279,
      "epoch": 5.2944952343901965,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002567886296615555,
      "loss": 4.6436,
      "mean_token_accuracy": 0.2363480031490326,
      "num_tokens": 118044634.0,
      "step": 68050
    },
    {
      "entropy": 5.33132791519165,
      "epoch": 5.294884263761914,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025676100514637523,
      "loss": 4.6156,
      "mean_token_accuracy": 0.23988511860370637,
      "num_tokens": 118053151.0,
      "step": 68055
    },
    {
      "entropy": 5.35179967880249,
      "epoch": 5.295273293133632,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025673338090794276,
      "loss": 4.708,
      "mean_token_accuracy": 0.23388061672449112,
      "num_tokens": 118061554.0,
      "step": 68060
    },
    {
      "entropy": 5.4061057567596436,
      "epoch": 5.295662322505349,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002567057569466771,
      "loss": 4.7306,
      "mean_token_accuracy": 0.23217600136995314,
      "num_tokens": 118070201.0,
      "step": 68065
    },
    {
      "entropy": 5.3677606105804445,
      "epoch": 5.296051351877066,
      "grad_norm": 1.203125,
      "learning_rate": 0.00025667813326299746,
      "loss": 4.6529,
      "mean_token_accuracy": 0.2347490206360817,
      "num_tokens": 118078397.0,
      "step": 68070
    },
    {
      "entropy": 5.3858870506286625,
      "epoch": 5.296440381248784,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025665050985732295,
      "loss": 4.6554,
      "mean_token_accuracy": 0.2424492210149765,
      "num_tokens": 118086127.0,
      "step": 68075
    },
    {
      "entropy": 5.31528902053833,
      "epoch": 5.296829410620502,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00025662288673007286,
      "loss": 4.6622,
      "mean_token_accuracy": 0.2249770477414131,
      "num_tokens": 118095466.0,
      "step": 68080
    },
    {
      "entropy": 5.375209856033325,
      "epoch": 5.2972184399922195,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00025659526388166617,
      "loss": 4.7286,
      "mean_token_accuracy": 0.222996324300766,
      "num_tokens": 118104452.0,
      "step": 68085
    },
    {
      "entropy": 5.4704029083251955,
      "epoch": 5.297607469363937,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025656764131252194,
      "loss": 4.7593,
      "mean_token_accuracy": 0.22569903880357742,
      "num_tokens": 118113347.0,
      "step": 68090
    },
    {
      "entropy": 5.446114826202392,
      "epoch": 5.297996498735655,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002565400190230595,
      "loss": 4.7698,
      "mean_token_accuracy": 0.2239377588033676,
      "num_tokens": 118122460.0,
      "step": 68095
    },
    {
      "entropy": 5.4446149349212645,
      "epoch": 5.298385528107372,
      "grad_norm": 1.25,
      "learning_rate": 0.0002565123970136978,
      "loss": 4.8144,
      "mean_token_accuracy": 0.22506532967090606,
      "num_tokens": 118131191.0,
      "step": 68100
    },
    {
      "entropy": 5.372435569763184,
      "epoch": 5.298774557479089,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002564847752848562,
      "loss": 4.7228,
      "mean_token_accuracy": 0.2330501824617386,
      "num_tokens": 118140202.0,
      "step": 68105
    },
    {
      "entropy": 5.4700723648071286,
      "epoch": 5.299163586850807,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025645715383695357,
      "loss": 4.764,
      "mean_token_accuracy": 0.22737912684679032,
      "num_tokens": 118149038.0,
      "step": 68110
    },
    {
      "entropy": 5.367864322662354,
      "epoch": 5.299552616222525,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002564295326704092,
      "loss": 4.6119,
      "mean_token_accuracy": 0.23903272151947022,
      "num_tokens": 118157524.0,
      "step": 68115
    },
    {
      "entropy": 5.372653245925903,
      "epoch": 5.2999416455942425,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002564019117856421,
      "loss": 4.6961,
      "mean_token_accuracy": 0.23593139201402663,
      "num_tokens": 118165399.0,
      "step": 68120
    },
    {
      "entropy": 5.334555816650391,
      "epoch": 5.30033067496596,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025637429118307133,
      "loss": 4.6275,
      "mean_token_accuracy": 0.24422341734170913,
      "num_tokens": 118173875.0,
      "step": 68125
    },
    {
      "entropy": 5.33554630279541,
      "epoch": 5.300719704337677,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002563466708631161,
      "loss": 4.7274,
      "mean_token_accuracy": 0.2325017660856247,
      "num_tokens": 118182854.0,
      "step": 68130
    },
    {
      "entropy": 5.368835735321045,
      "epoch": 5.301108733709395,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002563190508261954,
      "loss": 4.685,
      "mean_token_accuracy": 0.22467972636222838,
      "num_tokens": 118192409.0,
      "step": 68135
    },
    {
      "entropy": 5.447922849655152,
      "epoch": 5.301497763081112,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025629143107272837,
      "loss": 4.7521,
      "mean_token_accuracy": 0.22505049109458924,
      "num_tokens": 118200441.0,
      "step": 68140
    },
    {
      "entropy": 5.37948784828186,
      "epoch": 5.30188679245283,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002562638116031341,
      "loss": 4.7157,
      "mean_token_accuracy": 0.23519937694072723,
      "num_tokens": 118209618.0,
      "step": 68145
    },
    {
      "entropy": 5.405431890487671,
      "epoch": 5.302275821824548,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002562361924178317,
      "loss": 4.6858,
      "mean_token_accuracy": 0.22774650901556015,
      "num_tokens": 118218284.0,
      "step": 68150
    },
    {
      "entropy": 5.4235821723937985,
      "epoch": 5.3026648511962655,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025620857351724017,
      "loss": 4.7619,
      "mean_token_accuracy": 0.23083454519510269,
      "num_tokens": 118226760.0,
      "step": 68155
    },
    {
      "entropy": 5.331353092193604,
      "epoch": 5.303053880567983,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025618095490177864,
      "loss": 4.652,
      "mean_token_accuracy": 0.2351802483201027,
      "num_tokens": 118235511.0,
      "step": 68160
    },
    {
      "entropy": 5.392614841461182,
      "epoch": 5.303442909939701,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025615333657186615,
      "loss": 4.7404,
      "mean_token_accuracy": 0.22456002086400986,
      "num_tokens": 118243665.0,
      "step": 68165
    },
    {
      "entropy": 5.449662160873413,
      "epoch": 5.303831939311418,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002561257185279218,
      "loss": 4.7977,
      "mean_token_accuracy": 0.22283245474100113,
      "num_tokens": 118252996.0,
      "step": 68170
    },
    {
      "entropy": 5.4242912292480465,
      "epoch": 5.304220968683135,
      "grad_norm": 1.1875,
      "learning_rate": 0.00025609810077036467,
      "loss": 4.7157,
      "mean_token_accuracy": 0.23114178329706192,
      "num_tokens": 118261436.0,
      "step": 68175
    },
    {
      "entropy": 5.426201915740966,
      "epoch": 5.304609998054853,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002560704832996137,
      "loss": 4.7429,
      "mean_token_accuracy": 0.23235445469617844,
      "num_tokens": 118269537.0,
      "step": 68180
    },
    {
      "entropy": 5.379454755783081,
      "epoch": 5.304999027426571,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002560428661160881,
      "loss": 4.6869,
      "mean_token_accuracy": 0.23679289519786834,
      "num_tokens": 118277207.0,
      "step": 68185
    },
    {
      "entropy": 5.437486219406128,
      "epoch": 5.3053880567982885,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00025601524922020676,
      "loss": 4.7564,
      "mean_token_accuracy": 0.2364332288503647,
      "num_tokens": 118285210.0,
      "step": 68190
    },
    {
      "entropy": 5.4189800262451175,
      "epoch": 5.305777086170006,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002559876326123889,
      "loss": 4.7566,
      "mean_token_accuracy": 0.22733357399702073,
      "num_tokens": 118293668.0,
      "step": 68195
    },
    {
      "entropy": 5.517509031295776,
      "epoch": 5.306166115541723,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002559600162930534,
      "loss": 4.8368,
      "mean_token_accuracy": 0.22202365100383759,
      "num_tokens": 118302203.0,
      "step": 68200
    },
    {
      "entropy": 5.469154214859008,
      "epoch": 5.306555144913441,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002559324002626193,
      "loss": 4.7401,
      "mean_token_accuracy": 0.2301134094595909,
      "num_tokens": 118311252.0,
      "step": 68205
    },
    {
      "entropy": 5.392446708679199,
      "epoch": 5.306944174285158,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002559047845215057,
      "loss": 4.6874,
      "mean_token_accuracy": 0.2311457797884941,
      "num_tokens": 118320136.0,
      "step": 68210
    },
    {
      "entropy": 5.383842945098877,
      "epoch": 5.307333203656876,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002558771690701316,
      "loss": 4.6694,
      "mean_token_accuracy": 0.23015664368867875,
      "num_tokens": 118328871.0,
      "step": 68215
    },
    {
      "entropy": 5.420992994308472,
      "epoch": 5.307722233028594,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000255849553908916,
      "loss": 4.6759,
      "mean_token_accuracy": 0.23601814806461335,
      "num_tokens": 118337588.0,
      "step": 68220
    },
    {
      "entropy": 5.4108617305755615,
      "epoch": 5.3081112624003115,
      "grad_norm": 1.1875,
      "learning_rate": 0.00025582193903827795,
      "loss": 4.7184,
      "mean_token_accuracy": 0.22655766308307648,
      "num_tokens": 118347200.0,
      "step": 68225
    },
    {
      "entropy": 5.426698350906372,
      "epoch": 5.308500291772029,
      "grad_norm": 1.25,
      "learning_rate": 0.0002557943244586365,
      "loss": 4.6559,
      "mean_token_accuracy": 0.23621433079242707,
      "num_tokens": 118355226.0,
      "step": 68230
    },
    {
      "entropy": 5.374127340316773,
      "epoch": 5.308889321143746,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00025576671017041053,
      "loss": 4.6349,
      "mean_token_accuracy": 0.24040888845920563,
      "num_tokens": 118363491.0,
      "step": 68235
    },
    {
      "entropy": 5.416652250289917,
      "epoch": 5.309278350515464,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002557390961740192,
      "loss": 4.8012,
      "mean_token_accuracy": 0.22745504677295686,
      "num_tokens": 118372070.0,
      "step": 68240
    },
    {
      "entropy": 5.397065162658691,
      "epoch": 5.309667379887181,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00025571148246988137,
      "loss": 4.8041,
      "mean_token_accuracy": 0.2236845538020134,
      "num_tokens": 118380093.0,
      "step": 68245
    },
    {
      "entropy": 5.473998928070069,
      "epoch": 5.310056409258899,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025568386905841614,
      "loss": 4.7503,
      "mean_token_accuracy": 0.2338694825768471,
      "num_tokens": 118388795.0,
      "step": 68250
    },
    {
      "entropy": 5.497718715667725,
      "epoch": 5.310445438630617,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002556562559400424,
      "loss": 4.7594,
      "mean_token_accuracy": 0.23210667073726654,
      "num_tokens": 118397155.0,
      "step": 68255
    },
    {
      "entropy": 5.475577878952026,
      "epoch": 5.3108344680023345,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002556286431151792,
      "loss": 4.7716,
      "mean_token_accuracy": 0.2255820944905281,
      "num_tokens": 118405431.0,
      "step": 68260
    },
    {
      "entropy": 5.354866075515747,
      "epoch": 5.311223497374051,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002556010305842456,
      "loss": 4.6815,
      "mean_token_accuracy": 0.23072147369384766,
      "num_tokens": 118413882.0,
      "step": 68265
    },
    {
      "entropy": 5.4126180648803714,
      "epoch": 5.311612526745769,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002555734183476603,
      "loss": 4.727,
      "mean_token_accuracy": 0.2353020116686821,
      "num_tokens": 118422101.0,
      "step": 68270
    },
    {
      "entropy": 5.449949598312378,
      "epoch": 5.312001556117487,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002555458064058427,
      "loss": 4.7719,
      "mean_token_accuracy": 0.23055623322725297,
      "num_tokens": 118431100.0,
      "step": 68275
    },
    {
      "entropy": 5.447512578964234,
      "epoch": 5.312390585489204,
      "grad_norm": 1.203125,
      "learning_rate": 0.00025551819475921135,
      "loss": 4.8008,
      "mean_token_accuracy": 0.22128179371356965,
      "num_tokens": 118440398.0,
      "step": 68280
    },
    {
      "entropy": 5.419225120544434,
      "epoch": 5.312779614860922,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002554905834081854,
      "loss": 4.6395,
      "mean_token_accuracy": 0.23864019215106963,
      "num_tokens": 118448106.0,
      "step": 68285
    },
    {
      "entropy": 5.353057479858398,
      "epoch": 5.31316864423264,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002554629723531838,
      "loss": 4.6845,
      "mean_token_accuracy": 0.23960304111242295,
      "num_tokens": 118457381.0,
      "step": 68290
    },
    {
      "entropy": 5.323422288894653,
      "epoch": 5.3135576736043575,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002554353615946255,
      "loss": 4.6605,
      "mean_token_accuracy": 0.2350957587361336,
      "num_tokens": 118466064.0,
      "step": 68295
    },
    {
      "entropy": 5.424555730819702,
      "epoch": 5.313946702976074,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025540775113292943,
      "loss": 4.777,
      "mean_token_accuracy": 0.23124974071979523,
      "num_tokens": 118474781.0,
      "step": 68300
    },
    {
      "entropy": 5.5154975891113285,
      "epoch": 5.314335732347792,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002553801409685146,
      "loss": 4.7737,
      "mean_token_accuracy": 0.23112404048442842,
      "num_tokens": 118484090.0,
      "step": 68305
    },
    {
      "entropy": 5.383425760269165,
      "epoch": 5.31472476171951,
      "grad_norm": 1.203125,
      "learning_rate": 0.00025535253110179994,
      "loss": 4.6353,
      "mean_token_accuracy": 0.2288293421268463,
      "num_tokens": 118492238.0,
      "step": 68310
    },
    {
      "entropy": 5.409465551376343,
      "epoch": 5.315113791091227,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002553249215332043,
      "loss": 4.7616,
      "mean_token_accuracy": 0.22517294883728028,
      "num_tokens": 118501121.0,
      "step": 68315
    },
    {
      "entropy": 5.349946689605713,
      "epoch": 5.315502820462945,
      "grad_norm": 1.265625,
      "learning_rate": 0.00025529731226314663,
      "loss": 4.6686,
      "mean_token_accuracy": 0.24075792878866195,
      "num_tokens": 118509594.0,
      "step": 68320
    },
    {
      "entropy": 5.501828527450561,
      "epoch": 5.315891849834663,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002552697032920459,
      "loss": 4.8809,
      "mean_token_accuracy": 0.22262437641620636,
      "num_tokens": 118518261.0,
      "step": 68325
    },
    {
      "entropy": 5.449336147308349,
      "epoch": 5.3162808792063805,
      "grad_norm": 1.328125,
      "learning_rate": 0.00025524209462032105,
      "loss": 4.7749,
      "mean_token_accuracy": 0.22266142219305038,
      "num_tokens": 118527075.0,
      "step": 68330
    },
    {
      "entropy": 5.461490726470947,
      "epoch": 5.316669908578097,
      "grad_norm": 1.25,
      "learning_rate": 0.0002552144862483909,
      "loss": 4.7012,
      "mean_token_accuracy": 0.23407820910215377,
      "num_tokens": 118536235.0,
      "step": 68335
    },
    {
      "entropy": 5.393811988830566,
      "epoch": 5.317058937949815,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002551868781766745,
      "loss": 4.6954,
      "mean_token_accuracy": 0.23548172861337663,
      "num_tokens": 118544372.0,
      "step": 68340
    },
    {
      "entropy": 5.428950452804566,
      "epoch": 5.317447967321533,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002551592704055907,
      "loss": 4.7507,
      "mean_token_accuracy": 0.23705195039510726,
      "num_tokens": 118552719.0,
      "step": 68345
    },
    {
      "entropy": 5.35980486869812,
      "epoch": 5.31783699669325,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00025513166293555836,
      "loss": 4.6744,
      "mean_token_accuracy": 0.24016321450471878,
      "num_tokens": 118560976.0,
      "step": 68350
    },
    {
      "entropy": 5.430757808685303,
      "epoch": 5.318226026064968,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00025510405576699636,
      "loss": 4.8909,
      "mean_token_accuracy": 0.21354012489318847,
      "num_tokens": 118570003.0,
      "step": 68355
    },
    {
      "entropy": 5.368334484100342,
      "epoch": 5.318615055436686,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002550764489003237,
      "loss": 4.6548,
      "mean_token_accuracy": 0.23521031439304352,
      "num_tokens": 118578500.0,
      "step": 68360
    },
    {
      "entropy": 5.433859539031983,
      "epoch": 5.3190040848084035,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002550488423359592,
      "loss": 4.7115,
      "mean_token_accuracy": 0.2260461613535881,
      "num_tokens": 118586508.0,
      "step": 68365
    },
    {
      "entropy": 5.4215172290802,
      "epoch": 5.31939311418012,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025502123607432174,
      "loss": 4.8028,
      "mean_token_accuracy": 0.21999171376228333,
      "num_tokens": 118595395.0,
      "step": 68370
    },
    {
      "entropy": 5.357114171981811,
      "epoch": 5.319782143551838,
      "grad_norm": 1.203125,
      "learning_rate": 0.00025499363011583024,
      "loss": 4.6694,
      "mean_token_accuracy": 0.23116080164909364,
      "num_tokens": 118604698.0,
      "step": 68375
    },
    {
      "entropy": 5.399623107910156,
      "epoch": 5.320171172923556,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002549660244609036,
      "loss": 4.6855,
      "mean_token_accuracy": 0.23996977657079696,
      "num_tokens": 118613575.0,
      "step": 68380
    },
    {
      "entropy": 5.429344987869262,
      "epoch": 5.320560202295273,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002549384191099607,
      "loss": 4.7736,
      "mean_token_accuracy": 0.23215679824352264,
      "num_tokens": 118622709.0,
      "step": 68385
    },
    {
      "entropy": 5.404136228561401,
      "epoch": 5.320949231666991,
      "grad_norm": 1.34375,
      "learning_rate": 0.00025491081406342025,
      "loss": 4.6728,
      "mean_token_accuracy": 0.23223583102226258,
      "num_tokens": 118631744.0,
      "step": 68390
    },
    {
      "entropy": 5.462610816955566,
      "epoch": 5.321338261038709,
      "grad_norm": 1.25,
      "learning_rate": 0.00025488320932170127,
      "loss": 4.7388,
      "mean_token_accuracy": 0.2320067048072815,
      "num_tokens": 118639833.0,
      "step": 68395
    },
    {
      "entropy": 5.462297582626343,
      "epoch": 5.321727290410426,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002548556048852226,
      "loss": 4.7932,
      "mean_token_accuracy": 0.22314848005771637,
      "num_tokens": 118647870.0,
      "step": 68400
    },
    {
      "entropy": 5.345164203643799,
      "epoch": 5.322116319782143,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025482800075440296,
      "loss": 4.6762,
      "mean_token_accuracy": 0.2361781567335129,
      "num_tokens": 118656572.0,
      "step": 68405
    },
    {
      "entropy": 5.32938585281372,
      "epoch": 5.322505349153861,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002548003969296614,
      "loss": 4.664,
      "mean_token_accuracy": 0.23306829184293748,
      "num_tokens": 118665226.0,
      "step": 68410
    },
    {
      "entropy": 5.381419944763183,
      "epoch": 5.322894378525579,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025477279341141665,
      "loss": 4.7996,
      "mean_token_accuracy": 0.22352965623140336,
      "num_tokens": 118674736.0,
      "step": 68415
    },
    {
      "entropy": 5.370222282409668,
      "epoch": 5.323283407897296,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002547451902000875,
      "loss": 4.7068,
      "mean_token_accuracy": 0.2351436510682106,
      "num_tokens": 118682670.0,
      "step": 68420
    },
    {
      "entropy": 5.367035150527954,
      "epoch": 5.323672437269014,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025471758729609284,
      "loss": 4.7163,
      "mean_token_accuracy": 0.23232802152633666,
      "num_tokens": 118691758.0,
      "step": 68425
    },
    {
      "entropy": 5.3664836406707765,
      "epoch": 5.324061466640732,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00025468998469985155,
      "loss": 4.6175,
      "mean_token_accuracy": 0.23248442262411118,
      "num_tokens": 118700634.0,
      "step": 68430
    },
    {
      "entropy": 5.323926734924316,
      "epoch": 5.324450496012449,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002546623824117824,
      "loss": 4.5741,
      "mean_token_accuracy": 0.2437952384352684,
      "num_tokens": 118708968.0,
      "step": 68435
    },
    {
      "entropy": 5.420853614807129,
      "epoch": 5.324839525384166,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002546347804323043,
      "loss": 4.7697,
      "mean_token_accuracy": 0.23351432979106904,
      "num_tokens": 118717555.0,
      "step": 68440
    },
    {
      "entropy": 5.379099035263062,
      "epoch": 5.325228554755884,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002546071787618359,
      "loss": 4.7132,
      "mean_token_accuracy": 0.23423105776309966,
      "num_tokens": 118725815.0,
      "step": 68445
    },
    {
      "entropy": 5.444296550750733,
      "epoch": 5.325617584127602,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002545795774007962,
      "loss": 4.7259,
      "mean_token_accuracy": 0.23276938647031784,
      "num_tokens": 118734573.0,
      "step": 68450
    },
    {
      "entropy": 5.4277605533599855,
      "epoch": 5.326006613499319,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002545519763496038,
      "loss": 4.7288,
      "mean_token_accuracy": 0.22015426456928253,
      "num_tokens": 118743580.0,
      "step": 68455
    },
    {
      "entropy": 5.408220911026001,
      "epoch": 5.326395642871037,
      "grad_norm": 1.203125,
      "learning_rate": 0.00025452437560867774,
      "loss": 4.6802,
      "mean_token_accuracy": 0.2388938769698143,
      "num_tokens": 118751778.0,
      "step": 68460
    },
    {
      "entropy": 5.386774492263794,
      "epoch": 5.326784672242754,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002544967751784367,
      "loss": 4.7395,
      "mean_token_accuracy": 0.23511845022439956,
      "num_tokens": 118759756.0,
      "step": 68465
    },
    {
      "entropy": 5.387465143203736,
      "epoch": 5.327173701614472,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00025446917505929937,
      "loss": 4.7487,
      "mean_token_accuracy": 0.2242644503712654,
      "num_tokens": 118767971.0,
      "step": 68470
    },
    {
      "entropy": 5.460736989974976,
      "epoch": 5.327562730986189,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002544415752516846,
      "loss": 4.8147,
      "mean_token_accuracy": 0.21734969019889833,
      "num_tokens": 118776484.0,
      "step": 68475
    },
    {
      "entropy": 5.432301568984985,
      "epoch": 5.327951760357907,
      "grad_norm": 1.25,
      "learning_rate": 0.00025441397575601125,
      "loss": 4.7346,
      "mean_token_accuracy": 0.22482049763202666,
      "num_tokens": 118785967.0,
      "step": 68480
    },
    {
      "entropy": 5.423052215576172,
      "epoch": 5.328340789729625,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00025438637657269805,
      "loss": 4.7718,
      "mean_token_accuracy": 0.2262672781944275,
      "num_tokens": 118794663.0,
      "step": 68485
    },
    {
      "entropy": 5.380197906494141,
      "epoch": 5.328729819101342,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002543587777021636,
      "loss": 4.7041,
      "mean_token_accuracy": 0.23597897738218307,
      "num_tokens": 118803231.0,
      "step": 68490
    },
    {
      "entropy": 5.434101486206055,
      "epoch": 5.32911884847306,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002543311791448271,
      "loss": 4.7151,
      "mean_token_accuracy": 0.2352515786886215,
      "num_tokens": 118811499.0,
      "step": 68495
    },
    {
      "entropy": 5.412351083755493,
      "epoch": 5.329507877844778,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025430358090110705,
      "loss": 4.7532,
      "mean_token_accuracy": 0.22926274687051773,
      "num_tokens": 118820355.0,
      "step": 68500
    },
    {
      "entropy": 5.428201007843017,
      "epoch": 5.329896907216495,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025427598297142213,
      "loss": 4.7337,
      "mean_token_accuracy": 0.23664519786834717,
      "num_tokens": 118829509.0,
      "step": 68505
    },
    {
      "entropy": 5.425869369506836,
      "epoch": 5.330285936588212,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002542483853561912,
      "loss": 4.7289,
      "mean_token_accuracy": 0.23558319807052613,
      "num_tokens": 118837430.0,
      "step": 68510
    },
    {
      "entropy": 5.405182123184204,
      "epoch": 5.33067496595993,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000254220788055833,
      "loss": 4.6777,
      "mean_token_accuracy": 0.23449928611516951,
      "num_tokens": 118845512.0,
      "step": 68515
    },
    {
      "entropy": 5.450462102890015,
      "epoch": 5.331063995331648,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002541931910707662,
      "loss": 4.6985,
      "mean_token_accuracy": 0.22829007357358932,
      "num_tokens": 118854076.0,
      "step": 68520
    },
    {
      "entropy": 5.350256252288818,
      "epoch": 5.331453024703365,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002541655944014097,
      "loss": 4.6339,
      "mean_token_accuracy": 0.24013522416353225,
      "num_tokens": 118862383.0,
      "step": 68525
    },
    {
      "entropy": 5.409934759140015,
      "epoch": 5.331842054075083,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002541379980481821,
      "loss": 4.6616,
      "mean_token_accuracy": 0.23402135223150253,
      "num_tokens": 118870227.0,
      "step": 68530
    },
    {
      "entropy": 5.364732551574707,
      "epoch": 5.3322310834468,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002541104020115022,
      "loss": 4.7026,
      "mean_token_accuracy": 0.23599833250045776,
      "num_tokens": 118878760.0,
      "step": 68535
    },
    {
      "entropy": 5.3500714778900145,
      "epoch": 5.3326201128185176,
      "grad_norm": 1.25,
      "learning_rate": 0.00025408280629178864,
      "loss": 4.6691,
      "mean_token_accuracy": 0.23276754468679428,
      "num_tokens": 118887273.0,
      "step": 68540
    },
    {
      "entropy": 5.4352583408355715,
      "epoch": 5.333009142190235,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002540552108894603,
      "loss": 4.6839,
      "mean_token_accuracy": 0.2357409879565239,
      "num_tokens": 118896076.0,
      "step": 68545
    },
    {
      "entropy": 5.3710356712341305,
      "epoch": 5.333398171561953,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002540276158049358,
      "loss": 4.7254,
      "mean_token_accuracy": 0.2263556882739067,
      "num_tokens": 118905287.0,
      "step": 68550
    },
    {
      "entropy": 5.420057010650635,
      "epoch": 5.333787200933671,
      "grad_norm": 1.34375,
      "learning_rate": 0.00025400002103863367,
      "loss": 4.7671,
      "mean_token_accuracy": 0.22344884872436524,
      "num_tokens": 118913644.0,
      "step": 68555
    },
    {
      "entropy": 5.5147096633911135,
      "epoch": 5.334176230305388,
      "grad_norm": 1.25,
      "learning_rate": 0.00025397242659097293,
      "loss": 4.7805,
      "mean_token_accuracy": 0.23348459750413894,
      "num_tokens": 118922394.0,
      "step": 68560
    },
    {
      "entropy": 5.41204195022583,
      "epoch": 5.334565259677106,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002539448324623722,
      "loss": 4.6885,
      "mean_token_accuracy": 0.2329455643892288,
      "num_tokens": 118930164.0,
      "step": 68565
    },
    {
      "entropy": 5.434147310256958,
      "epoch": 5.334954289048823,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025391723865325,
      "loss": 4.7715,
      "mean_token_accuracy": 0.22877319306135177,
      "num_tokens": 118938711.0,
      "step": 68570
    },
    {
      "entropy": 5.389657354354858,
      "epoch": 5.3353433184205405,
      "grad_norm": 1.25,
      "learning_rate": 0.00025388964516402527,
      "loss": 4.7766,
      "mean_token_accuracy": 0.22858673930168152,
      "num_tokens": 118947334.0,
      "step": 68575
    },
    {
      "entropy": 5.434982109069824,
      "epoch": 5.335732347792258,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002538620519951165,
      "loss": 4.6716,
      "mean_token_accuracy": 0.24100470840930938,
      "num_tokens": 118955219.0,
      "step": 68580
    },
    {
      "entropy": 5.38073844909668,
      "epoch": 5.336121377163976,
      "grad_norm": 1.25,
      "learning_rate": 0.00025383445914694245,
      "loss": 4.7021,
      "mean_token_accuracy": 0.2291403204202652,
      "num_tokens": 118963723.0,
      "step": 68585
    },
    {
      "entropy": 5.447078037261963,
      "epoch": 5.336510406535694,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002538068666199218,
      "loss": 4.7665,
      "mean_token_accuracy": 0.222410124540329,
      "num_tokens": 118972024.0,
      "step": 68590
    },
    {
      "entropy": 5.458129739761352,
      "epoch": 5.336899435907411,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002537792744144732,
      "loss": 4.7973,
      "mean_token_accuracy": 0.22942695915699005,
      "num_tokens": 118980970.0,
      "step": 68595
    },
    {
      "entropy": 5.5080572128295895,
      "epoch": 5.337288465279128,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002537516825310154,
      "loss": 4.842,
      "mean_token_accuracy": 0.22439451068639754,
      "num_tokens": 118988883.0,
      "step": 68600
    },
    {
      "entropy": 5.441296720504761,
      "epoch": 5.337677494650846,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002537240909699669,
      "loss": 4.6828,
      "mean_token_accuracy": 0.23301975429058075,
      "num_tokens": 118997392.0,
      "step": 68605
    },
    {
      "entropy": 5.398757982254028,
      "epoch": 5.3380665240225635,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002536964997317465,
      "loss": 4.6686,
      "mean_token_accuracy": 0.2338730812072754,
      "num_tokens": 119005769.0,
      "step": 68610
    },
    {
      "entropy": 5.280376958847046,
      "epoch": 5.338455553394281,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025366890881677284,
      "loss": 4.5705,
      "mean_token_accuracy": 0.238440403342247,
      "num_tokens": 119014885.0,
      "step": 68615
    },
    {
      "entropy": 5.360273933410644,
      "epoch": 5.338844582765999,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002536413182254645,
      "loss": 4.7131,
      "mean_token_accuracy": 0.23360236436128617,
      "num_tokens": 119024209.0,
      "step": 68620
    },
    {
      "entropy": 5.359929513931275,
      "epoch": 5.339233612137717,
      "grad_norm": 1.21875,
      "learning_rate": 0.00025361372795824015,
      "loss": 4.6245,
      "mean_token_accuracy": 0.23610697239637374,
      "num_tokens": 119032977.0,
      "step": 68625
    },
    {
      "entropy": 5.401269769668579,
      "epoch": 5.339622641509434,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002535861380155185,
      "loss": 4.6733,
      "mean_token_accuracy": 0.23241161853075026,
      "num_tokens": 119041085.0,
      "step": 68630
    },
    {
      "entropy": 5.3947608947753904,
      "epoch": 5.340011670881151,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00025355854839771805,
      "loss": 4.6944,
      "mean_token_accuracy": 0.22551728636026383,
      "num_tokens": 119049407.0,
      "step": 68635
    },
    {
      "entropy": 5.3811286926269535,
      "epoch": 5.340400700252869,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00025353095910525753,
      "loss": 4.7352,
      "mean_token_accuracy": 0.23292427361011506,
      "num_tokens": 119057994.0,
      "step": 68640
    },
    {
      "entropy": 5.400046634674072,
      "epoch": 5.3407897296245865,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002535033701385555,
      "loss": 4.6607,
      "mean_token_accuracy": 0.23283186852931975,
      "num_tokens": 119066230.0,
      "step": 68645
    },
    {
      "entropy": 5.325264406204224,
      "epoch": 5.341178758996304,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002534757814980307,
      "loss": 4.6329,
      "mean_token_accuracy": 0.23990606367588044,
      "num_tokens": 119075146.0,
      "step": 68650
    },
    {
      "entropy": 5.452472972869873,
      "epoch": 5.341567788368022,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025344819318410167,
      "loss": 4.778,
      "mean_token_accuracy": 0.23047917932271958,
      "num_tokens": 119084297.0,
      "step": 68655
    },
    {
      "entropy": 5.47597484588623,
      "epoch": 5.34195681773974,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00025342060519718693,
      "loss": 4.7704,
      "mean_token_accuracy": 0.2309277355670929,
      "num_tokens": 119093430.0,
      "step": 68660
    },
    {
      "entropy": 5.402373266220093,
      "epoch": 5.342345847111457,
      "grad_norm": 1.34375,
      "learning_rate": 0.00025339301753770515,
      "loss": 4.831,
      "mean_token_accuracy": 0.22837504744529724,
      "num_tokens": 119103213.0,
      "step": 68665
    },
    {
      "entropy": 5.44883246421814,
      "epoch": 5.342734876483174,
      "grad_norm": 1.265625,
      "learning_rate": 0.000253365430206075,
      "loss": 4.7693,
      "mean_token_accuracy": 0.22818135917186738,
      "num_tokens": 119111412.0,
      "step": 68670
    },
    {
      "entropy": 5.526308870315551,
      "epoch": 5.343123905854892,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002533378432027149,
      "loss": 4.8157,
      "mean_token_accuracy": 0.2234542429447174,
      "num_tokens": 119120733.0,
      "step": 68675
    },
    {
      "entropy": 5.450880098342895,
      "epoch": 5.3435129352266095,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002533102565280437,
      "loss": 4.7185,
      "mean_token_accuracy": 0.2334365352988243,
      "num_tokens": 119129344.0,
      "step": 68680
    },
    {
      "entropy": 5.422966766357422,
      "epoch": 5.343901964598327,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002532826701824797,
      "loss": 4.7482,
      "mean_token_accuracy": 0.23017943054437637,
      "num_tokens": 119137520.0,
      "step": 68685
    },
    {
      "entropy": 5.423976087570191,
      "epoch": 5.344290993970045,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002532550841664416,
      "loss": 4.7841,
      "mean_token_accuracy": 0.2284732535481453,
      "num_tokens": 119146723.0,
      "step": 68690
    },
    {
      "entropy": 5.367533826828003,
      "epoch": 5.344680023341763,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002532274984803481,
      "loss": 4.5999,
      "mean_token_accuracy": 0.24107391238212586,
      "num_tokens": 119154937.0,
      "step": 68695
    },
    {
      "entropy": 5.394929647445679,
      "epoch": 5.34506905271348,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002531999131246176,
      "loss": 4.7298,
      "mean_token_accuracy": 0.2282811611890793,
      "num_tokens": 119164514.0,
      "step": 68700
    },
    {
      "entropy": 5.454681062698365,
      "epoch": 5.345458082085197,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002531723280996687,
      "loss": 4.8509,
      "mean_token_accuracy": 0.21593959778547286,
      "num_tokens": 119173449.0,
      "step": 68705
    },
    {
      "entropy": 5.432226467132568,
      "epoch": 5.345847111456915,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00025314474340592,
      "loss": 4.719,
      "mean_token_accuracy": 0.2242790177464485,
      "num_tokens": 119182390.0,
      "step": 68710
    },
    {
      "entropy": 5.451701068878174,
      "epoch": 5.3462361408286325,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025311715904379004,
      "loss": 4.7082,
      "mean_token_accuracy": 0.23050553798675538,
      "num_tokens": 119191595.0,
      "step": 68715
    },
    {
      "entropy": 5.378251075744629,
      "epoch": 5.34662517020035,
      "grad_norm": 1.1875,
      "learning_rate": 0.00025308957501369737,
      "loss": 4.6821,
      "mean_token_accuracy": 0.23130507320165633,
      "num_tokens": 119200678.0,
      "step": 68720
    },
    {
      "entropy": 5.400723600387574,
      "epoch": 5.347014199572068,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00025306199131606055,
      "loss": 4.7175,
      "mean_token_accuracy": 0.23373725712299348,
      "num_tokens": 119208916.0,
      "step": 68725
    },
    {
      "entropy": 5.426167583465576,
      "epoch": 5.347403228943786,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002530344079512981,
      "loss": 4.7932,
      "mean_token_accuracy": 0.21691125333309175,
      "num_tokens": 119218013.0,
      "step": 68730
    },
    {
      "entropy": 5.373668384552002,
      "epoch": 5.347792258315502,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025300682491982853,
      "loss": 4.67,
      "mean_token_accuracy": 0.23482376635074614,
      "num_tokens": 119226419.0,
      "step": 68735
    },
    {
      "entropy": 5.237364816665649,
      "epoch": 5.34818128768722,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002529792422220704,
      "loss": 4.5112,
      "mean_token_accuracy": 0.24689252227544783,
      "num_tokens": 119234979.0,
      "step": 68740
    },
    {
      "entropy": 5.407399415969849,
      "epoch": 5.348570317058938,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025295165985844215,
      "loss": 4.7603,
      "mean_token_accuracy": 0.23636563420295714,
      "num_tokens": 119243873.0,
      "step": 68745
    },
    {
      "entropy": 5.415172910690307,
      "epoch": 5.3489593464306555,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002529240778293624,
      "loss": 4.6909,
      "mean_token_accuracy": 0.23319346606731414,
      "num_tokens": 119252933.0,
      "step": 68750
    },
    {
      "entropy": 5.320423221588134,
      "epoch": 5.349348375802373,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025289649613524954,
      "loss": 4.564,
      "mean_token_accuracy": 0.24214782565832138,
      "num_tokens": 119261820.0,
      "step": 68755
    },
    {
      "entropy": 5.407308673858642,
      "epoch": 5.349737405174091,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002528689147765223,
      "loss": 4.785,
      "mean_token_accuracy": 0.21885529160499573,
      "num_tokens": 119270345.0,
      "step": 68760
    },
    {
      "entropy": 5.359142827987671,
      "epoch": 5.350126434545809,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025284133375359904,
      "loss": 4.6447,
      "mean_token_accuracy": 0.24187510311603547,
      "num_tokens": 119278093.0,
      "step": 68765
    },
    {
      "entropy": 5.424220085144043,
      "epoch": 5.350515463917525,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025281375306689824,
      "loss": 4.7757,
      "mean_token_accuracy": 0.2207789897918701,
      "num_tokens": 119286368.0,
      "step": 68770
    },
    {
      "entropy": 5.378355216979981,
      "epoch": 5.350904493289243,
      "grad_norm": 1.21875,
      "learning_rate": 0.00025278617271683846,
      "loss": 4.6811,
      "mean_token_accuracy": 0.22932713627815246,
      "num_tokens": 119295024.0,
      "step": 68775
    },
    {
      "entropy": 5.333453130722046,
      "epoch": 5.351293522660961,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002527585927038381,
      "loss": 4.6172,
      "mean_token_accuracy": 0.24416085481643676,
      "num_tokens": 119304120.0,
      "step": 68780
    },
    {
      "entropy": 5.338000679016114,
      "epoch": 5.3516825520326785,
      "grad_norm": 1.203125,
      "learning_rate": 0.00025273101302831575,
      "loss": 4.733,
      "mean_token_accuracy": 0.23005841225385665,
      "num_tokens": 119313528.0,
      "step": 68785
    },
    {
      "entropy": 5.370725393295288,
      "epoch": 5.352071581404396,
      "grad_norm": 1.125,
      "learning_rate": 0.00025270343369068976,
      "loss": 4.6786,
      "mean_token_accuracy": 0.23533975780010224,
      "num_tokens": 119322204.0,
      "step": 68790
    },
    {
      "entropy": 5.477041816711425,
      "epoch": 5.352460610776114,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002526758546913787,
      "loss": 4.8192,
      "mean_token_accuracy": 0.2263139456510544,
      "num_tokens": 119332060.0,
      "step": 68795
    },
    {
      "entropy": 5.309518241882325,
      "epoch": 5.352849640147831,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002526482760308011,
      "loss": 4.5299,
      "mean_token_accuracy": 0.24503831267356874,
      "num_tokens": 119340494.0,
      "step": 68800
    },
    {
      "entropy": 5.362177228927612,
      "epoch": 5.353238669519548,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025262069770937525,
      "loss": 4.7862,
      "mean_token_accuracy": 0.23022763282060624,
      "num_tokens": 119349859.0,
      "step": 68805
    },
    {
      "entropy": 5.403043127059936,
      "epoch": 5.353627698891266,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002525931197275197,
      "loss": 4.7801,
      "mean_token_accuracy": 0.225228151679039,
      "num_tokens": 119358775.0,
      "step": 68810
    },
    {
      "entropy": 5.440681838989258,
      "epoch": 5.354016728262984,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025256554208565296,
      "loss": 4.699,
      "mean_token_accuracy": 0.2298225149512291,
      "num_tokens": 119367319.0,
      "step": 68815
    },
    {
      "entropy": 5.374055051803589,
      "epoch": 5.3544057576347015,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00025253796478419323,
      "loss": 4.6068,
      "mean_token_accuracy": 0.24403384625911712,
      "num_tokens": 119376087.0,
      "step": 68820
    },
    {
      "entropy": 5.295899868011475,
      "epoch": 5.354794787006419,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002525103878235593,
      "loss": 4.5875,
      "mean_token_accuracy": 0.23225358128547668,
      "num_tokens": 119384552.0,
      "step": 68825
    },
    {
      "entropy": 5.394383239746094,
      "epoch": 5.355183816378137,
      "grad_norm": 1.25,
      "learning_rate": 0.0002524828112041694,
      "loss": 4.7081,
      "mean_token_accuracy": 0.2345711037516594,
      "num_tokens": 119392918.0,
      "step": 68830
    },
    {
      "entropy": 5.5039954662323,
      "epoch": 5.355572845749854,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00025245523492644197,
      "loss": 4.7922,
      "mean_token_accuracy": 0.23035633265972139,
      "num_tokens": 119401598.0,
      "step": 68835
    },
    {
      "entropy": 5.389233779907227,
      "epoch": 5.355961875121571,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002524276589907955,
      "loss": 4.7023,
      "mean_token_accuracy": 0.2388235807418823,
      "num_tokens": 119410576.0,
      "step": 68840
    },
    {
      "entropy": 5.404110670089722,
      "epoch": 5.356350904493289,
      "grad_norm": 1.203125,
      "learning_rate": 0.00025240008339764844,
      "loss": 4.779,
      "mean_token_accuracy": 0.23233670443296434,
      "num_tokens": 119419509.0,
      "step": 68845
    },
    {
      "entropy": 5.395064926147461,
      "epoch": 5.356739933865007,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000252372508147419,
      "loss": 4.6957,
      "mean_token_accuracy": 0.22916206568479539,
      "num_tokens": 119427505.0,
      "step": 68850
    },
    {
      "entropy": 5.397454023361206,
      "epoch": 5.3571289632367245,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002523449332405258,
      "loss": 4.6915,
      "mean_token_accuracy": 0.23323444724082948,
      "num_tokens": 119435637.0,
      "step": 68855
    },
    {
      "entropy": 5.443812561035156,
      "epoch": 5.357517992608442,
      "grad_norm": 1.25,
      "learning_rate": 0.0002523173586773872,
      "loss": 4.7011,
      "mean_token_accuracy": 0.22972651869058608,
      "num_tokens": 119443277.0,
      "step": 68860
    },
    {
      "entropy": 5.432233238220215,
      "epoch": 5.35790702198016,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002522897844584216,
      "loss": 4.7501,
      "mean_token_accuracy": 0.2290879711508751,
      "num_tokens": 119450623.0,
      "step": 68865
    },
    {
      "entropy": 5.395141363143921,
      "epoch": 5.358296051351877,
      "grad_norm": 1.25,
      "learning_rate": 0.00025226221058404733,
      "loss": 4.8018,
      "mean_token_accuracy": 0.22612121999263762,
      "num_tokens": 119459175.0,
      "step": 68870
    },
    {
      "entropy": 5.3890715599060055,
      "epoch": 5.358685080723594,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002522346370546828,
      "loss": 4.6834,
      "mean_token_accuracy": 0.23337047100067138,
      "num_tokens": 119468100.0,
      "step": 68875
    },
    {
      "entropy": 5.422239208221436,
      "epoch": 5.359074110095312,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002522070638707465,
      "loss": 4.7585,
      "mean_token_accuracy": 0.22786082923412324,
      "num_tokens": 119477097.0,
      "step": 68880
    },
    {
      "entropy": 5.3970122814178465,
      "epoch": 5.35946313946703,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025217949103265665,
      "loss": 4.6614,
      "mean_token_accuracy": 0.2400321453809738,
      "num_tokens": 119485197.0,
      "step": 68885
    },
    {
      "entropy": 5.377818441390991,
      "epoch": 5.3598521688387475,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00025215191854083166,
      "loss": 4.6636,
      "mean_token_accuracy": 0.24129363894462585,
      "num_tokens": 119495207.0,
      "step": 68890
    },
    {
      "entropy": 5.506122016906739,
      "epoch": 5.360241198210465,
      "grad_norm": 1.28125,
      "learning_rate": 0.00025212434639568996,
      "loss": 4.8191,
      "mean_token_accuracy": 0.22204598635435105,
      "num_tokens": 119503630.0,
      "step": 68895
    },
    {
      "entropy": 5.463771820068359,
      "epoch": 5.360630227582183,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00025209677459765,
      "loss": 4.7085,
      "mean_token_accuracy": 0.22663309276103974,
      "num_tokens": 119512425.0,
      "step": 68900
    },
    {
      "entropy": 5.436154365539551,
      "epoch": 5.3610192569539,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002520692031471299,
      "loss": 4.7858,
      "mean_token_accuracy": 0.2295249342918396,
      "num_tokens": 119521318.0,
      "step": 68905
    },
    {
      "entropy": 5.385901927947998,
      "epoch": 5.361408286325617,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025204163204454826,
      "loss": 4.6681,
      "mean_token_accuracy": 0.23272046595811843,
      "num_tokens": 119529953.0,
      "step": 68910
    },
    {
      "entropy": 5.4182476043701175,
      "epoch": 5.361797315697335,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002520140612903233,
      "loss": 4.784,
      "mean_token_accuracy": 0.22548335194587707,
      "num_tokens": 119539071.0,
      "step": 68915
    },
    {
      "entropy": 5.415350008010864,
      "epoch": 5.362186345069053,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002519864908848733,
      "loss": 4.6856,
      "mean_token_accuracy": 0.2372490480542183,
      "num_tokens": 119547548.0,
      "step": 68920
    },
    {
      "entropy": 5.363909196853638,
      "epoch": 5.3625753744407705,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025195892082861674,
      "loss": 4.6459,
      "mean_token_accuracy": 0.23845205157995225,
      "num_tokens": 119556113.0,
      "step": 68925
    },
    {
      "entropy": 5.4693207263946535,
      "epoch": 5.362964403812488,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002519313511219719,
      "loss": 4.8087,
      "mean_token_accuracy": 0.22240106910467147,
      "num_tokens": 119565510.0,
      "step": 68930
    },
    {
      "entropy": 5.382143211364746,
      "epoch": 5.363353433184205,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025190378176535703,
      "loss": 4.6991,
      "mean_token_accuracy": 0.233941014111042,
      "num_tokens": 119574433.0,
      "step": 68935
    },
    {
      "entropy": 5.473196649551392,
      "epoch": 5.363742462555923,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002518762127591905,
      "loss": 4.7106,
      "mean_token_accuracy": 0.2276538997888565,
      "num_tokens": 119582227.0,
      "step": 68940
    },
    {
      "entropy": 5.42279200553894,
      "epoch": 5.36413149192764,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002518486441038907,
      "loss": 4.723,
      "mean_token_accuracy": 0.22491416186094285,
      "num_tokens": 119590875.0,
      "step": 68945
    },
    {
      "entropy": 5.374335145950317,
      "epoch": 5.364520521299358,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002518210757998758,
      "loss": 4.7544,
      "mean_token_accuracy": 0.22617409825325013,
      "num_tokens": 119599306.0,
      "step": 68950
    },
    {
      "entropy": 5.450578784942627,
      "epoch": 5.364909550671076,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002517935078475642,
      "loss": 4.8067,
      "mean_token_accuracy": 0.22319098562002182,
      "num_tokens": 119608111.0,
      "step": 68955
    },
    {
      "entropy": 5.46391077041626,
      "epoch": 5.3652985800427935,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002517659402473742,
      "loss": 4.706,
      "mean_token_accuracy": 0.2312758147716522,
      "num_tokens": 119616559.0,
      "step": 68960
    },
    {
      "entropy": 5.415202379226685,
      "epoch": 5.365687609414511,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002517383729997241,
      "loss": 4.7217,
      "mean_token_accuracy": 0.22680216729640962,
      "num_tokens": 119626256.0,
      "step": 68965
    },
    {
      "entropy": 5.361013650894165,
      "epoch": 5.366076638786228,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002517108061050321,
      "loss": 4.7107,
      "mean_token_accuracy": 0.2294347271323204,
      "num_tokens": 119634716.0,
      "step": 68970
    },
    {
      "entropy": 5.462832736968994,
      "epoch": 5.366465668157946,
      "grad_norm": 1.375,
      "learning_rate": 0.00025168323956371657,
      "loss": 4.7711,
      "mean_token_accuracy": 0.2257754847407341,
      "num_tokens": 119642869.0,
      "step": 68975
    },
    {
      "entropy": 5.400960254669189,
      "epoch": 5.366854697529663,
      "grad_norm": 1.1875,
      "learning_rate": 0.00025165567337619583,
      "loss": 4.6587,
      "mean_token_accuracy": 0.23154679238796233,
      "num_tokens": 119651715.0,
      "step": 68980
    },
    {
      "entropy": 5.362062406539917,
      "epoch": 5.367243726901381,
      "grad_norm": 1.265625,
      "learning_rate": 0.000251628107542888,
      "loss": 4.6333,
      "mean_token_accuracy": 0.2406027004122734,
      "num_tokens": 119659782.0,
      "step": 68985
    },
    {
      "entropy": 5.43506350517273,
      "epoch": 5.367632756273099,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025160054206421143,
      "loss": 4.8085,
      "mean_token_accuracy": 0.22444754242897033,
      "num_tokens": 119668444.0,
      "step": 68990
    },
    {
      "entropy": 5.484446811676025,
      "epoch": 5.3680217856448165,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002515729769405845,
      "loss": 4.7528,
      "mean_token_accuracy": 0.2260366603732109,
      "num_tokens": 119677011.0,
      "step": 68995
    },
    {
      "entropy": 5.39534101486206,
      "epoch": 5.368410815016533,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002515454121724253,
      "loss": 4.7168,
      "mean_token_accuracy": 0.2317987471818924,
      "num_tokens": 119685398.0,
      "step": 69000
    },
    {
      "epoch": 5.368410815016533,
      "eval_entropy": 5.118699841274449,
      "eval_loss": 4.900583267211914,
      "eval_mean_token_accuracy": 0.2297408799665036,
      "eval_num_tokens": 119685398.0,
      "eval_runtime": 28.7309,
      "eval_samples_per_second": 1446.456,
      "eval_steps_per_second": 180.816,
      "step": 69000
    },
    {
      "entropy": 5.423506355285644,
      "epoch": 5.368799844388251,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00025151784776015216,
      "loss": 4.808,
      "mean_token_accuracy": 0.22472999691963197,
      "num_tokens": 119694815.0,
      "step": 69005
    },
    {
      "entropy": 5.427844572067261,
      "epoch": 5.369188873759969,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002514902837041833,
      "loss": 4.6851,
      "mean_token_accuracy": 0.23454775363206865,
      "num_tokens": 119703366.0,
      "step": 69010
    },
    {
      "entropy": 5.327227067947388,
      "epoch": 5.369577903131686,
      "grad_norm": 1.328125,
      "learning_rate": 0.00025146272000493695,
      "loss": 4.6433,
      "mean_token_accuracy": 0.23572032451629638,
      "num_tokens": 119711784.0,
      "step": 69015
    },
    {
      "entropy": 5.393718004226685,
      "epoch": 5.369966932503404,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002514351566628313,
      "loss": 4.7478,
      "mean_token_accuracy": 0.2223072960972786,
      "num_tokens": 119720579.0,
      "step": 69020
    },
    {
      "entropy": 5.4213700771331785,
      "epoch": 5.370355961875122,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002514075936782847,
      "loss": 4.7315,
      "mean_token_accuracy": 0.23256743848323821,
      "num_tokens": 119728512.0,
      "step": 69025
    },
    {
      "entropy": 5.456785249710083,
      "epoch": 5.3707449912468395,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002513800310517153,
      "loss": 4.7557,
      "mean_token_accuracy": 0.2401050701737404,
      "num_tokens": 119738867.0,
      "step": 69030
    },
    {
      "entropy": 5.470073652267456,
      "epoch": 5.371134020618557,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002513524687835414,
      "loss": 4.8383,
      "mean_token_accuracy": 0.22278754860162736,
      "num_tokens": 119748363.0,
      "step": 69035
    },
    {
      "entropy": 5.423344373703003,
      "epoch": 5.371523049990274,
      "grad_norm": 1.25,
      "learning_rate": 0.0002513249068741811,
      "loss": 4.7159,
      "mean_token_accuracy": 0.2281084507703781,
      "num_tokens": 119757032.0,
      "step": 69040
    },
    {
      "entropy": 5.399194622039795,
      "epoch": 5.371912079361992,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025129734532405265,
      "loss": 4.7174,
      "mean_token_accuracy": 0.23328703939914702,
      "num_tokens": 119766137.0,
      "step": 69045
    },
    {
      "entropy": 5.388381433486939,
      "epoch": 5.372301108733709,
      "grad_norm": 1.28125,
      "learning_rate": 0.00025126978413357424,
      "loss": 4.627,
      "mean_token_accuracy": 0.2414829507470131,
      "num_tokens": 119775487.0,
      "step": 69050
    },
    {
      "entropy": 5.3747858047485355,
      "epoch": 5.372690138105427,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002512422233031641,
      "loss": 4.7198,
      "mean_token_accuracy": 0.2361586272716522,
      "num_tokens": 119784398.0,
      "step": 69055
    },
    {
      "entropy": 5.377015113830566,
      "epoch": 5.373079167477145,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002512146628332404,
      "loss": 4.735,
      "mean_token_accuracy": 0.2317363739013672,
      "num_tokens": 119793249.0,
      "step": 69060
    },
    {
      "entropy": 5.398827743530274,
      "epoch": 5.3734681968488625,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002511871027242214,
      "loss": 4.7104,
      "mean_token_accuracy": 0.23233886659145356,
      "num_tokens": 119802513.0,
      "step": 69065
    },
    {
      "entropy": 5.458522653579712,
      "epoch": 5.373857226220579,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002511595429765251,
      "loss": 4.7209,
      "mean_token_accuracy": 0.23024237006902695,
      "num_tokens": 119811318.0,
      "step": 69070
    },
    {
      "entropy": 5.447855615615845,
      "epoch": 5.374246255592297,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025113198359056986,
      "loss": 4.6898,
      "mean_token_accuracy": 0.23032086938619614,
      "num_tokens": 119820057.0,
      "step": 69075
    },
    {
      "entropy": 5.4283051013946535,
      "epoch": 5.374635284964015,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002511044245667738,
      "loss": 4.6899,
      "mean_token_accuracy": 0.24170751869678497,
      "num_tokens": 119828931.0,
      "step": 69080
    },
    {
      "entropy": 5.43586802482605,
      "epoch": 5.375024314335732,
      "grad_norm": 1.171875,
      "learning_rate": 0.000251076865905555,
      "loss": 4.6954,
      "mean_token_accuracy": 0.2353301003575325,
      "num_tokens": 119837511.0,
      "step": 69085
    },
    {
      "entropy": 5.293330335617066,
      "epoch": 5.37541334370745,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00025104930760733163,
      "loss": 4.6154,
      "mean_token_accuracy": 0.23397128283977509,
      "num_tokens": 119845236.0,
      "step": 69090
    },
    {
      "entropy": 5.406250715255737,
      "epoch": 5.375802373079168,
      "grad_norm": 1.203125,
      "learning_rate": 0.000251021749672522,
      "loss": 4.7218,
      "mean_token_accuracy": 0.23249758034944534,
      "num_tokens": 119854465.0,
      "step": 69095
    },
    {
      "entropy": 5.5244039535522464,
      "epoch": 5.3761914024508854,
      "grad_norm": 1.09375,
      "learning_rate": 0.00025099419210154413,
      "loss": 4.774,
      "mean_token_accuracy": 0.22775956094264985,
      "num_tokens": 119863336.0,
      "step": 69100
    },
    {
      "entropy": 5.412106275558472,
      "epoch": 5.376580431822602,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002509666348948162,
      "loss": 4.7438,
      "mean_token_accuracy": 0.23257628828287125,
      "num_tokens": 119872394.0,
      "step": 69105
    },
    {
      "entropy": 5.374779891967774,
      "epoch": 5.37696946119432,
      "grad_norm": 1.34375,
      "learning_rate": 0.00025093907805275636,
      "loss": 4.7154,
      "mean_token_accuracy": 0.22919257432222367,
      "num_tokens": 119881663.0,
      "step": 69110
    },
    {
      "entropy": 5.317596101760865,
      "epoch": 5.377358490566038,
      "grad_norm": 1.171875,
      "learning_rate": 0.00025091152157578266,
      "loss": 4.619,
      "mean_token_accuracy": 0.23752924352884291,
      "num_tokens": 119891087.0,
      "step": 69115
    },
    {
      "entropy": 5.386262321472168,
      "epoch": 5.377747519937755,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002508839654643133,
      "loss": 4.6889,
      "mean_token_accuracy": 0.234139022231102,
      "num_tokens": 119899224.0,
      "step": 69120
    },
    {
      "entropy": 5.481901454925537,
      "epoch": 5.378136549309473,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002508564097187664,
      "loss": 4.7914,
      "mean_token_accuracy": 0.22064066380262376,
      "num_tokens": 119908520.0,
      "step": 69125
    },
    {
      "entropy": 5.468776512145996,
      "epoch": 5.378525578681191,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025082885433956007,
      "loss": 4.7747,
      "mean_token_accuracy": 0.22460367232561113,
      "num_tokens": 119917034.0,
      "step": 69130
    },
    {
      "entropy": 5.428314638137818,
      "epoch": 5.3789146080529076,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025080129932711234,
      "loss": 4.6752,
      "mean_token_accuracy": 0.23307447582483293,
      "num_tokens": 119925518.0,
      "step": 69135
    },
    {
      "entropy": 5.474184989929199,
      "epoch": 5.379303637424625,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002507737446818414,
      "loss": 4.7982,
      "mean_token_accuracy": 0.2275997966527939,
      "num_tokens": 119934280.0,
      "step": 69140
    },
    {
      "entropy": 5.301259613037109,
      "epoch": 5.379692666796343,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002507461904041653,
      "loss": 4.602,
      "mean_token_accuracy": 0.24738617688417436,
      "num_tokens": 119941847.0,
      "step": 69145
    },
    {
      "entropy": 5.402604341506958,
      "epoch": 5.380081696168061,
      "grad_norm": 1.25,
      "learning_rate": 0.00025071863649450216,
      "loss": 4.7578,
      "mean_token_accuracy": 0.23561530262231828,
      "num_tokens": 119950481.0,
      "step": 69150
    },
    {
      "entropy": 5.36326847076416,
      "epoch": 5.380470725539778,
      "grad_norm": 1.21875,
      "learning_rate": 0.00025069108295327006,
      "loss": 4.6007,
      "mean_token_accuracy": 0.2424105942249298,
      "num_tokens": 119959050.0,
      "step": 69155
    },
    {
      "entropy": 5.397031307220459,
      "epoch": 5.380859754911496,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025066352978088706,
      "loss": 4.6603,
      "mean_token_accuracy": 0.23416541665792465,
      "num_tokens": 119967827.0,
      "step": 69160
    },
    {
      "entropy": 5.358386039733887,
      "epoch": 5.381248784283214,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002506359769777713,
      "loss": 4.6522,
      "mean_token_accuracy": 0.2340443640947342,
      "num_tokens": 119975979.0,
      "step": 69165
    },
    {
      "entropy": 5.411128950119019,
      "epoch": 5.3816378136549305,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002506084245443408,
      "loss": 4.7216,
      "mean_token_accuracy": 0.23196171522140502,
      "num_tokens": 119984241.0,
      "step": 69170
    },
    {
      "entropy": 5.463791227340698,
      "epoch": 5.382026843026648,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002505808724810137,
      "loss": 4.7862,
      "mean_token_accuracy": 0.22687575370073318,
      "num_tokens": 119992728.0,
      "step": 69175
    },
    {
      "entropy": 5.399555015563965,
      "epoch": 5.382415872398366,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00025055332078820794,
      "loss": 4.6779,
      "mean_token_accuracy": 0.22688824385404588,
      "num_tokens": 120001417.0,
      "step": 69180
    },
    {
      "entropy": 5.493369770050049,
      "epoch": 5.382804901770084,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002505257694663416,
      "loss": 4.8086,
      "mean_token_accuracy": 0.23114608228206635,
      "num_tokens": 120010086.0,
      "step": 69185
    },
    {
      "entropy": 5.43845009803772,
      "epoch": 5.383193931141801,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002504982185158328,
      "loss": 4.7321,
      "mean_token_accuracy": 0.22903282940387726,
      "num_tokens": 120018854.0,
      "step": 69190
    },
    {
      "entropy": 5.459960699081421,
      "epoch": 5.383582960513519,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002504706679370995,
      "loss": 4.7208,
      "mean_token_accuracy": 0.23157539665699006,
      "num_tokens": 120027215.0,
      "step": 69195
    },
    {
      "entropy": 5.421270799636841,
      "epoch": 5.383971989885237,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002504431177305598,
      "loss": 4.6995,
      "mean_token_accuracy": 0.2412053257226944,
      "num_tokens": 120036251.0,
      "step": 69200
    },
    {
      "entropy": 5.393804502487183,
      "epoch": 5.3843610192569535,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002504155678966317,
      "loss": 4.6964,
      "mean_token_accuracy": 0.23098417222499848,
      "num_tokens": 120045453.0,
      "step": 69205
    },
    {
      "entropy": 5.380233955383301,
      "epoch": 5.384750048628671,
      "grad_norm": 1.25,
      "learning_rate": 0.0002503880184357333,
      "loss": 4.6639,
      "mean_token_accuracy": 0.23818280100822448,
      "num_tokens": 120055155.0,
      "step": 69210
    },
    {
      "entropy": 5.506730270385742,
      "epoch": 5.385139078000389,
      "grad_norm": 1.25,
      "learning_rate": 0.00025036046934828244,
      "loss": 4.8387,
      "mean_token_accuracy": 0.22549087107181548,
      "num_tokens": 120063642.0,
      "step": 69215
    },
    {
      "entropy": 5.469258117675781,
      "epoch": 5.385528107372107,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025033292063469714,
      "loss": 4.7285,
      "mean_token_accuracy": 0.22771794497966766,
      "num_tokens": 120072983.0,
      "step": 69220
    },
    {
      "entropy": 5.368207216262817,
      "epoch": 5.385917136743824,
      "grad_norm": 1.21875,
      "learning_rate": 0.00025030537229539567,
      "loss": 4.672,
      "mean_token_accuracy": 0.2337014004588127,
      "num_tokens": 120081382.0,
      "step": 69225
    },
    {
      "entropy": 5.437441205978393,
      "epoch": 5.386306166115542,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002502778243307958,
      "loss": 4.7096,
      "mean_token_accuracy": 0.23622626066207886,
      "num_tokens": 120089954.0,
      "step": 69230
    },
    {
      "entropy": 5.376938676834106,
      "epoch": 5.38669519548726,
      "grad_norm": 1.265625,
      "learning_rate": 0.00025025027674131567,
      "loss": 4.7538,
      "mean_token_accuracy": 0.23352345526218415,
      "num_tokens": 120098409.0,
      "step": 69235
    },
    {
      "entropy": 5.335484790802002,
      "epoch": 5.3870842248589765,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00025022272952737317,
      "loss": 4.6285,
      "mean_token_accuracy": 0.24056220352649688,
      "num_tokens": 120106554.0,
      "step": 69240
    },
    {
      "entropy": 5.3137535572052,
      "epoch": 5.387473254230694,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002501951826893863,
      "loss": 4.5644,
      "mean_token_accuracy": 0.25208657532930373,
      "num_tokens": 120116296.0,
      "step": 69245
    },
    {
      "entropy": 5.3971131324768065,
      "epoch": 5.387862283602412,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002501676362277731,
      "loss": 4.7464,
      "mean_token_accuracy": 0.23212269991636275,
      "num_tokens": 120126000.0,
      "step": 69250
    },
    {
      "entropy": 5.3636579513549805,
      "epoch": 5.38825131297413,
      "grad_norm": 1.265625,
      "learning_rate": 0.00025014009014295136,
      "loss": 4.7602,
      "mean_token_accuracy": 0.22827396541833878,
      "num_tokens": 120134869.0,
      "step": 69255
    },
    {
      "entropy": 5.33865180015564,
      "epoch": 5.388640342345847,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002501125444353393,
      "loss": 4.6616,
      "mean_token_accuracy": 0.23312715142965318,
      "num_tokens": 120144073.0,
      "step": 69260
    },
    {
      "entropy": 5.494708681106568,
      "epoch": 5.389029371717565,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002500849991053547,
      "loss": 4.7817,
      "mean_token_accuracy": 0.2263398826122284,
      "num_tokens": 120151561.0,
      "step": 69265
    },
    {
      "entropy": 5.46658182144165,
      "epoch": 5.389418401089282,
      "grad_norm": 1.171875,
      "learning_rate": 0.00025005745415341565,
      "loss": 4.8065,
      "mean_token_accuracy": 0.22815714478492738,
      "num_tokens": 120160915.0,
      "step": 69270
    },
    {
      "entropy": 5.4578447341918945,
      "epoch": 5.3898074304609995,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025002990957994,
      "loss": 4.7176,
      "mean_token_accuracy": 0.23403708040714263,
      "num_tokens": 120168723.0,
      "step": 69275
    },
    {
      "entropy": 5.313422775268554,
      "epoch": 5.390196459832717,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025000236538534577,
      "loss": 4.623,
      "mean_token_accuracy": 0.23663442134857177,
      "num_tokens": 120176804.0,
      "step": 69280
    },
    {
      "entropy": 5.448632860183716,
      "epoch": 5.390585489204435,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024997482157005074,
      "loss": 4.8243,
      "mean_token_accuracy": 0.22312912940979004,
      "num_tokens": 120185618.0,
      "step": 69285
    },
    {
      "entropy": 5.484459733963012,
      "epoch": 5.390974518576153,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000249947278134473,
      "loss": 4.8344,
      "mean_token_accuracy": 0.2237105891108513,
      "num_tokens": 120194303.0,
      "step": 69290
    },
    {
      "entropy": 5.45878963470459,
      "epoch": 5.39136354794787,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002499197350790305,
      "loss": 4.7803,
      "mean_token_accuracy": 0.2273254707455635,
      "num_tokens": 120202700.0,
      "step": 69295
    },
    {
      "entropy": 5.427468204498291,
      "epoch": 5.391752577319588,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024989219240414105,
      "loss": 4.7284,
      "mean_token_accuracy": 0.23186143040657042,
      "num_tokens": 120211028.0,
      "step": 69300
    },
    {
      "entropy": 5.345390653610229,
      "epoch": 5.392141606691305,
      "grad_norm": 1.25,
      "learning_rate": 0.00024986465011022263,
      "loss": 4.5625,
      "mean_token_accuracy": 0.2510059863328934,
      "num_tokens": 120219103.0,
      "step": 69305
    },
    {
      "entropy": 5.377266502380371,
      "epoch": 5.3925306360630225,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00024983710819769325,
      "loss": 4.7347,
      "mean_token_accuracy": 0.23217184394598006,
      "num_tokens": 120227910.0,
      "step": 69310
    },
    {
      "entropy": 5.433395051956177,
      "epoch": 5.39291966543474,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002498095666669706,
      "loss": 4.7031,
      "mean_token_accuracy": 0.23660168498754502,
      "num_tokens": 120236514.0,
      "step": 69315
    },
    {
      "entropy": 5.370156145095825,
      "epoch": 5.393308694806458,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002497820255184727,
      "loss": 4.6981,
      "mean_token_accuracy": 0.22830221503973008,
      "num_tokens": 120246073.0,
      "step": 69320
    },
    {
      "entropy": 5.451698732376099,
      "epoch": 5.393697724178176,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024975448475261737,
      "loss": 4.7788,
      "mean_token_accuracy": 0.22433886379003526,
      "num_tokens": 120254836.0,
      "step": 69325
    },
    {
      "entropy": 5.460968780517578,
      "epoch": 5.394086753549893,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00024972694436982255,
      "loss": 4.7949,
      "mean_token_accuracy": 0.2321621671319008,
      "num_tokens": 120264104.0,
      "step": 69330
    },
    {
      "entropy": 5.452217960357666,
      "epoch": 5.39447578292161,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002496994043705062,
      "loss": 4.7235,
      "mean_token_accuracy": 0.23301835358142853,
      "num_tokens": 120273021.0,
      "step": 69335
    },
    {
      "entropy": 5.388124084472656,
      "epoch": 5.394864812293328,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002496718647550861,
      "loss": 4.7161,
      "mean_token_accuracy": 0.22977873533964158,
      "num_tokens": 120281237.0,
      "step": 69340
    },
    {
      "entropy": 5.401782703399658,
      "epoch": 5.3952538416650455,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002496443255239801,
      "loss": 4.7364,
      "mean_token_accuracy": 0.23282259553670884,
      "num_tokens": 120289597.0,
      "step": 69345
    },
    {
      "entropy": 5.4439377784729,
      "epoch": 5.395642871036763,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002496167866776061,
      "loss": 4.7044,
      "mean_token_accuracy": 0.23319983333349228,
      "num_tokens": 120297451.0,
      "step": 69350
    },
    {
      "entropy": 5.388371706008911,
      "epoch": 5.396031900408481,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024958924821638195,
      "loss": 4.6613,
      "mean_token_accuracy": 0.23830080181360244,
      "num_tokens": 120306524.0,
      "step": 69355
    },
    {
      "entropy": 5.463131237030029,
      "epoch": 5.396420929780199,
      "grad_norm": 1.171875,
      "learning_rate": 0.00024956171014072555,
      "loss": 4.7498,
      "mean_token_accuracy": 0.22492137551307678,
      "num_tokens": 120315299.0,
      "step": 69360
    },
    {
      "entropy": 5.4397077560424805,
      "epoch": 5.396809959151916,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024953417245105467,
      "loss": 4.7659,
      "mean_token_accuracy": 0.22942063063383103,
      "num_tokens": 120323591.0,
      "step": 69365
    },
    {
      "entropy": 5.3833495616912845,
      "epoch": 5.397198988523634,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00024950663514778726,
      "loss": 4.7391,
      "mean_token_accuracy": 0.2292529821395874,
      "num_tokens": 120333280.0,
      "step": 69370
    },
    {
      "entropy": 5.4590222358703615,
      "epoch": 5.397588017895351,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024947909823134107,
      "loss": 4.8118,
      "mean_token_accuracy": 0.2257261887192726,
      "num_tokens": 120342063.0,
      "step": 69375
    },
    {
      "entropy": 5.386949729919434,
      "epoch": 5.3979770472670685,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002494515617021339,
      "loss": 4.6713,
      "mean_token_accuracy": 0.23412439674139024,
      "num_tokens": 120350179.0,
      "step": 69380
    },
    {
      "entropy": 5.361656713485718,
      "epoch": 5.398366076638786,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024942402556058364,
      "loss": 4.684,
      "mean_token_accuracy": 0.23524031937122344,
      "num_tokens": 120358888.0,
      "step": 69385
    },
    {
      "entropy": 5.44440279006958,
      "epoch": 5.398755106010504,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002493964898071081,
      "loss": 4.722,
      "mean_token_accuracy": 0.2364642933011055,
      "num_tokens": 120367077.0,
      "step": 69390
    },
    {
      "entropy": 5.369498348236084,
      "epoch": 5.399144135382222,
      "grad_norm": 1.125,
      "learning_rate": 0.0002493689544421251,
      "loss": 4.646,
      "mean_token_accuracy": 0.2398667111992836,
      "num_tokens": 120375964.0,
      "step": 69395
    },
    {
      "entropy": 5.383589506149292,
      "epoch": 5.399533164753939,
      "grad_norm": 1.15625,
      "learning_rate": 0.00024934141946605237,
      "loss": 4.6591,
      "mean_token_accuracy": 0.23259367048740387,
      "num_tokens": 120384812.0,
      "step": 69400
    },
    {
      "entropy": 5.434037399291992,
      "epoch": 5.399922194125656,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024931388487930777,
      "loss": 4.788,
      "mean_token_accuracy": 0.22993735522031783,
      "num_tokens": 120393659.0,
      "step": 69405
    },
    {
      "entropy": 5.357564449310303,
      "epoch": 5.400311223497374,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000249286350682309,
      "loss": 4.7033,
      "mean_token_accuracy": 0.2283765435218811,
      "num_tokens": 120401876.0,
      "step": 69410
    },
    {
      "entropy": 5.451650524139405,
      "epoch": 5.4007002528690915,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002492588168754741,
      "loss": 4.8434,
      "mean_token_accuracy": 0.23228127360343934,
      "num_tokens": 120410129.0,
      "step": 69415
    },
    {
      "entropy": 5.395188903808593,
      "epoch": 5.401089282240809,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024923128345922055,
      "loss": 4.6827,
      "mean_token_accuracy": 0.23758058845996857,
      "num_tokens": 120419525.0,
      "step": 69420
    },
    {
      "entropy": 5.436212348937988,
      "epoch": 5.401478311612527,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002492037504339663,
      "loss": 4.6982,
      "mean_token_accuracy": 0.23145072758197785,
      "num_tokens": 120427808.0,
      "step": 69425
    },
    {
      "entropy": 5.394135332107544,
      "epoch": 5.401867340984245,
      "grad_norm": 1.15625,
      "learning_rate": 0.00024917621780012905,
      "loss": 4.6999,
      "mean_token_accuracy": 0.2338148057460785,
      "num_tokens": 120436737.0,
      "step": 69430
    },
    {
      "entropy": 5.397472333908081,
      "epoch": 5.402256370355962,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024914868555812663,
      "loss": 4.7911,
      "mean_token_accuracy": 0.22699290364980698,
      "num_tokens": 120445547.0,
      "step": 69435
    },
    {
      "entropy": 5.462872076034546,
      "epoch": 5.402645399727679,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002491211537083768,
      "loss": 4.8303,
      "mean_token_accuracy": 0.2173090770840645,
      "num_tokens": 120453659.0,
      "step": 69440
    },
    {
      "entropy": 5.379418468475341,
      "epoch": 5.403034429099397,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002490936222512972,
      "loss": 4.6488,
      "mean_token_accuracy": 0.23608698844909667,
      "num_tokens": 120462828.0,
      "step": 69445
    },
    {
      "entropy": 5.464342021942139,
      "epoch": 5.4034234584711145,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002490660911873057,
      "loss": 4.7201,
      "mean_token_accuracy": 0.22766397297382354,
      "num_tokens": 120470625.0,
      "step": 69450
    },
    {
      "entropy": 5.335444355010987,
      "epoch": 5.403812487842832,
      "grad_norm": 1.109375,
      "learning_rate": 0.00024903856051681995,
      "loss": 4.6311,
      "mean_token_accuracy": 0.23547475337982177,
      "num_tokens": 120479555.0,
      "step": 69455
    },
    {
      "entropy": 5.2983945369720455,
      "epoch": 5.40420151721455,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002490110302402577,
      "loss": 4.6096,
      "mean_token_accuracy": 0.2378025010228157,
      "num_tokens": 120486929.0,
      "step": 69460
    },
    {
      "entropy": 5.353688383102417,
      "epoch": 5.404590546586268,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002489835003580367,
      "loss": 4.6256,
      "mean_token_accuracy": 0.23577210903167725,
      "num_tokens": 120495757.0,
      "step": 69465
    },
    {
      "entropy": 5.399839782714844,
      "epoch": 5.404979575957984,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024895597087057483,
      "loss": 4.6645,
      "mean_token_accuracy": 0.23598668426275254,
      "num_tokens": 120503598.0,
      "step": 69470
    },
    {
      "entropy": 5.299967861175537,
      "epoch": 5.405368605329702,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002489284417782895,
      "loss": 4.5772,
      "mean_token_accuracy": 0.24173648208379744,
      "num_tokens": 120511656.0,
      "step": 69475
    },
    {
      "entropy": 5.343464994430542,
      "epoch": 5.40575763470142,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002489009130815986,
      "loss": 4.7058,
      "mean_token_accuracy": 0.22958036214113237,
      "num_tokens": 120520861.0,
      "step": 69480
    },
    {
      "entropy": 5.275725984573365,
      "epoch": 5.4061466640731375,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002488733847809197,
      "loss": 4.5853,
      "mean_token_accuracy": 0.24319858103990555,
      "num_tokens": 120529910.0,
      "step": 69485
    },
    {
      "entropy": 5.378789186477661,
      "epoch": 5.406535693444855,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002488458568766707,
      "loss": 4.67,
      "mean_token_accuracy": 0.23692481964826584,
      "num_tokens": 120538649.0,
      "step": 69490
    },
    {
      "entropy": 5.386489295959473,
      "epoch": 5.406924722816573,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002488183293692692,
      "loss": 4.7515,
      "mean_token_accuracy": 0.2358615905046463,
      "num_tokens": 120547649.0,
      "step": 69495
    },
    {
      "entropy": 5.377122592926026,
      "epoch": 5.407313752188291,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002487908022591329,
      "loss": 4.706,
      "mean_token_accuracy": 0.2325155183672905,
      "num_tokens": 120556043.0,
      "step": 69500
    },
    {
      "entropy": 5.391704130172729,
      "epoch": 5.407702781560007,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00024876327554667943,
      "loss": 4.7342,
      "mean_token_accuracy": 0.23460321724414826,
      "num_tokens": 120564735.0,
      "step": 69505
    },
    {
      "entropy": 5.436771106719971,
      "epoch": 5.408091810931725,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002487357492323265,
      "loss": 4.8042,
      "mean_token_accuracy": 0.22642077803611754,
      "num_tokens": 120574203.0,
      "step": 69510
    },
    {
      "entropy": 5.489454984664917,
      "epoch": 5.408480840303443,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002487082233164918,
      "loss": 4.794,
      "mean_token_accuracy": 0.23582003265619278,
      "num_tokens": 120583396.0,
      "step": 69515
    },
    {
      "entropy": 5.435397052764893,
      "epoch": 5.4088698696751605,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002486806977995929,
      "loss": 4.779,
      "mean_token_accuracy": 0.22473386228084563,
      "num_tokens": 120592431.0,
      "step": 69520
    },
    {
      "entropy": 5.443766784667969,
      "epoch": 5.409258899046878,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002486531726820476,
      "loss": 4.7119,
      "mean_token_accuracy": 0.23093476742506028,
      "num_tokens": 120602632.0,
      "step": 69525
    },
    {
      "entropy": 5.458692216873169,
      "epoch": 5.409647928418596,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024862564796427337,
      "loss": 4.836,
      "mean_token_accuracy": 0.21941891610622405,
      "num_tokens": 120611475.0,
      "step": 69530
    },
    {
      "entropy": 5.558376979827881,
      "epoch": 5.410036957790314,
      "grad_norm": 1.15625,
      "learning_rate": 0.000248598123646688,
      "loss": 4.8709,
      "mean_token_accuracy": 0.21736570298671723,
      "num_tokens": 120620754.0,
      "step": 69535
    },
    {
      "entropy": 5.53126482963562,
      "epoch": 5.41042598716203,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002485705997297091,
      "loss": 4.7227,
      "mean_token_accuracy": 0.22985261976718901,
      "num_tokens": 120628606.0,
      "step": 69540
    },
    {
      "entropy": 5.3579840660095215,
      "epoch": 5.410815016533748,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024854307621375427,
      "loss": 4.6421,
      "mean_token_accuracy": 0.23879548758268357,
      "num_tokens": 120636064.0,
      "step": 69545
    },
    {
      "entropy": 5.36305570602417,
      "epoch": 5.411204045905466,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002485155530992411,
      "loss": 4.7353,
      "mean_token_accuracy": 0.23055816292762757,
      "num_tokens": 120644805.0,
      "step": 69550
    },
    {
      "entropy": 5.393668222427368,
      "epoch": 5.4115930752771835,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002484880303865873,
      "loss": 4.6885,
      "mean_token_accuracy": 0.23358451426029206,
      "num_tokens": 120653331.0,
      "step": 69555
    },
    {
      "entropy": 5.405296277999878,
      "epoch": 5.411982104648901,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002484605080762104,
      "loss": 4.6987,
      "mean_token_accuracy": 0.2383778989315033,
      "num_tokens": 120661927.0,
      "step": 69560
    },
    {
      "entropy": 5.378438425064087,
      "epoch": 5.412371134020619,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002484329861685281,
      "loss": 4.7357,
      "mean_token_accuracy": 0.2447812184691429,
      "num_tokens": 120669586.0,
      "step": 69565
    },
    {
      "entropy": 5.383087825775147,
      "epoch": 5.412760163392337,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024840546466395796,
      "loss": 4.6686,
      "mean_token_accuracy": 0.23115828037261962,
      "num_tokens": 120678776.0,
      "step": 69570
    },
    {
      "entropy": 5.48199462890625,
      "epoch": 5.413149192764053,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024837794356291756,
      "loss": 4.8083,
      "mean_token_accuracy": 0.22595032453536987,
      "num_tokens": 120686938.0,
      "step": 69575
    },
    {
      "entropy": 5.544227552413941,
      "epoch": 5.413538222135771,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002483504228658245,
      "loss": 4.819,
      "mean_token_accuracy": 0.2253738060593605,
      "num_tokens": 120694816.0,
      "step": 69580
    },
    {
      "entropy": 5.348663854598999,
      "epoch": 5.413927251507489,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002483229025730963,
      "loss": 4.6623,
      "mean_token_accuracy": 0.2359575092792511,
      "num_tokens": 120704172.0,
      "step": 69585
    },
    {
      "entropy": 5.47953953742981,
      "epoch": 5.4143162808792065,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002482953826851506,
      "loss": 4.8531,
      "mean_token_accuracy": 0.22059032171964646,
      "num_tokens": 120712914.0,
      "step": 69590
    },
    {
      "entropy": 5.458735036849975,
      "epoch": 5.414705310250924,
      "grad_norm": 1.296875,
      "learning_rate": 0.000248267863202405,
      "loss": 4.7828,
      "mean_token_accuracy": 0.2275280773639679,
      "num_tokens": 120721703.0,
      "step": 69595
    },
    {
      "entropy": 5.443139362335205,
      "epoch": 5.415094339622642,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024824034412527704,
      "loss": 4.692,
      "mean_token_accuracy": 0.23016002625226975,
      "num_tokens": 120729841.0,
      "step": 69600
    },
    {
      "entropy": 5.399218511581421,
      "epoch": 5.415483368994359,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002482128254541842,
      "loss": 4.6911,
      "mean_token_accuracy": 0.22917818129062653,
      "num_tokens": 120737416.0,
      "step": 69605
    },
    {
      "entropy": 5.397141551971435,
      "epoch": 5.415872398366076,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002481853071895442,
      "loss": 4.6578,
      "mean_token_accuracy": 0.23408349901437758,
      "num_tokens": 120746618.0,
      "step": 69610
    },
    {
      "entropy": 5.332381725311279,
      "epoch": 5.416261427737794,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002481577893317744,
      "loss": 4.6408,
      "mean_token_accuracy": 0.2368878483772278,
      "num_tokens": 120755259.0,
      "step": 69615
    },
    {
      "entropy": 5.413399028778076,
      "epoch": 5.416650457109512,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002481302718812924,
      "loss": 4.7387,
      "mean_token_accuracy": 0.23045077174901962,
      "num_tokens": 120763986.0,
      "step": 69620
    },
    {
      "entropy": 5.454868793487549,
      "epoch": 5.4170394864812295,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002481027548385158,
      "loss": 4.7489,
      "mean_token_accuracy": 0.23709761798381807,
      "num_tokens": 120771976.0,
      "step": 69625
    },
    {
      "entropy": 5.429926443099975,
      "epoch": 5.417428515852947,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024807523820386207,
      "loss": 4.7379,
      "mean_token_accuracy": 0.22935721427202224,
      "num_tokens": 120780426.0,
      "step": 69630
    },
    {
      "entropy": 5.47351336479187,
      "epoch": 5.417817545224665,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002480477219777488,
      "loss": 4.7664,
      "mean_token_accuracy": 0.22802679985761642,
      "num_tokens": 120789438.0,
      "step": 69635
    },
    {
      "entropy": 5.467291927337646,
      "epoch": 5.418206574596382,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002480202061605933,
      "loss": 4.7472,
      "mean_token_accuracy": 0.2314481645822525,
      "num_tokens": 120797705.0,
      "step": 69640
    },
    {
      "entropy": 5.43750524520874,
      "epoch": 5.418595603968099,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002479926907528134,
      "loss": 4.7702,
      "mean_token_accuracy": 0.2306249037384987,
      "num_tokens": 120806015.0,
      "step": 69645
    },
    {
      "entropy": 5.326909303665161,
      "epoch": 5.418984633339817,
      "grad_norm": 1.1875,
      "learning_rate": 0.00024796517575482633,
      "loss": 4.6782,
      "mean_token_accuracy": 0.23661180287599565,
      "num_tokens": 120815051.0,
      "step": 69650
    },
    {
      "entropy": 5.462689781188965,
      "epoch": 5.419373662711535,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002479376611670498,
      "loss": 4.6801,
      "mean_token_accuracy": 0.23674658983945845,
      "num_tokens": 120823108.0,
      "step": 69655
    },
    {
      "entropy": 5.428510618209839,
      "epoch": 5.4197626920832525,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024791014698990116,
      "loss": 4.7067,
      "mean_token_accuracy": 0.2435269370675087,
      "num_tokens": 120831049.0,
      "step": 69660
    },
    {
      "entropy": 5.361752080917358,
      "epoch": 5.42015172145497,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00024788263322379783,
      "loss": 4.757,
      "mean_token_accuracy": 0.22787317931652068,
      "num_tokens": 120840264.0,
      "step": 69665
    },
    {
      "entropy": 5.418995141983032,
      "epoch": 5.420540750826687,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024785511986915746,
      "loss": 4.7482,
      "mean_token_accuracy": 0.23244829177856446,
      "num_tokens": 120849006.0,
      "step": 69670
    },
    {
      "entropy": 5.425528526306152,
      "epoch": 5.420929780198405,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002478276069263974,
      "loss": 4.6785,
      "mean_token_accuracy": 0.23997917026281357,
      "num_tokens": 120857876.0,
      "step": 69675
    },
    {
      "entropy": 5.4137050151824955,
      "epoch": 5.421318809570122,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00024780009439593514,
      "loss": 4.7007,
      "mean_token_accuracy": 0.2324400097131729,
      "num_tokens": 120867462.0,
      "step": 69680
    },
    {
      "entropy": 5.427243423461914,
      "epoch": 5.42170783894184,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002477725822781881,
      "loss": 4.6932,
      "mean_token_accuracy": 0.23979664891958236,
      "num_tokens": 120876326.0,
      "step": 69685
    },
    {
      "entropy": 5.288679027557373,
      "epoch": 5.422096868313558,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002477450705735739,
      "loss": 4.5848,
      "mean_token_accuracy": 0.24881603717803955,
      "num_tokens": 120885339.0,
      "step": 69690
    },
    {
      "entropy": 5.385456371307373,
      "epoch": 5.4224858976852754,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024771755928250984,
      "loss": 4.685,
      "mean_token_accuracy": 0.233632892370224,
      "num_tokens": 120893337.0,
      "step": 69695
    },
    {
      "entropy": 5.409089612960815,
      "epoch": 5.422874927056993,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024769004840541337,
      "loss": 4.773,
      "mean_token_accuracy": 0.22381669282913208,
      "num_tokens": 120902267.0,
      "step": 69700
    },
    {
      "entropy": 5.337218713760376,
      "epoch": 5.42326395642871,
      "grad_norm": 1.265625,
      "learning_rate": 0.000247662537942702,
      "loss": 4.6915,
      "mean_token_accuracy": 0.23128504157066346,
      "num_tokens": 120910925.0,
      "step": 69705
    },
    {
      "entropy": 5.431468725204468,
      "epoch": 5.423652985800428,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002476350278947931,
      "loss": 4.7388,
      "mean_token_accuracy": 0.2329317718744278,
      "num_tokens": 120918965.0,
      "step": 69710
    },
    {
      "entropy": 5.436700057983399,
      "epoch": 5.424042015172145,
      "grad_norm": 1.171875,
      "learning_rate": 0.000247607518262104,
      "loss": 4.7466,
      "mean_token_accuracy": 0.23125744014978408,
      "num_tokens": 120927609.0,
      "step": 69715
    },
    {
      "entropy": 5.443602228164673,
      "epoch": 5.424431044543863,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002475800090450523,
      "loss": 4.7372,
      "mean_token_accuracy": 0.23425329625606536,
      "num_tokens": 120936334.0,
      "step": 69720
    },
    {
      "entropy": 5.418461465835572,
      "epoch": 5.424820073915581,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002475525002440553,
      "loss": 4.747,
      "mean_token_accuracy": 0.23060865253210067,
      "num_tokens": 120944753.0,
      "step": 69725
    },
    {
      "entropy": 5.418678522109985,
      "epoch": 5.425209103287298,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024752499185953043,
      "loss": 4.7056,
      "mean_token_accuracy": 0.23293332904577255,
      "num_tokens": 120953678.0,
      "step": 69730
    },
    {
      "entropy": 5.426810455322266,
      "epoch": 5.425598132659016,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002474974838918951,
      "loss": 4.7339,
      "mean_token_accuracy": 0.23009037226438522,
      "num_tokens": 120962000.0,
      "step": 69735
    },
    {
      "entropy": 5.4070981502532955,
      "epoch": 5.425987162030733,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002474699763415667,
      "loss": 4.68,
      "mean_token_accuracy": 0.23633635938167571,
      "num_tokens": 120970633.0,
      "step": 69740
    },
    {
      "entropy": 5.388733196258545,
      "epoch": 5.426376191402451,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024744246920896243,
      "loss": 4.7433,
      "mean_token_accuracy": 0.2228708028793335,
      "num_tokens": 120979404.0,
      "step": 69745
    },
    {
      "entropy": 5.381824254989624,
      "epoch": 5.426765220774168,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002474149624944999,
      "loss": 4.6391,
      "mean_token_accuracy": 0.2393185406923294,
      "num_tokens": 120988260.0,
      "step": 69750
    },
    {
      "entropy": 5.44990906715393,
      "epoch": 5.427154250145886,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024738745619859643,
      "loss": 4.8114,
      "mean_token_accuracy": 0.21840309202671052,
      "num_tokens": 120996880.0,
      "step": 69755
    },
    {
      "entropy": 5.420956230163574,
      "epoch": 5.427543279517604,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024735995032166934,
      "loss": 4.6596,
      "mean_token_accuracy": 0.2332982674241066,
      "num_tokens": 121005173.0,
      "step": 69760
    },
    {
      "entropy": 5.418930864334106,
      "epoch": 5.427932308889321,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002473324448641361,
      "loss": 4.7104,
      "mean_token_accuracy": 0.23531652688980104,
      "num_tokens": 121013945.0,
      "step": 69765
    },
    {
      "entropy": 5.496822547912598,
      "epoch": 5.428321338261039,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024730493982641396,
      "loss": 4.7292,
      "mean_token_accuracy": 0.23161191195249559,
      "num_tokens": 121023110.0,
      "step": 69770
    },
    {
      "entropy": 5.402050733566284,
      "epoch": 5.428710367632756,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002472774352089202,
      "loss": 4.7265,
      "mean_token_accuracy": 0.2261809378862381,
      "num_tokens": 121031542.0,
      "step": 69775
    },
    {
      "entropy": 5.516595315933228,
      "epoch": 5.429099397004474,
      "grad_norm": 1.375,
      "learning_rate": 0.0002472499310120723,
      "loss": 4.8246,
      "mean_token_accuracy": 0.21990877091884614,
      "num_tokens": 121040152.0,
      "step": 69780
    },
    {
      "entropy": 5.345261812210083,
      "epoch": 5.429488426376191,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002472224272362875,
      "loss": 4.613,
      "mean_token_accuracy": 0.24509314447641373,
      "num_tokens": 121048806.0,
      "step": 69785
    },
    {
      "entropy": 5.392209720611572,
      "epoch": 5.429877455747909,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024719492388198313,
      "loss": 4.7415,
      "mean_token_accuracy": 0.23578960299491883,
      "num_tokens": 121057477.0,
      "step": 69790
    },
    {
      "entropy": 5.476546955108643,
      "epoch": 5.430266485119627,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002471674209495765,
      "loss": 4.7556,
      "mean_token_accuracy": 0.22759706228971482,
      "num_tokens": 121066567.0,
      "step": 69795
    },
    {
      "entropy": 5.370533275604248,
      "epoch": 5.430655514491344,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024713991843948506,
      "loss": 4.6347,
      "mean_token_accuracy": 0.23870881050825118,
      "num_tokens": 121075962.0,
      "step": 69800
    },
    {
      "entropy": 5.33057861328125,
      "epoch": 5.431044543863061,
      "grad_norm": 1.1875,
      "learning_rate": 0.00024711241635212597,
      "loss": 4.6613,
      "mean_token_accuracy": 0.23056967705488204,
      "num_tokens": 121084603.0,
      "step": 69805
    },
    {
      "entropy": 5.376582908630371,
      "epoch": 5.431433573234779,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024708491468791656,
      "loss": 4.7561,
      "mean_token_accuracy": 0.23227910697460175,
      "num_tokens": 121093583.0,
      "step": 69810
    },
    {
      "entropy": 5.396458911895752,
      "epoch": 5.431822602606497,
      "grad_norm": 1.25,
      "learning_rate": 0.0002470574134472741,
      "loss": 4.6769,
      "mean_token_accuracy": 0.23346876800060273,
      "num_tokens": 121102012.0,
      "step": 69815
    },
    {
      "entropy": 5.4126996994018555,
      "epoch": 5.432211631978214,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024702991263061595,
      "loss": 4.8177,
      "mean_token_accuracy": 0.21887790709733962,
      "num_tokens": 121110366.0,
      "step": 69820
    },
    {
      "entropy": 5.431943082809449,
      "epoch": 5.432600661349932,
      "grad_norm": 1.171875,
      "learning_rate": 0.00024700241223835935,
      "loss": 4.8112,
      "mean_token_accuracy": 0.22494069039821624,
      "num_tokens": 121120132.0,
      "step": 69825
    },
    {
      "entropy": 5.340444564819336,
      "epoch": 5.43298969072165,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00024697491227092163,
      "loss": 4.6126,
      "mean_token_accuracy": 0.24079171121120452,
      "num_tokens": 121129093.0,
      "step": 69830
    },
    {
      "entropy": 5.436751842498779,
      "epoch": 5.433378720093367,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024694741272872,
      "loss": 4.7612,
      "mean_token_accuracy": 0.23306770622730255,
      "num_tokens": 121137332.0,
      "step": 69835
    },
    {
      "entropy": 5.497312021255493,
      "epoch": 5.433767749465084,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024691991361217166,
      "loss": 4.76,
      "mean_token_accuracy": 0.22322983741760255,
      "num_tokens": 121145933.0,
      "step": 69840
    },
    {
      "entropy": 5.401016902923584,
      "epoch": 5.434156778836802,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024689241492169403,
      "loss": 4.7414,
      "mean_token_accuracy": 0.23306696563959123,
      "num_tokens": 121154423.0,
      "step": 69845
    },
    {
      "entropy": 5.430587244033814,
      "epoch": 5.43454580820852,
      "grad_norm": 1.171875,
      "learning_rate": 0.00024686491665770424,
      "loss": 4.7063,
      "mean_token_accuracy": 0.23358284831047058,
      "num_tokens": 121162853.0,
      "step": 69850
    },
    {
      "entropy": 5.454870080947876,
      "epoch": 5.434934837580237,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024683741882061954,
      "loss": 4.7897,
      "mean_token_accuracy": 0.232746858894825,
      "num_tokens": 121171002.0,
      "step": 69855
    },
    {
      "entropy": 5.429724216461182,
      "epoch": 5.435323866951955,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002468099214108572,
      "loss": 4.8542,
      "mean_token_accuracy": 0.2178139865398407,
      "num_tokens": 121179537.0,
      "step": 69860
    },
    {
      "entropy": 5.458304786682129,
      "epoch": 5.435712896323673,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00024678242442883444,
      "loss": 4.7646,
      "mean_token_accuracy": 0.21891040056943895,
      "num_tokens": 121189225.0,
      "step": 69865
    },
    {
      "entropy": 5.344929504394531,
      "epoch": 5.43610192569539,
      "grad_norm": 1.25,
      "learning_rate": 0.00024675492787496844,
      "loss": 4.6952,
      "mean_token_accuracy": 0.2282920867204666,
      "num_tokens": 121198203.0,
      "step": 69870
    },
    {
      "entropy": 5.416469621658325,
      "epoch": 5.436490955067107,
      "grad_norm": 1.171875,
      "learning_rate": 0.00024672743174967645,
      "loss": 4.7152,
      "mean_token_accuracy": 0.23307677507400512,
      "num_tokens": 121206277.0,
      "step": 69875
    },
    {
      "entropy": 5.4137365341186525,
      "epoch": 5.436879984438825,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002466999360533757,
      "loss": 4.7148,
      "mean_token_accuracy": 0.2301306888461113,
      "num_tokens": 121214887.0,
      "step": 69880
    },
    {
      "entropy": 5.378177452087402,
      "epoch": 5.437269013810543,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002466724407864833,
      "loss": 4.671,
      "mean_token_accuracy": 0.2308019742369652,
      "num_tokens": 121223559.0,
      "step": 69885
    },
    {
      "entropy": 5.359039831161499,
      "epoch": 5.43765804318226,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024664494594941664,
      "loss": 4.6787,
      "mean_token_accuracy": 0.23452642410993577,
      "num_tokens": 121232936.0,
      "step": 69890
    },
    {
      "entropy": 5.382229518890381,
      "epoch": 5.438047072553978,
      "grad_norm": 1.25,
      "learning_rate": 0.00024661745154259273,
      "loss": 4.7242,
      "mean_token_accuracy": 0.22653417587280272,
      "num_tokens": 121241074.0,
      "step": 69895
    },
    {
      "entropy": 5.401311302185059,
      "epoch": 5.438436101925696,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002465899575664289,
      "loss": 4.709,
      "mean_token_accuracy": 0.23395226001739503,
      "num_tokens": 121249874.0,
      "step": 69900
    },
    {
      "entropy": 5.417638635635376,
      "epoch": 5.438825131297413,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024656246402134216,
      "loss": 4.7134,
      "mean_token_accuracy": 0.23267678320407867,
      "num_tokens": 121258925.0,
      "step": 69905
    },
    {
      "entropy": 5.365522146224976,
      "epoch": 5.43921416066913,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002465349709077498,
      "loss": 4.6694,
      "mean_token_accuracy": 0.24021781980991364,
      "num_tokens": 121268351.0,
      "step": 69910
    },
    {
      "entropy": 5.407857608795166,
      "epoch": 5.439603190040848,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002465074782260689,
      "loss": 4.7249,
      "mean_token_accuracy": 0.23254556804895402,
      "num_tokens": 121277018.0,
      "step": 69915
    },
    {
      "entropy": 5.402533864974975,
      "epoch": 5.439992219412566,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024647998597671674,
      "loss": 4.7161,
      "mean_token_accuracy": 0.23502183109521865,
      "num_tokens": 121285030.0,
      "step": 69920
    },
    {
      "entropy": 5.460343599319458,
      "epoch": 5.440381248784283,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002464524941601103,
      "loss": 4.7965,
      "mean_token_accuracy": 0.22272115349769592,
      "num_tokens": 121293487.0,
      "step": 69925
    },
    {
      "entropy": 5.413256359100342,
      "epoch": 5.440770278156001,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024642500277666696,
      "loss": 4.7631,
      "mean_token_accuracy": 0.2316405341029167,
      "num_tokens": 121301859.0,
      "step": 69930
    },
    {
      "entropy": 5.474071025848389,
      "epoch": 5.441159307527719,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024639751182680363,
      "loss": 4.7632,
      "mean_token_accuracy": 0.22675004750490188,
      "num_tokens": 121310178.0,
      "step": 69935
    },
    {
      "entropy": 5.351826477050781,
      "epoch": 5.4415483368994355,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002463700213109376,
      "loss": 4.7027,
      "mean_token_accuracy": 0.2316521629691124,
      "num_tokens": 121318479.0,
      "step": 69940
    },
    {
      "entropy": 5.353415584564209,
      "epoch": 5.441937366271153,
      "grad_norm": 1.21875,
      "learning_rate": 0.00024634253122948584,
      "loss": 4.6754,
      "mean_token_accuracy": 0.23054594844579696,
      "num_tokens": 121327477.0,
      "step": 69945
    },
    {
      "entropy": 5.4395331859588625,
      "epoch": 5.442326395642871,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024631504158286556,
      "loss": 4.7231,
      "mean_token_accuracy": 0.2321978658437729,
      "num_tokens": 121336470.0,
      "step": 69950
    },
    {
      "entropy": 5.457323741912842,
      "epoch": 5.442715425014589,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024628755237149385,
      "loss": 4.7612,
      "mean_token_accuracy": 0.2316087543964386,
      "num_tokens": 121344750.0,
      "step": 69955
    },
    {
      "entropy": 5.442560243606567,
      "epoch": 5.443104454386306,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024626006359578797,
      "loss": 4.7424,
      "mean_token_accuracy": 0.2308705136179924,
      "num_tokens": 121352999.0,
      "step": 69960
    },
    {
      "entropy": 5.394291114807129,
      "epoch": 5.443493483758024,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024623257525616484,
      "loss": 4.6968,
      "mean_token_accuracy": 0.23462053537368774,
      "num_tokens": 121361262.0,
      "step": 69965
    },
    {
      "entropy": 5.381725740432739,
      "epoch": 5.443882513129742,
      "grad_norm": 1.25,
      "learning_rate": 0.00024620508735304146,
      "loss": 4.6804,
      "mean_token_accuracy": 0.23458803743124007,
      "num_tokens": 121369730.0,
      "step": 69970
    },
    {
      "entropy": 5.449749517440796,
      "epoch": 5.4442715425014585,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002461775998868352,
      "loss": 4.7857,
      "mean_token_accuracy": 0.22504192888736724,
      "num_tokens": 121378282.0,
      "step": 69975
    },
    {
      "entropy": 5.372137928009034,
      "epoch": 5.444660571873176,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024615011285796293,
      "loss": 4.6626,
      "mean_token_accuracy": 0.23253389000892638,
      "num_tokens": 121387032.0,
      "step": 69980
    },
    {
      "entropy": 5.304735946655273,
      "epoch": 5.445049601244894,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00024612262626684176,
      "loss": 4.6351,
      "mean_token_accuracy": 0.23193720132112502,
      "num_tokens": 121395225.0,
      "step": 69985
    },
    {
      "entropy": 5.412107610702515,
      "epoch": 5.445438630616612,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002460951401138888,
      "loss": 4.7067,
      "mean_token_accuracy": 0.2321212559938431,
      "num_tokens": 121404662.0,
      "step": 69990
    },
    {
      "entropy": 5.444408273696899,
      "epoch": 5.445827659988329,
      "grad_norm": 1.171875,
      "learning_rate": 0.00024606765439952105,
      "loss": 4.7544,
      "mean_token_accuracy": 0.22917400151491166,
      "num_tokens": 121414671.0,
      "step": 69995
    },
    {
      "entropy": 5.51793475151062,
      "epoch": 5.446216689360047,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002460401691241556,
      "loss": 4.7658,
      "mean_token_accuracy": 0.22215323597192765,
      "num_tokens": 121423497.0,
      "step": 70000
    },
    {
      "entropy": 5.483133125305176,
      "epoch": 5.446605718731764,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024601268428820947,
      "loss": 4.7445,
      "mean_token_accuracy": 0.22411587685346604,
      "num_tokens": 121432303.0,
      "step": 70005
    },
    {
      "entropy": 5.419503784179687,
      "epoch": 5.4469947481034815,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024598519989209983,
      "loss": 4.7584,
      "mean_token_accuracy": 0.22890669405460357,
      "num_tokens": 121440799.0,
      "step": 70010
    },
    {
      "entropy": 5.4231503963470455,
      "epoch": 5.447383777475199,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002459577159362434,
      "loss": 4.8076,
      "mean_token_accuracy": 0.22493340075016022,
      "num_tokens": 121448754.0,
      "step": 70015
    },
    {
      "entropy": 5.39298906326294,
      "epoch": 5.447772806846917,
      "grad_norm": 1.25,
      "learning_rate": 0.0002459302324210575,
      "loss": 4.705,
      "mean_token_accuracy": 0.23102498203516006,
      "num_tokens": 121457097.0,
      "step": 70020
    },
    {
      "entropy": 5.478921937942505,
      "epoch": 5.448161836218635,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002459027493469591,
      "loss": 4.728,
      "mean_token_accuracy": 0.2270919233560562,
      "num_tokens": 121465936.0,
      "step": 70025
    },
    {
      "entropy": 5.415809965133667,
      "epoch": 5.448550865590352,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024587526671436517,
      "loss": 4.7353,
      "mean_token_accuracy": 0.23051173090934754,
      "num_tokens": 121474693.0,
      "step": 70030
    },
    {
      "entropy": 5.369022274017334,
      "epoch": 5.44893989496207,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024584778452369264,
      "loss": 4.6531,
      "mean_token_accuracy": 0.236466483771801,
      "num_tokens": 121483667.0,
      "step": 70035
    },
    {
      "entropy": 5.386430454254151,
      "epoch": 5.449328924333787,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024582030277535866,
      "loss": 4.7178,
      "mean_token_accuracy": 0.22982580810785294,
      "num_tokens": 121492866.0,
      "step": 70040
    },
    {
      "entropy": 5.440623378753662,
      "epoch": 5.4497179537055045,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00024579282146978016,
      "loss": 4.7028,
      "mean_token_accuracy": 0.22673131078481673,
      "num_tokens": 121502225.0,
      "step": 70045
    },
    {
      "entropy": 5.42845573425293,
      "epoch": 5.450106983077222,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000245765340607374,
      "loss": 4.7631,
      "mean_token_accuracy": 0.22806007117033006,
      "num_tokens": 121511430.0,
      "step": 70050
    },
    {
      "entropy": 5.35962815284729,
      "epoch": 5.45049601244894,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002457378601885573,
      "loss": 4.6992,
      "mean_token_accuracy": 0.2309756502509117,
      "num_tokens": 121520468.0,
      "step": 70055
    },
    {
      "entropy": 5.407279539108276,
      "epoch": 5.450885041820658,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024571038021374705,
      "loss": 4.7152,
      "mean_token_accuracy": 0.23406866639852525,
      "num_tokens": 121529071.0,
      "step": 70060
    },
    {
      "entropy": 5.417733764648437,
      "epoch": 5.451274071192375,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024568290068336014,
      "loss": 4.7746,
      "mean_token_accuracy": 0.22933125793933867,
      "num_tokens": 121538035.0,
      "step": 70065
    },
    {
      "entropy": 5.402890968322754,
      "epoch": 5.451663100564093,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002456554215978135,
      "loss": 4.7282,
      "mean_token_accuracy": 0.22448500990867615,
      "num_tokens": 121546987.0,
      "step": 70070
    },
    {
      "entropy": 5.46339602470398,
      "epoch": 5.45205212993581,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002456279429575242,
      "loss": 4.692,
      "mean_token_accuracy": 0.23416840881109238,
      "num_tokens": 121555872.0,
      "step": 70075
    },
    {
      "entropy": 5.470245695114135,
      "epoch": 5.4524411593075275,
      "grad_norm": 1.25,
      "learning_rate": 0.00024560046476290905,
      "loss": 4.7466,
      "mean_token_accuracy": 0.22516052573919296,
      "num_tokens": 121564873.0,
      "step": 70080
    },
    {
      "entropy": 5.418851518630982,
      "epoch": 5.452830188679245,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00024557298701438504,
      "loss": 4.7793,
      "mean_token_accuracy": 0.23250671923160554,
      "num_tokens": 121573479.0,
      "step": 70085
    },
    {
      "entropy": 5.341291284561157,
      "epoch": 5.453219218050963,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002455455097123691,
      "loss": 4.6511,
      "mean_token_accuracy": 0.23310232758522034,
      "num_tokens": 121581831.0,
      "step": 70090
    },
    {
      "entropy": 5.483042812347412,
      "epoch": 5.453608247422681,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002455180328572783,
      "loss": 4.7368,
      "mean_token_accuracy": 0.24039117246866226,
      "num_tokens": 121590028.0,
      "step": 70095
    },
    {
      "entropy": 5.414647150039673,
      "epoch": 5.453997276794398,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024549055644952926,
      "loss": 4.7229,
      "mean_token_accuracy": 0.22803862392902374,
      "num_tokens": 121599110.0,
      "step": 70100
    },
    {
      "entropy": 5.39364538192749,
      "epoch": 5.454386306166116,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002454630804895392,
      "loss": 4.7733,
      "mean_token_accuracy": 0.22963395714759827,
      "num_tokens": 121608355.0,
      "step": 70105
    },
    {
      "entropy": 5.3686693668365475,
      "epoch": 5.454775335537833,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002454356049777248,
      "loss": 4.6353,
      "mean_token_accuracy": 0.23543082177639008,
      "num_tokens": 121616415.0,
      "step": 70110
    },
    {
      "entropy": 5.444333076477051,
      "epoch": 5.4551643649095505,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024540812991450306,
      "loss": 4.776,
      "mean_token_accuracy": 0.22534285634756088,
      "num_tokens": 121624213.0,
      "step": 70115
    },
    {
      "entropy": 5.462574672698975,
      "epoch": 5.455553394281268,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024538065530029093,
      "loss": 4.7355,
      "mean_token_accuracy": 0.23439619988203048,
      "num_tokens": 121632866.0,
      "step": 70120
    },
    {
      "entropy": 5.433603715896607,
      "epoch": 5.455942423652986,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002453531811355051,
      "loss": 4.7684,
      "mean_token_accuracy": 0.2310313731431961,
      "num_tokens": 121641381.0,
      "step": 70125
    },
    {
      "entropy": 5.396178150177002,
      "epoch": 5.456331453024704,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00024532570742056257,
      "loss": 4.7217,
      "mean_token_accuracy": 0.23286139369010925,
      "num_tokens": 121650784.0,
      "step": 70130
    },
    {
      "entropy": 5.39460825920105,
      "epoch": 5.456720482396421,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024529823415588025,
      "loss": 4.7605,
      "mean_token_accuracy": 0.23404355198144913,
      "num_tokens": 121660355.0,
      "step": 70135
    },
    {
      "entropy": 5.400440835952759,
      "epoch": 5.457109511768138,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002452707613418749,
      "loss": 4.66,
      "mean_token_accuracy": 0.2350002110004425,
      "num_tokens": 121668725.0,
      "step": 70140
    },
    {
      "entropy": 5.47832760810852,
      "epoch": 5.457498541139856,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024524328897896345,
      "loss": 4.7688,
      "mean_token_accuracy": 0.2324103221297264,
      "num_tokens": 121677675.0,
      "step": 70145
    },
    {
      "entropy": 5.420437479019165,
      "epoch": 5.4578875705115735,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00024521581706756266,
      "loss": 4.756,
      "mean_token_accuracy": 0.2251098558306694,
      "num_tokens": 121686957.0,
      "step": 70150
    },
    {
      "entropy": 5.494581270217895,
      "epoch": 5.458276599883291,
      "grad_norm": 1.15625,
      "learning_rate": 0.00024518834560808954,
      "loss": 4.7399,
      "mean_token_accuracy": 0.22734930813312532,
      "num_tokens": 121694804.0,
      "step": 70155
    },
    {
      "entropy": 5.40276575088501,
      "epoch": 5.458665629255009,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024516087460096074,
      "loss": 4.6999,
      "mean_token_accuracy": 0.23059821724891663,
      "num_tokens": 121703461.0,
      "step": 70160
    },
    {
      "entropy": 5.395367383956909,
      "epoch": 5.459054658626727,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002451334040465933,
      "loss": 4.7217,
      "mean_token_accuracy": 0.22919812053442,
      "num_tokens": 121711762.0,
      "step": 70165
    },
    {
      "entropy": 5.368811988830567,
      "epoch": 5.459443687998444,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024510593394540383,
      "loss": 4.7505,
      "mean_token_accuracy": 0.23197755813598633,
      "num_tokens": 121720935.0,
      "step": 70170
    },
    {
      "entropy": 5.472686767578125,
      "epoch": 5.459832717370161,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002450784642978092,
      "loss": 4.8021,
      "mean_token_accuracy": 0.2263358026742935,
      "num_tokens": 121730120.0,
      "step": 70175
    },
    {
      "entropy": 5.491389369964599,
      "epoch": 5.460221746741879,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002450509951042263,
      "loss": 4.7047,
      "mean_token_accuracy": 0.23597283512353898,
      "num_tokens": 121738551.0,
      "step": 70180
    },
    {
      "entropy": 5.4011345386505125,
      "epoch": 5.4606107761135965,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024502352636507184,
      "loss": 4.7241,
      "mean_token_accuracy": 0.22926971763372422,
      "num_tokens": 121747079.0,
      "step": 70185
    },
    {
      "entropy": 5.385000276565552,
      "epoch": 5.460999805485314,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024499605808076273,
      "loss": 4.7644,
      "mean_token_accuracy": 0.22729700803756714,
      "num_tokens": 121755589.0,
      "step": 70190
    },
    {
      "entropy": 5.425689315795898,
      "epoch": 5.461388834857032,
      "grad_norm": 1.25,
      "learning_rate": 0.00024496859025171564,
      "loss": 4.7008,
      "mean_token_accuracy": 0.23764138668775558,
      "num_tokens": 121764320.0,
      "step": 70195
    },
    {
      "entropy": 5.440693998336792,
      "epoch": 5.46177786422875,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002449411228783474,
      "loss": 4.7272,
      "mean_token_accuracy": 0.22770379483699799,
      "num_tokens": 121773421.0,
      "step": 70200
    },
    {
      "entropy": 5.40166335105896,
      "epoch": 5.462166893600466,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024491365596107477,
      "loss": 4.7385,
      "mean_token_accuracy": 0.23650240451097487,
      "num_tokens": 121783636.0,
      "step": 70205
    },
    {
      "entropy": 5.34289140701294,
      "epoch": 5.462555922972184,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00024488618950031457,
      "loss": 4.6438,
      "mean_token_accuracy": 0.23462988883256913,
      "num_tokens": 121791731.0,
      "step": 70210
    },
    {
      "entropy": 5.347463226318359,
      "epoch": 5.462944952343902,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00024485872349648335,
      "loss": 4.6413,
      "mean_token_accuracy": 0.2441320076584816,
      "num_tokens": 121799876.0,
      "step": 70215
    },
    {
      "entropy": 5.302077436447144,
      "epoch": 5.4633339817156195,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024483125794999815,
      "loss": 4.6677,
      "mean_token_accuracy": 0.23624372333288193,
      "num_tokens": 121808342.0,
      "step": 70220
    },
    {
      "entropy": 5.408740043640137,
      "epoch": 5.463723011087337,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002448037928612756,
      "loss": 4.7222,
      "mean_token_accuracy": 0.23087164312601088,
      "num_tokens": 121816896.0,
      "step": 70225
    },
    {
      "entropy": 5.3666877269744875,
      "epoch": 5.464112040459055,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002447763282307324,
      "loss": 4.6583,
      "mean_token_accuracy": 0.2328501895070076,
      "num_tokens": 121825474.0,
      "step": 70230
    },
    {
      "entropy": 5.455490303039551,
      "epoch": 5.4645010698307726,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002447488640587854,
      "loss": 4.704,
      "mean_token_accuracy": 0.23164235800504684,
      "num_tokens": 121833604.0,
      "step": 70235
    },
    {
      "entropy": 5.339203500747681,
      "epoch": 5.46489009920249,
      "grad_norm": 1.25,
      "learning_rate": 0.00024472140034585113,
      "loss": 4.681,
      "mean_token_accuracy": 0.23588096350431442,
      "num_tokens": 121842610.0,
      "step": 70240
    },
    {
      "entropy": 5.386895179748535,
      "epoch": 5.465279128574207,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024469393709234647,
      "loss": 4.7607,
      "mean_token_accuracy": 0.22210632115602494,
      "num_tokens": 121851116.0,
      "step": 70245
    },
    {
      "entropy": 5.387976884841919,
      "epoch": 5.465668157945925,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024466647429868804,
      "loss": 4.6488,
      "mean_token_accuracy": 0.23826704919338226,
      "num_tokens": 121859674.0,
      "step": 70250
    },
    {
      "entropy": 5.409866189956665,
      "epoch": 5.4660571873176425,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002446390119652926,
      "loss": 4.6775,
      "mean_token_accuracy": 0.23887744992971421,
      "num_tokens": 121869528.0,
      "step": 70255
    },
    {
      "entropy": 5.471185731887817,
      "epoch": 5.46644621668936,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024461155009257684,
      "loss": 4.7873,
      "mean_token_accuracy": 0.22727449089288712,
      "num_tokens": 121878449.0,
      "step": 70260
    },
    {
      "entropy": 5.451321315765381,
      "epoch": 5.466835246061078,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024458408868095744,
      "loss": 4.7627,
      "mean_token_accuracy": 0.22335440516471863,
      "num_tokens": 121887255.0,
      "step": 70265
    },
    {
      "entropy": 5.416892099380493,
      "epoch": 5.4672242754327955,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024455662773085107,
      "loss": 4.7338,
      "mean_token_accuracy": 0.23399961292743682,
      "num_tokens": 121895424.0,
      "step": 70270
    },
    {
      "entropy": 5.421973609924317,
      "epoch": 5.467613304804512,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002445291672426744,
      "loss": 4.63,
      "mean_token_accuracy": 0.2390149265527725,
      "num_tokens": 121903114.0,
      "step": 70275
    },
    {
      "entropy": 5.409581661224365,
      "epoch": 5.46800233417623,
      "grad_norm": 1.375,
      "learning_rate": 0.00024450170721684416,
      "loss": 4.6927,
      "mean_token_accuracy": 0.23465573489665986,
      "num_tokens": 121911177.0,
      "step": 70280
    },
    {
      "entropy": 5.428075551986694,
      "epoch": 5.468391363547948,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002444742476537769,
      "loss": 4.7335,
      "mean_token_accuracy": 0.23092661499977113,
      "num_tokens": 121919541.0,
      "step": 70285
    },
    {
      "entropy": 5.323449897766113,
      "epoch": 5.4687803929196654,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024444678855388934,
      "loss": 4.6969,
      "mean_token_accuracy": 0.22235329151153566,
      "num_tokens": 121927939.0,
      "step": 70290
    },
    {
      "entropy": 5.398689317703247,
      "epoch": 5.469169422291383,
      "grad_norm": 1.1875,
      "learning_rate": 0.00024441932991759824,
      "loss": 4.7438,
      "mean_token_accuracy": 0.22741694152355194,
      "num_tokens": 121936831.0,
      "step": 70295
    },
    {
      "entropy": 5.443511819839477,
      "epoch": 5.469558451663101,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002443918717453201,
      "loss": 4.7433,
      "mean_token_accuracy": 0.2218385621905327,
      "num_tokens": 121945767.0,
      "step": 70300
    },
    {
      "entropy": 5.516142845153809,
      "epoch": 5.4699474810348185,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002443644140374715,
      "loss": 4.7963,
      "mean_token_accuracy": 0.22677680999040603,
      "num_tokens": 121954601.0,
      "step": 70305
    },
    {
      "entropy": 5.485755252838135,
      "epoch": 5.470336510406535,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024433695679446934,
      "loss": 4.7716,
      "mean_token_accuracy": 0.22823964208364486,
      "num_tokens": 121962886.0,
      "step": 70310
    },
    {
      "entropy": 5.378139543533325,
      "epoch": 5.470725539778253,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002443095000167299,
      "loss": 4.7487,
      "mean_token_accuracy": 0.22553160935640335,
      "num_tokens": 121971074.0,
      "step": 70315
    },
    {
      "entropy": 5.36142053604126,
      "epoch": 5.471114569149971,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024428204370467,
      "loss": 4.6616,
      "mean_token_accuracy": 0.23273782432079315,
      "num_tokens": 121978970.0,
      "step": 70320
    },
    {
      "entropy": 5.387151479721069,
      "epoch": 5.471503598521688,
      "grad_norm": 1.25,
      "learning_rate": 0.00024425458785870617,
      "loss": 4.7128,
      "mean_token_accuracy": 0.23110346347093583,
      "num_tokens": 121987520.0,
      "step": 70325
    },
    {
      "entropy": 5.393017530441284,
      "epoch": 5.471892627893406,
      "grad_norm": 1.25,
      "learning_rate": 0.00024422713247925507,
      "loss": 4.7119,
      "mean_token_accuracy": 0.23527584969997406,
      "num_tokens": 121996122.0,
      "step": 70330
    },
    {
      "entropy": 5.402265834808349,
      "epoch": 5.472281657265124,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002441996775667332,
      "loss": 4.6613,
      "mean_token_accuracy": 0.23618454486131668,
      "num_tokens": 122004857.0,
      "step": 70335
    },
    {
      "entropy": 5.390559148788452,
      "epoch": 5.472670686636841,
      "grad_norm": 1.375,
      "learning_rate": 0.0002441722231215572,
      "loss": 4.7234,
      "mean_token_accuracy": 0.23340353965759278,
      "num_tokens": 122013374.0,
      "step": 70340
    },
    {
      "entropy": 5.401590347290039,
      "epoch": 5.473059716008558,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00024414476914414373,
      "loss": 4.6754,
      "mean_token_accuracy": 0.23696849793195723,
      "num_tokens": 122022027.0,
      "step": 70345
    },
    {
      "entropy": 5.328091716766357,
      "epoch": 5.473448745380276,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024411731563490918,
      "loss": 4.6472,
      "mean_token_accuracy": 0.23487296104431152,
      "num_tokens": 122030942.0,
      "step": 70350
    },
    {
      "entropy": 5.428375911712647,
      "epoch": 5.473837774751994,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00024408986259427025,
      "loss": 4.7646,
      "mean_token_accuracy": 0.23261837363243104,
      "num_tokens": 122039675.0,
      "step": 70355
    },
    {
      "entropy": 5.526067113876342,
      "epoch": 5.474226804123711,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024406241002264346,
      "loss": 4.8488,
      "mean_token_accuracy": 0.22678534984588622,
      "num_tokens": 122047457.0,
      "step": 70360
    },
    {
      "entropy": 5.416643810272217,
      "epoch": 5.474615833495429,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00024403495792044532,
      "loss": 4.7007,
      "mean_token_accuracy": 0.22972971498966216,
      "num_tokens": 122056890.0,
      "step": 70365
    },
    {
      "entropy": 5.490651082992554,
      "epoch": 5.475004862867147,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024400750628809247,
      "loss": 4.7993,
      "mean_token_accuracy": 0.2294788032770157,
      "num_tokens": 122066144.0,
      "step": 70370
    },
    {
      "entropy": 5.401970958709716,
      "epoch": 5.475393892238864,
      "grad_norm": 1.1875,
      "learning_rate": 0.00024398005512600134,
      "loss": 4.7157,
      "mean_token_accuracy": 0.2314816564321518,
      "num_tokens": 122074719.0,
      "step": 70375
    },
    {
      "entropy": 5.424728965759277,
      "epoch": 5.475782921610581,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00024395260443458856,
      "loss": 4.793,
      "mean_token_accuracy": 0.22512896209955216,
      "num_tokens": 122083149.0,
      "step": 70380
    },
    {
      "entropy": 5.425227785110474,
      "epoch": 5.476171950982299,
      "grad_norm": 1.171875,
      "learning_rate": 0.00024392515421427054,
      "loss": 4.7518,
      "mean_token_accuracy": 0.22736071646213532,
      "num_tokens": 122092467.0,
      "step": 70385
    },
    {
      "entropy": 5.435410404205323,
      "epoch": 5.476560980354017,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024389770446546388,
      "loss": 4.7471,
      "mean_token_accuracy": 0.2288050591945648,
      "num_tokens": 122100885.0,
      "step": 70390
    },
    {
      "entropy": 5.414276361465454,
      "epoch": 5.476950009725734,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024387025518858503,
      "loss": 4.7745,
      "mean_token_accuracy": 0.23943827897310258,
      "num_tokens": 122109542.0,
      "step": 70395
    },
    {
      "entropy": 5.4028712749481205,
      "epoch": 5.477339039097452,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002438428063840505,
      "loss": 4.7126,
      "mean_token_accuracy": 0.2270074248313904,
      "num_tokens": 122117979.0,
      "step": 70400
    },
    {
      "entropy": 5.446815347671508,
      "epoch": 5.47772806846917,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024381535805227677,
      "loss": 4.8004,
      "mean_token_accuracy": 0.22397871166467667,
      "num_tokens": 122126559.0,
      "step": 70405
    },
    {
      "entropy": 5.39161524772644,
      "epoch": 5.478117097840887,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024378791019368036,
      "loss": 4.6477,
      "mean_token_accuracy": 0.23972657322883606,
      "num_tokens": 122135322.0,
      "step": 70410
    },
    {
      "entropy": 5.385728359222412,
      "epoch": 5.478506127212604,
      "grad_norm": 1.1875,
      "learning_rate": 0.00024376046280867765,
      "loss": 4.7188,
      "mean_token_accuracy": 0.23373919129371643,
      "num_tokens": 122143657.0,
      "step": 70415
    },
    {
      "entropy": 5.381788539886474,
      "epoch": 5.478895156584322,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024373301589768531,
      "loss": 4.7384,
      "mean_token_accuracy": 0.2314128577709198,
      "num_tokens": 122152577.0,
      "step": 70420
    },
    {
      "entropy": 5.409402704238891,
      "epoch": 5.47928418595604,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002437055694611197,
      "loss": 4.6855,
      "mean_token_accuracy": 0.2417663410305977,
      "num_tokens": 122161008.0,
      "step": 70425
    },
    {
      "entropy": 5.424704265594483,
      "epoch": 5.479673215327757,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00024367812349939723,
      "loss": 4.6959,
      "mean_token_accuracy": 0.23634953647851945,
      "num_tokens": 122169691.0,
      "step": 70430
    },
    {
      "entropy": 5.40051383972168,
      "epoch": 5.480062244699475,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024365067801293438,
      "loss": 4.6843,
      "mean_token_accuracy": 0.24186121076345443,
      "num_tokens": 122177652.0,
      "step": 70435
    },
    {
      "entropy": 5.393773412704467,
      "epoch": 5.480451274071193,
      "grad_norm": 1.25,
      "learning_rate": 0.00024362323300214755,
      "loss": 4.612,
      "mean_token_accuracy": 0.2360119953751564,
      "num_tokens": 122186748.0,
      "step": 70440
    },
    {
      "entropy": 5.343697118759155,
      "epoch": 5.48084030344291,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002435957884674533,
      "loss": 4.6826,
      "mean_token_accuracy": 0.23318090289831161,
      "num_tokens": 122194984.0,
      "step": 70445
    },
    {
      "entropy": 5.310951662063599,
      "epoch": 5.481229332814627,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002435683444092679,
      "loss": 4.5507,
      "mean_token_accuracy": 0.24514065980911254,
      "num_tokens": 122204137.0,
      "step": 70450
    },
    {
      "entropy": 5.467765522003174,
      "epoch": 5.481618362186345,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002435409008280079,
      "loss": 4.8528,
      "mean_token_accuracy": 0.2173578679561615,
      "num_tokens": 122212098.0,
      "step": 70455
    },
    {
      "entropy": 5.44714035987854,
      "epoch": 5.482007391558063,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002435134577240896,
      "loss": 4.7276,
      "mean_token_accuracy": 0.22981563508510588,
      "num_tokens": 122220359.0,
      "step": 70460
    },
    {
      "entropy": 5.418879127502441,
      "epoch": 5.48239642092978,
      "grad_norm": 1.125,
      "learning_rate": 0.00024348601509792956,
      "loss": 4.7439,
      "mean_token_accuracy": 0.22894397377967834,
      "num_tokens": 122229526.0,
      "step": 70465
    },
    {
      "entropy": 5.357851552963257,
      "epoch": 5.482785450301498,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00024345857294994405,
      "loss": 4.6833,
      "mean_token_accuracy": 0.2333046019077301,
      "num_tokens": 122239382.0,
      "step": 70470
    },
    {
      "entropy": 5.453622627258301,
      "epoch": 5.483174479673215,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024343113128054945,
      "loss": 4.7508,
      "mean_token_accuracy": 0.23433923721313477,
      "num_tokens": 122247820.0,
      "step": 70475
    },
    {
      "entropy": 5.4468272686004635,
      "epoch": 5.483563509044933,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002434036900901621,
      "loss": 4.7933,
      "mean_token_accuracy": 0.2309975102543831,
      "num_tokens": 122256016.0,
      "step": 70480
    },
    {
      "entropy": 5.385533952713013,
      "epoch": 5.48395253841665,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024337624937919856,
      "loss": 4.6546,
      "mean_token_accuracy": 0.2305649846792221,
      "num_tokens": 122264518.0,
      "step": 70485
    },
    {
      "entropy": 5.463404273986816,
      "epoch": 5.484341567788368,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024334880914807512,
      "loss": 4.7981,
      "mean_token_accuracy": 0.22577558904886247,
      "num_tokens": 122272828.0,
      "step": 70490
    },
    {
      "entropy": 5.3370969772338865,
      "epoch": 5.484730597160086,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00024332136939720806,
      "loss": 4.6563,
      "mean_token_accuracy": 0.24142964333295822,
      "num_tokens": 122281297.0,
      "step": 70495
    },
    {
      "entropy": 5.378442716598511,
      "epoch": 5.485119626531803,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00024329393012701396,
      "loss": 4.7559,
      "mean_token_accuracy": 0.2269502252340317,
      "num_tokens": 122290668.0,
      "step": 70500
    },
    {
      "entropy": 5.384352970123291,
      "epoch": 5.485508655903521,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024326649133790885,
      "loss": 4.6951,
      "mean_token_accuracy": 0.23127108812332153,
      "num_tokens": 122298998.0,
      "step": 70505
    },
    {
      "entropy": 5.310858917236328,
      "epoch": 5.485897685275238,
      "grad_norm": 1.375,
      "learning_rate": 0.00024323905303030925,
      "loss": 4.59,
      "mean_token_accuracy": 0.24498870968818665,
      "num_tokens": 122308341.0,
      "step": 70510
    },
    {
      "entropy": 5.394557666778565,
      "epoch": 5.486286714646956,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024321161520463147,
      "loss": 4.6702,
      "mean_token_accuracy": 0.23696606010198593,
      "num_tokens": 122316674.0,
      "step": 70515
    },
    {
      "entropy": 5.439527559280395,
      "epoch": 5.486675744018673,
      "grad_norm": 1.21875,
      "learning_rate": 0.00024318417786129188,
      "loss": 4.7193,
      "mean_token_accuracy": 0.23310763537883758,
      "num_tokens": 122326179.0,
      "step": 70520
    },
    {
      "entropy": 5.317496204376221,
      "epoch": 5.487064773390391,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00024315674100070672,
      "loss": 4.6531,
      "mean_token_accuracy": 0.24737457185983658,
      "num_tokens": 122334838.0,
      "step": 70525
    },
    {
      "entropy": 5.419668769836425,
      "epoch": 5.487453802762109,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002431293046232923,
      "loss": 4.7205,
      "mean_token_accuracy": 0.23586111068725585,
      "num_tokens": 122343204.0,
      "step": 70530
    },
    {
      "entropy": 5.288944101333618,
      "epoch": 5.487842832133826,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024310186872946498,
      "loss": 4.6164,
      "mean_token_accuracy": 0.23930510133504868,
      "num_tokens": 122352313.0,
      "step": 70535
    },
    {
      "entropy": 5.424410676956176,
      "epoch": 5.488231861505543,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024307443331964107,
      "loss": 4.6981,
      "mean_token_accuracy": 0.2280837044119835,
      "num_tokens": 122361468.0,
      "step": 70540
    },
    {
      "entropy": 5.420333909988403,
      "epoch": 5.488620890877261,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024304699839423677,
      "loss": 4.7465,
      "mean_token_accuracy": 0.22988288700580597,
      "num_tokens": 122369621.0,
      "step": 70545
    },
    {
      "entropy": 5.439290761947632,
      "epoch": 5.489009920248979,
      "grad_norm": 1.21875,
      "learning_rate": 0.00024301956395366838,
      "loss": 4.7468,
      "mean_token_accuracy": 0.2286800518631935,
      "num_tokens": 122377553.0,
      "step": 70550
    },
    {
      "entropy": 5.343237113952637,
      "epoch": 5.489398949620696,
      "grad_norm": 1.15625,
      "learning_rate": 0.00024299212999835223,
      "loss": 4.64,
      "mean_token_accuracy": 0.23968700915575028,
      "num_tokens": 122386322.0,
      "step": 70555
    },
    {
      "entropy": 5.429993724822998,
      "epoch": 5.489787978992414,
      "grad_norm": 1.21875,
      "learning_rate": 0.00024296469652870456,
      "loss": 4.8048,
      "mean_token_accuracy": 0.21803846061229706,
      "num_tokens": 122395196.0,
      "step": 70560
    },
    {
      "entropy": 5.475716924667358,
      "epoch": 5.490177008364132,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024293726354514166,
      "loss": 4.7456,
      "mean_token_accuracy": 0.23227256834506987,
      "num_tokens": 122403324.0,
      "step": 70565
    },
    {
      "entropy": 5.305783224105835,
      "epoch": 5.490566037735849,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002429098310480798,
      "loss": 4.6203,
      "mean_token_accuracy": 0.2392679914832115,
      "num_tokens": 122411651.0,
      "step": 70570
    },
    {
      "entropy": 5.437867307662964,
      "epoch": 5.490955067107567,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024288239903793508,
      "loss": 4.7814,
      "mean_token_accuracy": 0.2225402131676674,
      "num_tokens": 122419710.0,
      "step": 70575
    },
    {
      "entropy": 5.33840126991272,
      "epoch": 5.491344096479284,
      "grad_norm": 1.171875,
      "learning_rate": 0.00024285496751512393,
      "loss": 4.6687,
      "mean_token_accuracy": 0.238345967233181,
      "num_tokens": 122428213.0,
      "step": 70580
    },
    {
      "entropy": 5.3512654304504395,
      "epoch": 5.491733125851002,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002428275364800624,
      "loss": 4.6672,
      "mean_token_accuracy": 0.24225182235240936,
      "num_tokens": 122437376.0,
      "step": 70585
    },
    {
      "entropy": 5.378974962234497,
      "epoch": 5.492122155222719,
      "grad_norm": 1.109375,
      "learning_rate": 0.00024280010593316683,
      "loss": 4.7067,
      "mean_token_accuracy": 0.23174707144498824,
      "num_tokens": 122446581.0,
      "step": 70590
    },
    {
      "entropy": 5.421217346191407,
      "epoch": 5.492511184594437,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002427726758748534,
      "loss": 4.7507,
      "mean_token_accuracy": 0.22842974364757537,
      "num_tokens": 122455470.0,
      "step": 70595
    },
    {
      "entropy": 5.450832176208496,
      "epoch": 5.492900213966155,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002427452463055383,
      "loss": 4.7424,
      "mean_token_accuracy": 0.23095162361860275,
      "num_tokens": 122464943.0,
      "step": 70600
    },
    {
      "entropy": 5.415422439575195,
      "epoch": 5.493289243337872,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00024271781722563773,
      "loss": 4.7381,
      "mean_token_accuracy": 0.22880372852087022,
      "num_tokens": 122474058.0,
      "step": 70605
    },
    {
      "entropy": 5.346107864379883,
      "epoch": 5.493678272709589,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024269038863556798,
      "loss": 4.6511,
      "mean_token_accuracy": 0.23167358934879304,
      "num_tokens": 122482018.0,
      "step": 70610
    },
    {
      "entropy": 5.303422451019287,
      "epoch": 5.494067302081307,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024266296053574506,
      "loss": 4.6365,
      "mean_token_accuracy": 0.23482999801635743,
      "num_tokens": 122490419.0,
      "step": 70615
    },
    {
      "entropy": 5.410489559173584,
      "epoch": 5.494456331453025,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002426355329265853,
      "loss": 4.7613,
      "mean_token_accuracy": 0.22721103131771087,
      "num_tokens": 122500299.0,
      "step": 70620
    },
    {
      "entropy": 5.292138719558716,
      "epoch": 5.494845360824742,
      "grad_norm": 1.15625,
      "learning_rate": 0.00024260810580850479,
      "loss": 4.5829,
      "mean_token_accuracy": 0.24799180924892425,
      "num_tokens": 122509218.0,
      "step": 70625
    },
    {
      "entropy": 5.332752418518067,
      "epoch": 5.49523439019646,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024258067918191971,
      "loss": 4.669,
      "mean_token_accuracy": 0.22938060909509658,
      "num_tokens": 122517995.0,
      "step": 70630
    },
    {
      "entropy": 5.417664289474487,
      "epoch": 5.495623419568178,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002425532530472463,
      "loss": 4.766,
      "mean_token_accuracy": 0.22538333982229233,
      "num_tokens": 122526229.0,
      "step": 70635
    },
    {
      "entropy": 5.379167795181274,
      "epoch": 5.496012448939895,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024252582740490053,
      "loss": 4.6965,
      "mean_token_accuracy": 0.2346435844898224,
      "num_tokens": 122535197.0,
      "step": 70640
    },
    {
      "entropy": 5.389390230178833,
      "epoch": 5.496401478311612,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024249840225529874,
      "loss": 4.6594,
      "mean_token_accuracy": 0.23427533060312272,
      "num_tokens": 122543382.0,
      "step": 70645
    },
    {
      "entropy": 5.378939867019653,
      "epoch": 5.49679050768333,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002424709775988569,
      "loss": 4.6944,
      "mean_token_accuracy": 0.23746426403522491,
      "num_tokens": 122552001.0,
      "step": 70650
    },
    {
      "entropy": 5.414185190200806,
      "epoch": 5.497179537055048,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024244355343599116,
      "loss": 4.7336,
      "mean_token_accuracy": 0.23340862840414048,
      "num_tokens": 122560070.0,
      "step": 70655
    },
    {
      "entropy": 5.451605033874512,
      "epoch": 5.497568566426765,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024241612976711786,
      "loss": 4.7706,
      "mean_token_accuracy": 0.22580428719520568,
      "num_tokens": 122569097.0,
      "step": 70660
    },
    {
      "entropy": 5.526263809204101,
      "epoch": 5.497957595798483,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024238870659265278,
      "loss": 4.8124,
      "mean_token_accuracy": 0.23014591187238692,
      "num_tokens": 122577857.0,
      "step": 70665
    },
    {
      "entropy": 5.378426647186279,
      "epoch": 5.498346625170201,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024236128391301215,
      "loss": 4.6796,
      "mean_token_accuracy": 0.2276335760951042,
      "num_tokens": 122586587.0,
      "step": 70670
    },
    {
      "entropy": 5.3087965965271,
      "epoch": 5.4987356545419175,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024233386172861217,
      "loss": 4.5938,
      "mean_token_accuracy": 0.23619831651449202,
      "num_tokens": 122594789.0,
      "step": 70675
    },
    {
      "entropy": 5.380229377746582,
      "epoch": 5.499124683913635,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024230644003986869,
      "loss": 4.7391,
      "mean_token_accuracy": 0.22821490913629533,
      "num_tokens": 122603274.0,
      "step": 70680
    },
    {
      "entropy": 5.427090501785278,
      "epoch": 5.499513713285353,
      "grad_norm": 1.21875,
      "learning_rate": 0.00024227901884719806,
      "loss": 4.745,
      "mean_token_accuracy": 0.23368096649646758,
      "num_tokens": 122612192.0,
      "step": 70685
    },
    {
      "entropy": 5.406390571594239,
      "epoch": 5.499902742657071,
      "grad_norm": 1.25,
      "learning_rate": 0.00024225159815101626,
      "loss": 4.7125,
      "mean_token_accuracy": 0.23338521122932435,
      "num_tokens": 122620411.0,
      "step": 70690
    },
    {
      "entropy": 5.41592435836792,
      "epoch": 5.500291772028788,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00024222417795173929,
      "loss": 4.7421,
      "mean_token_accuracy": 0.22602510452270508,
      "num_tokens": 122629327.0,
      "step": 70695
    },
    {
      "entropy": 5.43199577331543,
      "epoch": 5.500680801400506,
      "grad_norm": 1.21875,
      "learning_rate": 0.00024219675824978332,
      "loss": 4.7891,
      "mean_token_accuracy": 0.22640966027975082,
      "num_tokens": 122638536.0,
      "step": 70700
    },
    {
      "entropy": 5.4074396133422855,
      "epoch": 5.501069830772224,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00024216933904556425,
      "loss": 4.6251,
      "mean_token_accuracy": 0.24110499620437623,
      "num_tokens": 122647144.0,
      "step": 70705
    },
    {
      "entropy": 5.418401145935059,
      "epoch": 5.501458860143941,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024214192033949827,
      "loss": 4.6796,
      "mean_token_accuracy": 0.2351738005876541,
      "num_tokens": 122656689.0,
      "step": 70710
    },
    {
      "entropy": 5.333641481399536,
      "epoch": 5.501847889515658,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024211450213200132,
      "loss": 4.692,
      "mean_token_accuracy": 0.23685812801122666,
      "num_tokens": 122665599.0,
      "step": 70715
    },
    {
      "entropy": 5.315492630004883,
      "epoch": 5.502236918887376,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024208708442348943,
      "loss": 4.6274,
      "mean_token_accuracy": 0.24509831219911576,
      "num_tokens": 122674048.0,
      "step": 70720
    },
    {
      "entropy": 5.404406213760376,
      "epoch": 5.502625948259094,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024205966721437867,
      "loss": 4.6836,
      "mean_token_accuracy": 0.2368836835026741,
      "num_tokens": 122682447.0,
      "step": 70725
    },
    {
      "entropy": 5.463923454284668,
      "epoch": 5.503014977630811,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024203225050508503,
      "loss": 4.7794,
      "mean_token_accuracy": 0.22529825419187546,
      "num_tokens": 122691221.0,
      "step": 70730
    },
    {
      "entropy": 5.3641749858856205,
      "epoch": 5.503404007002529,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002420048342960245,
      "loss": 4.6631,
      "mean_token_accuracy": 0.23389590084552764,
      "num_tokens": 122699728.0,
      "step": 70735
    },
    {
      "entropy": 5.417591381072998,
      "epoch": 5.503793036374246,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00024197741858761318,
      "loss": 4.7748,
      "mean_token_accuracy": 0.22758532464504241,
      "num_tokens": 122708858.0,
      "step": 70740
    },
    {
      "entropy": 5.33610258102417,
      "epoch": 5.5041820657459635,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002419500033802669,
      "loss": 4.6953,
      "mean_token_accuracy": 0.2365904375910759,
      "num_tokens": 122718027.0,
      "step": 70745
    },
    {
      "entropy": 5.4062799453735355,
      "epoch": 5.504571095117681,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002419225886744016,
      "loss": 4.6638,
      "mean_token_accuracy": 0.23261490166187287,
      "num_tokens": 122726090.0,
      "step": 70750
    },
    {
      "entropy": 5.389040899276734,
      "epoch": 5.504960124489399,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024189517447043352,
      "loss": 4.6957,
      "mean_token_accuracy": 0.23550100326538087,
      "num_tokens": 122734969.0,
      "step": 70755
    },
    {
      "entropy": 5.471482133865356,
      "epoch": 5.505349153861117,
      "grad_norm": 1.25,
      "learning_rate": 0.00024186776076877843,
      "loss": 4.8008,
      "mean_token_accuracy": 0.2269164577126503,
      "num_tokens": 122743759.0,
      "step": 70760
    },
    {
      "entropy": 5.356032466888427,
      "epoch": 5.505738183232834,
      "grad_norm": 1.1875,
      "learning_rate": 0.00024184034756985235,
      "loss": 4.6756,
      "mean_token_accuracy": 0.23498995900154113,
      "num_tokens": 122752586.0,
      "step": 70765
    },
    {
      "entropy": 5.341942834854126,
      "epoch": 5.506127212604552,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024181293487407125,
      "loss": 4.6077,
      "mean_token_accuracy": 0.23939125537872313,
      "num_tokens": 122761306.0,
      "step": 70770
    },
    {
      "entropy": 5.332425975799561,
      "epoch": 5.50651624197627,
      "grad_norm": 1.171875,
      "learning_rate": 0.000241785522681851,
      "loss": 4.6449,
      "mean_token_accuracy": 0.23213035315275193,
      "num_tokens": 122769676.0,
      "step": 70775
    },
    {
      "entropy": 5.430518722534179,
      "epoch": 5.5069052713479865,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024175811099360757,
      "loss": 4.8012,
      "mean_token_accuracy": 0.22427190840244293,
      "num_tokens": 122778035.0,
      "step": 70780
    },
    {
      "entropy": 5.33522744178772,
      "epoch": 5.507294300719704,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024173069980975691,
      "loss": 4.6404,
      "mean_token_accuracy": 0.23808291852474212,
      "num_tokens": 122786861.0,
      "step": 70785
    },
    {
      "entropy": 5.517639303207398,
      "epoch": 5.507683330091422,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024170328913071493,
      "loss": 4.8879,
      "mean_token_accuracy": 0.2208481639623642,
      "num_tokens": 122796022.0,
      "step": 70790
    },
    {
      "entropy": 5.369570112228393,
      "epoch": 5.5080723594631396,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024167587895689753,
      "loss": 4.605,
      "mean_token_accuracy": 0.23616992086172103,
      "num_tokens": 122804795.0,
      "step": 70795
    },
    {
      "entropy": 5.420101737976074,
      "epoch": 5.508461388834857,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024164846928872064,
      "loss": 4.7798,
      "mean_token_accuracy": 0.2234449177980423,
      "num_tokens": 122813570.0,
      "step": 70800
    },
    {
      "entropy": 5.415718698501587,
      "epoch": 5.508850418206575,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024162106012660012,
      "loss": 4.6709,
      "mean_token_accuracy": 0.2309708997607231,
      "num_tokens": 122822141.0,
      "step": 70805
    },
    {
      "entropy": 5.441112947463989,
      "epoch": 5.509239447578292,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002415936514709519,
      "loss": 4.7613,
      "mean_token_accuracy": 0.22063748091459273,
      "num_tokens": 122831443.0,
      "step": 70810
    },
    {
      "entropy": 5.404898977279663,
      "epoch": 5.5096284769500095,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002415662433221918,
      "loss": 4.7527,
      "mean_token_accuracy": 0.22742810398340224,
      "num_tokens": 122840424.0,
      "step": 70815
    },
    {
      "entropy": 5.404451608657837,
      "epoch": 5.510017506321727,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024153883568073576,
      "loss": 4.7332,
      "mean_token_accuracy": 0.22831309139728545,
      "num_tokens": 122848858.0,
      "step": 70820
    },
    {
      "entropy": 5.381207704544067,
      "epoch": 5.510406535693445,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002415114285469997,
      "loss": 4.6804,
      "mean_token_accuracy": 0.23861832022666932,
      "num_tokens": 122857576.0,
      "step": 70825
    },
    {
      "entropy": 5.3526207447052006,
      "epoch": 5.5107955650651625,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024148402192139935,
      "loss": 4.737,
      "mean_token_accuracy": 0.2295311525464058,
      "num_tokens": 122865968.0,
      "step": 70830
    },
    {
      "entropy": 5.452864360809326,
      "epoch": 5.51118459443688,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024145661580435063,
      "loss": 4.7583,
      "mean_token_accuracy": 0.23404245674610138,
      "num_tokens": 122874286.0,
      "step": 70835
    },
    {
      "entropy": 5.431307649612426,
      "epoch": 5.511573623808598,
      "grad_norm": 1.25,
      "learning_rate": 0.00024142921019626933,
      "loss": 4.709,
      "mean_token_accuracy": 0.23288986533880235,
      "num_tokens": 122882633.0,
      "step": 70840
    },
    {
      "entropy": 5.395624780654908,
      "epoch": 5.511962653180315,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024140180509757143,
      "loss": 4.7017,
      "mean_token_accuracy": 0.2306210994720459,
      "num_tokens": 122892065.0,
      "step": 70845
    },
    {
      "entropy": 5.392498016357422,
      "epoch": 5.5123516825520325,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00024137440050867265,
      "loss": 4.6823,
      "mean_token_accuracy": 0.23606627881526948,
      "num_tokens": 122900477.0,
      "step": 70850
    },
    {
      "entropy": 5.388213396072388,
      "epoch": 5.51274071192375,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024134699642998882,
      "loss": 4.7215,
      "mean_token_accuracy": 0.2273784190416336,
      "num_tokens": 122908868.0,
      "step": 70855
    },
    {
      "entropy": 5.325706338882446,
      "epoch": 5.513129741295468,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024131959286193572,
      "loss": 4.6502,
      "mean_token_accuracy": 0.2348801776766777,
      "num_tokens": 122917725.0,
      "step": 70860
    },
    {
      "entropy": 5.43540587425232,
      "epoch": 5.5135187706671855,
      "grad_norm": 1.25,
      "learning_rate": 0.00024129218980492917,
      "loss": 4.7374,
      "mean_token_accuracy": 0.22922678887844086,
      "num_tokens": 122927450.0,
      "step": 70865
    },
    {
      "entropy": 5.41433801651001,
      "epoch": 5.513907800038903,
      "grad_norm": 1.296875,
      "learning_rate": 0.000241264787259385,
      "loss": 4.7062,
      "mean_token_accuracy": 0.23392146974802017,
      "num_tokens": 122935766.0,
      "step": 70870
    },
    {
      "entropy": 5.444930744171143,
      "epoch": 5.51429682941062,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024123738522571902,
      "loss": 4.7241,
      "mean_token_accuracy": 0.23529625982046126,
      "num_tokens": 122943202.0,
      "step": 70875
    },
    {
      "entropy": 5.439007711410523,
      "epoch": 5.514685858782338,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002412099837043469,
      "loss": 4.7136,
      "mean_token_accuracy": 0.2285680204629898,
      "num_tokens": 122951543.0,
      "step": 70880
    },
    {
      "entropy": 5.37492561340332,
      "epoch": 5.515074888154055,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024118258269568456,
      "loss": 4.7229,
      "mean_token_accuracy": 0.22959576398134232,
      "num_tokens": 122960175.0,
      "step": 70885
    },
    {
      "entropy": 5.448202800750733,
      "epoch": 5.515463917525773,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024115518220014771,
      "loss": 4.7473,
      "mean_token_accuracy": 0.22924584299325942,
      "num_tokens": 122968779.0,
      "step": 70890
    },
    {
      "entropy": 5.458621311187744,
      "epoch": 5.515852946897491,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024112778221815206,
      "loss": 4.7343,
      "mean_token_accuracy": 0.22659550607204437,
      "num_tokens": 122977523.0,
      "step": 70895
    },
    {
      "entropy": 5.4293070316314695,
      "epoch": 5.5162419762692085,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024110038275011343,
      "loss": 4.7935,
      "mean_token_accuracy": 0.22974894493818282,
      "num_tokens": 122986260.0,
      "step": 70900
    },
    {
      "entropy": 5.408854961395264,
      "epoch": 5.516631005640926,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024107298379644754,
      "loss": 4.7316,
      "mean_token_accuracy": 0.23334001898765563,
      "num_tokens": 122995073.0,
      "step": 70905
    },
    {
      "entropy": 5.368807125091553,
      "epoch": 5.517020035012644,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002410455853575701,
      "loss": 4.6576,
      "mean_token_accuracy": 0.24093288332223892,
      "num_tokens": 123003913.0,
      "step": 70910
    },
    {
      "entropy": 5.348622608184814,
      "epoch": 5.517409064384361,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024101818743389687,
      "loss": 4.611,
      "mean_token_accuracy": 0.24176746159791945,
      "num_tokens": 123012851.0,
      "step": 70915
    },
    {
      "entropy": 5.436102819442749,
      "epoch": 5.517798093756078,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024099079002584355,
      "loss": 4.772,
      "mean_token_accuracy": 0.22975777089595795,
      "num_tokens": 123020912.0,
      "step": 70920
    },
    {
      "entropy": 5.328731632232666,
      "epoch": 5.518187123127796,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024096339313382586,
      "loss": 4.6528,
      "mean_token_accuracy": 0.23784373998641967,
      "num_tokens": 123029351.0,
      "step": 70925
    },
    {
      "entropy": 5.347149181365967,
      "epoch": 5.518576152499514,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024093599675825954,
      "loss": 4.5796,
      "mean_token_accuracy": 0.24292432367801667,
      "num_tokens": 123037238.0,
      "step": 70930
    },
    {
      "entropy": 5.43596248626709,
      "epoch": 5.5189651818712315,
      "grad_norm": 1.15625,
      "learning_rate": 0.00024090860089956024,
      "loss": 4.7763,
      "mean_token_accuracy": 0.22721119225025177,
      "num_tokens": 123046444.0,
      "step": 70935
    },
    {
      "entropy": 5.351435995101928,
      "epoch": 5.519354211242949,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002408812055581437,
      "loss": 4.6955,
      "mean_token_accuracy": 0.23849013596773147,
      "num_tokens": 123055485.0,
      "step": 70940
    },
    {
      "entropy": 5.429844903945923,
      "epoch": 5.519743240614666,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024085381073442536,
      "loss": 4.7454,
      "mean_token_accuracy": 0.2330682545900345,
      "num_tokens": 123063430.0,
      "step": 70945
    },
    {
      "entropy": 5.329717826843262,
      "epoch": 5.520132269986384,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024082641642882125,
      "loss": 4.5789,
      "mean_token_accuracy": 0.243253855407238,
      "num_tokens": 123071510.0,
      "step": 70950
    },
    {
      "entropy": 5.420927667617798,
      "epoch": 5.520521299358101,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024079902264174686,
      "loss": 4.7369,
      "mean_token_accuracy": 0.23120331317186354,
      "num_tokens": 123080263.0,
      "step": 70955
    },
    {
      "entropy": 5.343746566772461,
      "epoch": 5.520910328729819,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002407716293736179,
      "loss": 4.6178,
      "mean_token_accuracy": 0.24677411317825318,
      "num_tokens": 123089540.0,
      "step": 70960
    },
    {
      "entropy": 5.383777713775634,
      "epoch": 5.521299358101537,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00024074423662484996,
      "loss": 4.7357,
      "mean_token_accuracy": 0.2361590340733528,
      "num_tokens": 123098350.0,
      "step": 70965
    },
    {
      "entropy": 5.421955299377442,
      "epoch": 5.5216883874732545,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002407168443958587,
      "loss": 4.7612,
      "mean_token_accuracy": 0.23810844421386718,
      "num_tokens": 123107041.0,
      "step": 70970
    },
    {
      "entropy": 5.362704801559448,
      "epoch": 5.522077416844972,
      "grad_norm": 1.109375,
      "learning_rate": 0.00024068945268705978,
      "loss": 4.6385,
      "mean_token_accuracy": 0.24241780787706374,
      "num_tokens": 123115612.0,
      "step": 70975
    },
    {
      "entropy": 5.317593812942505,
      "epoch": 5.522466446216689,
      "grad_norm": 1.15625,
      "learning_rate": 0.00024066206149886876,
      "loss": 4.6186,
      "mean_token_accuracy": 0.2315910741686821,
      "num_tokens": 123124154.0,
      "step": 70980
    },
    {
      "entropy": 5.348397350311279,
      "epoch": 5.522855475588407,
      "grad_norm": 1.1875,
      "learning_rate": 0.00024063467083170132,
      "loss": 4.6422,
      "mean_token_accuracy": 0.24015880078077317,
      "num_tokens": 123133197.0,
      "step": 70985
    },
    {
      "entropy": 5.381210899353027,
      "epoch": 5.523244504960124,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002406072806859731,
      "loss": 4.7213,
      "mean_token_accuracy": 0.23230430632829666,
      "num_tokens": 123141153.0,
      "step": 70990
    },
    {
      "entropy": 5.393314933776855,
      "epoch": 5.523633534331842,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002405798910620996,
      "loss": 4.6779,
      "mean_token_accuracy": 0.234355691075325,
      "num_tokens": 123149015.0,
      "step": 70995
    },
    {
      "entropy": 5.406496858596801,
      "epoch": 5.52402256370356,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002405525019604965,
      "loss": 4.7095,
      "mean_token_accuracy": 0.2298208236694336,
      "num_tokens": 123156902.0,
      "step": 71000
    },
    {
      "entropy": 5.398468494415283,
      "epoch": 5.5244115930752775,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024052511338157928,
      "loss": 4.6881,
      "mean_token_accuracy": 0.23288017213344575,
      "num_tokens": 123166988.0,
      "step": 71005
    },
    {
      "entropy": 5.422676277160645,
      "epoch": 5.524800622446994,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024049772532576364,
      "loss": 4.7408,
      "mean_token_accuracy": 0.2305937498807907,
      "num_tokens": 123175287.0,
      "step": 71010
    },
    {
      "entropy": 5.388989400863648,
      "epoch": 5.525189651818712,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00024047033779346505,
      "loss": 4.74,
      "mean_token_accuracy": 0.22371324598789216,
      "num_tokens": 123183846.0,
      "step": 71015
    },
    {
      "entropy": 5.392907619476318,
      "epoch": 5.52557868119043,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002404429507850992,
      "loss": 4.6954,
      "mean_token_accuracy": 0.2265216514468193,
      "num_tokens": 123192448.0,
      "step": 71020
    },
    {
      "entropy": 5.3461230278015135,
      "epoch": 5.525967710562147,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002404155643010815,
      "loss": 4.7208,
      "mean_token_accuracy": 0.2307633489370346,
      "num_tokens": 123201523.0,
      "step": 71025
    },
    {
      "entropy": 5.402951240539551,
      "epoch": 5.526356739933865,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024038817834182762,
      "loss": 4.6923,
      "mean_token_accuracy": 0.2337244287133217,
      "num_tokens": 123210115.0,
      "step": 71030
    },
    {
      "entropy": 5.3267021656036375,
      "epoch": 5.526745769305583,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00024036079290775308,
      "loss": 4.6271,
      "mean_token_accuracy": 0.23818755149841309,
      "num_tokens": 123219045.0,
      "step": 71035
    },
    {
      "entropy": 5.3926914691925045,
      "epoch": 5.5271347986773005,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024033340799927338,
      "loss": 4.7622,
      "mean_token_accuracy": 0.23074618875980377,
      "num_tokens": 123228762.0,
      "step": 71040
    },
    {
      "entropy": 5.360033893585205,
      "epoch": 5.527523828049017,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000240306023616804,
      "loss": 4.6947,
      "mean_token_accuracy": 0.23364813774824142,
      "num_tokens": 123237596.0,
      "step": 71045
    },
    {
      "entropy": 5.3927061557769775,
      "epoch": 5.527912857420735,
      "grad_norm": 1.21875,
      "learning_rate": 0.00024027863976076047,
      "loss": 4.7655,
      "mean_token_accuracy": 0.23065024614334106,
      "num_tokens": 123246066.0,
      "step": 71050
    },
    {
      "entropy": 5.343384838104248,
      "epoch": 5.528301886792453,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002402512564315583,
      "loss": 4.6447,
      "mean_token_accuracy": 0.2433573231101036,
      "num_tokens": 123254337.0,
      "step": 71055
    },
    {
      "entropy": 5.411526441574097,
      "epoch": 5.52869091616417,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024022387362961306,
      "loss": 4.7311,
      "mean_token_accuracy": 0.2284692943096161,
      "num_tokens": 123262765.0,
      "step": 71060
    },
    {
      "entropy": 5.450392961502075,
      "epoch": 5.529079945535888,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00024019649135534015,
      "loss": 4.7418,
      "mean_token_accuracy": 0.23703196793794631,
      "num_tokens": 123271287.0,
      "step": 71065
    },
    {
      "entropy": 5.401127433776855,
      "epoch": 5.529468974907606,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024016910960915512,
      "loss": 4.7105,
      "mean_token_accuracy": 0.22997922003269194,
      "num_tokens": 123280629.0,
      "step": 71070
    },
    {
      "entropy": 5.458154582977295,
      "epoch": 5.529858004279323,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024014172839147336,
      "loss": 4.7178,
      "mean_token_accuracy": 0.23531197607517243,
      "num_tokens": 123289295.0,
      "step": 71075
    },
    {
      "entropy": 5.461092567443847,
      "epoch": 5.53024703365104,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024011434770271033,
      "loss": 4.8297,
      "mean_token_accuracy": 0.218813981115818,
      "num_tokens": 123298626.0,
      "step": 71080
    },
    {
      "entropy": 5.391166734695434,
      "epoch": 5.530636063022758,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002400869675432817,
      "loss": 4.7151,
      "mean_token_accuracy": 0.23841411769390106,
      "num_tokens": 123307449.0,
      "step": 71085
    },
    {
      "entropy": 5.386625909805298,
      "epoch": 5.531025092394476,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024005958791360266,
      "loss": 4.727,
      "mean_token_accuracy": 0.22424560487270356,
      "num_tokens": 123316285.0,
      "step": 71090
    },
    {
      "entropy": 5.411904573440552,
      "epoch": 5.531414121766193,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00024003220881408882,
      "loss": 4.7561,
      "mean_token_accuracy": 0.22617663592100143,
      "num_tokens": 123324866.0,
      "step": 71095
    },
    {
      "entropy": 5.431642246246338,
      "epoch": 5.531803151137911,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00024000483024515552,
      "loss": 4.7311,
      "mean_token_accuracy": 0.23143961876630784,
      "num_tokens": 123333304.0,
      "step": 71100
    },
    {
      "entropy": 5.429759740829468,
      "epoch": 5.532192180509629,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023997745220721823,
      "loss": 4.7347,
      "mean_token_accuracy": 0.23136481791734695,
      "num_tokens": 123341918.0,
      "step": 71105
    },
    {
      "entropy": 5.307168865203858,
      "epoch": 5.5325812098813465,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023995007470069236,
      "loss": 4.607,
      "mean_token_accuracy": 0.2420688271522522,
      "num_tokens": 123350323.0,
      "step": 71110
    },
    {
      "entropy": 5.384808158874511,
      "epoch": 5.532970239253063,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023992269772599335,
      "loss": 4.7881,
      "mean_token_accuracy": 0.22650383561849594,
      "num_tokens": 123359342.0,
      "step": 71115
    },
    {
      "entropy": 5.484365653991699,
      "epoch": 5.533359268624781,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023989532128353654,
      "loss": 4.7935,
      "mean_token_accuracy": 0.2248683899641037,
      "num_tokens": 123366924.0,
      "step": 71120
    },
    {
      "entropy": 5.277698183059693,
      "epoch": 5.533748297996499,
      "grad_norm": 1.25,
      "learning_rate": 0.0002398679453737374,
      "loss": 4.5963,
      "mean_token_accuracy": 0.24192337244749068,
      "num_tokens": 123376457.0,
      "step": 71125
    },
    {
      "entropy": 5.429205131530762,
      "epoch": 5.534137327368216,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023984056999701121,
      "loss": 4.7923,
      "mean_token_accuracy": 0.22591836154460906,
      "num_tokens": 123385174.0,
      "step": 71130
    },
    {
      "entropy": 5.388508558273315,
      "epoch": 5.534526356739934,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023981319515377347,
      "loss": 4.6366,
      "mean_token_accuracy": 0.24139496833086013,
      "num_tokens": 123393619.0,
      "step": 71135
    },
    {
      "entropy": 5.436376762390137,
      "epoch": 5.534915386111652,
      "grad_norm": 1.25,
      "learning_rate": 0.00023978582084443939,
      "loss": 4.6972,
      "mean_token_accuracy": 0.224521666765213,
      "num_tokens": 123401259.0,
      "step": 71140
    },
    {
      "entropy": 5.366864585876465,
      "epoch": 5.535304415483369,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002397584470694244,
      "loss": 4.6572,
      "mean_token_accuracy": 0.23841806650161743,
      "num_tokens": 123410067.0,
      "step": 71145
    },
    {
      "entropy": 5.396704530715942,
      "epoch": 5.535693444855086,
      "grad_norm": 1.203125,
      "learning_rate": 0.000239731073829144,
      "loss": 4.7211,
      "mean_token_accuracy": 0.22803351134061814,
      "num_tokens": 123418280.0,
      "step": 71150
    },
    {
      "entropy": 5.408065509796143,
      "epoch": 5.536082474226804,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002397037011240134,
      "loss": 4.7286,
      "mean_token_accuracy": 0.23040994107723237,
      "num_tokens": 123427531.0,
      "step": 71155
    },
    {
      "entropy": 5.436710500717163,
      "epoch": 5.536471503598522,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00023967632895444796,
      "loss": 4.7587,
      "mean_token_accuracy": 0.2270273357629776,
      "num_tokens": 123435976.0,
      "step": 71160
    },
    {
      "entropy": 5.399189376831055,
      "epoch": 5.536860532970239,
      "grad_norm": 1.34375,
      "learning_rate": 0.000239648957320863,
      "loss": 4.7092,
      "mean_token_accuracy": 0.23579172939062118,
      "num_tokens": 123444417.0,
      "step": 71165
    },
    {
      "entropy": 5.372974634170532,
      "epoch": 5.537249562341957,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002396215862236738,
      "loss": 4.7131,
      "mean_token_accuracy": 0.22912664860486984,
      "num_tokens": 123452528.0,
      "step": 71170
    },
    {
      "entropy": 5.437200355529785,
      "epoch": 5.537638591713675,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023959421566329574,
      "loss": 4.7427,
      "mean_token_accuracy": 0.23424499928951265,
      "num_tokens": 123461090.0,
      "step": 71175
    },
    {
      "entropy": 5.450372505187988,
      "epoch": 5.538027621085392,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023956684564014415,
      "loss": 4.717,
      "mean_token_accuracy": 0.23116310983896254,
      "num_tokens": 123469529.0,
      "step": 71180
    },
    {
      "entropy": 5.33239426612854,
      "epoch": 5.538416650457109,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023953947615463423,
      "loss": 4.5769,
      "mean_token_accuracy": 0.2445661410689354,
      "num_tokens": 123477816.0,
      "step": 71185
    },
    {
      "entropy": 5.378343725204468,
      "epoch": 5.538805679828827,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002395121072071813,
      "loss": 4.697,
      "mean_token_accuracy": 0.2306036412715912,
      "num_tokens": 123486805.0,
      "step": 71190
    },
    {
      "entropy": 5.420285367965699,
      "epoch": 5.539194709200545,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023948473879820072,
      "loss": 4.7457,
      "mean_token_accuracy": 0.2234325036406517,
      "num_tokens": 123495393.0,
      "step": 71195
    },
    {
      "entropy": 5.467308092117309,
      "epoch": 5.539583738572262,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023945737092810764,
      "loss": 4.7352,
      "mean_token_accuracy": 0.22717151045799255,
      "num_tokens": 123503319.0,
      "step": 71200
    },
    {
      "entropy": 5.360679769515992,
      "epoch": 5.53997276794398,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002394300035973175,
      "loss": 4.622,
      "mean_token_accuracy": 0.2428535759449005,
      "num_tokens": 123512051.0,
      "step": 71205
    },
    {
      "entropy": 5.406475830078125,
      "epoch": 5.540361797315697,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023940263680624535,
      "loss": 4.6747,
      "mean_token_accuracy": 0.23111325055360793,
      "num_tokens": 123520812.0,
      "step": 71210
    },
    {
      "entropy": 5.371293687820435,
      "epoch": 5.540750826687415,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023937527055530643,
      "loss": 4.6837,
      "mean_token_accuracy": 0.23089441359043122,
      "num_tokens": 123530079.0,
      "step": 71215
    },
    {
      "entropy": 5.390916776657105,
      "epoch": 5.541139856059132,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002393479048449162,
      "loss": 4.6909,
      "mean_token_accuracy": 0.2386585980653763,
      "num_tokens": 123538437.0,
      "step": 71220
    },
    {
      "entropy": 5.324668264389038,
      "epoch": 5.54152888543085,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023932053967548971,
      "loss": 4.6005,
      "mean_token_accuracy": 0.2368892699480057,
      "num_tokens": 123547565.0,
      "step": 71225
    },
    {
      "entropy": 5.390067625045776,
      "epoch": 5.541917914802568,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023929317504744235,
      "loss": 4.6801,
      "mean_token_accuracy": 0.2334108754992485,
      "num_tokens": 123555983.0,
      "step": 71230
    },
    {
      "entropy": 5.383025407791138,
      "epoch": 5.542306944174285,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023926581096118916,
      "loss": 4.6188,
      "mean_token_accuracy": 0.2459752917289734,
      "num_tokens": 123564440.0,
      "step": 71235
    },
    {
      "entropy": 5.398411512374878,
      "epoch": 5.542695973546003,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002392384474171454,
      "loss": 4.7225,
      "mean_token_accuracy": 0.23337140083312988,
      "num_tokens": 123573283.0,
      "step": 71240
    },
    {
      "entropy": 5.398371934890747,
      "epoch": 5.543085002917721,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023921108441572625,
      "loss": 4.6761,
      "mean_token_accuracy": 0.23508528918027877,
      "num_tokens": 123582002.0,
      "step": 71245
    },
    {
      "entropy": 5.393572044372559,
      "epoch": 5.543474032289438,
      "grad_norm": 1.40625,
      "learning_rate": 0.000239183721957347,
      "loss": 4.7707,
      "mean_token_accuracy": 0.2334441736340523,
      "num_tokens": 123590557.0,
      "step": 71250
    },
    {
      "entropy": 5.379699468612671,
      "epoch": 5.543863061661155,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00023915636004242274,
      "loss": 4.6352,
      "mean_token_accuracy": 0.23921067714691163,
      "num_tokens": 123599548.0,
      "step": 71255
    },
    {
      "entropy": 5.380928945541382,
      "epoch": 5.544252091032873,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002391289986713686,
      "loss": 4.6028,
      "mean_token_accuracy": 0.24641335904598236,
      "num_tokens": 123607958.0,
      "step": 71260
    },
    {
      "entropy": 5.50011887550354,
      "epoch": 5.544641120404591,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023910163784459986,
      "loss": 4.8626,
      "mean_token_accuracy": 0.21921878606081008,
      "num_tokens": 123615867.0,
      "step": 71265
    },
    {
      "entropy": 5.381432247161865,
      "epoch": 5.545030149776308,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002390742775625317,
      "loss": 4.6219,
      "mean_token_accuracy": 0.24499341398477553,
      "num_tokens": 123624133.0,
      "step": 71270
    },
    {
      "entropy": 5.352582216262817,
      "epoch": 5.545419179148025,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023904691782557908,
      "loss": 4.6117,
      "mean_token_accuracy": 0.2438999205827713,
      "num_tokens": 123632306.0,
      "step": 71275
    },
    {
      "entropy": 5.406718826293945,
      "epoch": 5.545808208519743,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023901955863415725,
      "loss": 4.7403,
      "mean_token_accuracy": 0.22616302818059922,
      "num_tokens": 123641225.0,
      "step": 71280
    },
    {
      "entropy": 5.38878288269043,
      "epoch": 5.546197237891461,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00023899219998868142,
      "loss": 4.7108,
      "mean_token_accuracy": 0.22584101259708406,
      "num_tokens": 123650291.0,
      "step": 71285
    },
    {
      "entropy": 5.350390529632568,
      "epoch": 5.546586267263178,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023896484188956664,
      "loss": 4.6463,
      "mean_token_accuracy": 0.23685629814863204,
      "num_tokens": 123659552.0,
      "step": 71290
    },
    {
      "entropy": 5.476536607742309,
      "epoch": 5.546975296634896,
      "grad_norm": 1.171875,
      "learning_rate": 0.000238937484337228,
      "loss": 4.7947,
      "mean_token_accuracy": 0.2296607404947281,
      "num_tokens": 123668440.0,
      "step": 71295
    },
    {
      "entropy": 5.358720684051514,
      "epoch": 5.547364326006614,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023891012733208062,
      "loss": 4.6843,
      "mean_token_accuracy": 0.22796618938446045,
      "num_tokens": 123677310.0,
      "step": 71300
    },
    {
      "entropy": 5.252505445480347,
      "epoch": 5.547753355378331,
      "grad_norm": 1.171875,
      "learning_rate": 0.00023888277087453962,
      "loss": 4.5642,
      "mean_token_accuracy": 0.24780309051275254,
      "num_tokens": 123686560.0,
      "step": 71305
    },
    {
      "entropy": 5.362309360504151,
      "epoch": 5.548142384750049,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023885541496502017,
      "loss": 4.739,
      "mean_token_accuracy": 0.23092172294855118,
      "num_tokens": 123695223.0,
      "step": 71310
    },
    {
      "entropy": 5.451550436019898,
      "epoch": 5.548531414121766,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023882805960393722,
      "loss": 4.7691,
      "mean_token_accuracy": 0.23090124726295472,
      "num_tokens": 123704415.0,
      "step": 71315
    },
    {
      "entropy": 5.516361665725708,
      "epoch": 5.548920443493484,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002388007047917059,
      "loss": 4.7619,
      "mean_token_accuracy": 0.23531751781702043,
      "num_tokens": 123713506.0,
      "step": 71320
    },
    {
      "entropy": 5.410640573501587,
      "epoch": 5.549309472865201,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023877335052874122,
      "loss": 4.7242,
      "mean_token_accuracy": 0.2308322086930275,
      "num_tokens": 123721999.0,
      "step": 71325
    },
    {
      "entropy": 5.357743549346924,
      "epoch": 5.549698502236919,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023874599681545833,
      "loss": 4.7005,
      "mean_token_accuracy": 0.23016363978385926,
      "num_tokens": 123731297.0,
      "step": 71330
    },
    {
      "entropy": 5.280794858932495,
      "epoch": 5.550087531608637,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002387186436522722,
      "loss": 4.5529,
      "mean_token_accuracy": 0.2422693818807602,
      "num_tokens": 123739704.0,
      "step": 71335
    },
    {
      "entropy": 5.4853137016296385,
      "epoch": 5.550476560980354,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002386912910395979,
      "loss": 4.7751,
      "mean_token_accuracy": 0.22573849111795424,
      "num_tokens": 123748811.0,
      "step": 71340
    },
    {
      "entropy": 5.4452026844024655,
      "epoch": 5.550865590352071,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023866393897785049,
      "loss": 4.7419,
      "mean_token_accuracy": 0.2291549935936928,
      "num_tokens": 123758226.0,
      "step": 71345
    },
    {
      "entropy": 5.410419368743897,
      "epoch": 5.551254619723789,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023863658746744494,
      "loss": 4.6806,
      "mean_token_accuracy": 0.22997353971004486,
      "num_tokens": 123766362.0,
      "step": 71350
    },
    {
      "entropy": 5.324558067321777,
      "epoch": 5.551643649095507,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023860923650879636,
      "loss": 4.6323,
      "mean_token_accuracy": 0.25093170553445815,
      "num_tokens": 123774238.0,
      "step": 71355
    },
    {
      "entropy": 5.396375608444214,
      "epoch": 5.552032678467224,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002385818861023197,
      "loss": 4.7054,
      "mean_token_accuracy": 0.23085001558065416,
      "num_tokens": 123782618.0,
      "step": 71360
    },
    {
      "entropy": 5.423918151855469,
      "epoch": 5.552421707838942,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023855453624842986,
      "loss": 4.7232,
      "mean_token_accuracy": 0.225795878469944,
      "num_tokens": 123790968.0,
      "step": 71365
    },
    {
      "entropy": 5.402528619766235,
      "epoch": 5.55281073721066,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000238527186947542,
      "loss": 4.6959,
      "mean_token_accuracy": 0.23518263548612595,
      "num_tokens": 123799933.0,
      "step": 71370
    },
    {
      "entropy": 5.404820728302002,
      "epoch": 5.553199766582377,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000238499838200071,
      "loss": 4.6337,
      "mean_token_accuracy": 0.2423001766204834,
      "num_tokens": 123808950.0,
      "step": 71375
    },
    {
      "entropy": 5.415762042999267,
      "epoch": 5.553588795954094,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023847249000643185,
      "loss": 4.7147,
      "mean_token_accuracy": 0.23493777215480804,
      "num_tokens": 123817200.0,
      "step": 71380
    },
    {
      "entropy": 5.3805653095245365,
      "epoch": 5.553977825325812,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023844514236703952,
      "loss": 4.6634,
      "mean_token_accuracy": 0.23659783601760864,
      "num_tokens": 123825459.0,
      "step": 71385
    },
    {
      "entropy": 5.439620542526245,
      "epoch": 5.5543668546975296,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000238417795282309,
      "loss": 4.7989,
      "mean_token_accuracy": 0.22537942230701447,
      "num_tokens": 123834880.0,
      "step": 71390
    },
    {
      "entropy": 5.511213874816894,
      "epoch": 5.554755884069247,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023839044875265526,
      "loss": 4.7774,
      "mean_token_accuracy": 0.2272901192307472,
      "num_tokens": 123842954.0,
      "step": 71395
    },
    {
      "entropy": 5.422222852706909,
      "epoch": 5.555144913440965,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002383631027784931,
      "loss": 4.6899,
      "mean_token_accuracy": 0.22925954610109328,
      "num_tokens": 123851863.0,
      "step": 71400
    },
    {
      "entropy": 5.394540071487427,
      "epoch": 5.555533942812683,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00023833575736023754,
      "loss": 4.7594,
      "mean_token_accuracy": 0.22967920303344727,
      "num_tokens": 123860796.0,
      "step": 71405
    },
    {
      "entropy": 5.411272811889648,
      "epoch": 5.5559229721843995,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00023830841249830348,
      "loss": 4.6535,
      "mean_token_accuracy": 0.23939443677663802,
      "num_tokens": 123869197.0,
      "step": 71410
    },
    {
      "entropy": 5.433346843719482,
      "epoch": 5.556312001556117,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002382810681931059,
      "loss": 4.7671,
      "mean_token_accuracy": 0.2311771109700203,
      "num_tokens": 123879172.0,
      "step": 71415
    },
    {
      "entropy": 5.4561436653137205,
      "epoch": 5.556701030927835,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023825372444505967,
      "loss": 4.7223,
      "mean_token_accuracy": 0.22813272029161452,
      "num_tokens": 123887776.0,
      "step": 71420
    },
    {
      "entropy": 5.348863029479981,
      "epoch": 5.5570900602995525,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023822638125457974,
      "loss": 4.6752,
      "mean_token_accuracy": 0.23891769349575043,
      "num_tokens": 123896280.0,
      "step": 71425
    },
    {
      "entropy": 5.360417175292969,
      "epoch": 5.55747908967127,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00023819903862208093,
      "loss": 4.6961,
      "mean_token_accuracy": 0.2308642566204071,
      "num_tokens": 123905658.0,
      "step": 71430
    },
    {
      "entropy": 5.365459728240967,
      "epoch": 5.557868119042988,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002381716965479781,
      "loss": 4.6778,
      "mean_token_accuracy": 0.23635804355144502,
      "num_tokens": 123914580.0,
      "step": 71435
    },
    {
      "entropy": 5.3354442596435545,
      "epoch": 5.558257148414706,
      "grad_norm": 1.25,
      "learning_rate": 0.00023814435503268617,
      "loss": 4.6211,
      "mean_token_accuracy": 0.2492464080452919,
      "num_tokens": 123923713.0,
      "step": 71440
    },
    {
      "entropy": 5.340963125228882,
      "epoch": 5.558646177786423,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023811701407661995,
      "loss": 4.611,
      "mean_token_accuracy": 0.2370154470205307,
      "num_tokens": 123932887.0,
      "step": 71445
    },
    {
      "entropy": 5.368391418457032,
      "epoch": 5.55903520715814,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00023808967368019434,
      "loss": 4.7249,
      "mean_token_accuracy": 0.2351517766714096,
      "num_tokens": 123941148.0,
      "step": 71450
    },
    {
      "entropy": 5.412996625900268,
      "epoch": 5.559424236529858,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002380623338438242,
      "loss": 4.6922,
      "mean_token_accuracy": 0.23314565271139145,
      "num_tokens": 123950150.0,
      "step": 71455
    },
    {
      "entropy": 5.472694969177246,
      "epoch": 5.5598132659015755,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002380349945679244,
      "loss": 4.782,
      "mean_token_accuracy": 0.2208203062415123,
      "num_tokens": 123958066.0,
      "step": 71460
    },
    {
      "entropy": 5.396882009506226,
      "epoch": 5.560202295273293,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023800765585290967,
      "loss": 4.7038,
      "mean_token_accuracy": 0.23330495059490203,
      "num_tokens": 123966958.0,
      "step": 71465
    },
    {
      "entropy": 5.450788259506226,
      "epoch": 5.560591324645011,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023798031769919488,
      "loss": 4.6988,
      "mean_token_accuracy": 0.23637136220932006,
      "num_tokens": 123975513.0,
      "step": 71470
    },
    {
      "entropy": 5.378251361846924,
      "epoch": 5.560980354016729,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002379529801071949,
      "loss": 4.6389,
      "mean_token_accuracy": 0.23775823712348937,
      "num_tokens": 123983741.0,
      "step": 71475
    },
    {
      "entropy": 5.4551694869995115,
      "epoch": 5.561369383388445,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002379256430773244,
      "loss": 4.8105,
      "mean_token_accuracy": 0.22398041188716888,
      "num_tokens": 123992949.0,
      "step": 71480
    },
    {
      "entropy": 5.368277740478516,
      "epoch": 5.561758412760163,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023789830660999822,
      "loss": 4.6441,
      "mean_token_accuracy": 0.23273210376501083,
      "num_tokens": 124001583.0,
      "step": 71485
    },
    {
      "entropy": 5.449559688568115,
      "epoch": 5.562147442131881,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002378709707056313,
      "loss": 4.7326,
      "mean_token_accuracy": 0.23272522687911987,
      "num_tokens": 124010101.0,
      "step": 71490
    },
    {
      "entropy": 5.375598335266114,
      "epoch": 5.5625364715035985,
      "grad_norm": 1.171875,
      "learning_rate": 0.00023784363536463827,
      "loss": 4.7059,
      "mean_token_accuracy": 0.2336138054728508,
      "num_tokens": 124019198.0,
      "step": 71495
    },
    {
      "entropy": 5.391961145401001,
      "epoch": 5.562925500875316,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023781630058743392,
      "loss": 4.7197,
      "mean_token_accuracy": 0.2336716741323471,
      "num_tokens": 124027476.0,
      "step": 71500
    },
    {
      "entropy": 5.371859884262085,
      "epoch": 5.563314530247034,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002377889663744331,
      "loss": 4.6378,
      "mean_token_accuracy": 0.23823229670524598,
      "num_tokens": 124035902.0,
      "step": 71505
    },
    {
      "entropy": 5.345307779312134,
      "epoch": 5.563703559618752,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002377616327260504,
      "loss": 4.5935,
      "mean_token_accuracy": 0.24612261354923248,
      "num_tokens": 124044413.0,
      "step": 71510
    },
    {
      "entropy": 5.3350670337677,
      "epoch": 5.564092588990468,
      "grad_norm": 1.375,
      "learning_rate": 0.00023773429964270065,
      "loss": 4.721,
      "mean_token_accuracy": 0.23531905114650725,
      "num_tokens": 124053655.0,
      "step": 71515
    },
    {
      "entropy": 5.394557666778565,
      "epoch": 5.564481618362186,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002377069671247986,
      "loss": 4.7424,
      "mean_token_accuracy": 0.22312032133340837,
      "num_tokens": 124062694.0,
      "step": 71520
    },
    {
      "entropy": 5.3451872825622555,
      "epoch": 5.564870647733904,
      "grad_norm": 1.25,
      "learning_rate": 0.00023767963517275903,
      "loss": 4.6427,
      "mean_token_accuracy": 0.23714347630739213,
      "num_tokens": 124072437.0,
      "step": 71525
    },
    {
      "entropy": 5.43124794960022,
      "epoch": 5.5652596771056215,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002376523037869965,
      "loss": 4.7004,
      "mean_token_accuracy": 0.22921424210071564,
      "num_tokens": 124081151.0,
      "step": 71530
    },
    {
      "entropy": 5.411799192428589,
      "epoch": 5.565648706477339,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023762497296792585,
      "loss": 4.7428,
      "mean_token_accuracy": 0.2230130761861801,
      "num_tokens": 124089242.0,
      "step": 71535
    },
    {
      "entropy": 5.48844575881958,
      "epoch": 5.566037735849057,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023759764271596173,
      "loss": 4.7856,
      "mean_token_accuracy": 0.22754015326499938,
      "num_tokens": 124097594.0,
      "step": 71540
    },
    {
      "entropy": 5.3827889919281,
      "epoch": 5.566426765220774,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023757031303151888,
      "loss": 4.677,
      "mean_token_accuracy": 0.2411901205778122,
      "num_tokens": 124106425.0,
      "step": 71545
    },
    {
      "entropy": 5.39788703918457,
      "epoch": 5.566815794592491,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023754298391501195,
      "loss": 4.7137,
      "mean_token_accuracy": 0.23502082824707032,
      "num_tokens": 124115320.0,
      "step": 71550
    },
    {
      "entropy": 5.366327238082886,
      "epoch": 5.567204823964209,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002375156553668556,
      "loss": 4.6975,
      "mean_token_accuracy": 0.23709276616573333,
      "num_tokens": 124124689.0,
      "step": 71555
    },
    {
      "entropy": 5.350233173370361,
      "epoch": 5.567593853335927,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002374883273874645,
      "loss": 4.7211,
      "mean_token_accuracy": 0.22450742572546006,
      "num_tokens": 124133540.0,
      "step": 71560
    },
    {
      "entropy": 5.453421497344971,
      "epoch": 5.5679828827076445,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002374609999772534,
      "loss": 4.7404,
      "mean_token_accuracy": 0.22793408781290053,
      "num_tokens": 124142089.0,
      "step": 71565
    },
    {
      "entropy": 5.53364839553833,
      "epoch": 5.568371912079362,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023743367313663682,
      "loss": 4.8821,
      "mean_token_accuracy": 0.21759896576404572,
      "num_tokens": 124151214.0,
      "step": 71570
    },
    {
      "entropy": 5.429281377792359,
      "epoch": 5.56876094145108,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023740634686602945,
      "loss": 4.68,
      "mean_token_accuracy": 0.22854895740747452,
      "num_tokens": 124159814.0,
      "step": 71575
    },
    {
      "entropy": 5.4113121509552,
      "epoch": 5.569149970822798,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023737902116584598,
      "loss": 4.7984,
      "mean_token_accuracy": 0.2253943130373955,
      "num_tokens": 124168562.0,
      "step": 71580
    },
    {
      "entropy": 5.4698254585266115,
      "epoch": 5.569539000194514,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023735169603650104,
      "loss": 4.8763,
      "mean_token_accuracy": 0.21940825283527374,
      "num_tokens": 124177519.0,
      "step": 71585
    },
    {
      "entropy": 5.362502670288086,
      "epoch": 5.569928029566232,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023732437147840906,
      "loss": 4.6525,
      "mean_token_accuracy": 0.2379278361797333,
      "num_tokens": 124185292.0,
      "step": 71590
    },
    {
      "entropy": 5.438634824752808,
      "epoch": 5.57031705893795,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023729704749198482,
      "loss": 4.752,
      "mean_token_accuracy": 0.22713001519441606,
      "num_tokens": 124194599.0,
      "step": 71595
    },
    {
      "entropy": 5.42065110206604,
      "epoch": 5.5707060883096675,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023726972407764286,
      "loss": 4.6712,
      "mean_token_accuracy": 0.23801439702510835,
      "num_tokens": 124202455.0,
      "step": 71600
    },
    {
      "entropy": 5.441595792770386,
      "epoch": 5.571095117681385,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023724240123579782,
      "loss": 4.7847,
      "mean_token_accuracy": 0.23056374937295915,
      "num_tokens": 124211115.0,
      "step": 71605
    },
    {
      "entropy": 5.402417135238648,
      "epoch": 5.571484147053102,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023721507896686412,
      "loss": 4.7176,
      "mean_token_accuracy": 0.2333522006869316,
      "num_tokens": 124220518.0,
      "step": 71610
    },
    {
      "entropy": 5.305482053756714,
      "epoch": 5.57187317642482,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023718775727125658,
      "loss": 4.5749,
      "mean_token_accuracy": 0.2447166308760643,
      "num_tokens": 124229293.0,
      "step": 71615
    },
    {
      "entropy": 5.432966852188111,
      "epoch": 5.572262205796537,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023716043614938965,
      "loss": 4.7491,
      "mean_token_accuracy": 0.2401465192437172,
      "num_tokens": 124238584.0,
      "step": 71620
    },
    {
      "entropy": 5.333869171142578,
      "epoch": 5.572651235168255,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023713311560167783,
      "loss": 4.7028,
      "mean_token_accuracy": 0.23468175679445266,
      "num_tokens": 124246386.0,
      "step": 71625
    },
    {
      "entropy": 5.289809465408325,
      "epoch": 5.573040264539973,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023710579562853573,
      "loss": 4.6751,
      "mean_token_accuracy": 0.23592764437198638,
      "num_tokens": 124254707.0,
      "step": 71630
    },
    {
      "entropy": 5.340475797653198,
      "epoch": 5.5734292939116905,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023707847623037783,
      "loss": 4.6377,
      "mean_token_accuracy": 0.24429748207330704,
      "num_tokens": 124262325.0,
      "step": 71635
    },
    {
      "entropy": 5.2711773872375485,
      "epoch": 5.573818323283408,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023705115740761874,
      "loss": 4.6033,
      "mean_token_accuracy": 0.24414844512939454,
      "num_tokens": 124270117.0,
      "step": 71640
    },
    {
      "entropy": 5.417718362808228,
      "epoch": 5.574207352655126,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023702383916067294,
      "loss": 4.7576,
      "mean_token_accuracy": 0.2212275579571724,
      "num_tokens": 124279192.0,
      "step": 71645
    },
    {
      "entropy": 5.339358234405518,
      "epoch": 5.574596382026843,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023699652148995488,
      "loss": 4.6326,
      "mean_token_accuracy": 0.23752682358026506,
      "num_tokens": 124288000.0,
      "step": 71650
    },
    {
      "entropy": 5.426206588745117,
      "epoch": 5.57498541139856,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023696920439587917,
      "loss": 4.7955,
      "mean_token_accuracy": 0.22723033279180527,
      "num_tokens": 124297804.0,
      "step": 71655
    },
    {
      "entropy": 5.355615520477295,
      "epoch": 5.575374440770278,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023694188787886023,
      "loss": 4.6585,
      "mean_token_accuracy": 0.24313735514879226,
      "num_tokens": 124306212.0,
      "step": 71660
    },
    {
      "entropy": 5.3077844142913815,
      "epoch": 5.575763470141996,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002369145719393126,
      "loss": 4.6048,
      "mean_token_accuracy": 0.23837098032236098,
      "num_tokens": 124314362.0,
      "step": 71665
    },
    {
      "entropy": 5.442368602752685,
      "epoch": 5.5761524995137135,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002368872565776507,
      "loss": 4.8215,
      "mean_token_accuracy": 0.22159159183502197,
      "num_tokens": 124323577.0,
      "step": 71670
    },
    {
      "entropy": 5.456831598281861,
      "epoch": 5.576541528885431,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00023685994179428887,
      "loss": 4.7513,
      "mean_token_accuracy": 0.22676711827516555,
      "num_tokens": 124332914.0,
      "step": 71675
    },
    {
      "entropy": 5.417503356933594,
      "epoch": 5.576930558257148,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023683262758964185,
      "loss": 4.6902,
      "mean_token_accuracy": 0.2387118011713028,
      "num_tokens": 124341285.0,
      "step": 71680
    },
    {
      "entropy": 5.454626607894897,
      "epoch": 5.577319587628866,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023680531396412397,
      "loss": 4.8088,
      "mean_token_accuracy": 0.22863685786724092,
      "num_tokens": 124349229.0,
      "step": 71685
    },
    {
      "entropy": 5.367154121398926,
      "epoch": 5.577708617000583,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023677800091814966,
      "loss": 4.7301,
      "mean_token_accuracy": 0.23102815300226212,
      "num_tokens": 124357420.0,
      "step": 71690
    },
    {
      "entropy": 5.502103805541992,
      "epoch": 5.578097646372301,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002367506884521334,
      "loss": 4.8707,
      "mean_token_accuracy": 0.22316930294036866,
      "num_tokens": 124366085.0,
      "step": 71695
    },
    {
      "entropy": 5.415379381179809,
      "epoch": 5.578486675744019,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023672337656648951,
      "loss": 4.6871,
      "mean_token_accuracy": 0.23156578540802003,
      "num_tokens": 124375046.0,
      "step": 71700
    },
    {
      "entropy": 5.52207760810852,
      "epoch": 5.5788757051157365,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002366960652616324,
      "loss": 4.7698,
      "mean_token_accuracy": 0.2249188542366028,
      "num_tokens": 124383557.0,
      "step": 71705
    },
    {
      "entropy": 5.35969557762146,
      "epoch": 5.579264734487454,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002366687545379766,
      "loss": 4.6849,
      "mean_token_accuracy": 0.23182230442762375,
      "num_tokens": 124392912.0,
      "step": 71710
    },
    {
      "entropy": 5.3644969940185545,
      "epoch": 5.579653763859171,
      "grad_norm": 1.375,
      "learning_rate": 0.00023664144439593644,
      "loss": 4.6811,
      "mean_token_accuracy": 0.2413771629333496,
      "num_tokens": 124402136.0,
      "step": 71715
    },
    {
      "entropy": 5.465112400054932,
      "epoch": 5.580042793230889,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00023661413483592626,
      "loss": 4.8315,
      "mean_token_accuracy": 0.22710364013910295,
      "num_tokens": 124412144.0,
      "step": 71720
    },
    {
      "entropy": 5.4108020782470705,
      "epoch": 5.580431822602606,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023658682585836049,
      "loss": 4.6934,
      "mean_token_accuracy": 0.23729199320077896,
      "num_tokens": 124420379.0,
      "step": 71725
    },
    {
      "entropy": 5.409643983840942,
      "epoch": 5.580820851974324,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023655951746365356,
      "loss": 4.6562,
      "mean_token_accuracy": 0.2363213613629341,
      "num_tokens": 124428708.0,
      "step": 71730
    },
    {
      "entropy": 5.275825166702271,
      "epoch": 5.581209881346042,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023653220965221973,
      "loss": 4.6297,
      "mean_token_accuracy": 0.2399386629462242,
      "num_tokens": 124437403.0,
      "step": 71735
    },
    {
      "entropy": 5.38487401008606,
      "epoch": 5.5815989107177595,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023650490242447338,
      "loss": 4.7186,
      "mean_token_accuracy": 0.22153247892856598,
      "num_tokens": 124446853.0,
      "step": 71740
    },
    {
      "entropy": 5.438062381744385,
      "epoch": 5.581987940089476,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023647759578082883,
      "loss": 4.7055,
      "mean_token_accuracy": 0.2302195832133293,
      "num_tokens": 124456414.0,
      "step": 71745
    },
    {
      "entropy": 5.46565203666687,
      "epoch": 5.582376969461194,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023645028972170046,
      "loss": 4.6761,
      "mean_token_accuracy": 0.2342332735657692,
      "num_tokens": 124464976.0,
      "step": 71750
    },
    {
      "entropy": 5.39146671295166,
      "epoch": 5.582765998832912,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023642298424750264,
      "loss": 4.773,
      "mean_token_accuracy": 0.22837668359279634,
      "num_tokens": 124473568.0,
      "step": 71755
    },
    {
      "entropy": 5.319103193283081,
      "epoch": 5.583155028204629,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023639567935864958,
      "loss": 4.6966,
      "mean_token_accuracy": 0.2299676477909088,
      "num_tokens": 124482395.0,
      "step": 71760
    },
    {
      "entropy": 5.389864492416382,
      "epoch": 5.583544057576347,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002363683750555557,
      "loss": 4.6848,
      "mean_token_accuracy": 0.2371253177523613,
      "num_tokens": 124490444.0,
      "step": 71765
    },
    {
      "entropy": 5.343293714523315,
      "epoch": 5.583933086948065,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002363410713386352,
      "loss": 4.6595,
      "mean_token_accuracy": 0.2299022912979126,
      "num_tokens": 124499068.0,
      "step": 71770
    },
    {
      "entropy": 5.376804447174072,
      "epoch": 5.5843221163197825,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023631376820830248,
      "loss": 4.6835,
      "mean_token_accuracy": 0.23131352961063384,
      "num_tokens": 124508071.0,
      "step": 71775
    },
    {
      "entropy": 5.372738361358643,
      "epoch": 5.5847111456915,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002362864656649717,
      "loss": 4.7037,
      "mean_token_accuracy": 0.2388909563422203,
      "num_tokens": 124516689.0,
      "step": 71780
    },
    {
      "entropy": 5.431550598144531,
      "epoch": 5.585100175063217,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023625916370905714,
      "loss": 4.7596,
      "mean_token_accuracy": 0.2341059237718582,
      "num_tokens": 124524752.0,
      "step": 71785
    },
    {
      "entropy": 5.4365678310394285,
      "epoch": 5.585489204434935,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023623186234097315,
      "loss": 4.7298,
      "mean_token_accuracy": 0.23439615815877915,
      "num_tokens": 124533605.0,
      "step": 71790
    },
    {
      "entropy": 5.383245897293091,
      "epoch": 5.585878233806652,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00023620456156113397,
      "loss": 4.7445,
      "mean_token_accuracy": 0.23875697702169418,
      "num_tokens": 124542910.0,
      "step": 71795
    },
    {
      "entropy": 5.446283149719238,
      "epoch": 5.58626726317837,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023617726136995377,
      "loss": 4.7145,
      "mean_token_accuracy": 0.23030093759298326,
      "num_tokens": 124551880.0,
      "step": 71800
    },
    {
      "entropy": 5.368617916107178,
      "epoch": 5.586656292550088,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023614996176784687,
      "loss": 4.6293,
      "mean_token_accuracy": 0.23986468613147735,
      "num_tokens": 124560455.0,
      "step": 71805
    },
    {
      "entropy": 5.377729654312134,
      "epoch": 5.5870453219218055,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023612266275522737,
      "loss": 4.6251,
      "mean_token_accuracy": 0.2405630975961685,
      "num_tokens": 124568925.0,
      "step": 71810
    },
    {
      "entropy": 5.40814242362976,
      "epoch": 5.587434351293522,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023609536433250966,
      "loss": 4.7692,
      "mean_token_accuracy": 0.227486352622509,
      "num_tokens": 124577162.0,
      "step": 71815
    },
    {
      "entropy": 5.4459450244903564,
      "epoch": 5.58782338066524,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023606806650010786,
      "loss": 4.7416,
      "mean_token_accuracy": 0.23575246334075928,
      "num_tokens": 124584892.0,
      "step": 71820
    },
    {
      "entropy": 5.369396591186524,
      "epoch": 5.588212410036958,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023604076925843622,
      "loss": 4.6529,
      "mean_token_accuracy": 0.2368028059601784,
      "num_tokens": 124594646.0,
      "step": 71825
    },
    {
      "entropy": 5.427077770233154,
      "epoch": 5.588601439408675,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023601347260790884,
      "loss": 4.7439,
      "mean_token_accuracy": 0.23060963600873946,
      "num_tokens": 124602047.0,
      "step": 71830
    },
    {
      "entropy": 5.392776203155518,
      "epoch": 5.588990468780393,
      "grad_norm": 1.171875,
      "learning_rate": 0.00023598617654894006,
      "loss": 4.7152,
      "mean_token_accuracy": 0.23863882273435594,
      "num_tokens": 124610783.0,
      "step": 71835
    },
    {
      "entropy": 5.412454271316529,
      "epoch": 5.589379498152111,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023595888108194385,
      "loss": 4.7489,
      "mean_token_accuracy": 0.23145599514245987,
      "num_tokens": 124619733.0,
      "step": 71840
    },
    {
      "entropy": 5.405062055587768,
      "epoch": 5.5897685275238285,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002359315862073345,
      "loss": 4.7231,
      "mean_token_accuracy": 0.23359524458646774,
      "num_tokens": 124629245.0,
      "step": 71845
    },
    {
      "entropy": 5.407244348526001,
      "epoch": 5.590157556895545,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002359042919255262,
      "loss": 4.6842,
      "mean_token_accuracy": 0.23674705177545546,
      "num_tokens": 124637542.0,
      "step": 71850
    },
    {
      "entropy": 5.383553600311279,
      "epoch": 5.590546586267263,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023587699823693304,
      "loss": 4.6593,
      "mean_token_accuracy": 0.23010266721248626,
      "num_tokens": 124645891.0,
      "step": 71855
    },
    {
      "entropy": 5.344635152816773,
      "epoch": 5.590935615638981,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023584970514196912,
      "loss": 4.6527,
      "mean_token_accuracy": 0.2364268496632576,
      "num_tokens": 124656210.0,
      "step": 71860
    },
    {
      "entropy": 5.449120807647705,
      "epoch": 5.591324645010698,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023582241264104864,
      "loss": 4.775,
      "mean_token_accuracy": 0.23100472390651702,
      "num_tokens": 124663854.0,
      "step": 71865
    },
    {
      "entropy": 5.484356498718261,
      "epoch": 5.591713674382416,
      "grad_norm": 1.140625,
      "learning_rate": 0.00023579512073458565,
      "loss": 4.8036,
      "mean_token_accuracy": 0.22345139533281327,
      "num_tokens": 124672339.0,
      "step": 71870
    },
    {
      "entropy": 5.3695759773254395,
      "epoch": 5.592102703754134,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023576782942299424,
      "loss": 4.7115,
      "mean_token_accuracy": 0.22701380848884584,
      "num_tokens": 124682231.0,
      "step": 71875
    },
    {
      "entropy": 5.373459005355835,
      "epoch": 5.592491733125851,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002357405387066886,
      "loss": 4.7173,
      "mean_token_accuracy": 0.23830377459526061,
      "num_tokens": 124691140.0,
      "step": 71880
    },
    {
      "entropy": 5.411927938461304,
      "epoch": 5.592880762497568,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002357132485860829,
      "loss": 4.7564,
      "mean_token_accuracy": 0.22921765297651292,
      "num_tokens": 124700388.0,
      "step": 71885
    },
    {
      "entropy": 5.496766138076782,
      "epoch": 5.593269791869286,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023568595906159107,
      "loss": 4.7573,
      "mean_token_accuracy": 0.23379006683826448,
      "num_tokens": 124709145.0,
      "step": 71890
    },
    {
      "entropy": 5.496967601776123,
      "epoch": 5.593658821241004,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023565867013362725,
      "loss": 4.7532,
      "mean_token_accuracy": 0.2251088798046112,
      "num_tokens": 124717463.0,
      "step": 71895
    },
    {
      "entropy": 5.4037847995758055,
      "epoch": 5.594047850612721,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023563138180260552,
      "loss": 4.8253,
      "mean_token_accuracy": 0.22761937975883484,
      "num_tokens": 124725812.0,
      "step": 71900
    },
    {
      "entropy": 5.402643775939941,
      "epoch": 5.594436879984439,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023560409406893986,
      "loss": 4.704,
      "mean_token_accuracy": 0.23644114136695862,
      "num_tokens": 124735210.0,
      "step": 71905
    },
    {
      "entropy": 5.443668699264526,
      "epoch": 5.594825909356157,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023557680693304435,
      "loss": 4.7585,
      "mean_token_accuracy": 0.23380009531974794,
      "num_tokens": 124744863.0,
      "step": 71910
    },
    {
      "entropy": 5.424369144439697,
      "epoch": 5.595214938727874,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023554952039533307,
      "loss": 4.7035,
      "mean_token_accuracy": 0.2284855216741562,
      "num_tokens": 124753519.0,
      "step": 71915
    },
    {
      "entropy": 5.36265230178833,
      "epoch": 5.595603968099591,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00023552223445622006,
      "loss": 4.6531,
      "mean_token_accuracy": 0.23201710879802703,
      "num_tokens": 124762313.0,
      "step": 71920
    },
    {
      "entropy": 5.352631616592407,
      "epoch": 5.595992997471309,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023549494911611924,
      "loss": 4.6932,
      "mean_token_accuracy": 0.24650082141160964,
      "num_tokens": 124771101.0,
      "step": 71925
    },
    {
      "entropy": 5.434217405319214,
      "epoch": 5.596382026843027,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00023546766437544477,
      "loss": 4.7672,
      "mean_token_accuracy": 0.2260288655757904,
      "num_tokens": 124780342.0,
      "step": 71930
    },
    {
      "entropy": 5.3966950416564945,
      "epoch": 5.596771056214744,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023544038023461057,
      "loss": 4.7061,
      "mean_token_accuracy": 0.23255095034837722,
      "num_tokens": 124788833.0,
      "step": 71935
    },
    {
      "entropy": 5.37835431098938,
      "epoch": 5.597160085586462,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002354130966940306,
      "loss": 4.6805,
      "mean_token_accuracy": 0.23502848893404008,
      "num_tokens": 124798247.0,
      "step": 71940
    },
    {
      "entropy": 5.415205001831055,
      "epoch": 5.597549114958179,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00023538581375411876,
      "loss": 4.7104,
      "mean_token_accuracy": 0.2310178682208061,
      "num_tokens": 124806800.0,
      "step": 71945
    },
    {
      "entropy": 5.48044056892395,
      "epoch": 5.597938144329897,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023535853141528923,
      "loss": 4.7611,
      "mean_token_accuracy": 0.23090974539518355,
      "num_tokens": 124815068.0,
      "step": 71950
    },
    {
      "entropy": 5.4402683734893795,
      "epoch": 5.598327173701614,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00023533124967795594,
      "loss": 4.7862,
      "mean_token_accuracy": 0.23309133052825928,
      "num_tokens": 124823660.0,
      "step": 71955
    },
    {
      "entropy": 5.288178205490112,
      "epoch": 5.598716203073332,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00023530396854253276,
      "loss": 4.5726,
      "mean_token_accuracy": 0.24579740762710572,
      "num_tokens": 124832409.0,
      "step": 71960
    },
    {
      "entropy": 5.434702062606812,
      "epoch": 5.59910523244505,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023527668800943374,
      "loss": 4.898,
      "mean_token_accuracy": 0.2217651829123497,
      "num_tokens": 124841691.0,
      "step": 71965
    },
    {
      "entropy": 5.475157785415649,
      "epoch": 5.599494261816767,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002352494080790727,
      "loss": 4.8094,
      "mean_token_accuracy": 0.22080370634794236,
      "num_tokens": 124850544.0,
      "step": 71970
    },
    {
      "entropy": 5.341800737380981,
      "epoch": 5.599883291188485,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002352221287518636,
      "loss": 4.5794,
      "mean_token_accuracy": 0.23962318301200866,
      "num_tokens": 124858516.0,
      "step": 71975
    },
    {
      "entropy": 5.487163352966308,
      "epoch": 5.600272320560203,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023519485002822038,
      "loss": 4.8442,
      "mean_token_accuracy": 0.21879047453403472,
      "num_tokens": 124867779.0,
      "step": 71980
    },
    {
      "entropy": 5.402089786529541,
      "epoch": 5.6006613499319196,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023516757190855705,
      "loss": 4.6863,
      "mean_token_accuracy": 0.23127496838569642,
      "num_tokens": 124877039.0,
      "step": 71985
    },
    {
      "entropy": 5.384178161621094,
      "epoch": 5.601050379303637,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023514029439328734,
      "loss": 4.6933,
      "mean_token_accuracy": 0.23888282775878905,
      "num_tokens": 124885743.0,
      "step": 71990
    },
    {
      "entropy": 5.370058059692383,
      "epoch": 5.601439408675355,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002351130174828252,
      "loss": 4.6477,
      "mean_token_accuracy": 0.23281227499246598,
      "num_tokens": 124894814.0,
      "step": 71995
    },
    {
      "entropy": 5.398576927185059,
      "epoch": 5.601828438047073,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023508574117758453,
      "loss": 4.7229,
      "mean_token_accuracy": 0.2389881432056427,
      "num_tokens": 124903261.0,
      "step": 72000
    },
    {
      "epoch": 5.601828438047073,
      "eval_entropy": 5.119988270742143,
      "eval_loss": 4.893167495727539,
      "eval_mean_token_accuracy": 0.23082889535027817,
      "eval_num_tokens": 124903261.0,
      "eval_runtime": 29.239,
      "eval_samples_per_second": 1421.319,
      "eval_steps_per_second": 177.673,
      "step": 72000
    },
    {
      "entropy": 5.347326612472534,
      "epoch": 5.60221746741879,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023505846547797928,
      "loss": 4.7149,
      "mean_token_accuracy": 0.22731718122959138,
      "num_tokens": 124912108.0,
      "step": 72005
    },
    {
      "entropy": 5.4922833919525145,
      "epoch": 5.602606496790508,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023503119038442312,
      "loss": 4.8176,
      "mean_token_accuracy": 0.22356500178575517,
      "num_tokens": 124920808.0,
      "step": 72010
    },
    {
      "entropy": 5.458099269866944,
      "epoch": 5.602995526162225,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002350039158973301,
      "loss": 4.6974,
      "mean_token_accuracy": 0.24435170441865922,
      "num_tokens": 124929920.0,
      "step": 72015
    },
    {
      "entropy": 5.4221728324890135,
      "epoch": 5.6033845555339425,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000234976642017114,
      "loss": 4.7212,
      "mean_token_accuracy": 0.23643734753131868,
      "num_tokens": 124938905.0,
      "step": 72020
    },
    {
      "entropy": 5.448312377929687,
      "epoch": 5.60377358490566,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023494936874418864,
      "loss": 4.7305,
      "mean_token_accuracy": 0.23173199445009232,
      "num_tokens": 124947413.0,
      "step": 72025
    },
    {
      "entropy": 5.420518398284912,
      "epoch": 5.604162614277378,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023492209607896793,
      "loss": 4.7379,
      "mean_token_accuracy": 0.22735135853290558,
      "num_tokens": 124955548.0,
      "step": 72030
    },
    {
      "entropy": 5.448617458343506,
      "epoch": 5.604551643649096,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023489482402186558,
      "loss": 4.749,
      "mean_token_accuracy": 0.23065790683031082,
      "num_tokens": 124963821.0,
      "step": 72035
    },
    {
      "entropy": 5.375195646286011,
      "epoch": 5.604940673020813,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002348675525732954,
      "loss": 4.6593,
      "mean_token_accuracy": 0.2345134809613228,
      "num_tokens": 124972436.0,
      "step": 72040
    },
    {
      "entropy": 5.385494565963745,
      "epoch": 5.605329702392531,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023484028173367127,
      "loss": 4.7249,
      "mean_token_accuracy": 0.23292153477668762,
      "num_tokens": 124981802.0,
      "step": 72045
    },
    {
      "entropy": 5.3511707305908205,
      "epoch": 5.605718731764248,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000234813011503407,
      "loss": 4.6965,
      "mean_token_accuracy": 0.23428622633218765,
      "num_tokens": 124990780.0,
      "step": 72050
    },
    {
      "entropy": 5.413948583602905,
      "epoch": 5.6061077611359655,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00023478574188291623,
      "loss": 4.7352,
      "mean_token_accuracy": 0.22952142506837844,
      "num_tokens": 124999892.0,
      "step": 72055
    },
    {
      "entropy": 5.489055013656616,
      "epoch": 5.606496790507683,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023475847287261282,
      "loss": 4.803,
      "mean_token_accuracy": 0.2266104504466057,
      "num_tokens": 125009319.0,
      "step": 72060
    },
    {
      "entropy": 5.4248046875,
      "epoch": 5.606885819879401,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023473120447291057,
      "loss": 4.6789,
      "mean_token_accuracy": 0.23381361961364747,
      "num_tokens": 125017825.0,
      "step": 72065
    },
    {
      "entropy": 5.4662433624267575,
      "epoch": 5.607274849251119,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002347039366842232,
      "loss": 4.8408,
      "mean_token_accuracy": 0.22155664563179017,
      "num_tokens": 125026171.0,
      "step": 72070
    },
    {
      "entropy": 5.395715951919556,
      "epoch": 5.607663878622836,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023467666950696442,
      "loss": 4.721,
      "mean_token_accuracy": 0.22939751148223878,
      "num_tokens": 125034931.0,
      "step": 72075
    },
    {
      "entropy": 5.431500577926636,
      "epoch": 5.608052907994553,
      "grad_norm": 1.25,
      "learning_rate": 0.00023464940294154796,
      "loss": 4.7361,
      "mean_token_accuracy": 0.2296561911702156,
      "num_tokens": 125044162.0,
      "step": 72080
    },
    {
      "entropy": 5.465833520889282,
      "epoch": 5.608441937366271,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023462213698838762,
      "loss": 4.7794,
      "mean_token_accuracy": 0.22939728200435638,
      "num_tokens": 125052081.0,
      "step": 72085
    },
    {
      "entropy": 5.416418600082397,
      "epoch": 5.6088309667379885,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00023459487164789717,
      "loss": 4.6817,
      "mean_token_accuracy": 0.24313343316316605,
      "num_tokens": 125061546.0,
      "step": 72090
    },
    {
      "entropy": 5.411910104751587,
      "epoch": 5.609219996109706,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002345676069204901,
      "loss": 4.7677,
      "mean_token_accuracy": 0.23238570988178253,
      "num_tokens": 125070861.0,
      "step": 72095
    },
    {
      "entropy": 5.381026935577393,
      "epoch": 5.609609025481424,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023454034280658036,
      "loss": 4.6155,
      "mean_token_accuracy": 0.24289755821228026,
      "num_tokens": 125079008.0,
      "step": 72100
    },
    {
      "entropy": 5.451247835159302,
      "epoch": 5.609998054853142,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023451307930658145,
      "loss": 4.7716,
      "mean_token_accuracy": 0.23669869601726531,
      "num_tokens": 125088028.0,
      "step": 72105
    },
    {
      "entropy": 5.413648366928101,
      "epoch": 5.610387084224859,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023448581642090716,
      "loss": 4.8091,
      "mean_token_accuracy": 0.22671463638544082,
      "num_tokens": 125096076.0,
      "step": 72110
    },
    {
      "entropy": 5.399301290512085,
      "epoch": 5.610776113596577,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023445855414997108,
      "loss": 4.7597,
      "mean_token_accuracy": 0.22801882773637772,
      "num_tokens": 125104986.0,
      "step": 72115
    },
    {
      "entropy": 5.402620315551758,
      "epoch": 5.611165142968294,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023443129249418694,
      "loss": 4.6437,
      "mean_token_accuracy": 0.23342441469430925,
      "num_tokens": 125113201.0,
      "step": 72120
    },
    {
      "entropy": 5.364388036727905,
      "epoch": 5.6115541723400115,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023440403145396838,
      "loss": 4.6208,
      "mean_token_accuracy": 0.24096485525369643,
      "num_tokens": 125122474.0,
      "step": 72125
    },
    {
      "entropy": 5.406528949737549,
      "epoch": 5.611943201711729,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023437677102972903,
      "loss": 4.7171,
      "mean_token_accuracy": 0.23398844450712203,
      "num_tokens": 125131151.0,
      "step": 72130
    },
    {
      "entropy": 5.344950389862061,
      "epoch": 5.612332231083447,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023434951122188247,
      "loss": 4.6778,
      "mean_token_accuracy": 0.2397386685013771,
      "num_tokens": 125140219.0,
      "step": 72135
    },
    {
      "entropy": 5.36824221611023,
      "epoch": 5.612721260455165,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002343222520308423,
      "loss": 4.7256,
      "mean_token_accuracy": 0.22888956367969512,
      "num_tokens": 125147642.0,
      "step": 72140
    },
    {
      "entropy": 5.47944746017456,
      "epoch": 5.613110289826882,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002342949934570223,
      "loss": 4.7851,
      "mean_token_accuracy": 0.2188117817044258,
      "num_tokens": 125156504.0,
      "step": 72145
    },
    {
      "entropy": 5.43713002204895,
      "epoch": 5.613499319198599,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023426773550083602,
      "loss": 4.6407,
      "mean_token_accuracy": 0.23788010478019714,
      "num_tokens": 125164400.0,
      "step": 72150
    },
    {
      "entropy": 5.438969707489013,
      "epoch": 5.613888348570317,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00023424047816269705,
      "loss": 4.7717,
      "mean_token_accuracy": 0.23499398827552795,
      "num_tokens": 125173255.0,
      "step": 72155
    },
    {
      "entropy": 5.3402317523956295,
      "epoch": 5.6142773779420345,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00023421322144301889,
      "loss": 4.6901,
      "mean_token_accuracy": 0.23190366178750993,
      "num_tokens": 125182403.0,
      "step": 72160
    },
    {
      "entropy": 5.443404960632324,
      "epoch": 5.614666407313752,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002341859653422152,
      "loss": 4.7909,
      "mean_token_accuracy": 0.22575951665639876,
      "num_tokens": 125190329.0,
      "step": 72165
    },
    {
      "entropy": 5.419959974288941,
      "epoch": 5.61505543668547,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023415870986069948,
      "loss": 4.6554,
      "mean_token_accuracy": 0.23340196311473846,
      "num_tokens": 125198580.0,
      "step": 72170
    },
    {
      "entropy": 5.366064405441284,
      "epoch": 5.615444466057188,
      "grad_norm": 1.140625,
      "learning_rate": 0.00023413145499888533,
      "loss": 4.6498,
      "mean_token_accuracy": 0.24136442691087723,
      "num_tokens": 125207528.0,
      "step": 72175
    },
    {
      "entropy": 5.406661415100098,
      "epoch": 5.615833495428905,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023410420075718635,
      "loss": 4.733,
      "mean_token_accuracy": 0.23382801413536072,
      "num_tokens": 125216060.0,
      "step": 72180
    },
    {
      "entropy": 5.441198492050171,
      "epoch": 5.616222524800622,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023407694713601602,
      "loss": 4.7506,
      "mean_token_accuracy": 0.2276070937514305,
      "num_tokens": 125224805.0,
      "step": 72185
    },
    {
      "entropy": 5.372823047637939,
      "epoch": 5.61661155417234,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023404969413578787,
      "loss": 4.6239,
      "mean_token_accuracy": 0.23438760936260222,
      "num_tokens": 125233046.0,
      "step": 72190
    },
    {
      "entropy": 5.381484174728394,
      "epoch": 5.6170005835440575,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002340224417569154,
      "loss": 4.6655,
      "mean_token_accuracy": 0.23331774324178695,
      "num_tokens": 125241601.0,
      "step": 72195
    },
    {
      "entropy": 5.393099451065064,
      "epoch": 5.617389612915775,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002339951899998122,
      "loss": 4.7189,
      "mean_token_accuracy": 0.22967860400676726,
      "num_tokens": 125250225.0,
      "step": 72200
    },
    {
      "entropy": 5.50399227142334,
      "epoch": 5.617778642287493,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023396793886489177,
      "loss": 4.7429,
      "mean_token_accuracy": 0.23074330687522887,
      "num_tokens": 125259110.0,
      "step": 72205
    },
    {
      "entropy": 5.478127527236938,
      "epoch": 5.618167671659211,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023394068835256738,
      "loss": 4.724,
      "mean_token_accuracy": 0.23487377911806107,
      "num_tokens": 125267070.0,
      "step": 72210
    },
    {
      "entropy": 5.3454516410827635,
      "epoch": 5.618556701030927,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023391343846325286,
      "loss": 4.7012,
      "mean_token_accuracy": 0.23785390257835387,
      "num_tokens": 125276007.0,
      "step": 72215
    },
    {
      "entropy": 5.488029193878174,
      "epoch": 5.618945730402645,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023388618919736143,
      "loss": 4.7806,
      "mean_token_accuracy": 0.23102326244115828,
      "num_tokens": 125284540.0,
      "step": 72220
    },
    {
      "entropy": 5.392019605636596,
      "epoch": 5.619334759774363,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023385894055530666,
      "loss": 4.7617,
      "mean_token_accuracy": 0.2277224063873291,
      "num_tokens": 125293522.0,
      "step": 72225
    },
    {
      "entropy": 5.396182584762573,
      "epoch": 5.6197237891460805,
      "grad_norm": 1.140625,
      "learning_rate": 0.00023383169253750203,
      "loss": 4.72,
      "mean_token_accuracy": 0.2376866579055786,
      "num_tokens": 125301947.0,
      "step": 72230
    },
    {
      "entropy": 5.4437408447265625,
      "epoch": 5.620112818517798,
      "grad_norm": 1.171875,
      "learning_rate": 0.00023380444514436093,
      "loss": 4.7323,
      "mean_token_accuracy": 0.22811394184827805,
      "num_tokens": 125310555.0,
      "step": 72235
    },
    {
      "entropy": 5.360163450241089,
      "epoch": 5.620501847889516,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002337771983762968,
      "loss": 4.6711,
      "mean_token_accuracy": 0.23130515664815904,
      "num_tokens": 125319077.0,
      "step": 72240
    },
    {
      "entropy": 5.401445055007935,
      "epoch": 5.620890877261234,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00023374995223372308,
      "loss": 4.719,
      "mean_token_accuracy": 0.23090377300977707,
      "num_tokens": 125327732.0,
      "step": 72245
    },
    {
      "entropy": 5.435986566543579,
      "epoch": 5.62127990663295,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023372270671705313,
      "loss": 4.7216,
      "mean_token_accuracy": 0.23113974928855896,
      "num_tokens": 125337098.0,
      "step": 72250
    },
    {
      "entropy": 5.3828071594238285,
      "epoch": 5.621668936004668,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023369546182670044,
      "loss": 4.7227,
      "mean_token_accuracy": 0.2283797562122345,
      "num_tokens": 125344848.0,
      "step": 72255
    },
    {
      "entropy": 5.415948009490966,
      "epoch": 5.622057965376386,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023366821756307837,
      "loss": 4.7351,
      "mean_token_accuracy": 0.22856331318616868,
      "num_tokens": 125354057.0,
      "step": 72260
    },
    {
      "entropy": 5.488615942001343,
      "epoch": 5.6224469947481035,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023364097392660033,
      "loss": 4.8012,
      "mean_token_accuracy": 0.22498073875904084,
      "num_tokens": 125362245.0,
      "step": 72265
    },
    {
      "entropy": 5.425380802154541,
      "epoch": 5.622836024119821,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023361373091767963,
      "loss": 4.66,
      "mean_token_accuracy": 0.2359096437692642,
      "num_tokens": 125371380.0,
      "step": 72270
    },
    {
      "entropy": 5.331160306930542,
      "epoch": 5.623225053491539,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002335864885367297,
      "loss": 4.6421,
      "mean_token_accuracy": 0.24448014199733734,
      "num_tokens": 125380554.0,
      "step": 72275
    },
    {
      "entropy": 5.428044986724854,
      "epoch": 5.623614082863256,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002335592467841639,
      "loss": 4.7804,
      "mean_token_accuracy": 0.2330235332250595,
      "num_tokens": 125389612.0,
      "step": 72280
    },
    {
      "entropy": 5.381367826461792,
      "epoch": 5.624003112234973,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023353200566039557,
      "loss": 4.701,
      "mean_token_accuracy": 0.2307939663529396,
      "num_tokens": 125397973.0,
      "step": 72285
    },
    {
      "entropy": 5.482031774520874,
      "epoch": 5.624392141606691,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023350476516583807,
      "loss": 4.8563,
      "mean_token_accuracy": 0.22164613008499146,
      "num_tokens": 125407380.0,
      "step": 72290
    },
    {
      "entropy": 5.3957549095153805,
      "epoch": 5.624781170978409,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023347752530090472,
      "loss": 4.6875,
      "mean_token_accuracy": 0.2306345134973526,
      "num_tokens": 125415917.0,
      "step": 72295
    },
    {
      "entropy": 5.411555814743042,
      "epoch": 5.6251702003501265,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002334502860660088,
      "loss": 4.647,
      "mean_token_accuracy": 0.23868130296468734,
      "num_tokens": 125424942.0,
      "step": 72300
    },
    {
      "entropy": 5.38042049407959,
      "epoch": 5.625559229721844,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00023342304746156367,
      "loss": 4.6393,
      "mean_token_accuracy": 0.23487719297409057,
      "num_tokens": 125434172.0,
      "step": 72305
    },
    {
      "entropy": 5.311665964126587,
      "epoch": 5.625948259093562,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023339580948798262,
      "loss": 4.6043,
      "mean_token_accuracy": 0.24396300166845322,
      "num_tokens": 125443509.0,
      "step": 72310
    },
    {
      "entropy": 5.409530258178711,
      "epoch": 5.62633728846528,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00023336857214567897,
      "loss": 4.7282,
      "mean_token_accuracy": 0.22486346513032912,
      "num_tokens": 125452556.0,
      "step": 72315
    },
    {
      "entropy": 5.372827529907227,
      "epoch": 5.626726317836996,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023334133543506598,
      "loss": 4.6717,
      "mean_token_accuracy": 0.23830690532922744,
      "num_tokens": 125461466.0,
      "step": 72320
    },
    {
      "entropy": 5.462872886657715,
      "epoch": 5.627115347208714,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002333140993565569,
      "loss": 4.8534,
      "mean_token_accuracy": 0.22249109745025636,
      "num_tokens": 125469212.0,
      "step": 72325
    },
    {
      "entropy": 5.431930780410767,
      "epoch": 5.627504376580432,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023328686391056497,
      "loss": 4.6652,
      "mean_token_accuracy": 0.23590313494205475,
      "num_tokens": 125478099.0,
      "step": 72330
    },
    {
      "entropy": 5.554479122161865,
      "epoch": 5.6278934059521495,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023325962909750349,
      "loss": 4.85,
      "mean_token_accuracy": 0.22178691029548644,
      "num_tokens": 125486617.0,
      "step": 72335
    },
    {
      "entropy": 5.41433801651001,
      "epoch": 5.628282435323867,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023323239491778564,
      "loss": 4.718,
      "mean_token_accuracy": 0.23739317804574966,
      "num_tokens": 125494323.0,
      "step": 72340
    },
    {
      "entropy": 5.420104598999023,
      "epoch": 5.628671464695585,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023320516137182484,
      "loss": 4.7562,
      "mean_token_accuracy": 0.2273334965109825,
      "num_tokens": 125502628.0,
      "step": 72345
    },
    {
      "entropy": 5.426651334762573,
      "epoch": 5.629060494067302,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00023317792846003416,
      "loss": 4.7681,
      "mean_token_accuracy": 0.22136711776256562,
      "num_tokens": 125511697.0,
      "step": 72350
    },
    {
      "entropy": 5.4469069480896,
      "epoch": 5.629449523439019,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00023315069618282686,
      "loss": 4.7847,
      "mean_token_accuracy": 0.2292821764945984,
      "num_tokens": 125520688.0,
      "step": 72355
    },
    {
      "entropy": 5.475170469284057,
      "epoch": 5.629838552810737,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002331234645406161,
      "loss": 4.78,
      "mean_token_accuracy": 0.22754317671060562,
      "num_tokens": 125528972.0,
      "step": 72360
    },
    {
      "entropy": 5.431169891357422,
      "epoch": 5.630227582182455,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023309623353381516,
      "loss": 4.6969,
      "mean_token_accuracy": 0.23814399987459184,
      "num_tokens": 125536794.0,
      "step": 72365
    },
    {
      "entropy": 5.402341079711914,
      "epoch": 5.6306166115541725,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023306900316283713,
      "loss": 4.79,
      "mean_token_accuracy": 0.22258255779743194,
      "num_tokens": 125546692.0,
      "step": 72370
    },
    {
      "entropy": 5.475697994232178,
      "epoch": 5.63100564092589,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002330417734280953,
      "loss": 4.7344,
      "mean_token_accuracy": 0.22960766553878784,
      "num_tokens": 125555113.0,
      "step": 72375
    },
    {
      "entropy": 5.400759077072143,
      "epoch": 5.631394670297608,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023301454433000268,
      "loss": 4.6594,
      "mean_token_accuracy": 0.2413504660129547,
      "num_tokens": 125563201.0,
      "step": 72380
    },
    {
      "entropy": 5.358392906188965,
      "epoch": 5.631783699669325,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002329873158689726,
      "loss": 4.5569,
      "mean_token_accuracy": 0.24572102576494217,
      "num_tokens": 125570523.0,
      "step": 72385
    },
    {
      "entropy": 5.339574766159058,
      "epoch": 5.632172729041042,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023296008804541812,
      "loss": 4.7489,
      "mean_token_accuracy": 0.2302490234375,
      "num_tokens": 125578553.0,
      "step": 72390
    },
    {
      "entropy": 5.410540580749512,
      "epoch": 5.63256175841276,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002329328608597524,
      "loss": 4.7398,
      "mean_token_accuracy": 0.23531377613544463,
      "num_tokens": 125587987.0,
      "step": 72395
    },
    {
      "entropy": 5.412122058868408,
      "epoch": 5.632950787784478,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023290563431238852,
      "loss": 4.7039,
      "mean_token_accuracy": 0.23779755234718322,
      "num_tokens": 125596187.0,
      "step": 72400
    },
    {
      "entropy": 5.470497417449951,
      "epoch": 5.6333398171561955,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002328784084037397,
      "loss": 4.8376,
      "mean_token_accuracy": 0.21667806357145308,
      "num_tokens": 125604481.0,
      "step": 72405
    },
    {
      "entropy": 5.437833404541015,
      "epoch": 5.633728846527913,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023285118313421887,
      "loss": 4.7203,
      "mean_token_accuracy": 0.2265430673956871,
      "num_tokens": 125612755.0,
      "step": 72410
    },
    {
      "entropy": 5.424828815460205,
      "epoch": 5.63411787589963,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002328239585042393,
      "loss": 4.702,
      "mean_token_accuracy": 0.24112111181020737,
      "num_tokens": 125621112.0,
      "step": 72415
    },
    {
      "entropy": 5.4787229061126705,
      "epoch": 5.634506905271348,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023279673451421406,
      "loss": 4.8012,
      "mean_token_accuracy": 0.22654237300157548,
      "num_tokens": 125630215.0,
      "step": 72420
    },
    {
      "entropy": 5.344723796844482,
      "epoch": 5.634895934643065,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023276951116455626,
      "loss": 4.6849,
      "mean_token_accuracy": 0.23177414238452912,
      "num_tokens": 125639187.0,
      "step": 72425
    },
    {
      "entropy": 5.425217199325561,
      "epoch": 5.635284964014783,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023274228845567886,
      "loss": 4.7025,
      "mean_token_accuracy": 0.23032159805297853,
      "num_tokens": 125647663.0,
      "step": 72430
    },
    {
      "entropy": 5.421003580093384,
      "epoch": 5.635673993386501,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023271506638799494,
      "loss": 4.7723,
      "mean_token_accuracy": 0.2195465698838234,
      "num_tokens": 125656184.0,
      "step": 72435
    },
    {
      "entropy": 5.346044111251831,
      "epoch": 5.6360630227582185,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00023268784496191758,
      "loss": 4.6634,
      "mean_token_accuracy": 0.2411385342478752,
      "num_tokens": 125664668.0,
      "step": 72440
    },
    {
      "entropy": 5.394611978530884,
      "epoch": 5.636452052129936,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023266062417785988,
      "loss": 4.6975,
      "mean_token_accuracy": 0.22927230298519136,
      "num_tokens": 125672466.0,
      "step": 72445
    },
    {
      "entropy": 5.426716470718384,
      "epoch": 5.636841081501654,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023263340403623477,
      "loss": 4.7659,
      "mean_token_accuracy": 0.22525952905416488,
      "num_tokens": 125682352.0,
      "step": 72450
    },
    {
      "entropy": 5.4373414516448975,
      "epoch": 5.637230110873371,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002326061845374553,
      "loss": 4.7033,
      "mean_token_accuracy": 0.2304536834359169,
      "num_tokens": 125691057.0,
      "step": 72455
    },
    {
      "entropy": 5.431629657745361,
      "epoch": 5.637619140245088,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023257896568193459,
      "loss": 4.7031,
      "mean_token_accuracy": 0.2360530525445938,
      "num_tokens": 125699792.0,
      "step": 72460
    },
    {
      "entropy": 5.419316959381104,
      "epoch": 5.638008169616806,
      "grad_norm": 1.375,
      "learning_rate": 0.0002325517474700855,
      "loss": 4.8048,
      "mean_token_accuracy": 0.2303388833999634,
      "num_tokens": 125709247.0,
      "step": 72465
    },
    {
      "entropy": 5.449718618392945,
      "epoch": 5.638397198988524,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002325245299023211,
      "loss": 4.7807,
      "mean_token_accuracy": 0.2269246682524681,
      "num_tokens": 125718574.0,
      "step": 72470
    },
    {
      "entropy": 5.410255813598633,
      "epoch": 5.6387862283602415,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023249731297905423,
      "loss": 4.7105,
      "mean_token_accuracy": 0.23775288164615632,
      "num_tokens": 125727379.0,
      "step": 72475
    },
    {
      "entropy": 5.483776330947876,
      "epoch": 5.639175257731958,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023247009670069812,
      "loss": 4.7771,
      "mean_token_accuracy": 0.22576013058423997,
      "num_tokens": 125735661.0,
      "step": 72480
    },
    {
      "entropy": 5.405849838256836,
      "epoch": 5.639564287103676,
      "grad_norm": 1.25,
      "learning_rate": 0.0002324428810676656,
      "loss": 4.7273,
      "mean_token_accuracy": 0.23246633261442184,
      "num_tokens": 125744178.0,
      "step": 72485
    },
    {
      "entropy": 5.375057649612427,
      "epoch": 5.639953316475394,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023241566608036957,
      "loss": 4.659,
      "mean_token_accuracy": 0.23629247695207595,
      "num_tokens": 125752212.0,
      "step": 72490
    },
    {
      "entropy": 5.375062274932861,
      "epoch": 5.640342345847111,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023238845173922308,
      "loss": 4.6932,
      "mean_token_accuracy": 0.2352395996451378,
      "num_tokens": 125761012.0,
      "step": 72495
    },
    {
      "entropy": 5.405709838867187,
      "epoch": 5.640731375218829,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023236123804463898,
      "loss": 4.71,
      "mean_token_accuracy": 0.23747551143169404,
      "num_tokens": 125770324.0,
      "step": 72500
    },
    {
      "entropy": 5.500503587722778,
      "epoch": 5.641120404590547,
      "grad_norm": 1.25,
      "learning_rate": 0.00023233402499703027,
      "loss": 4.7842,
      "mean_token_accuracy": 0.2330962300300598,
      "num_tokens": 125778607.0,
      "step": 72505
    },
    {
      "entropy": 5.359725427627564,
      "epoch": 5.6415094339622645,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002323068125968098,
      "loss": 4.6269,
      "mean_token_accuracy": 0.2376995399594307,
      "num_tokens": 125787647.0,
      "step": 72510
    },
    {
      "entropy": 5.387204837799072,
      "epoch": 5.641898463333982,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002322796008443905,
      "loss": 4.7016,
      "mean_token_accuracy": 0.2317735180258751,
      "num_tokens": 125796535.0,
      "step": 72515
    },
    {
      "entropy": 5.349104118347168,
      "epoch": 5.642287492705699,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002322523897401852,
      "loss": 4.6486,
      "mean_token_accuracy": 0.2381506845355034,
      "num_tokens": 125805373.0,
      "step": 72520
    },
    {
      "entropy": 5.405121517181397,
      "epoch": 5.642676522077417,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023222517928460685,
      "loss": 4.732,
      "mean_token_accuracy": 0.22640008628368377,
      "num_tokens": 125813680.0,
      "step": 72525
    },
    {
      "entropy": 5.405535316467285,
      "epoch": 5.643065551449134,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023219796947806842,
      "loss": 4.736,
      "mean_token_accuracy": 0.23710206151008606,
      "num_tokens": 125822486.0,
      "step": 72530
    },
    {
      "entropy": 5.3523273944854735,
      "epoch": 5.643454580820852,
      "grad_norm": 1.25,
      "learning_rate": 0.00023217076032098255,
      "loss": 4.6572,
      "mean_token_accuracy": 0.23039786219596864,
      "num_tokens": 125831976.0,
      "step": 72535
    },
    {
      "entropy": 5.358572292327881,
      "epoch": 5.64384361019257,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023214355181376222,
      "loss": 4.6865,
      "mean_token_accuracy": 0.23739436119794846,
      "num_tokens": 125840639.0,
      "step": 72540
    },
    {
      "entropy": 5.421099615097046,
      "epoch": 5.6442326395642874,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023211634395682042,
      "loss": 4.7037,
      "mean_token_accuracy": 0.22991849780082702,
      "num_tokens": 125848967.0,
      "step": 72545
    },
    {
      "entropy": 5.3703697681427,
      "epoch": 5.644621668936004,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00023208913675056976,
      "loss": 4.6884,
      "mean_token_accuracy": 0.23977784663438798,
      "num_tokens": 125858167.0,
      "step": 72550
    },
    {
      "entropy": 5.481612634658814,
      "epoch": 5.645010698307722,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002320619301954231,
      "loss": 4.7641,
      "mean_token_accuracy": 0.22612367272377015,
      "num_tokens": 125866928.0,
      "step": 72555
    },
    {
      "entropy": 5.400071907043457,
      "epoch": 5.64539972767944,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002320347242917934,
      "loss": 4.6739,
      "mean_token_accuracy": 0.240340693295002,
      "num_tokens": 125875675.0,
      "step": 72560
    },
    {
      "entropy": 5.454366445541382,
      "epoch": 5.645788757051157,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023200751904009332,
      "loss": 4.7152,
      "mean_token_accuracy": 0.23141793608665467,
      "num_tokens": 125885200.0,
      "step": 72565
    },
    {
      "entropy": 5.437732601165772,
      "epoch": 5.646177786422875,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023198031444073575,
      "loss": 4.7599,
      "mean_token_accuracy": 0.23076641261577607,
      "num_tokens": 125893416.0,
      "step": 72570
    },
    {
      "entropy": 5.32085952758789,
      "epoch": 5.646566815794593,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023195311049413337,
      "loss": 4.6721,
      "mean_token_accuracy": 0.24063338935375214,
      "num_tokens": 125902023.0,
      "step": 72575
    },
    {
      "entropy": 5.442850542068482,
      "epoch": 5.64695584516631,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023192590720069907,
      "loss": 4.7399,
      "mean_token_accuracy": 0.23738535195589067,
      "num_tokens": 125910667.0,
      "step": 72580
    },
    {
      "entropy": 5.341724109649658,
      "epoch": 5.647344874538027,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023189870456084561,
      "loss": 4.6416,
      "mean_token_accuracy": 0.23882248401641845,
      "num_tokens": 125919310.0,
      "step": 72585
    },
    {
      "entropy": 5.458663845062256,
      "epoch": 5.647733903909745,
      "grad_norm": 1.1875,
      "learning_rate": 0.00023187150257498567,
      "loss": 4.808,
      "mean_token_accuracy": 0.22109094262123108,
      "num_tokens": 125928437.0,
      "step": 72590
    },
    {
      "entropy": 5.447936105728149,
      "epoch": 5.648122933281463,
      "grad_norm": 1.234375,
      "learning_rate": 0.000231844301243532,
      "loss": 4.8006,
      "mean_token_accuracy": 0.22713441848754884,
      "num_tokens": 125937018.0,
      "step": 72595
    },
    {
      "entropy": 5.393181657791137,
      "epoch": 5.64851196265318,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002318171005668974,
      "loss": 4.6448,
      "mean_token_accuracy": 0.24214270412921907,
      "num_tokens": 125946816.0,
      "step": 72600
    },
    {
      "entropy": 5.358700513839722,
      "epoch": 5.648900992024898,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023178990054549447,
      "loss": 4.6639,
      "mean_token_accuracy": 0.2298433244228363,
      "num_tokens": 125955554.0,
      "step": 72605
    },
    {
      "entropy": 5.410617160797119,
      "epoch": 5.649290021396616,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023176270117973607,
      "loss": 4.7605,
      "mean_token_accuracy": 0.22394885569810868,
      "num_tokens": 125964506.0,
      "step": 72610
    },
    {
      "entropy": 5.457949829101563,
      "epoch": 5.6496790507683325,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000231735502470035,
      "loss": 4.7822,
      "mean_token_accuracy": 0.22689732164144516,
      "num_tokens": 125972244.0,
      "step": 72615
    },
    {
      "entropy": 5.351337528228759,
      "epoch": 5.65006808014005,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023170830441680368,
      "loss": 4.6607,
      "mean_token_accuracy": 0.2354472503066063,
      "num_tokens": 125980671.0,
      "step": 72620
    },
    {
      "entropy": 5.356338310241699,
      "epoch": 5.650457109511768,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00023168110702045496,
      "loss": 4.6429,
      "mean_token_accuracy": 0.23138432651758195,
      "num_tokens": 125989509.0,
      "step": 72625
    },
    {
      "entropy": 5.454546689987183,
      "epoch": 5.650846138883486,
      "grad_norm": 1.375,
      "learning_rate": 0.00023165391028140154,
      "loss": 4.8336,
      "mean_token_accuracy": 0.22842520773410796,
      "num_tokens": 125997315.0,
      "step": 72630
    },
    {
      "entropy": 5.351788330078125,
      "epoch": 5.651235168255203,
      "grad_norm": 1.4375,
      "learning_rate": 0.000231626714200056,
      "loss": 4.6172,
      "mean_token_accuracy": 0.24125102907419205,
      "num_tokens": 126005630.0,
      "step": 72635
    },
    {
      "entropy": 5.314880609512329,
      "epoch": 5.651624197626921,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00023159951877683099,
      "loss": 4.5325,
      "mean_token_accuracy": 0.24736319929361344,
      "num_tokens": 126014256.0,
      "step": 72640
    },
    {
      "entropy": 5.318935108184815,
      "epoch": 5.652013226998639,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002315723240121393,
      "loss": 4.6594,
      "mean_token_accuracy": 0.23997568488121032,
      "num_tokens": 126023259.0,
      "step": 72645
    },
    {
      "entropy": 5.363811874389649,
      "epoch": 5.652402256370356,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023154512990639336,
      "loss": 4.6583,
      "mean_token_accuracy": 0.23529096841812133,
      "num_tokens": 126031327.0,
      "step": 72650
    },
    {
      "entropy": 5.375575017929077,
      "epoch": 5.652791285742073,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023151793646000602,
      "loss": 4.6631,
      "mean_token_accuracy": 0.23313061743974686,
      "num_tokens": 126039349.0,
      "step": 72655
    },
    {
      "entropy": 5.382173871994018,
      "epoch": 5.653180315113791,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023149074367338974,
      "loss": 4.6812,
      "mean_token_accuracy": 0.2276252254843712,
      "num_tokens": 126049358.0,
      "step": 72660
    },
    {
      "entropy": 5.4068522453308105,
      "epoch": 5.653569344485509,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002314635515469572,
      "loss": 4.6605,
      "mean_token_accuracy": 0.23354194313287735,
      "num_tokens": 126057033.0,
      "step": 72665
    },
    {
      "entropy": 5.358712959289551,
      "epoch": 5.653958373857226,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023143636008112096,
      "loss": 4.6784,
      "mean_token_accuracy": 0.24706904739141464,
      "num_tokens": 126065690.0,
      "step": 72670
    },
    {
      "entropy": 5.433232021331787,
      "epoch": 5.654347403228944,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023140916927629352,
      "loss": 4.6879,
      "mean_token_accuracy": 0.23464627861976622,
      "num_tokens": 126073674.0,
      "step": 72675
    },
    {
      "entropy": 5.355509996414185,
      "epoch": 5.654736432600662,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002313819791328876,
      "loss": 4.6246,
      "mean_token_accuracy": 0.23583958446979522,
      "num_tokens": 126082280.0,
      "step": 72680
    },
    {
      "entropy": 5.448341321945191,
      "epoch": 5.6551254619723785,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023135478965131574,
      "loss": 4.8133,
      "mean_token_accuracy": 0.22511283904314042,
      "num_tokens": 126090524.0,
      "step": 72685
    },
    {
      "entropy": 5.357402992248535,
      "epoch": 5.655514491344096,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002313276008319905,
      "loss": 4.6872,
      "mean_token_accuracy": 0.24125647395849228,
      "num_tokens": 126099342.0,
      "step": 72690
    },
    {
      "entropy": 5.247012138366699,
      "epoch": 5.655903520715814,
      "grad_norm": 1.171875,
      "learning_rate": 0.00023130041267532443,
      "loss": 4.4909,
      "mean_token_accuracy": 0.24988892376422883,
      "num_tokens": 126107941.0,
      "step": 72695
    },
    {
      "entropy": 5.279316329956055,
      "epoch": 5.656292550087532,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023127322518173,
      "loss": 4.5807,
      "mean_token_accuracy": 0.23865536898374556,
      "num_tokens": 126116408.0,
      "step": 72700
    },
    {
      "entropy": 5.400543689727783,
      "epoch": 5.656681579459249,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002312460383516198,
      "loss": 4.7816,
      "mean_token_accuracy": 0.2230025351047516,
      "num_tokens": 126125611.0,
      "step": 72705
    },
    {
      "entropy": 5.410463428497314,
      "epoch": 5.657070608830967,
      "grad_norm": 1.25,
      "learning_rate": 0.00023121885218540627,
      "loss": 4.7119,
      "mean_token_accuracy": 0.23383795022964476,
      "num_tokens": 126133985.0,
      "step": 72710
    },
    {
      "entropy": 5.400418615341186,
      "epoch": 5.657459638202685,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023119166668350205,
      "loss": 4.679,
      "mean_token_accuracy": 0.2326354369521141,
      "num_tokens": 126141943.0,
      "step": 72715
    },
    {
      "entropy": 5.361264371871949,
      "epoch": 5.6578486675744015,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002311644818463195,
      "loss": 4.7609,
      "mean_token_accuracy": 0.23643314391374587,
      "num_tokens": 126150897.0,
      "step": 72720
    },
    {
      "entropy": 5.434242391586304,
      "epoch": 5.658237696946119,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002311372976742712,
      "loss": 4.6948,
      "mean_token_accuracy": 0.238344307243824,
      "num_tokens": 126159665.0,
      "step": 72725
    },
    {
      "entropy": 5.378707695007324,
      "epoch": 5.658626726317837,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002311101141677696,
      "loss": 4.6838,
      "mean_token_accuracy": 0.2381883829832077,
      "num_tokens": 126169695.0,
      "step": 72730
    },
    {
      "entropy": 5.371230411529541,
      "epoch": 5.659015755689555,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023108293132722714,
      "loss": 4.7341,
      "mean_token_accuracy": 0.23118516206741332,
      "num_tokens": 126178285.0,
      "step": 72735
    },
    {
      "entropy": 5.4683599948883055,
      "epoch": 5.659404785061272,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023105574915305627,
      "loss": 4.7999,
      "mean_token_accuracy": 0.22866797894239427,
      "num_tokens": 126186386.0,
      "step": 72740
    },
    {
      "entropy": 5.488535594940186,
      "epoch": 5.65979381443299,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002310285676456695,
      "loss": 4.829,
      "mean_token_accuracy": 0.22845159024000167,
      "num_tokens": 126194957.0,
      "step": 72745
    },
    {
      "entropy": 5.4724440574646,
      "epoch": 5.660182843804707,
      "grad_norm": 1.25,
      "learning_rate": 0.00023100138680547921,
      "loss": 4.773,
      "mean_token_accuracy": 0.22741208225488663,
      "num_tokens": 126204561.0,
      "step": 72750
    },
    {
      "entropy": 5.408730411529541,
      "epoch": 5.6605718731764245,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023097420663289786,
      "loss": 4.6883,
      "mean_token_accuracy": 0.2338494598865509,
      "num_tokens": 126212347.0,
      "step": 72755
    },
    {
      "entropy": 5.371700954437256,
      "epoch": 5.660960902548142,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023094702712833786,
      "loss": 4.6756,
      "mean_token_accuracy": 0.2347867399454117,
      "num_tokens": 126221043.0,
      "step": 72760
    },
    {
      "entropy": 5.462869310379029,
      "epoch": 5.66134993191986,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023091984829221157,
      "loss": 4.7574,
      "mean_token_accuracy": 0.22716951221227646,
      "num_tokens": 126229428.0,
      "step": 72765
    },
    {
      "entropy": 5.4716941833496096,
      "epoch": 5.661738961291578,
      "grad_norm": 1.140625,
      "learning_rate": 0.00023089267012493141,
      "loss": 4.7794,
      "mean_token_accuracy": 0.22727224230766296,
      "num_tokens": 126238398.0,
      "step": 72770
    },
    {
      "entropy": 5.408424425125122,
      "epoch": 5.662127990663295,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023086549262690993,
      "loss": 4.666,
      "mean_token_accuracy": 0.23357924520969392,
      "num_tokens": 126246617.0,
      "step": 72775
    },
    {
      "entropy": 5.360152292251587,
      "epoch": 5.662517020035013,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023083831579855918,
      "loss": 4.6787,
      "mean_token_accuracy": 0.23525431752204895,
      "num_tokens": 126255473.0,
      "step": 72780
    },
    {
      "entropy": 5.390235805511475,
      "epoch": 5.662906049406731,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023081113964029177,
      "loss": 4.7142,
      "mean_token_accuracy": 0.23409533351659775,
      "num_tokens": 126263789.0,
      "step": 72785
    },
    {
      "entropy": 5.431692218780517,
      "epoch": 5.6632950787784475,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023078396415251995,
      "loss": 4.8308,
      "mean_token_accuracy": 0.22908511012792587,
      "num_tokens": 126272981.0,
      "step": 72790
    },
    {
      "entropy": 5.387952136993408,
      "epoch": 5.663684108150165,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023075678933565615,
      "loss": 4.6926,
      "mean_token_accuracy": 0.2355685606598854,
      "num_tokens": 126281553.0,
      "step": 72795
    },
    {
      "entropy": 5.394940137863159,
      "epoch": 5.664073137521883,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002307296151901126,
      "loss": 4.6685,
      "mean_token_accuracy": 0.23252781182527543,
      "num_tokens": 126289825.0,
      "step": 72800
    },
    {
      "entropy": 5.309852504730225,
      "epoch": 5.664462166893601,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002307024417163016,
      "loss": 4.6256,
      "mean_token_accuracy": 0.23973995745182036,
      "num_tokens": 126298415.0,
      "step": 72805
    },
    {
      "entropy": 5.38768253326416,
      "epoch": 5.664851196265318,
      "grad_norm": 1.25,
      "learning_rate": 0.00023067526891463563,
      "loss": 4.6713,
      "mean_token_accuracy": 0.23421543836593628,
      "num_tokens": 126306938.0,
      "step": 72810
    },
    {
      "entropy": 5.445178556442261,
      "epoch": 5.665240225637035,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002306480967855269,
      "loss": 4.8115,
      "mean_token_accuracy": 0.22096768170595169,
      "num_tokens": 126315578.0,
      "step": 72815
    },
    {
      "entropy": 5.393955659866333,
      "epoch": 5.665629255008753,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023062092532938772,
      "loss": 4.7068,
      "mean_token_accuracy": 0.2346803814172745,
      "num_tokens": 126324688.0,
      "step": 72820
    },
    {
      "entropy": 5.482540845870972,
      "epoch": 5.6660182843804705,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023059375454663033,
      "loss": 4.7666,
      "mean_token_accuracy": 0.22929286658763887,
      "num_tokens": 126332848.0,
      "step": 72825
    },
    {
      "entropy": 5.393543767929077,
      "epoch": 5.666407313752188,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002305665844376671,
      "loss": 4.6586,
      "mean_token_accuracy": 0.24205812215805053,
      "num_tokens": 126342039.0,
      "step": 72830
    },
    {
      "entropy": 5.418577337265015,
      "epoch": 5.666796343123906,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023053941500291014,
      "loss": 4.7118,
      "mean_token_accuracy": 0.22892982214689256,
      "num_tokens": 126350471.0,
      "step": 72835
    },
    {
      "entropy": 5.432271480560303,
      "epoch": 5.667185372495624,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023051224624277185,
      "loss": 4.7197,
      "mean_token_accuracy": 0.2281401723623276,
      "num_tokens": 126359277.0,
      "step": 72840
    },
    {
      "entropy": 5.292291498184204,
      "epoch": 5.667574401867341,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002304850781576644,
      "loss": 4.5603,
      "mean_token_accuracy": 0.2403411567211151,
      "num_tokens": 126367198.0,
      "step": 72845
    },
    {
      "entropy": 5.375256776809692,
      "epoch": 5.667963431239059,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023045791074800004,
      "loss": 4.6998,
      "mean_token_accuracy": 0.23899523466825484,
      "num_tokens": 126375442.0,
      "step": 72850
    },
    {
      "entropy": 5.395594787597656,
      "epoch": 5.668352460610776,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000230430744014191,
      "loss": 4.7256,
      "mean_token_accuracy": 0.2299175128340721,
      "num_tokens": 126383591.0,
      "step": 72855
    },
    {
      "entropy": 5.425896406173706,
      "epoch": 5.6687414899824935,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023040357795664945,
      "loss": 4.7123,
      "mean_token_accuracy": 0.23457378000020981,
      "num_tokens": 126392090.0,
      "step": 72860
    },
    {
      "entropy": 5.4521399974823,
      "epoch": 5.669130519354211,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023037641257578768,
      "loss": 4.7886,
      "mean_token_accuracy": 0.22863523215055465,
      "num_tokens": 126400562.0,
      "step": 72865
    },
    {
      "entropy": 5.448321151733398,
      "epoch": 5.669519548725929,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023034924787201778,
      "loss": 4.7178,
      "mean_token_accuracy": 0.23340074867010116,
      "num_tokens": 126408925.0,
      "step": 72870
    },
    {
      "entropy": 5.454074907302856,
      "epoch": 5.669908578097647,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00023032208384575188,
      "loss": 4.7758,
      "mean_token_accuracy": 0.21978185921907425,
      "num_tokens": 126417662.0,
      "step": 72875
    },
    {
      "entropy": 5.406777954101562,
      "epoch": 5.670297607469364,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023029492049740232,
      "loss": 4.6676,
      "mean_token_accuracy": 0.24047560393810272,
      "num_tokens": 126426513.0,
      "step": 72880
    },
    {
      "entropy": 5.386661005020142,
      "epoch": 5.670686636841081,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023026775782738123,
      "loss": 4.6557,
      "mean_token_accuracy": 0.23544493764638902,
      "num_tokens": 126435161.0,
      "step": 72885
    },
    {
      "entropy": 5.446522188186646,
      "epoch": 5.671075666212799,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002302405958361007,
      "loss": 4.7285,
      "mean_token_accuracy": 0.2313833862543106,
      "num_tokens": 126443051.0,
      "step": 72890
    },
    {
      "entropy": 5.4804431915283205,
      "epoch": 5.6714646955845165,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023021343452397287,
      "loss": 4.8317,
      "mean_token_accuracy": 0.21914747655391692,
      "num_tokens": 126451764.0,
      "step": 72895
    },
    {
      "entropy": 5.504642486572266,
      "epoch": 5.671853724956234,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023018627389140989,
      "loss": 4.7712,
      "mean_token_accuracy": 0.22312382757663726,
      "num_tokens": 126460168.0,
      "step": 72900
    },
    {
      "entropy": 5.40142297744751,
      "epoch": 5.672242754327952,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002301591139388239,
      "loss": 4.7312,
      "mean_token_accuracy": 0.23557184189558028,
      "num_tokens": 126469867.0,
      "step": 72905
    },
    {
      "entropy": 5.3769838333129885,
      "epoch": 5.67263178369967,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023013195466662696,
      "loss": 4.6151,
      "mean_token_accuracy": 0.24287677854299544,
      "num_tokens": 126478668.0,
      "step": 72910
    },
    {
      "entropy": 5.444123983383179,
      "epoch": 5.673020813071387,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023010479607523123,
      "loss": 4.7646,
      "mean_token_accuracy": 0.22491187751293182,
      "num_tokens": 126488247.0,
      "step": 72915
    },
    {
      "entropy": 5.326867771148682,
      "epoch": 5.673409842443104,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00023007763816504866,
      "loss": 4.6005,
      "mean_token_accuracy": 0.2393566220998764,
      "num_tokens": 126497907.0,
      "step": 72920
    },
    {
      "entropy": 5.309686756134033,
      "epoch": 5.673798871814822,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023005048093649157,
      "loss": 4.5642,
      "mean_token_accuracy": 0.2420080006122589,
      "num_tokens": 126506482.0,
      "step": 72925
    },
    {
      "entropy": 5.378862810134888,
      "epoch": 5.6741879011865395,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023002332438997174,
      "loss": 4.7005,
      "mean_token_accuracy": 0.2350958377122879,
      "num_tokens": 126515312.0,
      "step": 72930
    },
    {
      "entropy": 5.415233373641968,
      "epoch": 5.674576930558257,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022999616852590148,
      "loss": 4.7374,
      "mean_token_accuracy": 0.2279706656932831,
      "num_tokens": 126522980.0,
      "step": 72935
    },
    {
      "entropy": 5.427444267272949,
      "epoch": 5.674965959929975,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022996901334469268,
      "loss": 4.7982,
      "mean_token_accuracy": 0.23050489872694016,
      "num_tokens": 126531721.0,
      "step": 72940
    },
    {
      "entropy": 5.4019331455230715,
      "epoch": 5.675354989301693,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00022994185884675738,
      "loss": 4.737,
      "mean_token_accuracy": 0.23016547858715058,
      "num_tokens": 126540433.0,
      "step": 72945
    },
    {
      "entropy": 5.46684160232544,
      "epoch": 5.675744018673409,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022991470503250778,
      "loss": 4.7524,
      "mean_token_accuracy": 0.235315765440464,
      "num_tokens": 126549520.0,
      "step": 72950
    },
    {
      "entropy": 5.483026027679443,
      "epoch": 5.676133048045127,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022988755190235573,
      "loss": 4.8103,
      "mean_token_accuracy": 0.22309497892856597,
      "num_tokens": 126558559.0,
      "step": 72955
    },
    {
      "entropy": 5.461029529571533,
      "epoch": 5.676522077416845,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022986039945671323,
      "loss": 4.7332,
      "mean_token_accuracy": 0.22894856929779053,
      "num_tokens": 126567675.0,
      "step": 72960
    },
    {
      "entropy": 5.444576930999756,
      "epoch": 5.6769111067885625,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00022983324769599239,
      "loss": 4.6636,
      "mean_token_accuracy": 0.23762525469064713,
      "num_tokens": 126576154.0,
      "step": 72965
    },
    {
      "entropy": 5.499536657333374,
      "epoch": 5.67730013616028,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002298060966206051,
      "loss": 4.8263,
      "mean_token_accuracy": 0.22560214400291442,
      "num_tokens": 126585042.0,
      "step": 72970
    },
    {
      "entropy": 5.388808107376098,
      "epoch": 5.677689165531998,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022977894623096336,
      "loss": 4.7062,
      "mean_token_accuracy": 0.23710624277591705,
      "num_tokens": 126594334.0,
      "step": 72975
    },
    {
      "entropy": 5.503040838241577,
      "epoch": 5.678078194903716,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022975179652747908,
      "loss": 4.7589,
      "mean_token_accuracy": 0.2315505936741829,
      "num_tokens": 126602139.0,
      "step": 72980
    },
    {
      "entropy": 5.529133272171021,
      "epoch": 5.678467224275433,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022972464751056432,
      "loss": 4.8763,
      "mean_token_accuracy": 0.2189146801829338,
      "num_tokens": 126610164.0,
      "step": 72985
    },
    {
      "entropy": 5.4394598484039305,
      "epoch": 5.67885625364715,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022969749918063092,
      "loss": 4.7649,
      "mean_token_accuracy": 0.22823501974344254,
      "num_tokens": 126619075.0,
      "step": 72990
    },
    {
      "entropy": 5.390476846694947,
      "epoch": 5.679245283018868,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022967035153809097,
      "loss": 4.71,
      "mean_token_accuracy": 0.2352984994649887,
      "num_tokens": 126627124.0,
      "step": 72995
    },
    {
      "entropy": 5.450757694244385,
      "epoch": 5.6796343123905855,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002296432045833562,
      "loss": 4.7729,
      "mean_token_accuracy": 0.23114334046840668,
      "num_tokens": 126636476.0,
      "step": 73000
    },
    {
      "entropy": 5.511966419219971,
      "epoch": 5.680023341762303,
      "grad_norm": 1.1875,
      "learning_rate": 0.00022961605831683856,
      "loss": 4.8022,
      "mean_token_accuracy": 0.22343357503414155,
      "num_tokens": 126644542.0,
      "step": 73005
    },
    {
      "entropy": 5.403992462158203,
      "epoch": 5.680412371134021,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00022958891273895004,
      "loss": 4.6306,
      "mean_token_accuracy": 0.2356755778193474,
      "num_tokens": 126652828.0,
      "step": 73010
    },
    {
      "entropy": 5.375300455093384,
      "epoch": 5.680801400505739,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022956176785010252,
      "loss": 4.6481,
      "mean_token_accuracy": 0.24217975586652757,
      "num_tokens": 126660941.0,
      "step": 73015
    },
    {
      "entropy": 5.430912971496582,
      "epoch": 5.681190429877455,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002295346236507079,
      "loss": 4.725,
      "mean_token_accuracy": 0.2348072946071625,
      "num_tokens": 126670171.0,
      "step": 73020
    },
    {
      "entropy": 5.367882251739502,
      "epoch": 5.681579459249173,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002295074801411779,
      "loss": 4.6681,
      "mean_token_accuracy": 0.2383909523487091,
      "num_tokens": 126678845.0,
      "step": 73025
    },
    {
      "entropy": 5.351116514205932,
      "epoch": 5.681968488620891,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022948033732192452,
      "loss": 4.5805,
      "mean_token_accuracy": 0.23803758919239043,
      "num_tokens": 126686977.0,
      "step": 73030
    },
    {
      "entropy": 5.389431190490723,
      "epoch": 5.6823575179926085,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022945319519335962,
      "loss": 4.6817,
      "mean_token_accuracy": 0.23209336698055266,
      "num_tokens": 126695411.0,
      "step": 73035
    },
    {
      "entropy": 5.319335651397705,
      "epoch": 5.682746547364326,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002294260537558949,
      "loss": 4.6514,
      "mean_token_accuracy": 0.23948210179805757,
      "num_tokens": 126704530.0,
      "step": 73040
    },
    {
      "entropy": 5.362835693359375,
      "epoch": 5.683135576736044,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022939891300994238,
      "loss": 4.6582,
      "mean_token_accuracy": 0.23780689835548402,
      "num_tokens": 126713027.0,
      "step": 73045
    },
    {
      "entropy": 5.420645761489868,
      "epoch": 5.683524606107762,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002293717729559137,
      "loss": 4.7901,
      "mean_token_accuracy": 0.22488992363214494,
      "num_tokens": 126721718.0,
      "step": 73050
    },
    {
      "entropy": 5.449434232711792,
      "epoch": 5.683913635479478,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022934463359422074,
      "loss": 4.7385,
      "mean_token_accuracy": 0.22876913249492645,
      "num_tokens": 126729658.0,
      "step": 73055
    },
    {
      "entropy": 5.4059131145477295,
      "epoch": 5.684302664851196,
      "grad_norm": 1.25,
      "learning_rate": 0.0002293174949252753,
      "loss": 4.6854,
      "mean_token_accuracy": 0.2340743750333786,
      "num_tokens": 126737936.0,
      "step": 73060
    },
    {
      "entropy": 5.372838687896729,
      "epoch": 5.684691694222914,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002292903569494892,
      "loss": 4.7,
      "mean_token_accuracy": 0.2338390365242958,
      "num_tokens": 126746544.0,
      "step": 73065
    },
    {
      "entropy": 5.361047887802124,
      "epoch": 5.6850807235946315,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022926321966727404,
      "loss": 4.6462,
      "mean_token_accuracy": 0.2376161888241768,
      "num_tokens": 126755627.0,
      "step": 73070
    },
    {
      "entropy": 5.385217809677124,
      "epoch": 5.685469752966349,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022923608307904192,
      "loss": 4.6316,
      "mean_token_accuracy": 0.2429195985198021,
      "num_tokens": 126763751.0,
      "step": 73075
    },
    {
      "entropy": 5.403415536880493,
      "epoch": 5.685858782338067,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002292089471852043,
      "loss": 4.7381,
      "mean_token_accuracy": 0.23219627588987352,
      "num_tokens": 126771959.0,
      "step": 73080
    },
    {
      "entropy": 5.367689371109009,
      "epoch": 5.686247811709784,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022918181198617306,
      "loss": 4.6621,
      "mean_token_accuracy": 0.2419017106294632,
      "num_tokens": 126779354.0,
      "step": 73085
    },
    {
      "entropy": 5.393605089187622,
      "epoch": 5.686636841081501,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00022915467748235984,
      "loss": 4.734,
      "mean_token_accuracy": 0.23124941140413285,
      "num_tokens": 126787812.0,
      "step": 73090
    },
    {
      "entropy": 5.423824119567871,
      "epoch": 5.687025870453219,
      "grad_norm": 1.1875,
      "learning_rate": 0.00022912754367417648,
      "loss": 4.7312,
      "mean_token_accuracy": 0.23051079362630844,
      "num_tokens": 126795919.0,
      "step": 73095
    },
    {
      "entropy": 5.406348657608032,
      "epoch": 5.687414899824937,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022910041056203458,
      "loss": 4.7427,
      "mean_token_accuracy": 0.24181899875402452,
      "num_tokens": 126804328.0,
      "step": 73100
    },
    {
      "entropy": 5.359303998947143,
      "epoch": 5.6878039291966545,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022907327814634593,
      "loss": 4.7343,
      "mean_token_accuracy": 0.22814283967018129,
      "num_tokens": 126812591.0,
      "step": 73105
    },
    {
      "entropy": 5.352538824081421,
      "epoch": 5.688192958568372,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022904614642752218,
      "loss": 4.7082,
      "mean_token_accuracy": 0.2286165550351143,
      "num_tokens": 126821783.0,
      "step": 73110
    },
    {
      "entropy": 5.511131381988525,
      "epoch": 5.68858198794009,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022901901540597498,
      "loss": 4.7904,
      "mean_token_accuracy": 0.22671881020069123,
      "num_tokens": 126830199.0,
      "step": 73115
    },
    {
      "entropy": 5.464781188964844,
      "epoch": 5.688971017311807,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002289918850821161,
      "loss": 4.7319,
      "mean_token_accuracy": 0.2273805946111679,
      "num_tokens": 126838971.0,
      "step": 73120
    },
    {
      "entropy": 5.414956092834473,
      "epoch": 5.689360046683524,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022896475545635704,
      "loss": 4.6539,
      "mean_token_accuracy": 0.2305045709013939,
      "num_tokens": 126847240.0,
      "step": 73125
    },
    {
      "entropy": 5.420384216308594,
      "epoch": 5.689749076055242,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00022893762652910965,
      "loss": 4.7635,
      "mean_token_accuracy": 0.2282832846045494,
      "num_tokens": 126856447.0,
      "step": 73130
    },
    {
      "entropy": 5.366698122024536,
      "epoch": 5.69013810542696,
      "grad_norm": 1.21875,
      "learning_rate": 0.00022891049830078543,
      "loss": 4.6769,
      "mean_token_accuracy": 0.23292567878961562,
      "num_tokens": 126865486.0,
      "step": 73135
    },
    {
      "entropy": 5.4012562274932865,
      "epoch": 5.6905271347986774,
      "grad_norm": 1.140625,
      "learning_rate": 0.00022888337077179595,
      "loss": 4.6931,
      "mean_token_accuracy": 0.22868673652410507,
      "num_tokens": 126874432.0,
      "step": 73140
    },
    {
      "entropy": 5.486894655227661,
      "epoch": 5.690916164170395,
      "grad_norm": 1.171875,
      "learning_rate": 0.00022885624394255294,
      "loss": 4.857,
      "mean_token_accuracy": 0.21694085448980333,
      "num_tokens": 126883247.0,
      "step": 73145
    },
    {
      "entropy": 5.460770082473755,
      "epoch": 5.691305193542112,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022882911781346805,
      "loss": 4.7567,
      "mean_token_accuracy": 0.22933626621961595,
      "num_tokens": 126892060.0,
      "step": 73150
    },
    {
      "entropy": 5.52426815032959,
      "epoch": 5.69169422291383,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002288019923849528,
      "loss": 4.8092,
      "mean_token_accuracy": 0.2238599732518196,
      "num_tokens": 126900960.0,
      "step": 73155
    },
    {
      "entropy": 5.479573011398315,
      "epoch": 5.692083252285547,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002287748676574188,
      "loss": 4.8221,
      "mean_token_accuracy": 0.2262978732585907,
      "num_tokens": 126911093.0,
      "step": 73160
    },
    {
      "entropy": 5.416458797454834,
      "epoch": 5.692472281657265,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002287477436312776,
      "loss": 4.7217,
      "mean_token_accuracy": 0.22899159938097,
      "num_tokens": 126920017.0,
      "step": 73165
    },
    {
      "entropy": 5.459509468078613,
      "epoch": 5.692861311028983,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022872062030694073,
      "loss": 4.8072,
      "mean_token_accuracy": 0.22413868457078934,
      "num_tokens": 126928272.0,
      "step": 73170
    },
    {
      "entropy": 5.423401021957398,
      "epoch": 5.6932503404007,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002286934976848198,
      "loss": 4.7231,
      "mean_token_accuracy": 0.2434844881296158,
      "num_tokens": 126937205.0,
      "step": 73175
    },
    {
      "entropy": 5.384331798553466,
      "epoch": 5.693639369772418,
      "grad_norm": 1.25,
      "learning_rate": 0.00022866637576532633,
      "loss": 4.6617,
      "mean_token_accuracy": 0.22852024734020232,
      "num_tokens": 126945865.0,
      "step": 73180
    },
    {
      "entropy": 5.332643795013428,
      "epoch": 5.694028399144136,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002286392545488719,
      "loss": 4.6912,
      "mean_token_accuracy": 0.22895937114953996,
      "num_tokens": 126954432.0,
      "step": 73185
    },
    {
      "entropy": 5.409226799011231,
      "epoch": 5.694417428515853,
      "grad_norm": 1.171875,
      "learning_rate": 0.00022861213403586802,
      "loss": 4.7124,
      "mean_token_accuracy": 0.23355409055948256,
      "num_tokens": 126962989.0,
      "step": 73190
    },
    {
      "entropy": 5.418423843383789,
      "epoch": 5.69480645788757,
      "grad_norm": 1.25,
      "learning_rate": 0.00022858501422672606,
      "loss": 4.688,
      "mean_token_accuracy": 0.23599325716495514,
      "num_tokens": 126970994.0,
      "step": 73195
    },
    {
      "entropy": 5.367571878433227,
      "epoch": 5.695195487259288,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002285578951218577,
      "loss": 4.7471,
      "mean_token_accuracy": 0.23247490972280502,
      "num_tokens": 126979387.0,
      "step": 73200
    },
    {
      "entropy": 5.411379146575928,
      "epoch": 5.695584516631006,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00022853077672167432,
      "loss": 4.7435,
      "mean_token_accuracy": 0.22461799532175064,
      "num_tokens": 126988317.0,
      "step": 73205
    },
    {
      "entropy": 5.403444147109985,
      "epoch": 5.695973546002723,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002285036590265875,
      "loss": 4.7101,
      "mean_token_accuracy": 0.23746386617422105,
      "num_tokens": 126997172.0,
      "step": 73210
    },
    {
      "entropy": 5.4229989528656,
      "epoch": 5.696362575374441,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022847654203700857,
      "loss": 4.7028,
      "mean_token_accuracy": 0.23142081499099731,
      "num_tokens": 127006251.0,
      "step": 73215
    },
    {
      "entropy": 5.401953268051147,
      "epoch": 5.696751604746158,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022844942575334914,
      "loss": 4.6555,
      "mean_token_accuracy": 0.23450588881969453,
      "num_tokens": 127014471.0,
      "step": 73220
    },
    {
      "entropy": 5.478708744049072,
      "epoch": 5.697140634117876,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022842231017602055,
      "loss": 4.8084,
      "mean_token_accuracy": 0.22336014956235886,
      "num_tokens": 127023987.0,
      "step": 73225
    },
    {
      "entropy": 5.547585725784302,
      "epoch": 5.697529663489593,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022839519530543423,
      "loss": 4.9336,
      "mean_token_accuracy": 0.2143992677330971,
      "num_tokens": 127033287.0,
      "step": 73230
    },
    {
      "entropy": 5.4086614608764645,
      "epoch": 5.697918692861311,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002283680811420017,
      "loss": 4.6685,
      "mean_token_accuracy": 0.24128328263759613,
      "num_tokens": 127042074.0,
      "step": 73235
    },
    {
      "entropy": 5.373089456558228,
      "epoch": 5.698307722233029,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022834096768613437,
      "loss": 4.6077,
      "mean_token_accuracy": 0.2423490032553673,
      "num_tokens": 127049852.0,
      "step": 73240
    },
    {
      "entropy": 5.35197491645813,
      "epoch": 5.698696751604746,
      "grad_norm": 1.171875,
      "learning_rate": 0.00022831385493824353,
      "loss": 4.7068,
      "mean_token_accuracy": 0.235366028547287,
      "num_tokens": 127058812.0,
      "step": 73245
    },
    {
      "entropy": 5.35046763420105,
      "epoch": 5.699085780976464,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00022828674289874064,
      "loss": 4.6346,
      "mean_token_accuracy": 0.23686148822307587,
      "num_tokens": 127067657.0,
      "step": 73250
    },
    {
      "entropy": 5.40112829208374,
      "epoch": 5.699474810348181,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022825963156803704,
      "loss": 4.677,
      "mean_token_accuracy": 0.23847858607769012,
      "num_tokens": 127075779.0,
      "step": 73255
    },
    {
      "entropy": 5.340543603897094,
      "epoch": 5.699863839719899,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022823252094654413,
      "loss": 4.6522,
      "mean_token_accuracy": 0.23514725863933564,
      "num_tokens": 127084384.0,
      "step": 73260
    },
    {
      "entropy": 5.371319961547852,
      "epoch": 5.700252869091616,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002282054110346733,
      "loss": 4.6637,
      "mean_token_accuracy": 0.2378919690847397,
      "num_tokens": 127092686.0,
      "step": 73265
    },
    {
      "entropy": 5.445283651351929,
      "epoch": 5.700641898463334,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022817830183283582,
      "loss": 4.7717,
      "mean_token_accuracy": 0.2296251967549324,
      "num_tokens": 127100588.0,
      "step": 73270
    },
    {
      "entropy": 5.365663385391235,
      "epoch": 5.701030927835052,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022815119334144312,
      "loss": 4.6646,
      "mean_token_accuracy": 0.23461905866861343,
      "num_tokens": 127109474.0,
      "step": 73275
    },
    {
      "entropy": 5.416236734390258,
      "epoch": 5.701419957206769,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022812408556090657,
      "loss": 4.6703,
      "mean_token_accuracy": 0.23859499543905258,
      "num_tokens": 127118166.0,
      "step": 73280
    },
    {
      "entropy": 5.41538553237915,
      "epoch": 5.701808986578486,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00022809697849163736,
      "loss": 4.7229,
      "mean_token_accuracy": 0.23413300961256028,
      "num_tokens": 127127060.0,
      "step": 73285
    },
    {
      "entropy": 5.318397331237793,
      "epoch": 5.702198015950204,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022806987213404685,
      "loss": 4.6117,
      "mean_token_accuracy": 0.23752814084291457,
      "num_tokens": 127135030.0,
      "step": 73290
    },
    {
      "entropy": 5.384644079208374,
      "epoch": 5.702587045321922,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022804276648854632,
      "loss": 4.7284,
      "mean_token_accuracy": 0.23244843780994415,
      "num_tokens": 127142872.0,
      "step": 73295
    },
    {
      "entropy": 5.399064397811889,
      "epoch": 5.702976074693639,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022801566155554708,
      "loss": 4.6652,
      "mean_token_accuracy": 0.2374661907553673,
      "num_tokens": 127151242.0,
      "step": 73300
    },
    {
      "entropy": 5.433575248718261,
      "epoch": 5.703365104065357,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022798855733546038,
      "loss": 4.7523,
      "mean_token_accuracy": 0.22547928988933563,
      "num_tokens": 127160168.0,
      "step": 73305
    },
    {
      "entropy": 5.457707786560059,
      "epoch": 5.703754133437075,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002279614538286975,
      "loss": 4.7997,
      "mean_token_accuracy": 0.2203005760908127,
      "num_tokens": 127168568.0,
      "step": 73310
    },
    {
      "entropy": 5.513027477264404,
      "epoch": 5.704143162808792,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022793435103566972,
      "loss": 4.8018,
      "mean_token_accuracy": 0.22134671211242676,
      "num_tokens": 127177394.0,
      "step": 73315
    },
    {
      "entropy": 5.449186229705811,
      "epoch": 5.70453219218051,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022790724895678822,
      "loss": 4.7342,
      "mean_token_accuracy": 0.2303750455379486,
      "num_tokens": 127185533.0,
      "step": 73320
    },
    {
      "entropy": 5.406860303878784,
      "epoch": 5.704921221552227,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022788014759246429,
      "loss": 4.7318,
      "mean_token_accuracy": 0.233727864921093,
      "num_tokens": 127193917.0,
      "step": 73325
    },
    {
      "entropy": 5.260691785812378,
      "epoch": 5.705310250923945,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002278530469431091,
      "loss": 4.5603,
      "mean_token_accuracy": 0.2437954366207123,
      "num_tokens": 127204164.0,
      "step": 73330
    },
    {
      "entropy": 5.400282669067383,
      "epoch": 5.705699280295662,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002278259470091339,
      "loss": 4.6946,
      "mean_token_accuracy": 0.23294628858566285,
      "num_tokens": 127212780.0,
      "step": 73335
    },
    {
      "entropy": 5.4572361469268795,
      "epoch": 5.70608830966738,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022779884779094972,
      "loss": 4.7409,
      "mean_token_accuracy": 0.2317083641886711,
      "num_tokens": 127221022.0,
      "step": 73340
    },
    {
      "entropy": 5.512049674987793,
      "epoch": 5.706477339039098,
      "grad_norm": 1.140625,
      "learning_rate": 0.00022777174928896804,
      "loss": 4.7884,
      "mean_token_accuracy": 0.2269298255443573,
      "num_tokens": 127229742.0,
      "step": 73345
    },
    {
      "entropy": 5.399073505401612,
      "epoch": 5.706866368410815,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002277446515035999,
      "loss": 4.66,
      "mean_token_accuracy": 0.2347613051533699,
      "num_tokens": 127238008.0,
      "step": 73350
    },
    {
      "entropy": 5.421376037597656,
      "epoch": 5.707255397782532,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022771755443525644,
      "loss": 4.6475,
      "mean_token_accuracy": 0.2414591506123543,
      "num_tokens": 127246824.0,
      "step": 73355
    },
    {
      "entropy": 5.437095880508423,
      "epoch": 5.70764442715425,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022769045808434884,
      "loss": 4.7357,
      "mean_token_accuracy": 0.22985170483589173,
      "num_tokens": 127255174.0,
      "step": 73360
    },
    {
      "entropy": 5.409363794326782,
      "epoch": 5.708033456525968,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002276633624512882,
      "loss": 4.7152,
      "mean_token_accuracy": 0.2317097082734108,
      "num_tokens": 127263888.0,
      "step": 73365
    },
    {
      "entropy": 5.413269567489624,
      "epoch": 5.708422485897685,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022763626753648575,
      "loss": 4.7492,
      "mean_token_accuracy": 0.23427136838436127,
      "num_tokens": 127272259.0,
      "step": 73370
    },
    {
      "entropy": 5.404621267318726,
      "epoch": 5.708811515269403,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022760917334035252,
      "loss": 4.6344,
      "mean_token_accuracy": 0.24147583693265914,
      "num_tokens": 127280018.0,
      "step": 73375
    },
    {
      "entropy": 5.315544891357422,
      "epoch": 5.709200544641121,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022758207986329964,
      "loss": 4.5822,
      "mean_token_accuracy": 0.24037724286317824,
      "num_tokens": 127287879.0,
      "step": 73380
    },
    {
      "entropy": 5.457150983810425,
      "epoch": 5.709589574012838,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022755498710573825,
      "loss": 4.7585,
      "mean_token_accuracy": 0.22876127511262895,
      "num_tokens": 127295872.0,
      "step": 73385
    },
    {
      "entropy": 5.419118452072143,
      "epoch": 5.709978603384555,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022752789506807935,
      "loss": 4.7468,
      "mean_token_accuracy": 0.23326620310544968,
      "num_tokens": 127304250.0,
      "step": 73390
    },
    {
      "entropy": 5.369149589538575,
      "epoch": 5.710367632756273,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022750080375073418,
      "loss": 4.6341,
      "mean_token_accuracy": 0.23289757072925568,
      "num_tokens": 127312668.0,
      "step": 73395
    },
    {
      "entropy": 5.418279218673706,
      "epoch": 5.710756662127991,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022747371315411368,
      "loss": 4.7385,
      "mean_token_accuracy": 0.23330868631601334,
      "num_tokens": 127321691.0,
      "step": 73400
    },
    {
      "entropy": 5.416310405731201,
      "epoch": 5.711145691499708,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022744662327862887,
      "loss": 4.6868,
      "mean_token_accuracy": 0.23808684200048447,
      "num_tokens": 127331005.0,
      "step": 73405
    },
    {
      "entropy": 5.3581915378570555,
      "epoch": 5.711534720871426,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002274195341246909,
      "loss": 4.6118,
      "mean_token_accuracy": 0.23715014308691024,
      "num_tokens": 127339103.0,
      "step": 73410
    },
    {
      "entropy": 5.3850304126739506,
      "epoch": 5.711923750243144,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022739244569271077,
      "loss": 4.7263,
      "mean_token_accuracy": 0.2280083566904068,
      "num_tokens": 127347455.0,
      "step": 73415
    },
    {
      "entropy": 5.468974351882935,
      "epoch": 5.7123127796148605,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00022736535798309953,
      "loss": 4.7939,
      "mean_token_accuracy": 0.2233325406908989,
      "num_tokens": 127356940.0,
      "step": 73420
    },
    {
      "entropy": 5.448297786712646,
      "epoch": 5.712701808986578,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022733827099626815,
      "loss": 4.6818,
      "mean_token_accuracy": 0.24207652509212493,
      "num_tokens": 127365479.0,
      "step": 73425
    },
    {
      "entropy": 5.483569717407226,
      "epoch": 5.713090838358296,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002273111847326277,
      "loss": 4.7378,
      "mean_token_accuracy": 0.22649875581264495,
      "num_tokens": 127375076.0,
      "step": 73430
    },
    {
      "entropy": 5.490187168121338,
      "epoch": 5.713479867730014,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022728409919258908,
      "loss": 4.7656,
      "mean_token_accuracy": 0.22399675995111465,
      "num_tokens": 127383544.0,
      "step": 73435
    },
    {
      "entropy": 5.396808910369873,
      "epoch": 5.713868897101731,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022725701437656328,
      "loss": 4.6449,
      "mean_token_accuracy": 0.2402901291847229,
      "num_tokens": 127391405.0,
      "step": 73440
    },
    {
      "entropy": 5.402274179458618,
      "epoch": 5.714257926473449,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002272299302849613,
      "loss": 4.7307,
      "mean_token_accuracy": 0.23221128135919572,
      "num_tokens": 127399580.0,
      "step": 73445
    },
    {
      "entropy": 5.3394567489624025,
      "epoch": 5.714646955845167,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002272028469181941,
      "loss": 4.6495,
      "mean_token_accuracy": 0.23874157071113586,
      "num_tokens": 127408670.0,
      "step": 73450
    },
    {
      "entropy": 5.383472728729248,
      "epoch": 5.7150359852168835,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00022717576427667258,
      "loss": 4.6686,
      "mean_token_accuracy": 0.2384016200900078,
      "num_tokens": 127417217.0,
      "step": 73455
    },
    {
      "entropy": 5.422609186172485,
      "epoch": 5.715425014588601,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022714868236080776,
      "loss": 4.6273,
      "mean_token_accuracy": 0.24762362539768218,
      "num_tokens": 127425398.0,
      "step": 73460
    },
    {
      "entropy": 5.456642818450928,
      "epoch": 5.715814043960319,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022712160117101054,
      "loss": 4.8015,
      "mean_token_accuracy": 0.2252141758799553,
      "num_tokens": 127433478.0,
      "step": 73465
    },
    {
      "entropy": 5.415934038162232,
      "epoch": 5.716203073332037,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022709452070769176,
      "loss": 4.6851,
      "mean_token_accuracy": 0.2337421640753746,
      "num_tokens": 127441752.0,
      "step": 73470
    },
    {
      "entropy": 5.473828744888306,
      "epoch": 5.716592102703754,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022706744097126242,
      "loss": 4.7496,
      "mean_token_accuracy": 0.23114845752716065,
      "num_tokens": 127450731.0,
      "step": 73475
    },
    {
      "entropy": 5.380221605300903,
      "epoch": 5.716981132075472,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022704036196213336,
      "loss": 4.6524,
      "mean_token_accuracy": 0.23269561678171158,
      "num_tokens": 127459556.0,
      "step": 73480
    },
    {
      "entropy": 5.378362131118775,
      "epoch": 5.717370161447189,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022701328368071545,
      "loss": 4.6748,
      "mean_token_accuracy": 0.23598323464393617,
      "num_tokens": 127467418.0,
      "step": 73485
    },
    {
      "entropy": 5.428715467453003,
      "epoch": 5.7177591908189065,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00022698620612741962,
      "loss": 4.7865,
      "mean_token_accuracy": 0.227668334543705,
      "num_tokens": 127476461.0,
      "step": 73490
    },
    {
      "entropy": 5.367031764984131,
      "epoch": 5.718148220190624,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022695912930265666,
      "loss": 4.6857,
      "mean_token_accuracy": 0.23061956465244293,
      "num_tokens": 127485476.0,
      "step": 73495
    },
    {
      "entropy": 5.417669439315796,
      "epoch": 5.718537249562342,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002269320532068375,
      "loss": 4.6713,
      "mean_token_accuracy": 0.23877793699502944,
      "num_tokens": 127493783.0,
      "step": 73500
    },
    {
      "entropy": 5.37351393699646,
      "epoch": 5.71892627893406,
      "grad_norm": 1.375,
      "learning_rate": 0.0002269049778403729,
      "loss": 4.6636,
      "mean_token_accuracy": 0.23703791052103043,
      "num_tokens": 127502243.0,
      "step": 73505
    },
    {
      "entropy": 5.399152708053589,
      "epoch": 5.719315308305777,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022687790320367375,
      "loss": 4.7393,
      "mean_token_accuracy": 0.22813319861888887,
      "num_tokens": 127510879.0,
      "step": 73510
    },
    {
      "entropy": 5.417322397232056,
      "epoch": 5.719704337677495,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022685082929715074,
      "loss": 4.6766,
      "mean_token_accuracy": 0.23632766157388688,
      "num_tokens": 127518918.0,
      "step": 73515
    },
    {
      "entropy": 5.415064954757691,
      "epoch": 5.720093367049213,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00022682375612121481,
      "loss": 4.7544,
      "mean_token_accuracy": 0.22950664907693863,
      "num_tokens": 127527782.0,
      "step": 73520
    },
    {
      "entropy": 5.383284950256348,
      "epoch": 5.7204823964209295,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002267966836762767,
      "loss": 4.6631,
      "mean_token_accuracy": 0.234574593603611,
      "num_tokens": 127537015.0,
      "step": 73525
    },
    {
      "entropy": 5.385509729385376,
      "epoch": 5.720871425792647,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00022676961196274714,
      "loss": 4.7061,
      "mean_token_accuracy": 0.23799406439065934,
      "num_tokens": 127545634.0,
      "step": 73530
    },
    {
      "entropy": 5.4751664161682125,
      "epoch": 5.721260455164365,
      "grad_norm": 1.1875,
      "learning_rate": 0.00022674254098103692,
      "loss": 4.7887,
      "mean_token_accuracy": 0.23128096163272857,
      "num_tokens": 127554019.0,
      "step": 73535
    },
    {
      "entropy": 5.465732908248901,
      "epoch": 5.721649484536083,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002267154707315569,
      "loss": 4.8241,
      "mean_token_accuracy": 0.2262078508734703,
      "num_tokens": 127562996.0,
      "step": 73540
    },
    {
      "entropy": 5.471660995483399,
      "epoch": 5.7220385139078,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002266884012147178,
      "loss": 4.6736,
      "mean_token_accuracy": 0.23983193337917327,
      "num_tokens": 127572603.0,
      "step": 73545
    },
    {
      "entropy": 5.460847520828247,
      "epoch": 5.722427543279518,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002266613324309303,
      "loss": 4.7384,
      "mean_token_accuracy": 0.22942037880420685,
      "num_tokens": 127580708.0,
      "step": 73550
    },
    {
      "entropy": 5.370373821258545,
      "epoch": 5.722816572651235,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022663426438060508,
      "loss": 4.6651,
      "mean_token_accuracy": 0.2355080172419548,
      "num_tokens": 127589062.0,
      "step": 73555
    },
    {
      "entropy": 5.431367492675781,
      "epoch": 5.7232056020229525,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022660719706415295,
      "loss": 4.6845,
      "mean_token_accuracy": 0.24026960581541063,
      "num_tokens": 127597322.0,
      "step": 73560
    },
    {
      "entropy": 5.474208784103394,
      "epoch": 5.72359463139467,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022658013048198455,
      "loss": 4.7671,
      "mean_token_accuracy": 0.22355885356664656,
      "num_tokens": 127605698.0,
      "step": 73565
    },
    {
      "entropy": 5.417050218582153,
      "epoch": 5.723983660766388,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022655306463451063,
      "loss": 4.6844,
      "mean_token_accuracy": 0.2399067685008049,
      "num_tokens": 127614178.0,
      "step": 73570
    },
    {
      "entropy": 5.385628032684326,
      "epoch": 5.724372690138106,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00022652599952214186,
      "loss": 4.6277,
      "mean_token_accuracy": 0.2406775489449501,
      "num_tokens": 127623069.0,
      "step": 73575
    },
    {
      "entropy": 5.441081619262695,
      "epoch": 5.724761719509823,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022649893514528892,
      "loss": 4.7361,
      "mean_token_accuracy": 0.22981414794921876,
      "num_tokens": 127631284.0,
      "step": 73580
    },
    {
      "entropy": 5.371806240081787,
      "epoch": 5.725150748881541,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002264718715043624,
      "loss": 4.6667,
      "mean_token_accuracy": 0.2376416563987732,
      "num_tokens": 127639810.0,
      "step": 73585
    },
    {
      "entropy": 5.501563882827758,
      "epoch": 5.725539778253258,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000226444808599773,
      "loss": 4.8531,
      "mean_token_accuracy": 0.23133372366428376,
      "num_tokens": 127648298.0,
      "step": 73590
    },
    {
      "entropy": 5.466367435455322,
      "epoch": 5.7259288076249755,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022641774643193137,
      "loss": 4.7174,
      "mean_token_accuracy": 0.23105482757091522,
      "num_tokens": 127656990.0,
      "step": 73595
    },
    {
      "entropy": 5.40721378326416,
      "epoch": 5.726317836996693,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022639068500124803,
      "loss": 4.5878,
      "mean_token_accuracy": 0.24179897159337999,
      "num_tokens": 127665284.0,
      "step": 73600
    },
    {
      "entropy": 5.397235965728759,
      "epoch": 5.726706866368411,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022636362430813363,
      "loss": 4.6838,
      "mean_token_accuracy": 0.24110848754644393,
      "num_tokens": 127674770.0,
      "step": 73605
    },
    {
      "entropy": 5.506020641326904,
      "epoch": 5.727095895740129,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00022633656435299893,
      "loss": 4.8489,
      "mean_token_accuracy": 0.22367931455373763,
      "num_tokens": 127683409.0,
      "step": 73610
    },
    {
      "entropy": 5.435642862319947,
      "epoch": 5.727484925111846,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022630950513625436,
      "loss": 4.6905,
      "mean_token_accuracy": 0.23249044120311738,
      "num_tokens": 127692699.0,
      "step": 73615
    },
    {
      "entropy": 5.532311201095581,
      "epoch": 5.727873954483563,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022628244665831067,
      "loss": 4.7883,
      "mean_token_accuracy": 0.22562696486711503,
      "num_tokens": 127701720.0,
      "step": 73620
    },
    {
      "entropy": 5.493682432174682,
      "epoch": 5.728262983855281,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002262553889195782,
      "loss": 4.7378,
      "mean_token_accuracy": 0.22634753137826918,
      "num_tokens": 127710420.0,
      "step": 73625
    },
    {
      "entropy": 5.386697769165039,
      "epoch": 5.7286520132269985,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022622833192046765,
      "loss": 4.6504,
      "mean_token_accuracy": 0.2333774283528328,
      "num_tokens": 127718505.0,
      "step": 73630
    },
    {
      "entropy": 5.435186862945557,
      "epoch": 5.729041042598716,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022620127566138948,
      "loss": 4.7943,
      "mean_token_accuracy": 0.22910430878400803,
      "num_tokens": 127727372.0,
      "step": 73635
    },
    {
      "entropy": 5.444840145111084,
      "epoch": 5.729430071970434,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022617422014275436,
      "loss": 4.7554,
      "mean_token_accuracy": 0.2180034562945366,
      "num_tokens": 127735731.0,
      "step": 73640
    },
    {
      "entropy": 5.391874551773071,
      "epoch": 5.7298191013421516,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002261471653649727,
      "loss": 4.7146,
      "mean_token_accuracy": 0.22967735975980758,
      "num_tokens": 127744442.0,
      "step": 73645
    },
    {
      "entropy": 5.380767917633056,
      "epoch": 5.730208130713869,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022612011132845506,
      "loss": 4.6737,
      "mean_token_accuracy": 0.2314346179366112,
      "num_tokens": 127753548.0,
      "step": 73650
    },
    {
      "entropy": 5.4210546016693115,
      "epoch": 5.730597160085587,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002260930580336119,
      "loss": 4.7364,
      "mean_token_accuracy": 0.22596848011016846,
      "num_tokens": 127762734.0,
      "step": 73655
    },
    {
      "entropy": 5.39854588508606,
      "epoch": 5.730986189457304,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00022606600548085375,
      "loss": 4.7481,
      "mean_token_accuracy": 0.2308316186070442,
      "num_tokens": 127772032.0,
      "step": 73660
    },
    {
      "entropy": 5.518133115768433,
      "epoch": 5.7313752188290215,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00022603895367059109,
      "loss": 4.7917,
      "mean_token_accuracy": 0.2305429458618164,
      "num_tokens": 127780731.0,
      "step": 73665
    },
    {
      "entropy": 5.483555126190185,
      "epoch": 5.731764248200739,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022601190260323434,
      "loss": 4.6596,
      "mean_token_accuracy": 0.23248251676559448,
      "num_tokens": 127789404.0,
      "step": 73670
    },
    {
      "entropy": 5.461701583862305,
      "epoch": 5.732153277572457,
      "grad_norm": 1.265625,
      "learning_rate": 0.000225984852279194,
      "loss": 4.7672,
      "mean_token_accuracy": 0.22734092324972152,
      "num_tokens": 127798232.0,
      "step": 73675
    },
    {
      "entropy": 5.367747640609741,
      "epoch": 5.7325423069441745,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022595780269888055,
      "loss": 4.6567,
      "mean_token_accuracy": 0.23423434048891068,
      "num_tokens": 127806952.0,
      "step": 73680
    },
    {
      "entropy": 5.445312070846557,
      "epoch": 5.732931336315891,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022593075386270438,
      "loss": 4.782,
      "mean_token_accuracy": 0.22917819321155547,
      "num_tokens": 127816178.0,
      "step": 73685
    },
    {
      "entropy": 5.413422155380249,
      "epoch": 5.733320365687609,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022590370577107595,
      "loss": 4.703,
      "mean_token_accuracy": 0.23152417689561844,
      "num_tokens": 127825381.0,
      "step": 73690
    },
    {
      "entropy": 5.4223192691802975,
      "epoch": 5.733709395059327,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002258766584244056,
      "loss": 4.6732,
      "mean_token_accuracy": 0.23360444903373717,
      "num_tokens": 127834232.0,
      "step": 73695
    },
    {
      "entropy": 5.429973793029785,
      "epoch": 5.7340984244310444,
      "grad_norm": 1.25,
      "learning_rate": 0.00022584961182310382,
      "loss": 4.7126,
      "mean_token_accuracy": 0.241447576880455,
      "num_tokens": 127842677.0,
      "step": 73700
    },
    {
      "entropy": 5.444335699081421,
      "epoch": 5.734487453802762,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022582256596758094,
      "loss": 4.6391,
      "mean_token_accuracy": 0.24439713209867478,
      "num_tokens": 127851021.0,
      "step": 73705
    },
    {
      "entropy": 5.450626945495605,
      "epoch": 5.73487648317448,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022579552085824728,
      "loss": 4.7583,
      "mean_token_accuracy": 0.2255527749657631,
      "num_tokens": 127859393.0,
      "step": 73710
    },
    {
      "entropy": 5.408960342407227,
      "epoch": 5.7352655125461975,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002257684764955133,
      "loss": 4.7515,
      "mean_token_accuracy": 0.22435106039047242,
      "num_tokens": 127867967.0,
      "step": 73715
    },
    {
      "entropy": 5.424421501159668,
      "epoch": 5.735654541917915,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022574143287978937,
      "loss": 4.7115,
      "mean_token_accuracy": 0.23482483327388765,
      "num_tokens": 127876761.0,
      "step": 73720
    },
    {
      "entropy": 5.422431468963623,
      "epoch": 5.736043571289632,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022571439001148574,
      "loss": 4.7027,
      "mean_token_accuracy": 0.2413535788655281,
      "num_tokens": 127885326.0,
      "step": 73725
    },
    {
      "entropy": 5.403384733200073,
      "epoch": 5.73643260066135,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022568734789101287,
      "loss": 4.7057,
      "mean_token_accuracy": 0.23271361440420152,
      "num_tokens": 127894404.0,
      "step": 73730
    },
    {
      "entropy": 5.401701974868774,
      "epoch": 5.736821630033067,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002256603065187809,
      "loss": 4.6396,
      "mean_token_accuracy": 0.23960778266191482,
      "num_tokens": 127903790.0,
      "step": 73735
    },
    {
      "entropy": 5.414132928848266,
      "epoch": 5.737210659404785,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00022563326589520027,
      "loss": 4.7151,
      "mean_token_accuracy": 0.23212529569864274,
      "num_tokens": 127911588.0,
      "step": 73740
    },
    {
      "entropy": 5.389629793167114,
      "epoch": 5.737599688776503,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022560622602068125,
      "loss": 4.725,
      "mean_token_accuracy": 0.23009524792432784,
      "num_tokens": 127919773.0,
      "step": 73745
    },
    {
      "entropy": 5.370613145828247,
      "epoch": 5.7379887181482205,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022557918689563423,
      "loss": 4.6567,
      "mean_token_accuracy": 0.23385123461484908,
      "num_tokens": 127927895.0,
      "step": 73750
    },
    {
      "entropy": 5.417871189117432,
      "epoch": 5.738377747519937,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002255521485204693,
      "loss": 4.7176,
      "mean_token_accuracy": 0.23000911325216294,
      "num_tokens": 127937245.0,
      "step": 73755
    },
    {
      "entropy": 5.418044662475586,
      "epoch": 5.738766776891655,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022552511089559684,
      "loss": 4.686,
      "mean_token_accuracy": 0.23142379969358445,
      "num_tokens": 127946256.0,
      "step": 73760
    },
    {
      "entropy": 5.481625127792358,
      "epoch": 5.739155806263373,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002254980740214271,
      "loss": 4.858,
      "mean_token_accuracy": 0.2227955400943756,
      "num_tokens": 127954916.0,
      "step": 73765
    },
    {
      "entropy": 5.3974974155426025,
      "epoch": 5.73954483563509,
      "grad_norm": 1.25,
      "learning_rate": 0.0002254710378983702,
      "loss": 4.6953,
      "mean_token_accuracy": 0.23457067161798478,
      "num_tokens": 127963442.0,
      "step": 73770
    },
    {
      "entropy": 5.464633989334106,
      "epoch": 5.739933865006808,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022544400252683656,
      "loss": 4.7986,
      "mean_token_accuracy": 0.22650066167116165,
      "num_tokens": 127972586.0,
      "step": 73775
    },
    {
      "entropy": 5.377392578125,
      "epoch": 5.740322894378526,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002254169679072363,
      "loss": 4.5663,
      "mean_token_accuracy": 0.2482578918337822,
      "num_tokens": 127980895.0,
      "step": 73780
    },
    {
      "entropy": 5.391235399246216,
      "epoch": 5.7407119237502435,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00022538993403997966,
      "loss": 4.5942,
      "mean_token_accuracy": 0.24795944690704347,
      "num_tokens": 127989350.0,
      "step": 73785
    },
    {
      "entropy": 5.378514814376831,
      "epoch": 5.74110095312196,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022536290092547672,
      "loss": 4.7042,
      "mean_token_accuracy": 0.23347381204366685,
      "num_tokens": 127997621.0,
      "step": 73790
    },
    {
      "entropy": 5.512717151641846,
      "epoch": 5.741489982493678,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002253358685641378,
      "loss": 4.792,
      "mean_token_accuracy": 0.22682008594274522,
      "num_tokens": 128006749.0,
      "step": 73795
    },
    {
      "entropy": 5.408950662612915,
      "epoch": 5.741879011865396,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000225308836956373,
      "loss": 4.6566,
      "mean_token_accuracy": 0.23567890375852585,
      "num_tokens": 128015786.0,
      "step": 73800
    },
    {
      "entropy": 5.4045709609985355,
      "epoch": 5.742268041237113,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022528180610259247,
      "loss": 4.6649,
      "mean_token_accuracy": 0.23491364419460298,
      "num_tokens": 128024175.0,
      "step": 73805
    },
    {
      "entropy": 5.425813865661621,
      "epoch": 5.742657070608831,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002252547760032065,
      "loss": 4.7364,
      "mean_token_accuracy": 0.22760885059833527,
      "num_tokens": 128034492.0,
      "step": 73810
    },
    {
      "entropy": 5.446305513381958,
      "epoch": 5.743046099980549,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022522774665862506,
      "loss": 4.8118,
      "mean_token_accuracy": 0.22012476921081542,
      "num_tokens": 128043562.0,
      "step": 73815
    },
    {
      "entropy": 5.360762739181519,
      "epoch": 5.743435129352266,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002252007180692584,
      "loss": 4.6433,
      "mean_token_accuracy": 0.2385942667722702,
      "num_tokens": 128051964.0,
      "step": 73820
    },
    {
      "entropy": 5.367166185379029,
      "epoch": 5.743824158723983,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022517369023551654,
      "loss": 4.6443,
      "mean_token_accuracy": 0.24060625731945037,
      "num_tokens": 128060907.0,
      "step": 73825
    },
    {
      "entropy": 5.396235752105713,
      "epoch": 5.744213188095701,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022514666315780963,
      "loss": 4.721,
      "mean_token_accuracy": 0.227887424826622,
      "num_tokens": 128070157.0,
      "step": 73830
    },
    {
      "entropy": 5.391544437408447,
      "epoch": 5.744602217467419,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022511963683654774,
      "loss": 4.6756,
      "mean_token_accuracy": 0.23627124428749086,
      "num_tokens": 128078204.0,
      "step": 73835
    },
    {
      "entropy": 5.434022092819214,
      "epoch": 5.744991246839136,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022509261127214098,
      "loss": 4.711,
      "mean_token_accuracy": 0.22877957671880722,
      "num_tokens": 128087137.0,
      "step": 73840
    },
    {
      "entropy": 5.374085140228272,
      "epoch": 5.745380276210854,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002250655864649994,
      "loss": 4.6551,
      "mean_token_accuracy": 0.23768645524978638,
      "num_tokens": 128096225.0,
      "step": 73845
    },
    {
      "entropy": 5.364320182800293,
      "epoch": 5.745769305582572,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002250385624155331,
      "loss": 4.6715,
      "mean_token_accuracy": 0.23089954107999802,
      "num_tokens": 128104511.0,
      "step": 73850
    },
    {
      "entropy": 5.45472993850708,
      "epoch": 5.7461583349542895,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00022501153912415206,
      "loss": 4.7477,
      "mean_token_accuracy": 0.23879833221435548,
      "num_tokens": 128113925.0,
      "step": 73855
    },
    {
      "entropy": 5.486845874786377,
      "epoch": 5.746547364326006,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022498451659126639,
      "loss": 4.8043,
      "mean_token_accuracy": 0.22685846090316772,
      "num_tokens": 128122325.0,
      "step": 73860
    },
    {
      "entropy": 5.409062051773072,
      "epoch": 5.746936393697724,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022495749481728604,
      "loss": 4.6891,
      "mean_token_accuracy": 0.23553712666034698,
      "num_tokens": 128130547.0,
      "step": 73865
    },
    {
      "entropy": 5.410422706604004,
      "epoch": 5.747325423069442,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000224930473802621,
      "loss": 4.6992,
      "mean_token_accuracy": 0.2317163109779358,
      "num_tokens": 128138602.0,
      "step": 73870
    },
    {
      "entropy": 5.439382410049438,
      "epoch": 5.747714452441159,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022490345354768138,
      "loss": 4.6878,
      "mean_token_accuracy": 0.23932355046272277,
      "num_tokens": 128147094.0,
      "step": 73875
    },
    {
      "entropy": 5.326736259460449,
      "epoch": 5.748103481812877,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002248764340528771,
      "loss": 4.6641,
      "mean_token_accuracy": 0.24019404649734497,
      "num_tokens": 128156360.0,
      "step": 73880
    },
    {
      "entropy": 5.417441940307617,
      "epoch": 5.748492511184595,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002248494153186182,
      "loss": 4.7166,
      "mean_token_accuracy": 0.23182264864444732,
      "num_tokens": 128165566.0,
      "step": 73885
    },
    {
      "entropy": 5.392944765090943,
      "epoch": 5.748881540556312,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022482239734531462,
      "loss": 4.7017,
      "mean_token_accuracy": 0.22837583720684052,
      "num_tokens": 128174533.0,
      "step": 73890
    },
    {
      "entropy": 5.370966386795044,
      "epoch": 5.749270569928029,
      "grad_norm": 1.25,
      "learning_rate": 0.00022479538013337624,
      "loss": 4.6524,
      "mean_token_accuracy": 0.2363009497523308,
      "num_tokens": 128182707.0,
      "step": 73895
    },
    {
      "entropy": 5.447098445892334,
      "epoch": 5.749659599299747,
      "grad_norm": 1.171875,
      "learning_rate": 0.00022476836368321302,
      "loss": 4.7808,
      "mean_token_accuracy": 0.22568032294511794,
      "num_tokens": 128191327.0,
      "step": 73900
    },
    {
      "entropy": 5.4696861743927006,
      "epoch": 5.750048628671465,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022474134799523499,
      "loss": 4.7725,
      "mean_token_accuracy": 0.23010928630828859,
      "num_tokens": 128199317.0,
      "step": 73905
    },
    {
      "entropy": 5.468022871017456,
      "epoch": 5.750437658043182,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022471433306985202,
      "loss": 4.7717,
      "mean_token_accuracy": 0.22585202157497405,
      "num_tokens": 128207936.0,
      "step": 73910
    },
    {
      "entropy": 5.47581114768982,
      "epoch": 5.7508266874149,
      "grad_norm": 1.21875,
      "learning_rate": 0.00022468731890747397,
      "loss": 4.7523,
      "mean_token_accuracy": 0.2323294073343277,
      "num_tokens": 128216355.0,
      "step": 73915
    },
    {
      "entropy": 5.445416641235352,
      "epoch": 5.751215716786618,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002246603055085108,
      "loss": 4.6943,
      "mean_token_accuracy": 0.2362087622284889,
      "num_tokens": 128224418.0,
      "step": 73920
    },
    {
      "entropy": 5.374938201904297,
      "epoch": 5.751604746158335,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00022463329287337235,
      "loss": 4.7128,
      "mean_token_accuracy": 0.23209499716758727,
      "num_tokens": 128233165.0,
      "step": 73925
    },
    {
      "entropy": 5.442474269866944,
      "epoch": 5.751993775530052,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002246062810024685,
      "loss": 4.7899,
      "mean_token_accuracy": 0.22759284675121308,
      "num_tokens": 128242225.0,
      "step": 73930
    },
    {
      "entropy": 5.432229328155517,
      "epoch": 5.75238280490177,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022457926989620915,
      "loss": 4.688,
      "mean_token_accuracy": 0.23734041899442673,
      "num_tokens": 128251145.0,
      "step": 73935
    },
    {
      "entropy": 5.46351466178894,
      "epoch": 5.752771834273488,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022455225955500408,
      "loss": 4.7498,
      "mean_token_accuracy": 0.2310709297657013,
      "num_tokens": 128259089.0,
      "step": 73940
    },
    {
      "entropy": 5.4063269138336185,
      "epoch": 5.753160863645205,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00022452524997926322,
      "loss": 4.6978,
      "mean_token_accuracy": 0.23250336945056915,
      "num_tokens": 128267919.0,
      "step": 73945
    },
    {
      "entropy": 5.4855663776397705,
      "epoch": 5.753549893016923,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002244982411693964,
      "loss": 4.8055,
      "mean_token_accuracy": 0.2275300294160843,
      "num_tokens": 128277095.0,
      "step": 73950
    },
    {
      "entropy": 5.356501340866089,
      "epoch": 5.75393892238864,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002244712331258133,
      "loss": 4.6186,
      "mean_token_accuracy": 0.24129323810338973,
      "num_tokens": 128285579.0,
      "step": 73955
    },
    {
      "entropy": 5.427411937713623,
      "epoch": 5.754327951760358,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002244442258489238,
      "loss": 4.6994,
      "mean_token_accuracy": 0.23572994023561478,
      "num_tokens": 128294436.0,
      "step": 73960
    },
    {
      "entropy": 5.423990392684937,
      "epoch": 5.754716981132075,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022441721933913778,
      "loss": 4.7973,
      "mean_token_accuracy": 0.22555950433015823,
      "num_tokens": 128304019.0,
      "step": 73965
    },
    {
      "entropy": 5.452146530151367,
      "epoch": 5.755106010503793,
      "grad_norm": 1.1875,
      "learning_rate": 0.000224390213596865,
      "loss": 4.7322,
      "mean_token_accuracy": 0.22729984819889068,
      "num_tokens": 128313121.0,
      "step": 73970
    },
    {
      "entropy": 5.390659475326538,
      "epoch": 5.755495039875511,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002243632086225151,
      "loss": 4.6646,
      "mean_token_accuracy": 0.2346356064081192,
      "num_tokens": 128321965.0,
      "step": 73975
    },
    {
      "entropy": 5.435939025878906,
      "epoch": 5.755884069247228,
      "grad_norm": 1.25,
      "learning_rate": 0.0002243362044164979,
      "loss": 4.7291,
      "mean_token_accuracy": 0.2328208401799202,
      "num_tokens": 128331163.0,
      "step": 73980
    },
    {
      "entropy": 5.441748237609863,
      "epoch": 5.756273098618946,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022430920097922313,
      "loss": 4.7748,
      "mean_token_accuracy": 0.2238151967525482,
      "num_tokens": 128339475.0,
      "step": 73985
    },
    {
      "entropy": 5.403398418426514,
      "epoch": 5.756662127990664,
      "grad_norm": 1.25,
      "learning_rate": 0.00022428219831110063,
      "loss": 4.69,
      "mean_token_accuracy": 0.23782752603292465,
      "num_tokens": 128347470.0,
      "step": 73990
    },
    {
      "entropy": 5.428991746902466,
      "epoch": 5.757051157362381,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022425519641254,
      "loss": 4.7555,
      "mean_token_accuracy": 0.22936067581176758,
      "num_tokens": 128356149.0,
      "step": 73995
    },
    {
      "entropy": 5.482574939727783,
      "epoch": 5.757440186734098,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022422819528395094,
      "loss": 4.793,
      "mean_token_accuracy": 0.22717421501874924,
      "num_tokens": 128364377.0,
      "step": 74000
    },
    {
      "entropy": 5.378600788116455,
      "epoch": 5.757829216105816,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022420119492574326,
      "loss": 4.6563,
      "mean_token_accuracy": 0.2377389192581177,
      "num_tokens": 128373019.0,
      "step": 74005
    },
    {
      "entropy": 5.396333360671997,
      "epoch": 5.758218245477534,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022417419533832662,
      "loss": 4.6073,
      "mean_token_accuracy": 0.24184005856513976,
      "num_tokens": 128380591.0,
      "step": 74010
    },
    {
      "entropy": 5.4169658660888675,
      "epoch": 5.758607274849251,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00022414719652211063,
      "loss": 4.7051,
      "mean_token_accuracy": 0.23891893327236174,
      "num_tokens": 128389826.0,
      "step": 74015
    },
    {
      "entropy": 5.343662738800049,
      "epoch": 5.758996304220968,
      "grad_norm": 1.25,
      "learning_rate": 0.00022412019847750498,
      "loss": 4.5654,
      "mean_token_accuracy": 0.24480109959840773,
      "num_tokens": 128397868.0,
      "step": 74020
    },
    {
      "entropy": 5.435055541992187,
      "epoch": 5.759385333592686,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002240932012049194,
      "loss": 4.7136,
      "mean_token_accuracy": 0.23906700164079667,
      "num_tokens": 128405314.0,
      "step": 74025
    },
    {
      "entropy": 5.319104909896851,
      "epoch": 5.759774362964404,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002240662047047634,
      "loss": 4.6029,
      "mean_token_accuracy": 0.24366777390241623,
      "num_tokens": 128414267.0,
      "step": 74030
    },
    {
      "entropy": 5.426264762878418,
      "epoch": 5.760163392336121,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022403920897744672,
      "loss": 4.7773,
      "mean_token_accuracy": 0.22767980992794037,
      "num_tokens": 128422492.0,
      "step": 74035
    },
    {
      "entropy": 5.419209718704224,
      "epoch": 5.760552421707839,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002240122140233789,
      "loss": 4.6753,
      "mean_token_accuracy": 0.23965973556041717,
      "num_tokens": 128430841.0,
      "step": 74040
    },
    {
      "entropy": 5.349329900741577,
      "epoch": 5.760941451079557,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002239852198429696,
      "loss": 4.662,
      "mean_token_accuracy": 0.23828287720680236,
      "num_tokens": 128439055.0,
      "step": 74045
    },
    {
      "entropy": 5.3890949249267575,
      "epoch": 5.761330480451274,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022395822643662844,
      "loss": 4.6844,
      "mean_token_accuracy": 0.23531799763441086,
      "num_tokens": 128447929.0,
      "step": 74050
    },
    {
      "entropy": 5.339063310623169,
      "epoch": 5.761719509822992,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022393123380476488,
      "loss": 4.6153,
      "mean_token_accuracy": 0.23798087388277053,
      "num_tokens": 128456111.0,
      "step": 74055
    },
    {
      "entropy": 5.373596334457398,
      "epoch": 5.762108539194709,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022390424194778858,
      "loss": 4.7139,
      "mean_token_accuracy": 0.23050627559423448,
      "num_tokens": 128464594.0,
      "step": 74060
    },
    {
      "entropy": 5.4263232231140135,
      "epoch": 5.762497568566427,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022387725086610904,
      "loss": 4.659,
      "mean_token_accuracy": 0.23774437010288238,
      "num_tokens": 128472620.0,
      "step": 74065
    },
    {
      "entropy": 5.433831262588501,
      "epoch": 5.762886597938144,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002238502605601358,
      "loss": 4.7137,
      "mean_token_accuracy": 0.23045556098222733,
      "num_tokens": 128481570.0,
      "step": 74070
    },
    {
      "entropy": 5.349554538726807,
      "epoch": 5.763275627309862,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022382327103027854,
      "loss": 4.6208,
      "mean_token_accuracy": 0.23337680101394653,
      "num_tokens": 128489703.0,
      "step": 74075
    },
    {
      "entropy": 5.459225034713745,
      "epoch": 5.76366465668158,
      "grad_norm": 1.375,
      "learning_rate": 0.00022379628227694675,
      "loss": 4.7086,
      "mean_token_accuracy": 0.23332745879888533,
      "num_tokens": 128497172.0,
      "step": 74080
    },
    {
      "entropy": 5.386168050765991,
      "epoch": 5.764053686053297,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022376929430054982,
      "loss": 4.6125,
      "mean_token_accuracy": 0.24308489710092546,
      "num_tokens": 128504890.0,
      "step": 74085
    },
    {
      "entropy": 5.39875340461731,
      "epoch": 5.764442715425014,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022374230710149723,
      "loss": 4.6249,
      "mean_token_accuracy": 0.24026311188936234,
      "num_tokens": 128512410.0,
      "step": 74090
    },
    {
      "entropy": 5.437826108932495,
      "epoch": 5.764831744796732,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022371532068019863,
      "loss": 4.8167,
      "mean_token_accuracy": 0.22937958985567092,
      "num_tokens": 128521556.0,
      "step": 74095
    },
    {
      "entropy": 5.397477293014527,
      "epoch": 5.76522077416845,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022368833503706338,
      "loss": 4.6703,
      "mean_token_accuracy": 0.23650281578302385,
      "num_tokens": 128530226.0,
      "step": 74100
    },
    {
      "entropy": 5.371507072448731,
      "epoch": 5.765609803540167,
      "grad_norm": 1.296875,
      "learning_rate": 0.000223661350172501,
      "loss": 4.5868,
      "mean_token_accuracy": 0.24629130959510803,
      "num_tokens": 128538178.0,
      "step": 74105
    },
    {
      "entropy": 5.371761751174927,
      "epoch": 5.765998832911885,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022363436608692085,
      "loss": 4.6705,
      "mean_token_accuracy": 0.23747970014810563,
      "num_tokens": 128546371.0,
      "step": 74110
    },
    {
      "entropy": 5.41437554359436,
      "epoch": 5.766387862283603,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002236073827807325,
      "loss": 4.6771,
      "mean_token_accuracy": 0.23324495404958726,
      "num_tokens": 128554764.0,
      "step": 74115
    },
    {
      "entropy": 5.397982835769653,
      "epoch": 5.76677689165532,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002235804002543453,
      "loss": 4.6587,
      "mean_token_accuracy": 0.244511678814888,
      "num_tokens": 128562759.0,
      "step": 74120
    },
    {
      "entropy": 5.326992034912109,
      "epoch": 5.767165921027037,
      "grad_norm": 1.25,
      "learning_rate": 0.0002235534185081687,
      "loss": 4.6663,
      "mean_token_accuracy": 0.2325431987643242,
      "num_tokens": 128571656.0,
      "step": 74125
    },
    {
      "entropy": 5.41493706703186,
      "epoch": 5.767554950398755,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002235264375426121,
      "loss": 4.7169,
      "mean_token_accuracy": 0.2352779671549797,
      "num_tokens": 128581202.0,
      "step": 74130
    },
    {
      "entropy": 5.433605623245239,
      "epoch": 5.767943979770473,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022349945735808475,
      "loss": 4.6487,
      "mean_token_accuracy": 0.24327562749385834,
      "num_tokens": 128589188.0,
      "step": 74135
    },
    {
      "entropy": 5.408305025100708,
      "epoch": 5.76833300914219,
      "grad_norm": 1.1875,
      "learning_rate": 0.00022347247795499625,
      "loss": 4.6922,
      "mean_token_accuracy": 0.23074866235256195,
      "num_tokens": 128597923.0,
      "step": 74140
    },
    {
      "entropy": 5.3879234790802,
      "epoch": 5.768722038513908,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022344549933375597,
      "loss": 4.6759,
      "mean_token_accuracy": 0.23637087941169738,
      "num_tokens": 128606478.0,
      "step": 74145
    },
    {
      "entropy": 5.359304618835449,
      "epoch": 5.769111067885626,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022341852149477309,
      "loss": 4.5827,
      "mean_token_accuracy": 0.24708991050720214,
      "num_tokens": 128614707.0,
      "step": 74150
    },
    {
      "entropy": 5.442174005508423,
      "epoch": 5.7695000972573425,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022339154443845705,
      "loss": 4.8125,
      "mean_token_accuracy": 0.22557737678289413,
      "num_tokens": 128623182.0,
      "step": 74155
    },
    {
      "entropy": 5.4692711353302,
      "epoch": 5.76988912662906,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022336456816521722,
      "loss": 4.7858,
      "mean_token_accuracy": 0.234621761739254,
      "num_tokens": 128631924.0,
      "step": 74160
    },
    {
      "entropy": 5.5559131622314455,
      "epoch": 5.770278156000778,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002233375926754629,
      "loss": 4.8156,
      "mean_token_accuracy": 0.2298200473189354,
      "num_tokens": 128640196.0,
      "step": 74165
    },
    {
      "entropy": 5.493968296051025,
      "epoch": 5.770667185372496,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00022331061796960328,
      "loss": 4.7266,
      "mean_token_accuracy": 0.2375857263803482,
      "num_tokens": 128648842.0,
      "step": 74170
    },
    {
      "entropy": 5.37688570022583,
      "epoch": 5.771056214744213,
      "grad_norm": 1.171875,
      "learning_rate": 0.00022328364404804784,
      "loss": 4.6355,
      "mean_token_accuracy": 0.24683040827512742,
      "num_tokens": 128657678.0,
      "step": 74175
    },
    {
      "entropy": 5.458963298797608,
      "epoch": 5.771445244115931,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022325667091120576,
      "loss": 4.7233,
      "mean_token_accuracy": 0.23484363108873368,
      "num_tokens": 128665655.0,
      "step": 74180
    },
    {
      "entropy": 5.405974674224853,
      "epoch": 5.771834273487649,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022322969855948638,
      "loss": 4.7287,
      "mean_token_accuracy": 0.22520939856767655,
      "num_tokens": 128674858.0,
      "step": 74185
    },
    {
      "entropy": 5.393363904953003,
      "epoch": 5.772223302859366,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022320272699329885,
      "loss": 4.5886,
      "mean_token_accuracy": 0.23660751879215242,
      "num_tokens": 128683541.0,
      "step": 74190
    },
    {
      "entropy": 5.442883729934692,
      "epoch": 5.772612332231083,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022317575621305254,
      "loss": 4.7516,
      "mean_token_accuracy": 0.2295124813914299,
      "num_tokens": 128692031.0,
      "step": 74195
    },
    {
      "entropy": 5.369556999206543,
      "epoch": 5.773001361602801,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022314878621915656,
      "loss": 4.6361,
      "mean_token_accuracy": 0.23340573012828827,
      "num_tokens": 128701352.0,
      "step": 74200
    },
    {
      "entropy": 5.46612515449524,
      "epoch": 5.773390390974519,
      "grad_norm": 1.25,
      "learning_rate": 0.00022312181701202033,
      "loss": 4.7678,
      "mean_token_accuracy": 0.23648154735565186,
      "num_tokens": 128710025.0,
      "step": 74205
    },
    {
      "entropy": 5.416579961776733,
      "epoch": 5.773779420346236,
      "grad_norm": 1.21875,
      "learning_rate": 0.00022309484859205293,
      "loss": 4.7182,
      "mean_token_accuracy": 0.22983547300100327,
      "num_tokens": 128719243.0,
      "step": 74210
    },
    {
      "entropy": 5.431369209289551,
      "epoch": 5.774168449717954,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022306788095966356,
      "loss": 4.6966,
      "mean_token_accuracy": 0.23154416382312776,
      "num_tokens": 128727591.0,
      "step": 74215
    },
    {
      "entropy": 5.428254652023315,
      "epoch": 5.774557479089672,
      "grad_norm": 1.1875,
      "learning_rate": 0.00022304091411526144,
      "loss": 4.8233,
      "mean_token_accuracy": 0.2282763496041298,
      "num_tokens": 128736712.0,
      "step": 74220
    },
    {
      "entropy": 5.4492286205291744,
      "epoch": 5.7749465084613885,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002230139480592558,
      "loss": 4.6757,
      "mean_token_accuracy": 0.23108662366867067,
      "num_tokens": 128745347.0,
      "step": 74225
    },
    {
      "entropy": 5.424688482284546,
      "epoch": 5.775335537833106,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022298698279205571,
      "loss": 4.7002,
      "mean_token_accuracy": 0.23164417296648027,
      "num_tokens": 128753605.0,
      "step": 74230
    },
    {
      "entropy": 5.464305973052978,
      "epoch": 5.775724567204824,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00022296001831407036,
      "loss": 4.7734,
      "mean_token_accuracy": 0.22795771062374115,
      "num_tokens": 128762206.0,
      "step": 74235
    },
    {
      "entropy": 5.409697675704956,
      "epoch": 5.7761135965765416,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022293305462570896,
      "loss": 4.6824,
      "mean_token_accuracy": 0.2267390936613083,
      "num_tokens": 128770673.0,
      "step": 74240
    },
    {
      "entropy": 5.474594926834106,
      "epoch": 5.776502625948259,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022290609172738058,
      "loss": 4.767,
      "mean_token_accuracy": 0.22980809062719346,
      "num_tokens": 128779162.0,
      "step": 74245
    },
    {
      "entropy": 5.399310111999512,
      "epoch": 5.776891655319977,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00022287912961949425,
      "loss": 4.6485,
      "mean_token_accuracy": 0.24672498255968095,
      "num_tokens": 128788156.0,
      "step": 74250
    },
    {
      "entropy": 5.443849039077759,
      "epoch": 5.777280684691695,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022285216830245924,
      "loss": 4.7092,
      "mean_token_accuracy": 0.2387464836239815,
      "num_tokens": 128796061.0,
      "step": 74255
    },
    {
      "entropy": 5.371252012252808,
      "epoch": 5.7776697140634115,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002228252077766846,
      "loss": 4.621,
      "mean_token_accuracy": 0.24149923175573348,
      "num_tokens": 128804629.0,
      "step": 74260
    },
    {
      "entropy": 5.387916183471679,
      "epoch": 5.778058743435129,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002227982480425793,
      "loss": 4.6451,
      "mean_token_accuracy": 0.23638035506010055,
      "num_tokens": 128812626.0,
      "step": 74265
    },
    {
      "entropy": 5.458868980407715,
      "epoch": 5.778447772806847,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002227712891005525,
      "loss": 4.7805,
      "mean_token_accuracy": 0.23028348833322526,
      "num_tokens": 128822471.0,
      "step": 74270
    },
    {
      "entropy": 5.48684549331665,
      "epoch": 5.7788368021785645,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022274433095101332,
      "loss": 4.769,
      "mean_token_accuracy": 0.2296008959412575,
      "num_tokens": 128831980.0,
      "step": 74275
    },
    {
      "entropy": 5.491603660583496,
      "epoch": 5.779225831550282,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022271737359437073,
      "loss": 4.7682,
      "mean_token_accuracy": 0.23590102791786194,
      "num_tokens": 128840090.0,
      "step": 74280
    },
    {
      "entropy": 5.360780239105225,
      "epoch": 5.779614860922,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022269041703103375,
      "loss": 4.5644,
      "mean_token_accuracy": 0.2381490260362625,
      "num_tokens": 128848101.0,
      "step": 74285
    },
    {
      "entropy": 5.3654265880584715,
      "epoch": 5.780003890293717,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022266346126141145,
      "loss": 4.66,
      "mean_token_accuracy": 0.23792554289102555,
      "num_tokens": 128856873.0,
      "step": 74290
    },
    {
      "entropy": 5.382408666610718,
      "epoch": 5.7803929196654344,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002226365062859128,
      "loss": 4.7238,
      "mean_token_accuracy": 0.23125093430280685,
      "num_tokens": 128865519.0,
      "step": 74295
    },
    {
      "entropy": 5.383880472183227,
      "epoch": 5.780781949037152,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022260955210494688,
      "loss": 4.6633,
      "mean_token_accuracy": 0.23730054646730422,
      "num_tokens": 128874487.0,
      "step": 74300
    },
    {
      "entropy": 5.48652024269104,
      "epoch": 5.78117097840887,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022258259871892256,
      "loss": 4.8522,
      "mean_token_accuracy": 0.22080710977315904,
      "num_tokens": 128883696.0,
      "step": 74305
    },
    {
      "entropy": 5.461329936981201,
      "epoch": 5.7815600077805875,
      "grad_norm": 1.25,
      "learning_rate": 0.0002225556461282489,
      "loss": 4.713,
      "mean_token_accuracy": 0.2350350558757782,
      "num_tokens": 128892117.0,
      "step": 74310
    },
    {
      "entropy": 5.479112577438355,
      "epoch": 5.781949037152305,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022252869433333483,
      "loss": 4.7489,
      "mean_token_accuracy": 0.23064574748277664,
      "num_tokens": 128900660.0,
      "step": 74315
    },
    {
      "entropy": 5.309270429611206,
      "epoch": 5.782338066524023,
      "grad_norm": 1.25,
      "learning_rate": 0.0002225017433345894,
      "loss": 4.5595,
      "mean_token_accuracy": 0.2465588390827179,
      "num_tokens": 128909491.0,
      "step": 74320
    },
    {
      "entropy": 5.43945541381836,
      "epoch": 5.78272709589574,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022247479313242135,
      "loss": 4.7892,
      "mean_token_accuracy": 0.22952382117509842,
      "num_tokens": 128918476.0,
      "step": 74325
    },
    {
      "entropy": 5.445417881011963,
      "epoch": 5.783116125267457,
      "grad_norm": 1.171875,
      "learning_rate": 0.00022244784372723976,
      "loss": 4.6828,
      "mean_token_accuracy": 0.24167085886001588,
      "num_tokens": 128927192.0,
      "step": 74330
    },
    {
      "entropy": 5.437497091293335,
      "epoch": 5.783505154639175,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022242089511945336,
      "loss": 4.715,
      "mean_token_accuracy": 0.23190619349479674,
      "num_tokens": 128935467.0,
      "step": 74335
    },
    {
      "entropy": 5.394255781173706,
      "epoch": 5.783894184010893,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022239394730947127,
      "loss": 4.7223,
      "mean_token_accuracy": 0.22848013639450074,
      "num_tokens": 128943440.0,
      "step": 74340
    },
    {
      "entropy": 5.519095134735108,
      "epoch": 5.7842832133826105,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022236700029770235,
      "loss": 4.8066,
      "mean_token_accuracy": 0.23265400826931,
      "num_tokens": 128952020.0,
      "step": 74345
    },
    {
      "entropy": 5.32225866317749,
      "epoch": 5.784672242754328,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022234005408455543,
      "loss": 4.5446,
      "mean_token_accuracy": 0.24647454917430878,
      "num_tokens": 128960799.0,
      "step": 74350
    },
    {
      "entropy": 5.410928010940552,
      "epoch": 5.785061272126045,
      "grad_norm": 1.25,
      "learning_rate": 0.00022231310867043934,
      "loss": 4.7463,
      "mean_token_accuracy": 0.23711289018392562,
      "num_tokens": 128969331.0,
      "step": 74355
    },
    {
      "entropy": 5.431405878067016,
      "epoch": 5.785450301497763,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022228616405576296,
      "loss": 4.7089,
      "mean_token_accuracy": 0.22777220755815505,
      "num_tokens": 128978014.0,
      "step": 74360
    },
    {
      "entropy": 5.3731261730194095,
      "epoch": 5.78583933086948,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022225922024093515,
      "loss": 4.6813,
      "mean_token_accuracy": 0.23613603115081788,
      "num_tokens": 128986678.0,
      "step": 74365
    },
    {
      "entropy": 5.381593179702759,
      "epoch": 5.786228360241198,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00022223227722636474,
      "loss": 4.6506,
      "mean_token_accuracy": 0.23816826939582825,
      "num_tokens": 128995292.0,
      "step": 74370
    },
    {
      "entropy": 5.3938836574554445,
      "epoch": 5.786617389612916,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022220533501246049,
      "loss": 4.6873,
      "mean_token_accuracy": 0.2386918395757675,
      "num_tokens": 129003886.0,
      "step": 74375
    },
    {
      "entropy": 5.347758817672729,
      "epoch": 5.7870064189846335,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022217839359963133,
      "loss": 4.6517,
      "mean_token_accuracy": 0.24117647856473923,
      "num_tokens": 129012548.0,
      "step": 74380
    },
    {
      "entropy": 5.473883390426636,
      "epoch": 5.787395448356351,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00022215145298828587,
      "loss": 4.837,
      "mean_token_accuracy": 0.22882620096206666,
      "num_tokens": 129020882.0,
      "step": 74385
    },
    {
      "entropy": 5.396356058120728,
      "epoch": 5.787784477728069,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002221245131788331,
      "loss": 4.7069,
      "mean_token_accuracy": 0.2335044413805008,
      "num_tokens": 129029317.0,
      "step": 74390
    },
    {
      "entropy": 5.4961690425872805,
      "epoch": 5.788173507099786,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002220975741716816,
      "loss": 4.7547,
      "mean_token_accuracy": 0.23673991709947587,
      "num_tokens": 129036731.0,
      "step": 74395
    },
    {
      "entropy": 5.433673524856568,
      "epoch": 5.788562536471503,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002220706359672402,
      "loss": 4.8183,
      "mean_token_accuracy": 0.22737989872694014,
      "num_tokens": 129045868.0,
      "step": 74400
    },
    {
      "entropy": 5.458584403991699,
      "epoch": 5.788951565843221,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022204369856591765,
      "loss": 4.7754,
      "mean_token_accuracy": 0.22906596809625626,
      "num_tokens": 129054380.0,
      "step": 74405
    },
    {
      "entropy": 5.36266679763794,
      "epoch": 5.789340595214939,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022201676196812275,
      "loss": 4.6357,
      "mean_token_accuracy": 0.24031752794981004,
      "num_tokens": 129062571.0,
      "step": 74410
    },
    {
      "entropy": 5.371146631240845,
      "epoch": 5.7897296245866565,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022198982617426417,
      "loss": 4.6854,
      "mean_token_accuracy": 0.24225378036499023,
      "num_tokens": 129070997.0,
      "step": 74415
    },
    {
      "entropy": 5.429295015335083,
      "epoch": 5.790118653958374,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022196289118475056,
      "loss": 4.7203,
      "mean_token_accuracy": 0.23503382056951522,
      "num_tokens": 129078494.0,
      "step": 74420
    },
    {
      "entropy": 5.355884265899658,
      "epoch": 5.790507683330091,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022193595699999063,
      "loss": 4.6476,
      "mean_token_accuracy": 0.24502550065517426,
      "num_tokens": 129086925.0,
      "step": 74425
    },
    {
      "entropy": 5.416491222381592,
      "epoch": 5.790896712701809,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022190902362039316,
      "loss": 4.7291,
      "mean_token_accuracy": 0.2254323199391365,
      "num_tokens": 129094765.0,
      "step": 74430
    },
    {
      "entropy": 5.465409612655639,
      "epoch": 5.791285742073526,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022188209104636675,
      "loss": 4.8012,
      "mean_token_accuracy": 0.22590084373950958,
      "num_tokens": 129103098.0,
      "step": 74435
    },
    {
      "entropy": 5.403447055816651,
      "epoch": 5.791674771445244,
      "grad_norm": 1.21875,
      "learning_rate": 0.00022185515927832,
      "loss": 4.7397,
      "mean_token_accuracy": 0.232652048766613,
      "num_tokens": 129112315.0,
      "step": 74440
    },
    {
      "entropy": 5.514015960693359,
      "epoch": 5.792063800816962,
      "grad_norm": 1.25,
      "learning_rate": 0.00022182822831666167,
      "loss": 4.8215,
      "mean_token_accuracy": 0.22247550636529922,
      "num_tokens": 129121111.0,
      "step": 74445
    },
    {
      "entropy": 5.426935195922852,
      "epoch": 5.7924528301886795,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00022180129816180033,
      "loss": 4.7084,
      "mean_token_accuracy": 0.23662593066692353,
      "num_tokens": 129129974.0,
      "step": 74450
    },
    {
      "entropy": 5.475278902053833,
      "epoch": 5.792841859560397,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022177436881414458,
      "loss": 4.7038,
      "mean_token_accuracy": 0.23413096964359284,
      "num_tokens": 129138786.0,
      "step": 74455
    },
    {
      "entropy": 5.397673654556274,
      "epoch": 5.793230888932114,
      "grad_norm": 1.25,
      "learning_rate": 0.00022174744027410315,
      "loss": 4.67,
      "mean_token_accuracy": 0.23068359792232512,
      "num_tokens": 129147244.0,
      "step": 74460
    },
    {
      "entropy": 5.4286301612854,
      "epoch": 5.793619918303832,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00022172051254208442,
      "loss": 4.7399,
      "mean_token_accuracy": 0.23031911253929138,
      "num_tokens": 129156613.0,
      "step": 74465
    },
    {
      "entropy": 5.410142612457276,
      "epoch": 5.794008947675549,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022169358561849724,
      "loss": 4.6695,
      "mean_token_accuracy": 0.23844918459653855,
      "num_tokens": 129165361.0,
      "step": 74470
    },
    {
      "entropy": 5.514263486862182,
      "epoch": 5.794397977047267,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022166665950374998,
      "loss": 4.7906,
      "mean_token_accuracy": 0.22912862300872802,
      "num_tokens": 129173479.0,
      "step": 74475
    },
    {
      "entropy": 5.342743062973023,
      "epoch": 5.794787006418985,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022163973419825128,
      "loss": 4.6053,
      "mean_token_accuracy": 0.23137644231319426,
      "num_tokens": 129182294.0,
      "step": 74480
    },
    {
      "entropy": 5.404165840148925,
      "epoch": 5.7951760357907025,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002216128097024097,
      "loss": 4.7456,
      "mean_token_accuracy": 0.23067908585071564,
      "num_tokens": 129190518.0,
      "step": 74485
    },
    {
      "entropy": 5.340090322494507,
      "epoch": 5.795565065162419,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022158588601663376,
      "loss": 4.5905,
      "mean_token_accuracy": 0.24503228068351746,
      "num_tokens": 129199465.0,
      "step": 74490
    },
    {
      "entropy": 5.365250444412231,
      "epoch": 5.795954094534137,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00022155896314133194,
      "loss": 4.5999,
      "mean_token_accuracy": 0.24214914739131926,
      "num_tokens": 129208302.0,
      "step": 74495
    },
    {
      "entropy": 5.399610233306885,
      "epoch": 5.796343123905855,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022153204107691287,
      "loss": 4.6345,
      "mean_token_accuracy": 0.23436587750911714,
      "num_tokens": 129216827.0,
      "step": 74500
    },
    {
      "entropy": 5.39179425239563,
      "epoch": 5.796732153277572,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022150511982378491,
      "loss": 4.6874,
      "mean_token_accuracy": 0.23448649048805237,
      "num_tokens": 129225097.0,
      "step": 74505
    },
    {
      "entropy": 5.354122972488403,
      "epoch": 5.79712118264929,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002214781993823567,
      "loss": 4.6383,
      "mean_token_accuracy": 0.2402149721980095,
      "num_tokens": 129233289.0,
      "step": 74510
    },
    {
      "entropy": 5.445846128463745,
      "epoch": 5.797510212021008,
      "grad_norm": 1.1875,
      "learning_rate": 0.00022145127975303658,
      "loss": 4.8242,
      "mean_token_accuracy": 0.22951031625270843,
      "num_tokens": 129242261.0,
      "step": 74515
    },
    {
      "entropy": 5.421792507171631,
      "epoch": 5.7978992413927255,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022142436093623304,
      "loss": 4.7742,
      "mean_token_accuracy": 0.2321198508143425,
      "num_tokens": 129251408.0,
      "step": 74520
    },
    {
      "entropy": 5.478861474990845,
      "epoch": 5.798288270764443,
      "grad_norm": 1.1875,
      "learning_rate": 0.00022139744293235453,
      "loss": 4.8017,
      "mean_token_accuracy": 0.22436314225196838,
      "num_tokens": 129260950.0,
      "step": 74525
    },
    {
      "entropy": 5.486075162887573,
      "epoch": 5.79867730013616,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022137052574180955,
      "loss": 4.7252,
      "mean_token_accuracy": 0.2276096075773239,
      "num_tokens": 129269136.0,
      "step": 74530
    },
    {
      "entropy": 5.457859134674072,
      "epoch": 5.799066329507878,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00022134360936500636,
      "loss": 4.7982,
      "mean_token_accuracy": 0.220009046792984,
      "num_tokens": 129278443.0,
      "step": 74535
    },
    {
      "entropy": 5.337863969802856,
      "epoch": 5.799455358879595,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022131669380235364,
      "loss": 4.5996,
      "mean_token_accuracy": 0.23430920094251634,
      "num_tokens": 129287423.0,
      "step": 74540
    },
    {
      "entropy": 5.316799163818359,
      "epoch": 5.799844388251313,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022128977905425955,
      "loss": 4.6197,
      "mean_token_accuracy": 0.24310948848724365,
      "num_tokens": 129295974.0,
      "step": 74545
    },
    {
      "entropy": 5.429731798171997,
      "epoch": 5.800233417623031,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022126286512113265,
      "loss": 4.7767,
      "mean_token_accuracy": 0.22923233807086946,
      "num_tokens": 129304269.0,
      "step": 74550
    },
    {
      "entropy": 5.454990816116333,
      "epoch": 5.800622446994748,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002212359520033812,
      "loss": 4.7799,
      "mean_token_accuracy": 0.22976387441158294,
      "num_tokens": 129313493.0,
      "step": 74555
    },
    {
      "entropy": 5.4604839324951175,
      "epoch": 5.801011476366465,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022120903970141364,
      "loss": 4.7058,
      "mean_token_accuracy": 0.23522233664989473,
      "num_tokens": 129321634.0,
      "step": 74560
    },
    {
      "entropy": 5.36354022026062,
      "epoch": 5.801400505738183,
      "grad_norm": 1.140625,
      "learning_rate": 0.00022118212821563827,
      "loss": 4.6927,
      "mean_token_accuracy": 0.22870127409696578,
      "num_tokens": 129332285.0,
      "step": 74565
    },
    {
      "entropy": 5.4068482398986815,
      "epoch": 5.801789535109901,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022115521754646346,
      "loss": 4.6765,
      "mean_token_accuracy": 0.24199732393026352,
      "num_tokens": 129340890.0,
      "step": 74570
    },
    {
      "entropy": 5.435033559799194,
      "epoch": 5.802178564481618,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022112830769429744,
      "loss": 4.7345,
      "mean_token_accuracy": 0.2355238229036331,
      "num_tokens": 129349327.0,
      "step": 74575
    },
    {
      "entropy": 5.484268522262573,
      "epoch": 5.802567593853336,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022110139865954865,
      "loss": 4.7307,
      "mean_token_accuracy": 0.2255270451307297,
      "num_tokens": 129357588.0,
      "step": 74580
    },
    {
      "entropy": 5.4908465385437015,
      "epoch": 5.802956623225054,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002210744904426254,
      "loss": 4.7311,
      "mean_token_accuracy": 0.226112200319767,
      "num_tokens": 129366473.0,
      "step": 74585
    },
    {
      "entropy": 5.534784603118896,
      "epoch": 5.8033456525967715,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002210475830439359,
      "loss": 4.8348,
      "mean_token_accuracy": 0.21948008984327316,
      "num_tokens": 129374639.0,
      "step": 74590
    },
    {
      "entropy": 5.392705249786377,
      "epoch": 5.803734681968488,
      "grad_norm": 1.21875,
      "learning_rate": 0.00022102067646388845,
      "loss": 4.6725,
      "mean_token_accuracy": 0.2400367259979248,
      "num_tokens": 129383334.0,
      "step": 74595
    },
    {
      "entropy": 5.324016475677491,
      "epoch": 5.804123711340206,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022099377070289124,
      "loss": 4.6695,
      "mean_token_accuracy": 0.23413713425397872,
      "num_tokens": 129391267.0,
      "step": 74600
    },
    {
      "entropy": 5.374117231369018,
      "epoch": 5.804512740711924,
      "grad_norm": 1.171875,
      "learning_rate": 0.00022096686576135268,
      "loss": 4.6645,
      "mean_token_accuracy": 0.23549238890409468,
      "num_tokens": 129399631.0,
      "step": 74605
    },
    {
      "entropy": 5.508076810836792,
      "epoch": 5.804901770083641,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022093996163968088,
      "loss": 4.8077,
      "mean_token_accuracy": 0.22782034426927567,
      "num_tokens": 129408742.0,
      "step": 74610
    },
    {
      "entropy": 5.450793266296387,
      "epoch": 5.805290799455359,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022091305833828412,
      "loss": 4.7224,
      "mean_token_accuracy": 0.2326617181301117,
      "num_tokens": 129416793.0,
      "step": 74615
    },
    {
      "entropy": 5.386793851852417,
      "epoch": 5.805679828827077,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00022088615585757072,
      "loss": 4.685,
      "mean_token_accuracy": 0.23962218016386033,
      "num_tokens": 129425785.0,
      "step": 74620
    },
    {
      "entropy": 5.367297744750976,
      "epoch": 5.806068858198794,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00022085925419794872,
      "loss": 4.6607,
      "mean_token_accuracy": 0.2411056786775589,
      "num_tokens": 129434962.0,
      "step": 74625
    },
    {
      "entropy": 5.401784038543701,
      "epoch": 5.806457887570511,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022083235335982633,
      "loss": 4.7868,
      "mean_token_accuracy": 0.22850004583597183,
      "num_tokens": 129444034.0,
      "step": 74630
    },
    {
      "entropy": 5.346003913879395,
      "epoch": 5.806846916942229,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00022080545334361175,
      "loss": 4.6402,
      "mean_token_accuracy": 0.23843393176794053,
      "num_tokens": 129453278.0,
      "step": 74635
    },
    {
      "entropy": 5.435863971710205,
      "epoch": 5.807235946313947,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002207785541497132,
      "loss": 4.7527,
      "mean_token_accuracy": 0.2252323567867279,
      "num_tokens": 129461541.0,
      "step": 74640
    },
    {
      "entropy": 5.4055928707122805,
      "epoch": 5.807624975685664,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00022075165577853872,
      "loss": 4.7088,
      "mean_token_accuracy": 0.23647991120815276,
      "num_tokens": 129470098.0,
      "step": 74645
    },
    {
      "entropy": 5.353036642074585,
      "epoch": 5.808014005057382,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002207247582304966,
      "loss": 4.6058,
      "mean_token_accuracy": 0.23366523087024688,
      "num_tokens": 129477970.0,
      "step": 74650
    },
    {
      "entropy": 5.418648958206177,
      "epoch": 5.8084030344291,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022069786150599486,
      "loss": 4.6587,
      "mean_token_accuracy": 0.23883021175861358,
      "num_tokens": 129486115.0,
      "step": 74655
    },
    {
      "entropy": 5.401842832565308,
      "epoch": 5.808792063800817,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022067096560544165,
      "loss": 4.6821,
      "mean_token_accuracy": 0.23765720576047897,
      "num_tokens": 129494360.0,
      "step": 74660
    },
    {
      "entropy": 5.382536792755127,
      "epoch": 5.809181093172534,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022064407052924494,
      "loss": 4.7044,
      "mean_token_accuracy": 0.24210364371538162,
      "num_tokens": 129503547.0,
      "step": 74665
    },
    {
      "entropy": 5.351253938674927,
      "epoch": 5.809570122544252,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022061717627781308,
      "loss": 4.6619,
      "mean_token_accuracy": 0.2442904904484749,
      "num_tokens": 129511723.0,
      "step": 74670
    },
    {
      "entropy": 5.420757102966308,
      "epoch": 5.80995915191597,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022059028285155397,
      "loss": 4.6625,
      "mean_token_accuracy": 0.23985886871814727,
      "num_tokens": 129519965.0,
      "step": 74675
    },
    {
      "entropy": 5.613866567611694,
      "epoch": 5.810348181287687,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022056339025087573,
      "loss": 4.872,
      "mean_token_accuracy": 0.22071827352046966,
      "num_tokens": 129529059.0,
      "step": 74680
    },
    {
      "entropy": 5.406025314331055,
      "epoch": 5.810737210659405,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002205364984761864,
      "loss": 4.6648,
      "mean_token_accuracy": 0.24102855920791627,
      "num_tokens": 129537536.0,
      "step": 74685
    },
    {
      "entropy": 5.355253410339356,
      "epoch": 5.811126240031122,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000220509607527894,
      "loss": 4.7019,
      "mean_token_accuracy": 0.23133934140205384,
      "num_tokens": 129546579.0,
      "step": 74690
    },
    {
      "entropy": 5.311376285552979,
      "epoch": 5.81151526940284,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002204827174064066,
      "loss": 4.5972,
      "mean_token_accuracy": 0.2452852413058281,
      "num_tokens": 129555878.0,
      "step": 74695
    },
    {
      "entropy": 5.347614002227783,
      "epoch": 5.811904298774557,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022045582811213221,
      "loss": 4.6539,
      "mean_token_accuracy": 0.23980513364076614,
      "num_tokens": 129564612.0,
      "step": 74700
    },
    {
      "entropy": 5.478641080856323,
      "epoch": 5.812293328146275,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00022042893964547877,
      "loss": 4.7506,
      "mean_token_accuracy": 0.2272763356566429,
      "num_tokens": 129573084.0,
      "step": 74705
    },
    {
      "entropy": 5.4538696765899655,
      "epoch": 5.812682357517993,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002204020520068543,
      "loss": 4.7387,
      "mean_token_accuracy": 0.23256502002477647,
      "num_tokens": 129581446.0,
      "step": 74710
    },
    {
      "entropy": 5.42230257987976,
      "epoch": 5.81307138688971,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022037516519666684,
      "loss": 4.7568,
      "mean_token_accuracy": 0.2264723852276802,
      "num_tokens": 129590935.0,
      "step": 74715
    },
    {
      "entropy": 5.47996335029602,
      "epoch": 5.813460416261428,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002203482792153243,
      "loss": 4.7578,
      "mean_token_accuracy": 0.2241360515356064,
      "num_tokens": 129599237.0,
      "step": 74720
    },
    {
      "entropy": 5.4172422885894775,
      "epoch": 5.813849445633146,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002203213940632346,
      "loss": 4.6106,
      "mean_token_accuracy": 0.2370064750313759,
      "num_tokens": 129607915.0,
      "step": 74725
    },
    {
      "entropy": 5.4401463031768795,
      "epoch": 5.814238475004863,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022029450974080572,
      "loss": 4.7376,
      "mean_token_accuracy": 0.2277721256017685,
      "num_tokens": 129616659.0,
      "step": 74730
    },
    {
      "entropy": 5.386662817001342,
      "epoch": 5.81462750437658,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022026762624844555,
      "loss": 4.6854,
      "mean_token_accuracy": 0.23934656977653504,
      "num_tokens": 129624950.0,
      "step": 74735
    },
    {
      "entropy": 5.3544025897979735,
      "epoch": 5.815016533748298,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022024074358656205,
      "loss": 4.644,
      "mean_token_accuracy": 0.23759970515966417,
      "num_tokens": 129633977.0,
      "step": 74740
    },
    {
      "entropy": 5.3817283630371096,
      "epoch": 5.815405563120016,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022021386175556318,
      "loss": 4.6426,
      "mean_token_accuracy": 0.23941560089588165,
      "num_tokens": 129641926.0,
      "step": 74745
    },
    {
      "entropy": 5.3997111320495605,
      "epoch": 5.815794592491733,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002201869807558567,
      "loss": 4.7394,
      "mean_token_accuracy": 0.22222796976566314,
      "num_tokens": 129651030.0,
      "step": 74750
    },
    {
      "entropy": 5.455397605895996,
      "epoch": 5.816183621863451,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022016010058785053,
      "loss": 4.7558,
      "mean_token_accuracy": 0.2309670552611351,
      "num_tokens": 129658701.0,
      "step": 74755
    },
    {
      "entropy": 5.4305548667907715,
      "epoch": 5.816572651235168,
      "grad_norm": 1.21875,
      "learning_rate": 0.00022013322125195256,
      "loss": 4.6775,
      "mean_token_accuracy": 0.23988669067621232,
      "num_tokens": 129666641.0,
      "step": 74760
    },
    {
      "entropy": 5.384586048126221,
      "epoch": 5.816961680606886,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002201063427485706,
      "loss": 4.7764,
      "mean_token_accuracy": 0.23116834163665773,
      "num_tokens": 129674575.0,
      "step": 74765
    },
    {
      "entropy": 5.366006517410279,
      "epoch": 5.817350709978603,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002200794650781126,
      "loss": 4.7118,
      "mean_token_accuracy": 0.23996674865484238,
      "num_tokens": 129683593.0,
      "step": 74770
    },
    {
      "entropy": 5.367778539657593,
      "epoch": 5.817739739350321,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022005258824098624,
      "loss": 4.698,
      "mean_token_accuracy": 0.23109169453382492,
      "num_tokens": 129692009.0,
      "step": 74775
    },
    {
      "entropy": 5.399206018447876,
      "epoch": 5.818128768722039,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002200257122375995,
      "loss": 4.6402,
      "mean_token_accuracy": 0.24286833852529527,
      "num_tokens": 129701156.0,
      "step": 74780
    },
    {
      "entropy": 5.4480749607086185,
      "epoch": 5.818517798093756,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00021999883706835994,
      "loss": 4.7337,
      "mean_token_accuracy": 0.23289646506309508,
      "num_tokens": 129709985.0,
      "step": 74785
    },
    {
      "entropy": 5.40141487121582,
      "epoch": 5.818906827465474,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021997196273367553,
      "loss": 4.6788,
      "mean_token_accuracy": 0.23765258938074113,
      "num_tokens": 129718723.0,
      "step": 74790
    },
    {
      "entropy": 5.358666276931762,
      "epoch": 5.819295856837191,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000219945089233954,
      "loss": 4.6325,
      "mean_token_accuracy": 0.23483209311962128,
      "num_tokens": 129727557.0,
      "step": 74795
    },
    {
      "entropy": 5.457809066772461,
      "epoch": 5.819684886208909,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021991821656960303,
      "loss": 4.8202,
      "mean_token_accuracy": 0.22609094977378846,
      "num_tokens": 129736514.0,
      "step": 74800
    },
    {
      "entropy": 5.408131694793701,
      "epoch": 5.820073915580626,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021989134474103056,
      "loss": 4.6778,
      "mean_token_accuracy": 0.23687091171741487,
      "num_tokens": 129745327.0,
      "step": 74805
    },
    {
      "entropy": 5.335020542144775,
      "epoch": 5.820462944952344,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021986447374864422,
      "loss": 4.6585,
      "mean_token_accuracy": 0.2289414331316948,
      "num_tokens": 129753843.0,
      "step": 74810
    },
    {
      "entropy": 5.533915328979492,
      "epoch": 5.820851974324062,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021983760359285171,
      "loss": 4.8504,
      "mean_token_accuracy": 0.2298211023211479,
      "num_tokens": 129762438.0,
      "step": 74815
    },
    {
      "entropy": 5.425169467926025,
      "epoch": 5.821241003695779,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002198107342740608,
      "loss": 4.6794,
      "mean_token_accuracy": 0.23923430740833282,
      "num_tokens": 129770996.0,
      "step": 74820
    },
    {
      "entropy": 5.402015447616577,
      "epoch": 5.821630033067496,
      "grad_norm": 1.25,
      "learning_rate": 0.00021978386579267907,
      "loss": 4.6607,
      "mean_token_accuracy": 0.23781231194734573,
      "num_tokens": 129779379.0,
      "step": 74825
    },
    {
      "entropy": 5.419774150848388,
      "epoch": 5.822019062439214,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021975699814911432,
      "loss": 4.7018,
      "mean_token_accuracy": 0.23461648374795913,
      "num_tokens": 129788320.0,
      "step": 74830
    },
    {
      "entropy": 5.451441144943237,
      "epoch": 5.8224080918109316,
      "grad_norm": 1.140625,
      "learning_rate": 0.00021973013134377422,
      "loss": 4.795,
      "mean_token_accuracy": 0.228517609834671,
      "num_tokens": 129798004.0,
      "step": 74835
    },
    {
      "entropy": 5.38828673362732,
      "epoch": 5.822797121182649,
      "grad_norm": 1.171875,
      "learning_rate": 0.00021970326537706643,
      "loss": 4.6374,
      "mean_token_accuracy": 0.24212099313735963,
      "num_tokens": 129807504.0,
      "step": 74840
    },
    {
      "entropy": 5.47853946685791,
      "epoch": 5.823186150554367,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021967640024939851,
      "loss": 4.7609,
      "mean_token_accuracy": 0.23351117521524428,
      "num_tokens": 129815552.0,
      "step": 74845
    },
    {
      "entropy": 5.359498596191406,
      "epoch": 5.823575179926085,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002196495359611782,
      "loss": 4.6618,
      "mean_token_accuracy": 0.2350616231560707,
      "num_tokens": 129824522.0,
      "step": 74850
    },
    {
      "entropy": 5.327143287658691,
      "epoch": 5.823964209297802,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002196226725128131,
      "loss": 4.6484,
      "mean_token_accuracy": 0.23251783400774,
      "num_tokens": 129833682.0,
      "step": 74855
    },
    {
      "entropy": 5.470075988769532,
      "epoch": 5.82435323866952,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00021959580990471073,
      "loss": 4.8327,
      "mean_token_accuracy": 0.22469857037067414,
      "num_tokens": 129842090.0,
      "step": 74860
    },
    {
      "entropy": 5.541165447235107,
      "epoch": 5.824742268041237,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002195689481372788,
      "loss": 4.8214,
      "mean_token_accuracy": 0.2250499337911606,
      "num_tokens": 129850636.0,
      "step": 74865
    },
    {
      "entropy": 5.415022945404052,
      "epoch": 5.8251312974129545,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021954208721092482,
      "loss": 4.6825,
      "mean_token_accuracy": 0.24056483507156373,
      "num_tokens": 129859501.0,
      "step": 74870
    },
    {
      "entropy": 5.491297960281372,
      "epoch": 5.825520326784672,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021951522712605648,
      "loss": 4.8737,
      "mean_token_accuracy": 0.21909617632627487,
      "num_tokens": 129868064.0,
      "step": 74875
    },
    {
      "entropy": 5.450510263442993,
      "epoch": 5.82590935615639,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021948836788308117,
      "loss": 4.7974,
      "mean_token_accuracy": 0.230095711350441,
      "num_tokens": 129877322.0,
      "step": 74880
    },
    {
      "entropy": 5.440655326843261,
      "epoch": 5.826298385528108,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021946150948240657,
      "loss": 4.7157,
      "mean_token_accuracy": 0.22622565031051636,
      "num_tokens": 129887319.0,
      "step": 74885
    },
    {
      "entropy": 5.413228511810303,
      "epoch": 5.8266874148998244,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021943465192444023,
      "loss": 4.6673,
      "mean_token_accuracy": 0.24003024995326996,
      "num_tokens": 129895726.0,
      "step": 74890
    },
    {
      "entropy": 5.519196891784668,
      "epoch": 5.827076444271542,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021940779520958952,
      "loss": 4.8341,
      "mean_token_accuracy": 0.221943299472332,
      "num_tokens": 129905011.0,
      "step": 74895
    },
    {
      "entropy": 5.285239362716675,
      "epoch": 5.82746547364326,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002193809393382621,
      "loss": 4.6113,
      "mean_token_accuracy": 0.23783642649650574,
      "num_tokens": 129914912.0,
      "step": 74900
    },
    {
      "entropy": 5.401373481750488,
      "epoch": 5.8278545030149775,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021935408431086528,
      "loss": 4.7177,
      "mean_token_accuracy": 0.2349007546901703,
      "num_tokens": 129923602.0,
      "step": 74905
    },
    {
      "entropy": 5.351676750183105,
      "epoch": 5.828243532386695,
      "grad_norm": 1.375,
      "learning_rate": 0.00021932723012780677,
      "loss": 4.6181,
      "mean_token_accuracy": 0.23882330060005189,
      "num_tokens": 129931626.0,
      "step": 74910
    },
    {
      "entropy": 5.4146496772766115,
      "epoch": 5.828632561758413,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021930037678949388,
      "loss": 4.7239,
      "mean_token_accuracy": 0.236628982424736,
      "num_tokens": 129940097.0,
      "step": 74915
    },
    {
      "entropy": 5.350550508499145,
      "epoch": 5.829021591130131,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021927352429633413,
      "loss": 4.719,
      "mean_token_accuracy": 0.23776727318763732,
      "num_tokens": 129948500.0,
      "step": 74920
    },
    {
      "entropy": 5.441208171844482,
      "epoch": 5.829410620501848,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021924667264873492,
      "loss": 4.7547,
      "mean_token_accuracy": 0.2314123332500458,
      "num_tokens": 129957224.0,
      "step": 74925
    },
    {
      "entropy": 5.485840463638306,
      "epoch": 5.829799649873565,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002192198218471037,
      "loss": 4.7617,
      "mean_token_accuracy": 0.2331481471657753,
      "num_tokens": 129965962.0,
      "step": 74930
    },
    {
      "entropy": 5.3584291458129885,
      "epoch": 5.830188679245283,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021919297189184795,
      "loss": 4.6123,
      "mean_token_accuracy": 0.23515749871730804,
      "num_tokens": 129973984.0,
      "step": 74935
    },
    {
      "entropy": 5.466229486465454,
      "epoch": 5.8305777086170005,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000219166122783375,
      "loss": 4.8219,
      "mean_token_accuracy": 0.22972820401191713,
      "num_tokens": 129982509.0,
      "step": 74940
    },
    {
      "entropy": 5.384138059616089,
      "epoch": 5.830966737988718,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021913927452209232,
      "loss": 4.7097,
      "mean_token_accuracy": 0.23329777121543885,
      "num_tokens": 129990682.0,
      "step": 74945
    },
    {
      "entropy": 5.4293050289154055,
      "epoch": 5.831355767360436,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021911242710840717,
      "loss": 4.7058,
      "mean_token_accuracy": 0.2347875416278839,
      "num_tokens": 129999727.0,
      "step": 74950
    },
    {
      "entropy": 5.378530359268188,
      "epoch": 5.831744796732154,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021908558054272703,
      "loss": 4.6285,
      "mean_token_accuracy": 0.24019158780574798,
      "num_tokens": 130008584.0,
      "step": 74955
    },
    {
      "entropy": 5.352772378921509,
      "epoch": 5.83213382610387,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002190587348254592,
      "loss": 4.6592,
      "mean_token_accuracy": 0.2342011421918869,
      "num_tokens": 130017277.0,
      "step": 74960
    },
    {
      "entropy": 5.307932138442993,
      "epoch": 5.832522855475588,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000219031889957011,
      "loss": 4.6304,
      "mean_token_accuracy": 0.2356787458062172,
      "num_tokens": 130025863.0,
      "step": 74965
    },
    {
      "entropy": 5.458473777770996,
      "epoch": 5.832911884847306,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021900504593778985,
      "loss": 4.8697,
      "mean_token_accuracy": 0.21456712037324904,
      "num_tokens": 130034475.0,
      "step": 74970
    },
    {
      "entropy": 5.3968682289123535,
      "epoch": 5.8333009142190235,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021897820276820296,
      "loss": 4.6614,
      "mean_token_accuracy": 0.24137951880693437,
      "num_tokens": 130043332.0,
      "step": 74975
    },
    {
      "entropy": 5.471196794509888,
      "epoch": 5.833689943590741,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002189513604486577,
      "loss": 4.7754,
      "mean_token_accuracy": 0.22156031876802446,
      "num_tokens": 130051988.0,
      "step": 74980
    },
    {
      "entropy": 5.386880874633789,
      "epoch": 5.834078972962459,
      "grad_norm": 1.25,
      "learning_rate": 0.0002189245189795613,
      "loss": 4.6835,
      "mean_token_accuracy": 0.23673636168241502,
      "num_tokens": 130060136.0,
      "step": 74985
    },
    {
      "entropy": 5.378296184539795,
      "epoch": 5.834468002334177,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021889767836132107,
      "loss": 4.6871,
      "mean_token_accuracy": 0.23277265429496766,
      "num_tokens": 130068653.0,
      "step": 74990
    },
    {
      "entropy": 5.427887630462647,
      "epoch": 5.834857031705893,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021887083859434433,
      "loss": 4.7747,
      "mean_token_accuracy": 0.22601681798696518,
      "num_tokens": 130077681.0,
      "step": 74995
    },
    {
      "entropy": 5.420094966888428,
      "epoch": 5.835246061077611,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021884399967903818,
      "loss": 4.6979,
      "mean_token_accuracy": 0.23909737914800644,
      "num_tokens": 130085977.0,
      "step": 75000
    },
    {
      "epoch": 5.835246061077611,
      "eval_entropy": 5.189041754279261,
      "eval_loss": 4.8822550773620605,
      "eval_mean_token_accuracy": 0.23228316240378363,
      "eval_num_tokens": 130085977.0,
      "eval_runtime": 28.6477,
      "eval_samples_per_second": 1450.656,
      "eval_steps_per_second": 181.341,
      "step": 75000
    },
    {
      "entropy": 5.440436220169067,
      "epoch": 5.835635090449329,
      "grad_norm": 1.25,
      "learning_rate": 0.00021881716161581,
      "loss": 4.771,
      "mean_token_accuracy": 0.22901004999876023,
      "num_tokens": 130093954.0,
      "step": 75005
    },
    {
      "entropy": 5.393560552597046,
      "epoch": 5.8360241198210465,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021879032440506697,
      "loss": 4.7215,
      "mean_token_accuracy": 0.22607511132955552,
      "num_tokens": 130102268.0,
      "step": 75010
    },
    {
      "entropy": 5.4361615657806395,
      "epoch": 5.836413149192764,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002187634880472163,
      "loss": 4.6851,
      "mean_token_accuracy": 0.2343365117907524,
      "num_tokens": 130109988.0,
      "step": 75015
    },
    {
      "entropy": 5.353677415847779,
      "epoch": 5.836802178564482,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00021873665254266517,
      "loss": 4.7159,
      "mean_token_accuracy": 0.235113725066185,
      "num_tokens": 130119140.0,
      "step": 75020
    },
    {
      "entropy": 5.389383220672608,
      "epoch": 5.837191207936199,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021870981789182082,
      "loss": 4.6796,
      "mean_token_accuracy": 0.23319031149148942,
      "num_tokens": 130127956.0,
      "step": 75025
    },
    {
      "entropy": 5.417613315582275,
      "epoch": 5.837580237307916,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002186829840950904,
      "loss": 4.7685,
      "mean_token_accuracy": 0.23745323419570924,
      "num_tokens": 130137405.0,
      "step": 75030
    },
    {
      "entropy": 5.492853450775146,
      "epoch": 5.837969266679634,
      "grad_norm": 1.3125,
      "learning_rate": 0.000218656151152881,
      "loss": 4.7212,
      "mean_token_accuracy": 0.23608200252056122,
      "num_tokens": 130145972.0,
      "step": 75035
    },
    {
      "entropy": 5.443893098831177,
      "epoch": 5.838358296051352,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00021862931906559992,
      "loss": 4.7775,
      "mean_token_accuracy": 0.22974815368652343,
      "num_tokens": 130155575.0,
      "step": 75040
    },
    {
      "entropy": 5.403016376495361,
      "epoch": 5.8387473254230695,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00021860248783365412,
      "loss": 4.6462,
      "mean_token_accuracy": 0.23377186805009842,
      "num_tokens": 130164737.0,
      "step": 75045
    },
    {
      "entropy": 5.369895124435425,
      "epoch": 5.839136354794787,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00021857565745745083,
      "loss": 4.7085,
      "mean_token_accuracy": 0.2369548961520195,
      "num_tokens": 130174376.0,
      "step": 75050
    },
    {
      "entropy": 5.432428932189941,
      "epoch": 5.839525384166505,
      "grad_norm": 1.140625,
      "learning_rate": 0.00021854882793739716,
      "loss": 4.7835,
      "mean_token_accuracy": 0.2221780851483345,
      "num_tokens": 130183321.0,
      "step": 75055
    },
    {
      "entropy": 5.312546873092652,
      "epoch": 5.839914413538223,
      "grad_norm": 1.15625,
      "learning_rate": 0.00021852199927390015,
      "loss": 4.589,
      "mean_token_accuracy": 0.23595236241817474,
      "num_tokens": 130191766.0,
      "step": 75060
    },
    {
      "entropy": 5.384715223312378,
      "epoch": 5.840303442909939,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002184951714673668,
      "loss": 4.7476,
      "mean_token_accuracy": 0.23173545002937318,
      "num_tokens": 130201033.0,
      "step": 75065
    },
    {
      "entropy": 5.374744987487793,
      "epoch": 5.840692472281657,
      "grad_norm": 1.171875,
      "learning_rate": 0.00021846834451820437,
      "loss": 4.6638,
      "mean_token_accuracy": 0.23468036353588104,
      "num_tokens": 130210397.0,
      "step": 75070
    },
    {
      "entropy": 5.333154249191284,
      "epoch": 5.841081501653375,
      "grad_norm": 1.25,
      "learning_rate": 0.00021844151842681986,
      "loss": 4.5843,
      "mean_token_accuracy": 0.2460404485464096,
      "num_tokens": 130219297.0,
      "step": 75075
    },
    {
      "entropy": 5.411053466796875,
      "epoch": 5.8414705310250925,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021841469319362034,
      "loss": 4.7437,
      "mean_token_accuracy": 0.23074395209550858,
      "num_tokens": 130227606.0,
      "step": 75080
    },
    {
      "entropy": 5.371033191680908,
      "epoch": 5.84185956039681,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021838786881901267,
      "loss": 4.6881,
      "mean_token_accuracy": 0.24003113955259323,
      "num_tokens": 130236636.0,
      "step": 75085
    },
    {
      "entropy": 5.412061786651611,
      "epoch": 5.842248589768528,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021836104530340407,
      "loss": 4.6553,
      "mean_token_accuracy": 0.2395092263817787,
      "num_tokens": 130244705.0,
      "step": 75090
    },
    {
      "entropy": 5.511723852157592,
      "epoch": 5.842637619140245,
      "grad_norm": 1.21875,
      "learning_rate": 0.00021833422264720136,
      "loss": 4.8694,
      "mean_token_accuracy": 0.21953065097332,
      "num_tokens": 130252914.0,
      "step": 75095
    },
    {
      "entropy": 5.414542961120605,
      "epoch": 5.843026648511962,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021830740085081169,
      "loss": 4.7495,
      "mean_token_accuracy": 0.23152572959661483,
      "num_tokens": 130262533.0,
      "step": 75100
    },
    {
      "entropy": 5.406823253631591,
      "epoch": 5.84341567788368,
      "grad_norm": 1.34375,
      "learning_rate": 0.000218280579914642,
      "loss": 4.6301,
      "mean_token_accuracy": 0.23029024600982667,
      "num_tokens": 130270104.0,
      "step": 75105
    },
    {
      "entropy": 5.3930768966674805,
      "epoch": 5.843804707255398,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021825375983909918,
      "loss": 4.706,
      "mean_token_accuracy": 0.2328567162156105,
      "num_tokens": 130279101.0,
      "step": 75110
    },
    {
      "entropy": 5.423156023025513,
      "epoch": 5.8441937366271155,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002182269406245903,
      "loss": 4.7127,
      "mean_token_accuracy": 0.2351074695587158,
      "num_tokens": 130288343.0,
      "step": 75115
    },
    {
      "entropy": 5.412784624099731,
      "epoch": 5.844582765998833,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021820012227152214,
      "loss": 4.6644,
      "mean_token_accuracy": 0.24363360702991485,
      "num_tokens": 130297755.0,
      "step": 75120
    },
    {
      "entropy": 5.379198694229126,
      "epoch": 5.844971795370551,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021817330478030177,
      "loss": 4.6914,
      "mean_token_accuracy": 0.23391299843788146,
      "num_tokens": 130306517.0,
      "step": 75125
    },
    {
      "entropy": 5.3564064502716064,
      "epoch": 5.845360824742268,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00021814648815133604,
      "loss": 4.723,
      "mean_token_accuracy": 0.23891696184873581,
      "num_tokens": 130315633.0,
      "step": 75130
    },
    {
      "entropy": 5.373842477798462,
      "epoch": 5.845749854113985,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021811967238503184,
      "loss": 4.6155,
      "mean_token_accuracy": 0.23208567202091218,
      "num_tokens": 130324241.0,
      "step": 75135
    },
    {
      "entropy": 5.449653196334839,
      "epoch": 5.846138883485703,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021809285748179612,
      "loss": 4.7778,
      "mean_token_accuracy": 0.22367812991142272,
      "num_tokens": 130332569.0,
      "step": 75140
    },
    {
      "entropy": 5.420026731491089,
      "epoch": 5.846527912857421,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002180660434420357,
      "loss": 4.6786,
      "mean_token_accuracy": 0.23103420287370682,
      "num_tokens": 130341667.0,
      "step": 75145
    },
    {
      "entropy": 5.472999000549317,
      "epoch": 5.8469169422291385,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021803923026615746,
      "loss": 4.7945,
      "mean_token_accuracy": 0.22467969506978988,
      "num_tokens": 130350659.0,
      "step": 75150
    },
    {
      "entropy": 5.435426235198975,
      "epoch": 5.847305971600856,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002180124179545682,
      "loss": 4.689,
      "mean_token_accuracy": 0.23886821717023848,
      "num_tokens": 130359010.0,
      "step": 75155
    },
    {
      "entropy": 5.384161520004272,
      "epoch": 5.847695000972573,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021798560650767486,
      "loss": 4.7649,
      "mean_token_accuracy": 0.23188586831092833,
      "num_tokens": 130368649.0,
      "step": 75160
    },
    {
      "entropy": 5.328509712219239,
      "epoch": 5.848084030344291,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00021795879592588418,
      "loss": 4.6639,
      "mean_token_accuracy": 0.23339553624391557,
      "num_tokens": 130377123.0,
      "step": 75165
    },
    {
      "entropy": 5.38571720123291,
      "epoch": 5.848473059716008,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021793198620960293,
      "loss": 4.6094,
      "mean_token_accuracy": 0.24373237937688827,
      "num_tokens": 130385408.0,
      "step": 75170
    },
    {
      "entropy": 5.493203735351562,
      "epoch": 5.848862089087726,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021790517735923798,
      "loss": 4.8478,
      "mean_token_accuracy": 0.21712367981672287,
      "num_tokens": 130394359.0,
      "step": 75175
    },
    {
      "entropy": 5.362399530410767,
      "epoch": 5.849251118459444,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002178783693751961,
      "loss": 4.5868,
      "mean_token_accuracy": 0.24853430092334747,
      "num_tokens": 130403155.0,
      "step": 75180
    },
    {
      "entropy": 5.407753562927246,
      "epoch": 5.8496401478311615,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000217851562257884,
      "loss": 4.7515,
      "mean_token_accuracy": 0.23204280585050582,
      "num_tokens": 130411281.0,
      "step": 75185
    },
    {
      "entropy": 5.430237531661987,
      "epoch": 5.850029177202879,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021782475600770856,
      "loss": 4.7906,
      "mean_token_accuracy": 0.22574642300605774,
      "num_tokens": 130419310.0,
      "step": 75190
    },
    {
      "entropy": 5.459553909301758,
      "epoch": 5.850418206574597,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021779795062507634,
      "loss": 4.6999,
      "mean_token_accuracy": 0.23933560252189637,
      "num_tokens": 130427668.0,
      "step": 75195
    },
    {
      "entropy": 5.400187587738037,
      "epoch": 5.850807235946314,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00021777114611039422,
      "loss": 4.7116,
      "mean_token_accuracy": 0.23006826341152192,
      "num_tokens": 130436727.0,
      "step": 75200
    },
    {
      "entropy": 5.433020114898682,
      "epoch": 5.851196265318031,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021774434246406887,
      "loss": 4.7671,
      "mean_token_accuracy": 0.23111289441585542,
      "num_tokens": 130445347.0,
      "step": 75205
    },
    {
      "entropy": 5.4675311088562015,
      "epoch": 5.851585294689749,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021771753968650705,
      "loss": 4.7775,
      "mean_token_accuracy": 0.23570704609155654,
      "num_tokens": 130454410.0,
      "step": 75210
    },
    {
      "entropy": 5.4735596656799315,
      "epoch": 5.851974324061467,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021769073777811533,
      "loss": 4.7999,
      "mean_token_accuracy": 0.22724099457263947,
      "num_tokens": 130463289.0,
      "step": 75215
    },
    {
      "entropy": 5.377410745620727,
      "epoch": 5.8523633534331845,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021766393673930046,
      "loss": 4.6233,
      "mean_token_accuracy": 0.2403624624013901,
      "num_tokens": 130471645.0,
      "step": 75220
    },
    {
      "entropy": 5.397852087020874,
      "epoch": 5.852752382804901,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021763713657046906,
      "loss": 4.6725,
      "mean_token_accuracy": 0.23484694063663483,
      "num_tokens": 130480246.0,
      "step": 75225
    },
    {
      "entropy": 5.341460943222046,
      "epoch": 5.853141412176619,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021761033727202782,
      "loss": 4.7109,
      "mean_token_accuracy": 0.23132773190736772,
      "num_tokens": 130488960.0,
      "step": 75230
    },
    {
      "entropy": 5.26563081741333,
      "epoch": 5.853530441548337,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00021758353884438343,
      "loss": 4.5422,
      "mean_token_accuracy": 0.24834733009338378,
      "num_tokens": 130497776.0,
      "step": 75235
    },
    {
      "entropy": 5.3771710872650145,
      "epoch": 5.853919470920054,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002175567412879424,
      "loss": 4.6718,
      "mean_token_accuracy": 0.2354313999414444,
      "num_tokens": 130507108.0,
      "step": 75240
    },
    {
      "entropy": 5.452317571640014,
      "epoch": 5.854308500291772,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00021752994460311144,
      "loss": 4.7389,
      "mean_token_accuracy": 0.23730547726154327,
      "num_tokens": 130515675.0,
      "step": 75245
    },
    {
      "entropy": 5.482429695129395,
      "epoch": 5.85469752966349,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021750314879029708,
      "loss": 4.794,
      "mean_token_accuracy": 0.23259347826242446,
      "num_tokens": 130524528.0,
      "step": 75250
    },
    {
      "entropy": 5.456248760223389,
      "epoch": 5.8550865590352075,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002174763538499059,
      "loss": 4.7622,
      "mean_token_accuracy": 0.22153889685869216,
      "num_tokens": 130533126.0,
      "step": 75255
    },
    {
      "entropy": 5.409974908828735,
      "epoch": 5.855475588406925,
      "grad_norm": 1.25,
      "learning_rate": 0.00021744955978234449,
      "loss": 4.7303,
      "mean_token_accuracy": 0.2326360210776329,
      "num_tokens": 130541820.0,
      "step": 75260
    },
    {
      "entropy": 5.308184719085693,
      "epoch": 5.855864617778642,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002174227665880193,
      "loss": 4.4939,
      "mean_token_accuracy": 0.24678035080432892,
      "num_tokens": 130550438.0,
      "step": 75265
    },
    {
      "entropy": 5.401589870452881,
      "epoch": 5.85625364715036,
      "grad_norm": 1.21875,
      "learning_rate": 0.00021739597426733714,
      "loss": 4.7928,
      "mean_token_accuracy": 0.23371152728796005,
      "num_tokens": 130559694.0,
      "step": 75270
    },
    {
      "entropy": 5.373457527160644,
      "epoch": 5.856642676522077,
      "grad_norm": 1.21875,
      "learning_rate": 0.00021736918282070435,
      "loss": 4.7087,
      "mean_token_accuracy": 0.23313440829515458,
      "num_tokens": 130568442.0,
      "step": 75275
    },
    {
      "entropy": 5.487116050720215,
      "epoch": 5.857031705893795,
      "grad_norm": 1.1875,
      "learning_rate": 0.00021734239224852742,
      "loss": 4.6914,
      "mean_token_accuracy": 0.22669809460639953,
      "num_tokens": 130577934.0,
      "step": 75280
    },
    {
      "entropy": 5.378036594390869,
      "epoch": 5.857420735265513,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021731560255121296,
      "loss": 4.597,
      "mean_token_accuracy": 0.245254448056221,
      "num_tokens": 130586083.0,
      "step": 75285
    },
    {
      "entropy": 5.425135135650635,
      "epoch": 5.8578097646372305,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021728881372916743,
      "loss": 4.6723,
      "mean_token_accuracy": 0.2345619887113571,
      "num_tokens": 130594645.0,
      "step": 75290
    },
    {
      "entropy": 5.345851230621338,
      "epoch": 5.858198794008947,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002172620257827973,
      "loss": 4.7421,
      "mean_token_accuracy": 0.23869220167398453,
      "num_tokens": 130602626.0,
      "step": 75295
    },
    {
      "entropy": 5.333785581588745,
      "epoch": 5.858587823380665,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021723523871250905,
      "loss": 4.723,
      "mean_token_accuracy": 0.2388553574681282,
      "num_tokens": 130611204.0,
      "step": 75300
    },
    {
      "entropy": 5.457787704467774,
      "epoch": 5.858976852752383,
      "grad_norm": 1.21875,
      "learning_rate": 0.00021720845251870907,
      "loss": 4.8097,
      "mean_token_accuracy": 0.22408759742975234,
      "num_tokens": 130619989.0,
      "step": 75305
    },
    {
      "entropy": 5.400807809829712,
      "epoch": 5.8593658821241,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021718166720180384,
      "loss": 4.659,
      "mean_token_accuracy": 0.24193836003541946,
      "num_tokens": 130628288.0,
      "step": 75310
    },
    {
      "entropy": 5.413465309143066,
      "epoch": 5.859754911495818,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002171548827621998,
      "loss": 4.7394,
      "mean_token_accuracy": 0.23456821739673614,
      "num_tokens": 130636682.0,
      "step": 75315
    },
    {
      "entropy": 5.376766300201416,
      "epoch": 5.860143940867536,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002171280992003033,
      "loss": 4.7574,
      "mean_token_accuracy": 0.231772518157959,
      "num_tokens": 130645702.0,
      "step": 75320
    },
    {
      "entropy": 5.452128648757935,
      "epoch": 5.8605329702392535,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021710131651652087,
      "loss": 4.7965,
      "mean_token_accuracy": 0.2315473288297653,
      "num_tokens": 130655254.0,
      "step": 75325
    },
    {
      "entropy": 5.452722692489624,
      "epoch": 5.86092199961097,
      "grad_norm": 1.25,
      "learning_rate": 0.00021707453471125866,
      "loss": 4.6241,
      "mean_token_accuracy": 0.23871950209140777,
      "num_tokens": 130663411.0,
      "step": 75330
    },
    {
      "entropy": 5.451143074035644,
      "epoch": 5.861311028982688,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021704775378492326,
      "loss": 4.7535,
      "mean_token_accuracy": 0.23164176642894746,
      "num_tokens": 130671731.0,
      "step": 75335
    },
    {
      "entropy": 5.384535551071167,
      "epoch": 5.861700058354406,
      "grad_norm": 1.171875,
      "learning_rate": 0.00021702097373792095,
      "loss": 4.7315,
      "mean_token_accuracy": 0.2344660699367523,
      "num_tokens": 130681095.0,
      "step": 75340
    },
    {
      "entropy": 5.452794361114502,
      "epoch": 5.862089087726123,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021699419457065812,
      "loss": 4.7586,
      "mean_token_accuracy": 0.23082228600978852,
      "num_tokens": 130690005.0,
      "step": 75345
    },
    {
      "entropy": 5.37485785484314,
      "epoch": 5.862478117097841,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021696741628354104,
      "loss": 4.5965,
      "mean_token_accuracy": 0.24523642808198928,
      "num_tokens": 130697788.0,
      "step": 75350
    },
    {
      "entropy": 5.284783029556275,
      "epoch": 5.862867146469559,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021694063887697614,
      "loss": 4.5576,
      "mean_token_accuracy": 0.24906431436538695,
      "num_tokens": 130706968.0,
      "step": 75355
    },
    {
      "entropy": 5.429599571228027,
      "epoch": 5.863256175841276,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00021691386235136951,
      "loss": 4.7126,
      "mean_token_accuracy": 0.23352719098329544,
      "num_tokens": 130715428.0,
      "step": 75360
    },
    {
      "entropy": 5.307359218597412,
      "epoch": 5.863645205212993,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021688708670712764,
      "loss": 4.652,
      "mean_token_accuracy": 0.23202627450227736,
      "num_tokens": 130724101.0,
      "step": 75365
    },
    {
      "entropy": 5.358681583404541,
      "epoch": 5.864034234584711,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00021686031194465666,
      "loss": 4.7232,
      "mean_token_accuracy": 0.22714757174253464,
      "num_tokens": 130732549.0,
      "step": 75370
    },
    {
      "entropy": 5.357511234283447,
      "epoch": 5.864423263956429,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021683353806436296,
      "loss": 4.644,
      "mean_token_accuracy": 0.23871582746505737,
      "num_tokens": 130741011.0,
      "step": 75375
    },
    {
      "entropy": 5.423283815383911,
      "epoch": 5.864812293328146,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002168067650666527,
      "loss": 4.6659,
      "mean_token_accuracy": 0.23612017929553986,
      "num_tokens": 130749722.0,
      "step": 75380
    },
    {
      "entropy": 5.432904005050659,
      "epoch": 5.865201322699864,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021677999295193224,
      "loss": 4.7371,
      "mean_token_accuracy": 0.2311698317527771,
      "num_tokens": 130757984.0,
      "step": 75385
    },
    {
      "entropy": 5.413618659973144,
      "epoch": 5.865590352071582,
      "grad_norm": 1.1875,
      "learning_rate": 0.00021675322172060764,
      "loss": 4.7181,
      "mean_token_accuracy": 0.2324267655611038,
      "num_tokens": 130766341.0,
      "step": 75390
    },
    {
      "entropy": 5.444217109680176,
      "epoch": 5.8659793814432994,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021672645137308516,
      "loss": 4.6791,
      "mean_token_accuracy": 0.23747870028018953,
      "num_tokens": 130774711.0,
      "step": 75395
    },
    {
      "entropy": 5.3784956455230715,
      "epoch": 5.866368410815016,
      "grad_norm": 1.25,
      "learning_rate": 0.00021669968190977107,
      "loss": 4.7325,
      "mean_token_accuracy": 0.23173762261867523,
      "num_tokens": 130784297.0,
      "step": 75400
    },
    {
      "entropy": 5.309084081649781,
      "epoch": 5.866757440186734,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021667291333107153,
      "loss": 4.5827,
      "mean_token_accuracy": 0.24605681896209716,
      "num_tokens": 130792562.0,
      "step": 75405
    },
    {
      "entropy": 5.36782054901123,
      "epoch": 5.867146469558452,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00021664614563739266,
      "loss": 4.6837,
      "mean_token_accuracy": 0.23487478792667388,
      "num_tokens": 130801089.0,
      "step": 75410
    },
    {
      "entropy": 5.483813810348511,
      "epoch": 5.867535498930169,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021661937882914067,
      "loss": 4.8506,
      "mean_token_accuracy": 0.22688038945198058,
      "num_tokens": 130810047.0,
      "step": 75415
    },
    {
      "entropy": 5.439763689041138,
      "epoch": 5.867924528301887,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021659261290672168,
      "loss": 4.8066,
      "mean_token_accuracy": 0.22324874848127366,
      "num_tokens": 130819997.0,
      "step": 75420
    },
    {
      "entropy": 5.428563117980957,
      "epoch": 5.868313557673605,
      "grad_norm": 1.21875,
      "learning_rate": 0.00021656584787054184,
      "loss": 4.6611,
      "mean_token_accuracy": 0.24274247735738755,
      "num_tokens": 130828513.0,
      "step": 75425
    },
    {
      "entropy": 5.357526779174805,
      "epoch": 5.8687025870453216,
      "grad_norm": 1.375,
      "learning_rate": 0.00021653908372100728,
      "loss": 4.6485,
      "mean_token_accuracy": 0.23672694265842437,
      "num_tokens": 130837371.0,
      "step": 75430
    },
    {
      "entropy": 5.411957263946533,
      "epoch": 5.869091616417039,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00021651232045852403,
      "loss": 4.7296,
      "mean_token_accuracy": 0.2305314436554909,
      "num_tokens": 130846270.0,
      "step": 75435
    },
    {
      "entropy": 5.422478199005127,
      "epoch": 5.869480645788757,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002164855580834983,
      "loss": 4.7404,
      "mean_token_accuracy": 0.22998669147491455,
      "num_tokens": 130855228.0,
      "step": 75440
    },
    {
      "entropy": 5.3998371124267575,
      "epoch": 5.869869675160475,
      "grad_norm": 1.234375,
      "learning_rate": 0.000216458796596336,
      "loss": 4.6729,
      "mean_token_accuracy": 0.23512174040079117,
      "num_tokens": 130863795.0,
      "step": 75445
    },
    {
      "entropy": 5.35286774635315,
      "epoch": 5.870258704532192,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00021643203599744327,
      "loss": 4.6411,
      "mean_token_accuracy": 0.23935913890600205,
      "num_tokens": 130871561.0,
      "step": 75450
    },
    {
      "entropy": 5.440907764434814,
      "epoch": 5.87064773390391,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021640527628722622,
      "loss": 4.7872,
      "mean_token_accuracy": 0.23571144491434098,
      "num_tokens": 130881142.0,
      "step": 75455
    },
    {
      "entropy": 5.3262245655059814,
      "epoch": 5.871036763275628,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021637851746609082,
      "loss": 4.5578,
      "mean_token_accuracy": 0.2419270396232605,
      "num_tokens": 130889959.0,
      "step": 75460
    },
    {
      "entropy": 5.415669107437134,
      "epoch": 5.8714257926473445,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021635175953444314,
      "loss": 4.7009,
      "mean_token_accuracy": 0.23700659424066545,
      "num_tokens": 130899404.0,
      "step": 75465
    },
    {
      "entropy": 5.429981231689453,
      "epoch": 5.871814822019062,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002163250024926891,
      "loss": 4.7639,
      "mean_token_accuracy": 0.22510949671268463,
      "num_tokens": 130908428.0,
      "step": 75470
    },
    {
      "entropy": 5.352370071411133,
      "epoch": 5.87220385139078,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021629824634123479,
      "loss": 4.6429,
      "mean_token_accuracy": 0.23536140620708465,
      "num_tokens": 130917100.0,
      "step": 75475
    },
    {
      "entropy": 5.353133583068848,
      "epoch": 5.872592880762498,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021627149108048615,
      "loss": 4.6668,
      "mean_token_accuracy": 0.24075983017683028,
      "num_tokens": 130926600.0,
      "step": 75480
    },
    {
      "entropy": 5.375202560424805,
      "epoch": 5.872981910134215,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002162447367108492,
      "loss": 4.7541,
      "mean_token_accuracy": 0.230471433699131,
      "num_tokens": 130935726.0,
      "step": 75485
    },
    {
      "entropy": 5.522343158721924,
      "epoch": 5.873370939505933,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021621798323272985,
      "loss": 4.7978,
      "mean_token_accuracy": 0.2298094928264618,
      "num_tokens": 130944543.0,
      "step": 75490
    },
    {
      "entropy": 5.407302570343018,
      "epoch": 5.87375996887765,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000216191230646534,
      "loss": 4.6611,
      "mean_token_accuracy": 0.24088895469903945,
      "num_tokens": 130952872.0,
      "step": 75495
    },
    {
      "entropy": 5.413905906677246,
      "epoch": 5.8741489982493675,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021616447895266762,
      "loss": 4.7554,
      "mean_token_accuracy": 0.23421289771795273,
      "num_tokens": 130961552.0,
      "step": 75500
    },
    {
      "entropy": 5.384846210479736,
      "epoch": 5.874538027621085,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021613772815153666,
      "loss": 4.6331,
      "mean_token_accuracy": 0.23659877181053163,
      "num_tokens": 130969985.0,
      "step": 75505
    },
    {
      "entropy": 5.407803153991699,
      "epoch": 5.874927056992803,
      "grad_norm": 1.171875,
      "learning_rate": 0.00021611097824354697,
      "loss": 4.6736,
      "mean_token_accuracy": 0.2402425229549408,
      "num_tokens": 130978631.0,
      "step": 75510
    },
    {
      "entropy": 5.4554932594299315,
      "epoch": 5.875316086364521,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002160842292291045,
      "loss": 4.7071,
      "mean_token_accuracy": 0.22992434054613115,
      "num_tokens": 130987049.0,
      "step": 75515
    },
    {
      "entropy": 5.411276912689209,
      "epoch": 5.875705115736238,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021605748110861507,
      "loss": 4.6565,
      "mean_token_accuracy": 0.23444025665521623,
      "num_tokens": 130995440.0,
      "step": 75520
    },
    {
      "entropy": 5.440057277679443,
      "epoch": 5.876094145107956,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021603073388248452,
      "loss": 4.7202,
      "mean_token_accuracy": 0.23721881508827208,
      "num_tokens": 131003812.0,
      "step": 75525
    },
    {
      "entropy": 5.342504501342773,
      "epoch": 5.876483174479673,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002160039875511186,
      "loss": 4.6789,
      "mean_token_accuracy": 0.23969455659389496,
      "num_tokens": 131012719.0,
      "step": 75530
    },
    {
      "entropy": 5.336475276947022,
      "epoch": 5.8768722038513905,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021597724211492336,
      "loss": 4.6158,
      "mean_token_accuracy": 0.24498770385980606,
      "num_tokens": 131020982.0,
      "step": 75535
    },
    {
      "entropy": 5.497298336029052,
      "epoch": 5.877261233223108,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021595049757430458,
      "loss": 4.7876,
      "mean_token_accuracy": 0.22875746190547944,
      "num_tokens": 131029772.0,
      "step": 75540
    },
    {
      "entropy": 5.442143058776855,
      "epoch": 5.877650262594826,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021592375392966802,
      "loss": 4.7701,
      "mean_token_accuracy": 0.23075332641601562,
      "num_tokens": 131038050.0,
      "step": 75545
    },
    {
      "entropy": 5.306444835662842,
      "epoch": 5.878039291966544,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021589701118141942,
      "loss": 4.5394,
      "mean_token_accuracy": 0.2478276401758194,
      "num_tokens": 131046969.0,
      "step": 75550
    },
    {
      "entropy": 5.443571424484253,
      "epoch": 5.878428321338261,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021587026932996456,
      "loss": 4.7914,
      "mean_token_accuracy": 0.226687516272068,
      "num_tokens": 131055670.0,
      "step": 75555
    },
    {
      "entropy": 5.327049922943115,
      "epoch": 5.878817350709978,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021584352837570936,
      "loss": 4.6028,
      "mean_token_accuracy": 0.24279688894748688,
      "num_tokens": 131064596.0,
      "step": 75560
    },
    {
      "entropy": 5.414186000823975,
      "epoch": 5.879206380081696,
      "grad_norm": 1.21875,
      "learning_rate": 0.00021581678831905939,
      "loss": 4.6956,
      "mean_token_accuracy": 0.2394741579890251,
      "num_tokens": 131073853.0,
      "step": 75565
    },
    {
      "entropy": 5.404716205596924,
      "epoch": 5.8795954094534135,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021579004916042044,
      "loss": 4.7353,
      "mean_token_accuracy": 0.2250706285238266,
      "num_tokens": 131082562.0,
      "step": 75570
    },
    {
      "entropy": 5.463085651397705,
      "epoch": 5.879984438825131,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021576331090019825,
      "loss": 4.8081,
      "mean_token_accuracy": 0.2226671114563942,
      "num_tokens": 131091649.0,
      "step": 75575
    },
    {
      "entropy": 5.437637567520142,
      "epoch": 5.880373468196849,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021573657353879855,
      "loss": 4.793,
      "mean_token_accuracy": 0.23455215692520143,
      "num_tokens": 131100095.0,
      "step": 75580
    },
    {
      "entropy": 5.449171304702759,
      "epoch": 5.880762497568567,
      "grad_norm": 1.21875,
      "learning_rate": 0.00021570983707662705,
      "loss": 4.7126,
      "mean_token_accuracy": 0.23134711980819703,
      "num_tokens": 131108882.0,
      "step": 75585
    },
    {
      "entropy": 5.364608716964722,
      "epoch": 5.881151526940284,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021568310151408939,
      "loss": 4.606,
      "mean_token_accuracy": 0.23858586549758912,
      "num_tokens": 131117181.0,
      "step": 75590
    },
    {
      "entropy": 5.418085050582886,
      "epoch": 5.881540556312002,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021565636685159117,
      "loss": 4.704,
      "mean_token_accuracy": 0.23170534819364547,
      "num_tokens": 131125524.0,
      "step": 75595
    },
    {
      "entropy": 5.456203269958496,
      "epoch": 5.881929585683719,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002156296330895382,
      "loss": 4.8395,
      "mean_token_accuracy": 0.21729145348072051,
      "num_tokens": 131134158.0,
      "step": 75600
    },
    {
      "entropy": 5.321143674850464,
      "epoch": 5.8823186150554365,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00021560290022833606,
      "loss": 4.5967,
      "mean_token_accuracy": 0.2384444311261177,
      "num_tokens": 131142778.0,
      "step": 75605
    },
    {
      "entropy": 5.393635702133179,
      "epoch": 5.882707644427154,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002155761682683904,
      "loss": 4.7073,
      "mean_token_accuracy": 0.23169841468334199,
      "num_tokens": 131151534.0,
      "step": 75610
    },
    {
      "entropy": 5.42937445640564,
      "epoch": 5.883096673798872,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021554943721010673,
      "loss": 4.7576,
      "mean_token_accuracy": 0.22836636751890182,
      "num_tokens": 131159487.0,
      "step": 75615
    },
    {
      "entropy": 5.475813913345337,
      "epoch": 5.88348570317059,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002155227070538907,
      "loss": 4.815,
      "mean_token_accuracy": 0.22410892993211745,
      "num_tokens": 131167854.0,
      "step": 75620
    },
    {
      "entropy": 5.357760000228882,
      "epoch": 5.883874732542307,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021549597780014807,
      "loss": 4.6124,
      "mean_token_accuracy": 0.23711044937372208,
      "num_tokens": 131176359.0,
      "step": 75625
    },
    {
      "entropy": 5.423770427703857,
      "epoch": 5.884263761914024,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002154692494492842,
      "loss": 4.736,
      "mean_token_accuracy": 0.22697033435106279,
      "num_tokens": 131185277.0,
      "step": 75630
    },
    {
      "entropy": 5.47053165435791,
      "epoch": 5.884652791285742,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021544252200170465,
      "loss": 4.7376,
      "mean_token_accuracy": 0.22495833784341812,
      "num_tokens": 131193397.0,
      "step": 75635
    },
    {
      "entropy": 5.478902387619018,
      "epoch": 5.8850418206574595,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002154157954578151,
      "loss": 4.644,
      "mean_token_accuracy": 0.23610741794109344,
      "num_tokens": 131201902.0,
      "step": 75640
    },
    {
      "entropy": 5.353312587738037,
      "epoch": 5.885430850029177,
      "grad_norm": 1.21875,
      "learning_rate": 0.000215389069818021,
      "loss": 4.6493,
      "mean_token_accuracy": 0.23242338448762895,
      "num_tokens": 131210794.0,
      "step": 75645
    },
    {
      "entropy": 5.380631160736084,
      "epoch": 5.885819879400895,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00021536234508272783,
      "loss": 4.654,
      "mean_token_accuracy": 0.24049709588289261,
      "num_tokens": 131219607.0,
      "step": 75650
    },
    {
      "entropy": 5.42310528755188,
      "epoch": 5.886208908772613,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00021533562125234125,
      "loss": 4.7497,
      "mean_token_accuracy": 0.2324309080839157,
      "num_tokens": 131228480.0,
      "step": 75655
    },
    {
      "entropy": 5.320685529708863,
      "epoch": 5.88659793814433,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00021530889832726652,
      "loss": 4.65,
      "mean_token_accuracy": 0.23750785291194915,
      "num_tokens": 131237727.0,
      "step": 75660
    },
    {
      "entropy": 5.377548933029175,
      "epoch": 5.886986967516047,
      "grad_norm": 1.171875,
      "learning_rate": 0.00021528217630790937,
      "loss": 4.6256,
      "mean_token_accuracy": 0.24403776824474335,
      "num_tokens": 131246490.0,
      "step": 75665
    },
    {
      "entropy": 5.393870162963867,
      "epoch": 5.887375996887765,
      "grad_norm": 1.21875,
      "learning_rate": 0.00021525545519467506,
      "loss": 4.6625,
      "mean_token_accuracy": 0.23279063105583192,
      "num_tokens": 131255443.0,
      "step": 75670
    },
    {
      "entropy": 5.447005653381348,
      "epoch": 5.8877650262594825,
      "grad_norm": 1.21875,
      "learning_rate": 0.00021522873498796912,
      "loss": 4.7181,
      "mean_token_accuracy": 0.23437806069850922,
      "num_tokens": 131264153.0,
      "step": 75675
    },
    {
      "entropy": 5.416633129119873,
      "epoch": 5.8881540556312,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00021520201568819708,
      "loss": 4.6961,
      "mean_token_accuracy": 0.23607893884181977,
      "num_tokens": 131272732.0,
      "step": 75680
    },
    {
      "entropy": 5.353087949752807,
      "epoch": 5.888543085002918,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021517529729576425,
      "loss": 4.6613,
      "mean_token_accuracy": 0.2362336665391922,
      "num_tokens": 131281305.0,
      "step": 75685
    },
    {
      "entropy": 5.4908417701721195,
      "epoch": 5.888932114374636,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021514857981107604,
      "loss": 4.8235,
      "mean_token_accuracy": 0.21835297495126724,
      "num_tokens": 131290091.0,
      "step": 75690
    },
    {
      "entropy": 5.4522357940673825,
      "epoch": 5.889321143746352,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021512186323453786,
      "loss": 4.7301,
      "mean_token_accuracy": 0.23350744545459748,
      "num_tokens": 131298086.0,
      "step": 75695
    },
    {
      "entropy": 5.389719724655151,
      "epoch": 5.88971017311807,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021509514756655514,
      "loss": 4.5941,
      "mean_token_accuracy": 0.24608222097158433,
      "num_tokens": 131306209.0,
      "step": 75700
    },
    {
      "entropy": 5.447935152053833,
      "epoch": 5.890099202489788,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021506843280753318,
      "loss": 4.8163,
      "mean_token_accuracy": 0.22604831159114838,
      "num_tokens": 131315059.0,
      "step": 75705
    },
    {
      "entropy": 5.380401134490967,
      "epoch": 5.8904882318615055,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021504171895787738,
      "loss": 4.6649,
      "mean_token_accuracy": 0.2372438684105873,
      "num_tokens": 131323086.0,
      "step": 75710
    },
    {
      "entropy": 5.403443813323975,
      "epoch": 5.890877261233223,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000215015006017993,
      "loss": 4.6912,
      "mean_token_accuracy": 0.23732939809560777,
      "num_tokens": 131331883.0,
      "step": 75715
    },
    {
      "entropy": 5.42533016204834,
      "epoch": 5.891266290604941,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002149882939882855,
      "loss": 4.75,
      "mean_token_accuracy": 0.2265559196472168,
      "num_tokens": 131340404.0,
      "step": 75720
    },
    {
      "entropy": 5.383144283294678,
      "epoch": 5.891655319976659,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021496158286916011,
      "loss": 4.6753,
      "mean_token_accuracy": 0.236497662961483,
      "num_tokens": 131348974.0,
      "step": 75725
    },
    {
      "entropy": 5.360157775878906,
      "epoch": 5.892044349348376,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00021493487266102202,
      "loss": 4.7116,
      "mean_token_accuracy": 0.23610188961029052,
      "num_tokens": 131357671.0,
      "step": 75730
    },
    {
      "entropy": 5.359032440185547,
      "epoch": 5.892433378720093,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00021490816336427676,
      "loss": 4.6784,
      "mean_token_accuracy": 0.23339841961860658,
      "num_tokens": 131366997.0,
      "step": 75735
    },
    {
      "entropy": 5.400457239151001,
      "epoch": 5.892822408091811,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021488145497932949,
      "loss": 4.7014,
      "mean_token_accuracy": 0.23129364401102065,
      "num_tokens": 131375371.0,
      "step": 75740
    },
    {
      "entropy": 5.5018024921417235,
      "epoch": 5.8932114374635285,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021485474750658536,
      "loss": 4.8227,
      "mean_token_accuracy": 0.2217098668217659,
      "num_tokens": 131384599.0,
      "step": 75745
    },
    {
      "entropy": 5.41101450920105,
      "epoch": 5.893600466835246,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021482804094644976,
      "loss": 4.669,
      "mean_token_accuracy": 0.2319781646132469,
      "num_tokens": 131393699.0,
      "step": 75750
    },
    {
      "entropy": 5.460923099517823,
      "epoch": 5.893989496206964,
      "grad_norm": 1.25,
      "learning_rate": 0.00021480133529932789,
      "loss": 4.7437,
      "mean_token_accuracy": 0.23407427668571473,
      "num_tokens": 131401988.0,
      "step": 75755
    },
    {
      "entropy": 5.3761261940002445,
      "epoch": 5.894378525578681,
      "grad_norm": 1.25,
      "learning_rate": 0.0002147746305656249,
      "loss": 4.6771,
      "mean_token_accuracy": 0.2330164596438408,
      "num_tokens": 131410717.0,
      "step": 75760
    },
    {
      "entropy": 5.311336946487427,
      "epoch": 5.894767554950398,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000214747926745746,
      "loss": 4.5466,
      "mean_token_accuracy": 0.25152658224105834,
      "num_tokens": 131419622.0,
      "step": 75765
    },
    {
      "entropy": 5.404670524597168,
      "epoch": 5.895156584322116,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021472122384009646,
      "loss": 4.7314,
      "mean_token_accuracy": 0.22854116559028625,
      "num_tokens": 131428273.0,
      "step": 75770
    },
    {
      "entropy": 5.457834005355835,
      "epoch": 5.895545613693834,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002146945218490814,
      "loss": 4.7753,
      "mean_token_accuracy": 0.22752442806959153,
      "num_tokens": 131437025.0,
      "step": 75775
    },
    {
      "entropy": 5.358960866928101,
      "epoch": 5.8959346430655515,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021466782077310598,
      "loss": 4.6605,
      "mean_token_accuracy": 0.24315745383501053,
      "num_tokens": 131445363.0,
      "step": 75780
    },
    {
      "entropy": 5.447675132751465,
      "epoch": 5.896323672437269,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002146411206125754,
      "loss": 4.7423,
      "mean_token_accuracy": 0.22962729930877684,
      "num_tokens": 131454823.0,
      "step": 75785
    },
    {
      "entropy": 5.487676382064819,
      "epoch": 5.896712701808987,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021461442136789472,
      "loss": 4.773,
      "mean_token_accuracy": 0.23070111572742463,
      "num_tokens": 131462836.0,
      "step": 75790
    },
    {
      "entropy": 5.449763059616089,
      "epoch": 5.897101731180705,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021458772303946895,
      "loss": 4.7819,
      "mean_token_accuracy": 0.22161486446857454,
      "num_tokens": 131471755.0,
      "step": 75795
    },
    {
      "entropy": 5.459450435638428,
      "epoch": 5.897490760552421,
      "grad_norm": 1.21875,
      "learning_rate": 0.00021456102562770346,
      "loss": 4.6997,
      "mean_token_accuracy": 0.23135231733322142,
      "num_tokens": 131480902.0,
      "step": 75800
    },
    {
      "entropy": 5.416858386993408,
      "epoch": 5.897879789924139,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002145343291330032,
      "loss": 4.7771,
      "mean_token_accuracy": 0.22894878834486007,
      "num_tokens": 131489705.0,
      "step": 75805
    },
    {
      "entropy": 5.397819948196411,
      "epoch": 5.898268819295857,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021450763355577318,
      "loss": 4.7305,
      "mean_token_accuracy": 0.22951875180006026,
      "num_tokens": 131497501.0,
      "step": 75810
    },
    {
      "entropy": 5.463372898101807,
      "epoch": 5.8986578486675745,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021448093889641867,
      "loss": 4.7419,
      "mean_token_accuracy": 0.23163956254720688,
      "num_tokens": 131506412.0,
      "step": 75815
    },
    {
      "entropy": 5.438165760040283,
      "epoch": 5.899046878039292,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002144542451553445,
      "loss": 4.7322,
      "mean_token_accuracy": 0.23713373988866807,
      "num_tokens": 131515473.0,
      "step": 75820
    },
    {
      "entropy": 5.279882431030273,
      "epoch": 5.89943590741101,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002144275523329558,
      "loss": 4.5713,
      "mean_token_accuracy": 0.24541660100221635,
      "num_tokens": 131524224.0,
      "step": 75825
    },
    {
      "entropy": 5.410733366012574,
      "epoch": 5.899824936782727,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021440086042965762,
      "loss": 4.719,
      "mean_token_accuracy": 0.2402533173561096,
      "num_tokens": 131532916.0,
      "step": 75830
    },
    {
      "entropy": 5.387822771072388,
      "epoch": 5.900213966154444,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021437416944585487,
      "loss": 4.7066,
      "mean_token_accuracy": 0.2368634134531021,
      "num_tokens": 131541466.0,
      "step": 75835
    },
    {
      "entropy": 5.286342477798462,
      "epoch": 5.900602995526162,
      "grad_norm": 1.25,
      "learning_rate": 0.00021434747938195266,
      "loss": 4.5514,
      "mean_token_accuracy": 0.24460933059453965,
      "num_tokens": 131550830.0,
      "step": 75840
    },
    {
      "entropy": 5.380221796035767,
      "epoch": 5.90099202489788,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021432079023835587,
      "loss": 4.7384,
      "mean_token_accuracy": 0.22526040226221083,
      "num_tokens": 131559113.0,
      "step": 75845
    },
    {
      "entropy": 5.392773962020874,
      "epoch": 5.9013810542695975,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002142941020154695,
      "loss": 4.7277,
      "mean_token_accuracy": 0.23377591669559478,
      "num_tokens": 131567102.0,
      "step": 75850
    },
    {
      "entropy": 5.483459568023681,
      "epoch": 5.901770083641315,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021426741471369855,
      "loss": 4.7294,
      "mean_token_accuracy": 0.22994182407855987,
      "num_tokens": 131576177.0,
      "step": 75855
    },
    {
      "entropy": 5.43617615699768,
      "epoch": 5.902159113013033,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021424072833344781,
      "loss": 4.7179,
      "mean_token_accuracy": 0.23620351999998093,
      "num_tokens": 131584860.0,
      "step": 75860
    },
    {
      "entropy": 5.403309726715088,
      "epoch": 5.90254814238475,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002142140428751224,
      "loss": 4.7296,
      "mean_token_accuracy": 0.23249084055423735,
      "num_tokens": 131594464.0,
      "step": 75865
    },
    {
      "entropy": 5.3663276672363285,
      "epoch": 5.902937171756467,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002141873583391271,
      "loss": 4.6963,
      "mean_token_accuracy": 0.23175112754106522,
      "num_tokens": 131603116.0,
      "step": 75870
    },
    {
      "entropy": 5.318330144882202,
      "epoch": 5.903326201128185,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021416067472586692,
      "loss": 4.5783,
      "mean_token_accuracy": 0.2420200377702713,
      "num_tokens": 131611273.0,
      "step": 75875
    },
    {
      "entropy": 5.400623798370361,
      "epoch": 5.903715230499903,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021413399203574658,
      "loss": 4.7705,
      "mean_token_accuracy": 0.2247158706188202,
      "num_tokens": 131620141.0,
      "step": 75880
    },
    {
      "entropy": 5.410860776901245,
      "epoch": 5.9041042598716205,
      "grad_norm": 1.25,
      "learning_rate": 0.00021410731026917106,
      "loss": 4.7516,
      "mean_token_accuracy": 0.2289738655090332,
      "num_tokens": 131629020.0,
      "step": 75885
    },
    {
      "entropy": 5.418546152114868,
      "epoch": 5.904493289243338,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002140806294265452,
      "loss": 4.62,
      "mean_token_accuracy": 0.23874539285898208,
      "num_tokens": 131636982.0,
      "step": 75890
    },
    {
      "entropy": 5.418768501281738,
      "epoch": 5.904882318615055,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002140539495082739,
      "loss": 4.7298,
      "mean_token_accuracy": 0.2272329270839691,
      "num_tokens": 131646124.0,
      "step": 75895
    },
    {
      "entropy": 5.3883003234863285,
      "epoch": 5.905271347986773,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021402727051476184,
      "loss": 4.668,
      "mean_token_accuracy": 0.23712324351072311,
      "num_tokens": 131655021.0,
      "step": 75900
    },
    {
      "entropy": 5.401298475265503,
      "epoch": 5.90566037735849,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002140005924464139,
      "loss": 4.7033,
      "mean_token_accuracy": 0.23961022049188613,
      "num_tokens": 131664146.0,
      "step": 75905
    },
    {
      "entropy": 5.439096260070801,
      "epoch": 5.906049406730208,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021397391530363487,
      "loss": 4.7166,
      "mean_token_accuracy": 0.2325331449508667,
      "num_tokens": 131671861.0,
      "step": 75910
    },
    {
      "entropy": 5.431197881698608,
      "epoch": 5.906438436101926,
      "grad_norm": 1.375,
      "learning_rate": 0.00021394723908682955,
      "loss": 4.6797,
      "mean_token_accuracy": 0.2432627186179161,
      "num_tokens": 131679936.0,
      "step": 75915
    },
    {
      "entropy": 5.354474592208862,
      "epoch": 5.9068274654736435,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021392056379640267,
      "loss": 4.6726,
      "mean_token_accuracy": 0.23982449769973754,
      "num_tokens": 131687902.0,
      "step": 75920
    },
    {
      "entropy": 5.318656492233276,
      "epoch": 5.907216494845361,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00021389388943275907,
      "loss": 4.569,
      "mean_token_accuracy": 0.25642589628696444,
      "num_tokens": 131696386.0,
      "step": 75925
    },
    {
      "entropy": 5.347502088546753,
      "epoch": 5.907605524217079,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021386721599630333,
      "loss": 4.6253,
      "mean_token_accuracy": 0.24499854296445847,
      "num_tokens": 131704763.0,
      "step": 75930
    },
    {
      "entropy": 5.43164644241333,
      "epoch": 5.907994553588796,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021384054348744035,
      "loss": 4.8059,
      "mean_token_accuracy": 0.22562328726053238,
      "num_tokens": 131713272.0,
      "step": 75935
    },
    {
      "entropy": 5.428429079055786,
      "epoch": 5.908383582960513,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002138138719065748,
      "loss": 4.7243,
      "mean_token_accuracy": 0.22657574564218522,
      "num_tokens": 131722084.0,
      "step": 75940
    },
    {
      "entropy": 5.5224555969238285,
      "epoch": 5.908772612332231,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002137872012541114,
      "loss": 4.8029,
      "mean_token_accuracy": 0.22344074547290801,
      "num_tokens": 131730576.0,
      "step": 75945
    },
    {
      "entropy": 5.494440698623658,
      "epoch": 5.909161641703949,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021376053153045473,
      "loss": 4.797,
      "mean_token_accuracy": 0.22645127177238464,
      "num_tokens": 131739715.0,
      "step": 75950
    },
    {
      "entropy": 5.397754144668579,
      "epoch": 5.9095506710756665,
      "grad_norm": 1.21875,
      "learning_rate": 0.00021373386273600953,
      "loss": 4.6735,
      "mean_token_accuracy": 0.2362038552761078,
      "num_tokens": 131749006.0,
      "step": 75955
    },
    {
      "entropy": 5.451133823394775,
      "epoch": 5.909939700447384,
      "grad_norm": 1.171875,
      "learning_rate": 0.00021370719487118046,
      "loss": 4.8163,
      "mean_token_accuracy": 0.22346639782190322,
      "num_tokens": 131757909.0,
      "step": 75960
    },
    {
      "entropy": 5.370549249649048,
      "epoch": 5.910328729819101,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002136805279363721,
      "loss": 4.6771,
      "mean_token_accuracy": 0.23761456608772277,
      "num_tokens": 131766932.0,
      "step": 75965
    },
    {
      "entropy": 5.395207834243775,
      "epoch": 5.910717759190819,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00021365386193198917,
      "loss": 4.7383,
      "mean_token_accuracy": 0.23079489022493363,
      "num_tokens": 131775765.0,
      "step": 75970
    },
    {
      "entropy": 5.465735673904419,
      "epoch": 5.911106788562536,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002136271968584363,
      "loss": 4.7305,
      "mean_token_accuracy": 0.2312104359269142,
      "num_tokens": 131784266.0,
      "step": 75975
    },
    {
      "entropy": 5.461711502075195,
      "epoch": 5.911495817934254,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021360053271611795,
      "loss": 4.7226,
      "mean_token_accuracy": 0.23457106500864028,
      "num_tokens": 131792345.0,
      "step": 75980
    },
    {
      "entropy": 5.368264150619507,
      "epoch": 5.911884847305972,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021357386950543884,
      "loss": 4.6613,
      "mean_token_accuracy": 0.23398571461439133,
      "num_tokens": 131801213.0,
      "step": 75985
    },
    {
      "entropy": 5.363194513320923,
      "epoch": 5.912273876677689,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021354720722680347,
      "loss": 4.6587,
      "mean_token_accuracy": 0.24018124341964722,
      "num_tokens": 131809530.0,
      "step": 75990
    },
    {
      "entropy": 5.3555890083312985,
      "epoch": 5.912662906049407,
      "grad_norm": 1.25,
      "learning_rate": 0.00021352054588061632,
      "loss": 4.6523,
      "mean_token_accuracy": 0.23964514285326005,
      "num_tokens": 131817686.0,
      "step": 75995
    },
    {
      "entropy": 5.408930921554566,
      "epoch": 5.913051935421124,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002134938854672821,
      "loss": 4.6699,
      "mean_token_accuracy": 0.23864445984363555,
      "num_tokens": 131827199.0,
      "step": 76000
    },
    {
      "entropy": 5.390394926071167,
      "epoch": 5.913440964792842,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021346722598720536,
      "loss": 4.742,
      "mean_token_accuracy": 0.23177618980407716,
      "num_tokens": 131835899.0,
      "step": 76005
    },
    {
      "entropy": 5.317753505706787,
      "epoch": 5.913829994164559,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021344056744079044,
      "loss": 4.5299,
      "mean_token_accuracy": 0.24599709063768388,
      "num_tokens": 131843862.0,
      "step": 76010
    },
    {
      "entropy": 5.381081247329712,
      "epoch": 5.914219023536277,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021341390982844194,
      "loss": 4.6797,
      "mean_token_accuracy": 0.24037142992019653,
      "num_tokens": 131852970.0,
      "step": 76015
    },
    {
      "entropy": 5.4305901527404785,
      "epoch": 5.914608052907995,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021338725315056434,
      "loss": 4.7662,
      "mean_token_accuracy": 0.2277293771505356,
      "num_tokens": 131861623.0,
      "step": 76020
    },
    {
      "entropy": 5.360603475570679,
      "epoch": 5.914997082279712,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021336059740756214,
      "loss": 4.6498,
      "mean_token_accuracy": 0.23802496641874313,
      "num_tokens": 131869813.0,
      "step": 76025
    },
    {
      "entropy": 5.436095476150513,
      "epoch": 5.915386111651429,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00021333394259983975,
      "loss": 4.7624,
      "mean_token_accuracy": 0.2325201690196991,
      "num_tokens": 131879634.0,
      "step": 76030
    },
    {
      "entropy": 5.384984874725342,
      "epoch": 5.915775141023147,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021330728872780158,
      "loss": 4.6649,
      "mean_token_accuracy": 0.2393465295433998,
      "num_tokens": 131888962.0,
      "step": 76035
    },
    {
      "entropy": 5.394105243682861,
      "epoch": 5.916164170394865,
      "grad_norm": 1.25,
      "learning_rate": 0.00021328063579185213,
      "loss": 4.6441,
      "mean_token_accuracy": 0.2384956106543541,
      "num_tokens": 131897356.0,
      "step": 76040
    },
    {
      "entropy": 5.4822228908538815,
      "epoch": 5.916553199766582,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002132539837923958,
      "loss": 4.8027,
      "mean_token_accuracy": 0.2314058557152748,
      "num_tokens": 131905466.0,
      "step": 76045
    },
    {
      "entropy": 5.421515798568725,
      "epoch": 5.9169422291383,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021322733272983703,
      "loss": 4.7862,
      "mean_token_accuracy": 0.2252401739358902,
      "num_tokens": 131914098.0,
      "step": 76050
    },
    {
      "entropy": 5.503602981567383,
      "epoch": 5.917331258510018,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021320068260458015,
      "loss": 4.7247,
      "mean_token_accuracy": 0.2342439815402031,
      "num_tokens": 131921736.0,
      "step": 76055
    },
    {
      "entropy": 5.424953269958496,
      "epoch": 5.917720287881735,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021317403341702945,
      "loss": 4.6984,
      "mean_token_accuracy": 0.23516819924116134,
      "num_tokens": 131929748.0,
      "step": 76060
    },
    {
      "entropy": 5.39015154838562,
      "epoch": 5.918109317253453,
      "grad_norm": 1.21875,
      "learning_rate": 0.00021314738516758946,
      "loss": 4.7531,
      "mean_token_accuracy": 0.22663253247737886,
      "num_tokens": 131938924.0,
      "step": 76065
    },
    {
      "entropy": 5.467780113220215,
      "epoch": 5.91849834662517,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002131207378566644,
      "loss": 4.7703,
      "mean_token_accuracy": 0.23378603011369706,
      "num_tokens": 131946960.0,
      "step": 76070
    },
    {
      "entropy": 5.496523809432984,
      "epoch": 5.918887375996888,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00021309409148465876,
      "loss": 4.8002,
      "mean_token_accuracy": 0.23000468760728837,
      "num_tokens": 131955340.0,
      "step": 76075
    },
    {
      "entropy": 5.500981855392456,
      "epoch": 5.919276405368605,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002130674460519767,
      "loss": 4.7705,
      "mean_token_accuracy": 0.22325754463672637,
      "num_tokens": 131963573.0,
      "step": 76080
    },
    {
      "entropy": 5.369329118728638,
      "epoch": 5.919665434740323,
      "grad_norm": 1.1875,
      "learning_rate": 0.00021304080155902256,
      "loss": 4.6837,
      "mean_token_accuracy": 0.23269959390163422,
      "num_tokens": 131971753.0,
      "step": 76085
    },
    {
      "entropy": 5.458973979949951,
      "epoch": 5.920054464112041,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021301415800620066,
      "loss": 4.7915,
      "mean_token_accuracy": 0.22604616731405258,
      "num_tokens": 131980221.0,
      "step": 76090
    },
    {
      "entropy": 5.42647910118103,
      "epoch": 5.9204434934837575,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00021298751539391525,
      "loss": 4.8037,
      "mean_token_accuracy": 0.22318702787160874,
      "num_tokens": 131989104.0,
      "step": 76095
    },
    {
      "entropy": 5.417198276519775,
      "epoch": 5.920832522855475,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002129608737225706,
      "loss": 4.7112,
      "mean_token_accuracy": 0.2349577948451042,
      "num_tokens": 131997596.0,
      "step": 76100
    },
    {
      "entropy": 5.463526630401612,
      "epoch": 5.921221552227193,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021293423299257092,
      "loss": 4.7261,
      "mean_token_accuracy": 0.23496688455343245,
      "num_tokens": 132006056.0,
      "step": 76105
    },
    {
      "entropy": 5.451560306549072,
      "epoch": 5.921610581598911,
      "grad_norm": 1.375,
      "learning_rate": 0.00021290759320432046,
      "loss": 4.7201,
      "mean_token_accuracy": 0.23233759105205537,
      "num_tokens": 132014068.0,
      "step": 76110
    },
    {
      "entropy": 5.400394296646118,
      "epoch": 5.921999610970628,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002128809543582234,
      "loss": 4.6837,
      "mean_token_accuracy": 0.2356068804860115,
      "num_tokens": 132023567.0,
      "step": 76115
    },
    {
      "entropy": 5.386044263839722,
      "epoch": 5.922388640342346,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021285431645468405,
      "loss": 4.7099,
      "mean_token_accuracy": 0.23647297620773317,
      "num_tokens": 132032993.0,
      "step": 76120
    },
    {
      "entropy": 5.475244951248169,
      "epoch": 5.922777669714064,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021282767949410653,
      "loss": 4.7874,
      "mean_token_accuracy": 0.22518738806247712,
      "num_tokens": 132041625.0,
      "step": 76125
    },
    {
      "entropy": 5.544586896896362,
      "epoch": 5.923166699085781,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021280104347689495,
      "loss": 4.7748,
      "mean_token_accuracy": 0.231954625248909,
      "num_tokens": 132050227.0,
      "step": 76130
    },
    {
      "entropy": 5.405128288269043,
      "epoch": 5.923555728457498,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021277440840345363,
      "loss": 4.7015,
      "mean_token_accuracy": 0.23236709237098693,
      "num_tokens": 132058813.0,
      "step": 76135
    },
    {
      "entropy": 5.364491176605225,
      "epoch": 5.923944757829216,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021274777427418657,
      "loss": 4.6956,
      "mean_token_accuracy": 0.23926969319581987,
      "num_tokens": 132066485.0,
      "step": 76140
    },
    {
      "entropy": 5.383483362197876,
      "epoch": 5.924333787200934,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021272114108949797,
      "loss": 4.6822,
      "mean_token_accuracy": 0.23116630911827088,
      "num_tokens": 132075531.0,
      "step": 76145
    },
    {
      "entropy": 5.360352897644043,
      "epoch": 5.924722816572651,
      "grad_norm": 1.25,
      "learning_rate": 0.0002126945088497919,
      "loss": 4.6548,
      "mean_token_accuracy": 0.2338838368654251,
      "num_tokens": 132084531.0,
      "step": 76150
    },
    {
      "entropy": 5.373699045181274,
      "epoch": 5.925111845944369,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002126678775554725,
      "loss": 4.7207,
      "mean_token_accuracy": 0.22971969544887544,
      "num_tokens": 132093462.0,
      "step": 76155
    },
    {
      "entropy": 5.4652563571929935,
      "epoch": 5.925500875316087,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002126412472069439,
      "loss": 4.7205,
      "mean_token_accuracy": 0.22945373505353928,
      "num_tokens": 132102035.0,
      "step": 76160
    },
    {
      "entropy": 5.352635431289673,
      "epoch": 5.9258899046878035,
      "grad_norm": 1.15625,
      "learning_rate": 0.00021261461780461006,
      "loss": 4.6302,
      "mean_token_accuracy": 0.2402480885386467,
      "num_tokens": 132110488.0,
      "step": 76165
    },
    {
      "entropy": 5.3912210941314695,
      "epoch": 5.926278934059521,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021258798934887513,
      "loss": 4.6484,
      "mean_token_accuracy": 0.24137549996376037,
      "num_tokens": 132118459.0,
      "step": 76170
    },
    {
      "entropy": 5.369686269760132,
      "epoch": 5.926667963431239,
      "grad_norm": 1.1875,
      "learning_rate": 0.00021256136184014313,
      "loss": 4.6567,
      "mean_token_accuracy": 0.2321525976061821,
      "num_tokens": 132126768.0,
      "step": 76175
    },
    {
      "entropy": 5.392707014083863,
      "epoch": 5.927056992802957,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021253473527881806,
      "loss": 4.6854,
      "mean_token_accuracy": 0.2317980170249939,
      "num_tokens": 132135295.0,
      "step": 76180
    },
    {
      "entropy": 5.416296005249023,
      "epoch": 5.927446022174674,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00021250810966530392,
      "loss": 4.7414,
      "mean_token_accuracy": 0.23518838584423066,
      "num_tokens": 132144805.0,
      "step": 76185
    },
    {
      "entropy": 5.344917297363281,
      "epoch": 5.927835051546392,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002124814850000048,
      "loss": 4.6545,
      "mean_token_accuracy": 0.2364877536892891,
      "num_tokens": 132153359.0,
      "step": 76190
    },
    {
      "entropy": 5.355239295959473,
      "epoch": 5.92822408091811,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00021245486128332458,
      "loss": 4.6292,
      "mean_token_accuracy": 0.23918001651763915,
      "num_tokens": 132162390.0,
      "step": 76195
    },
    {
      "entropy": 5.383667898178101,
      "epoch": 5.9286131102898265,
      "grad_norm": 1.21875,
      "learning_rate": 0.00021242823851566733,
      "loss": 4.6677,
      "mean_token_accuracy": 0.22867198288440704,
      "num_tokens": 132170880.0,
      "step": 76200
    },
    {
      "entropy": 5.3855204582214355,
      "epoch": 5.929002139661544,
      "grad_norm": 1.1875,
      "learning_rate": 0.00021240161669743696,
      "loss": 4.7512,
      "mean_token_accuracy": 0.2263859450817108,
      "num_tokens": 132180966.0,
      "step": 76205
    },
    {
      "entropy": 5.41846981048584,
      "epoch": 5.929391169033262,
      "grad_norm": 1.25,
      "learning_rate": 0.0002123749958290374,
      "loss": 4.6571,
      "mean_token_accuracy": 0.23393280804157257,
      "num_tokens": 132189339.0,
      "step": 76210
    },
    {
      "entropy": 5.391187906265259,
      "epoch": 5.92978019840498,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002123483759108727,
      "loss": 4.6677,
      "mean_token_accuracy": 0.24157705008983613,
      "num_tokens": 132197654.0,
      "step": 76215
    },
    {
      "entropy": 5.451472568511963,
      "epoch": 5.930169227776697,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021232175694334653,
      "loss": 4.7025,
      "mean_token_accuracy": 0.2315573662519455,
      "num_tokens": 132206479.0,
      "step": 76220
    },
    {
      "entropy": 5.350254535675049,
      "epoch": 5.930558257148415,
      "grad_norm": 1.171875,
      "learning_rate": 0.00021229513892686303,
      "loss": 4.6751,
      "mean_token_accuracy": 0.23381597101688384,
      "num_tokens": 132215930.0,
      "step": 76225
    },
    {
      "entropy": 5.359291934967041,
      "epoch": 5.930947286520132,
      "grad_norm": 1.203125,
      "learning_rate": 0.000212268521861826,
      "loss": 4.6158,
      "mean_token_accuracy": 0.23722731322050095,
      "num_tokens": 132224133.0,
      "step": 76230
    },
    {
      "entropy": 5.428776454925537,
      "epoch": 5.9313363158918495,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002122419057486393,
      "loss": 4.7052,
      "mean_token_accuracy": 0.23858193159103394,
      "num_tokens": 132233171.0,
      "step": 76235
    },
    {
      "entropy": 5.441717529296875,
      "epoch": 5.931725345263567,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021221529058770673,
      "loss": 4.8059,
      "mean_token_accuracy": 0.2207982137799263,
      "num_tokens": 132242111.0,
      "step": 76240
    },
    {
      "entropy": 5.406039762496948,
      "epoch": 5.932114374635285,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021218867637943224,
      "loss": 4.7367,
      "mean_token_accuracy": 0.23015722930431365,
      "num_tokens": 132250817.0,
      "step": 76245
    },
    {
      "entropy": 5.303554534912109,
      "epoch": 5.932503404007003,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002121620631242196,
      "loss": 4.5595,
      "mean_token_accuracy": 0.25084811449050903,
      "num_tokens": 132259081.0,
      "step": 76250
    },
    {
      "entropy": 5.36400785446167,
      "epoch": 5.93289243337872,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021213545082247266,
      "loss": 4.6544,
      "mean_token_accuracy": 0.23282447308301926,
      "num_tokens": 132268174.0,
      "step": 76255
    },
    {
      "entropy": 5.311514949798584,
      "epoch": 5.933281462750438,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021210883947459509,
      "loss": 4.6593,
      "mean_token_accuracy": 0.24024641811847686,
      "num_tokens": 132276834.0,
      "step": 76260
    },
    {
      "entropy": 5.4077270984649655,
      "epoch": 5.933670492122156,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021208222908099085,
      "loss": 4.6969,
      "mean_token_accuracy": 0.22875261157751084,
      "num_tokens": 132285384.0,
      "step": 76265
    },
    {
      "entropy": 5.427400207519531,
      "epoch": 5.9340595214938725,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002120556196420636,
      "loss": 4.754,
      "mean_token_accuracy": 0.2294132798910141,
      "num_tokens": 132294485.0,
      "step": 76270
    },
    {
      "entropy": 5.434949970245361,
      "epoch": 5.93444855086559,
      "grad_norm": 1.25,
      "learning_rate": 0.00021202901115821728,
      "loss": 4.7643,
      "mean_token_accuracy": 0.23428494930267335,
      "num_tokens": 132303277.0,
      "step": 76275
    },
    {
      "entropy": 5.421936988830566,
      "epoch": 5.934837580237308,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002120024036298554,
      "loss": 4.7202,
      "mean_token_accuracy": 0.23176118433475495,
      "num_tokens": 132311795.0,
      "step": 76280
    },
    {
      "entropy": 5.393312740325928,
      "epoch": 5.935226609609026,
      "grad_norm": 1.25,
      "learning_rate": 0.00021197579705738173,
      "loss": 4.7372,
      "mean_token_accuracy": 0.23162569105625153,
      "num_tokens": 132320349.0,
      "step": 76285
    },
    {
      "entropy": 5.369066858291626,
      "epoch": 5.935615638980743,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021194919144120006,
      "loss": 4.6097,
      "mean_token_accuracy": 0.2417198449373245,
      "num_tokens": 132328595.0,
      "step": 76290
    },
    {
      "entropy": 5.44274115562439,
      "epoch": 5.936004668352461,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00021192258678171406,
      "loss": 4.7651,
      "mean_token_accuracy": 0.23599553257226943,
      "num_tokens": 132338782.0,
      "step": 76295
    },
    {
      "entropy": 5.506654262542725,
      "epoch": 5.936393697724178,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002118959830793274,
      "loss": 4.8028,
      "mean_token_accuracy": 0.2305076688528061,
      "num_tokens": 132347222.0,
      "step": 76300
    },
    {
      "entropy": 5.507441759109497,
      "epoch": 5.9367827270958955,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021186938033444382,
      "loss": 4.8367,
      "mean_token_accuracy": 0.2293297231197357,
      "num_tokens": 132355881.0,
      "step": 76305
    },
    {
      "entropy": 5.491884374618531,
      "epoch": 5.937171756467613,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021184277854746682,
      "loss": 4.7392,
      "mean_token_accuracy": 0.22916317731142044,
      "num_tokens": 132365951.0,
      "step": 76310
    },
    {
      "entropy": 5.41453070640564,
      "epoch": 5.937560785839331,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00021181617771880013,
      "loss": 4.7717,
      "mean_token_accuracy": 0.22971153110265732,
      "num_tokens": 132374522.0,
      "step": 76315
    },
    {
      "entropy": 5.395865392684937,
      "epoch": 5.937949815211049,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021178957784884745,
      "loss": 4.6952,
      "mean_token_accuracy": 0.2322625383734703,
      "num_tokens": 132383242.0,
      "step": 76320
    },
    {
      "entropy": 5.4336583614349365,
      "epoch": 5.938338844582766,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021176297893801222,
      "loss": 4.6561,
      "mean_token_accuracy": 0.24170120805501938,
      "num_tokens": 132391894.0,
      "step": 76325
    },
    {
      "entropy": 5.463323354721069,
      "epoch": 5.938727873954484,
      "grad_norm": 1.21875,
      "learning_rate": 0.00021173638098669818,
      "loss": 4.7663,
      "mean_token_accuracy": 0.231906895339489,
      "num_tokens": 132400260.0,
      "step": 76330
    },
    {
      "entropy": 5.4831788539886475,
      "epoch": 5.939116903326201,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002117097839953089,
      "loss": 4.8219,
      "mean_token_accuracy": 0.22252528071403505,
      "num_tokens": 132408341.0,
      "step": 76335
    },
    {
      "entropy": 5.438129663467407,
      "epoch": 5.9395059326979185,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00021168318796424784,
      "loss": 4.7608,
      "mean_token_accuracy": 0.23347238451242447,
      "num_tokens": 132417507.0,
      "step": 76340
    },
    {
      "entropy": 5.4105963706970215,
      "epoch": 5.939894962069636,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021165659289391865,
      "loss": 4.7314,
      "mean_token_accuracy": 0.22635268568992614,
      "num_tokens": 132426440.0,
      "step": 76345
    },
    {
      "entropy": 5.3511568069458,
      "epoch": 5.940283991441354,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002116299987847249,
      "loss": 4.6548,
      "mean_token_accuracy": 0.23678378909826278,
      "num_tokens": 132434740.0,
      "step": 76350
    },
    {
      "entropy": 5.453312778472901,
      "epoch": 5.940673020813072,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021160340563707004,
      "loss": 4.7697,
      "mean_token_accuracy": 0.2272647276520729,
      "num_tokens": 132443070.0,
      "step": 76355
    },
    {
      "entropy": 5.432166719436646,
      "epoch": 5.941062050184789,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00021157681345135754,
      "loss": 4.7353,
      "mean_token_accuracy": 0.2346034973859787,
      "num_tokens": 132451587.0,
      "step": 76360
    },
    {
      "entropy": 5.367437314987183,
      "epoch": 5.941451079556506,
      "grad_norm": 1.25,
      "learning_rate": 0.00021155022222799097,
      "loss": 4.6647,
      "mean_token_accuracy": 0.23171533197164534,
      "num_tokens": 132460017.0,
      "step": 76365
    },
    {
      "entropy": 5.3677257061004635,
      "epoch": 5.941840108928224,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021152363196737378,
      "loss": 4.6695,
      "mean_token_accuracy": 0.24204727858304978,
      "num_tokens": 132468470.0,
      "step": 76370
    },
    {
      "entropy": 5.372550201416016,
      "epoch": 5.9422291382999415,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002114970426699095,
      "loss": 4.6713,
      "mean_token_accuracy": 0.23977320194244384,
      "num_tokens": 132476827.0,
      "step": 76375
    },
    {
      "entropy": 5.363492059707641,
      "epoch": 5.942618167671659,
      "grad_norm": 1.25,
      "learning_rate": 0.00021147045433600142,
      "loss": 4.6788,
      "mean_token_accuracy": 0.23303119391202926,
      "num_tokens": 132485346.0,
      "step": 76380
    },
    {
      "entropy": 5.490091466903687,
      "epoch": 5.943007197043377,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021144386696605316,
      "loss": 4.7844,
      "mean_token_accuracy": 0.22473648935556412,
      "num_tokens": 132493470.0,
      "step": 76385
    },
    {
      "entropy": 5.403557300567627,
      "epoch": 5.943396226415095,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021141728056046804,
      "loss": 4.6176,
      "mean_token_accuracy": 0.24426782727241517,
      "num_tokens": 132502228.0,
      "step": 76390
    },
    {
      "entropy": 5.46008415222168,
      "epoch": 5.943785255786812,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021139069511964942,
      "loss": 4.7546,
      "mean_token_accuracy": 0.23142188936471939,
      "num_tokens": 132510489.0,
      "step": 76395
    },
    {
      "entropy": 5.436754369735718,
      "epoch": 5.944174285158529,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002113641106440008,
      "loss": 4.7202,
      "mean_token_accuracy": 0.23291650265455247,
      "num_tokens": 132519372.0,
      "step": 76400
    },
    {
      "entropy": 5.443526554107666,
      "epoch": 5.944563314530247,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002113375271339255,
      "loss": 4.7669,
      "mean_token_accuracy": 0.23147513419389726,
      "num_tokens": 132527924.0,
      "step": 76405
    },
    {
      "entropy": 5.431633710861206,
      "epoch": 5.9449523439019645,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002113109445898269,
      "loss": 4.6948,
      "mean_token_accuracy": 0.23597197979688644,
      "num_tokens": 132537440.0,
      "step": 76410
    },
    {
      "entropy": 5.49529185295105,
      "epoch": 5.945341373273682,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021128436301210842,
      "loss": 4.812,
      "mean_token_accuracy": 0.22199621498584748,
      "num_tokens": 132546134.0,
      "step": 76415
    },
    {
      "entropy": 5.466610765457153,
      "epoch": 5.9457304026454,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021125778240117327,
      "loss": 4.7213,
      "mean_token_accuracy": 0.23033841997385024,
      "num_tokens": 132554590.0,
      "step": 76420
    },
    {
      "entropy": 5.424131631851196,
      "epoch": 5.946119432017118,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002112312027574248,
      "loss": 4.6885,
      "mean_token_accuracy": 0.23711908757686614,
      "num_tokens": 132563934.0,
      "step": 76425
    },
    {
      "entropy": 5.446535921096801,
      "epoch": 5.946508461388834,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021120462408126634,
      "loss": 4.794,
      "mean_token_accuracy": 0.22354500889778137,
      "num_tokens": 132572581.0,
      "step": 76430
    },
    {
      "entropy": 5.436050748825073,
      "epoch": 5.946897490760552,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021117804637310123,
      "loss": 4.8246,
      "mean_token_accuracy": 0.22417698204517364,
      "num_tokens": 132580363.0,
      "step": 76435
    },
    {
      "entropy": 5.392775249481201,
      "epoch": 5.94728652013227,
      "grad_norm": 1.25,
      "learning_rate": 0.0002111514696333326,
      "loss": 4.7042,
      "mean_token_accuracy": 0.23352446854114534,
      "num_tokens": 132589116.0,
      "step": 76440
    },
    {
      "entropy": 5.384840774536133,
      "epoch": 5.9476755495039875,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00021112489386236383,
      "loss": 4.6617,
      "mean_token_accuracy": 0.2400180831551552,
      "num_tokens": 132598114.0,
      "step": 76445
    },
    {
      "entropy": 5.355743169784546,
      "epoch": 5.948064578875705,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00021109831906059807,
      "loss": 4.6663,
      "mean_token_accuracy": 0.23696862310171127,
      "num_tokens": 132608219.0,
      "step": 76450
    },
    {
      "entropy": 5.404343461990356,
      "epoch": 5.948453608247423,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002110717452284387,
      "loss": 4.7239,
      "mean_token_accuracy": 0.2336655542254448,
      "num_tokens": 132617469.0,
      "step": 76455
    },
    {
      "entropy": 5.355747222900391,
      "epoch": 5.948842637619141,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021104517236628868,
      "loss": 4.6149,
      "mean_token_accuracy": 0.2373715177178383,
      "num_tokens": 132626696.0,
      "step": 76460
    },
    {
      "entropy": 5.397546625137329,
      "epoch": 5.949231666990858,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002110186004745515,
      "loss": 4.7557,
      "mean_token_accuracy": 0.23237808644771576,
      "num_tokens": 132635392.0,
      "step": 76465
    },
    {
      "entropy": 5.4194964408874515,
      "epoch": 5.949620696362575,
      "grad_norm": 1.140625,
      "learning_rate": 0.00021099202955363023,
      "loss": 4.7628,
      "mean_token_accuracy": 0.22070899605751038,
      "num_tokens": 132644538.0,
      "step": 76470
    },
    {
      "entropy": 5.382121753692627,
      "epoch": 5.950009725734293,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021096545960392805,
      "loss": 4.6722,
      "mean_token_accuracy": 0.23862248957157134,
      "num_tokens": 132653663.0,
      "step": 76475
    },
    {
      "entropy": 5.330176877975464,
      "epoch": 5.9503987551060105,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00021093889062584803,
      "loss": 4.6393,
      "mean_token_accuracy": 0.24172260314226152,
      "num_tokens": 132662044.0,
      "step": 76480
    },
    {
      "entropy": 5.286471748352051,
      "epoch": 5.950787784477728,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021091232261979343,
      "loss": 4.6335,
      "mean_token_accuracy": 0.24145150780677796,
      "num_tokens": 132670654.0,
      "step": 76485
    },
    {
      "entropy": 5.429275560379028,
      "epoch": 5.951176813849446,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002108857555861673,
      "loss": 4.7741,
      "mean_token_accuracy": 0.22701884657144547,
      "num_tokens": 132680060.0,
      "step": 76490
    },
    {
      "entropy": 5.398393201828003,
      "epoch": 5.9515658432211636,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021085918952537275,
      "loss": 4.709,
      "mean_token_accuracy": 0.23249875903129577,
      "num_tokens": 132689385.0,
      "step": 76495
    },
    {
      "entropy": 5.370587921142578,
      "epoch": 5.95195487259288,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00021083262443781297,
      "loss": 4.6185,
      "mean_token_accuracy": 0.23504396378993989,
      "num_tokens": 132698009.0,
      "step": 76500
    },
    {
      "entropy": 5.449375343322754,
      "epoch": 5.952343901964598,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021080606032389094,
      "loss": 4.7952,
      "mean_token_accuracy": 0.235028176009655,
      "num_tokens": 132705889.0,
      "step": 76505
    },
    {
      "entropy": 5.406833791732788,
      "epoch": 5.952732931336316,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002107794971840098,
      "loss": 4.7406,
      "mean_token_accuracy": 0.23015243262052537,
      "num_tokens": 132714964.0,
      "step": 76510
    },
    {
      "entropy": 5.395500564575196,
      "epoch": 5.9531219607080335,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021075293501857264,
      "loss": 4.6896,
      "mean_token_accuracy": 0.2335689917206764,
      "num_tokens": 132724347.0,
      "step": 76515
    },
    {
      "entropy": 5.437651014328003,
      "epoch": 5.953510990079751,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021072637382798232,
      "loss": 4.7015,
      "mean_token_accuracy": 0.2274057000875473,
      "num_tokens": 132732236.0,
      "step": 76520
    },
    {
      "entropy": 5.441507053375244,
      "epoch": 5.953900019451469,
      "grad_norm": 1.375,
      "learning_rate": 0.00021069981361264195,
      "loss": 4.7653,
      "mean_token_accuracy": 0.22785140573978424,
      "num_tokens": 132740164.0,
      "step": 76525
    },
    {
      "entropy": 5.4347742080688475,
      "epoch": 5.9542890488231865,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002106732543729546,
      "loss": 4.7282,
      "mean_token_accuracy": 0.22833696007728577,
      "num_tokens": 132748446.0,
      "step": 76530
    },
    {
      "entropy": 5.394736623764038,
      "epoch": 5.954678078194903,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021064669610932325,
      "loss": 4.7759,
      "mean_token_accuracy": 0.23221419602632523,
      "num_tokens": 132757428.0,
      "step": 76535
    },
    {
      "entropy": 5.416598224639893,
      "epoch": 5.955067107566621,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021062013882215086,
      "loss": 4.7506,
      "mean_token_accuracy": 0.22751812934875487,
      "num_tokens": 132765854.0,
      "step": 76540
    },
    {
      "entropy": 5.453408432006836,
      "epoch": 5.955456136938339,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021059358251184046,
      "loss": 4.7558,
      "mean_token_accuracy": 0.22911751568317412,
      "num_tokens": 132773576.0,
      "step": 76545
    },
    {
      "entropy": 5.401673126220703,
      "epoch": 5.9558451663100564,
      "grad_norm": 1.25,
      "learning_rate": 0.00021056702717879482,
      "loss": 4.7385,
      "mean_token_accuracy": 0.23613151460886,
      "num_tokens": 132782371.0,
      "step": 76550
    },
    {
      "entropy": 5.438835763931275,
      "epoch": 5.956234195681774,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021054047282341703,
      "loss": 4.7682,
      "mean_token_accuracy": 0.22956449985504152,
      "num_tokens": 132790991.0,
      "step": 76555
    },
    {
      "entropy": 5.419137287139892,
      "epoch": 5.956623225053492,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00021051391944610997,
      "loss": 4.6743,
      "mean_token_accuracy": 0.24255043566226958,
      "num_tokens": 132799817.0,
      "step": 76560
    },
    {
      "entropy": 5.3667065620422365,
      "epoch": 5.957012254425209,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021048736704727656,
      "loss": 4.6427,
      "mean_token_accuracy": 0.23503957390785218,
      "num_tokens": 132807523.0,
      "step": 76565
    },
    {
      "entropy": 5.388102865219116,
      "epoch": 5.957401283796926,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00021046081562731966,
      "loss": 4.6925,
      "mean_token_accuracy": 0.24372862726449968,
      "num_tokens": 132816314.0,
      "step": 76570
    },
    {
      "entropy": 5.455353593826294,
      "epoch": 5.957790313168644,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021043426518664216,
      "loss": 4.8079,
      "mean_token_accuracy": 0.21809467673301697,
      "num_tokens": 132824738.0,
      "step": 76575
    },
    {
      "entropy": 5.430813026428223,
      "epoch": 5.958179342540362,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021040771572564693,
      "loss": 4.713,
      "mean_token_accuracy": 0.237574964761734,
      "num_tokens": 132833395.0,
      "step": 76580
    },
    {
      "entropy": 5.431329584121704,
      "epoch": 5.958568371912079,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021038116724473677,
      "loss": 4.6647,
      "mean_token_accuracy": 0.23489255011081694,
      "num_tokens": 132841798.0,
      "step": 76585
    },
    {
      "entropy": 5.465257692337036,
      "epoch": 5.958957401283797,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021035461974431446,
      "loss": 4.7852,
      "mean_token_accuracy": 0.21890774518251419,
      "num_tokens": 132850087.0,
      "step": 76590
    },
    {
      "entropy": 5.436032629013061,
      "epoch": 5.959346430655515,
      "grad_norm": 1.375,
      "learning_rate": 0.00021032807322478303,
      "loss": 4.7851,
      "mean_token_accuracy": 0.2263456255197525,
      "num_tokens": 132858012.0,
      "step": 76595
    },
    {
      "entropy": 5.387731218338013,
      "epoch": 5.9597354600272325,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021030152768654515,
      "loss": 4.6757,
      "mean_token_accuracy": 0.23943857699632645,
      "num_tokens": 132866521.0,
      "step": 76600
    },
    {
      "entropy": 5.44189944267273,
      "epoch": 5.960124489398949,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021027498313000358,
      "loss": 4.6983,
      "mean_token_accuracy": 0.23528296649456024,
      "num_tokens": 132874997.0,
      "step": 76605
    },
    {
      "entropy": 5.475102281570434,
      "epoch": 5.960513518770667,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021024843955556112,
      "loss": 4.8038,
      "mean_token_accuracy": 0.21828186959028245,
      "num_tokens": 132883030.0,
      "step": 76610
    },
    {
      "entropy": 5.3733603954315186,
      "epoch": 5.960902548142385,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021022189696362053,
      "loss": 4.6543,
      "mean_token_accuracy": 0.23544911891222,
      "num_tokens": 132891283.0,
      "step": 76615
    },
    {
      "entropy": 5.3570554733276365,
      "epoch": 5.961291577514102,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002101953553545845,
      "loss": 4.6557,
      "mean_token_accuracy": 0.2361129567027092,
      "num_tokens": 132899652.0,
      "step": 76620
    },
    {
      "entropy": 5.398224592208862,
      "epoch": 5.96168060688582,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002101688147288559,
      "loss": 4.7106,
      "mean_token_accuracy": 0.23126392662525178,
      "num_tokens": 132908127.0,
      "step": 76625
    },
    {
      "entropy": 5.419182252883911,
      "epoch": 5.962069636257538,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002101422750868373,
      "loss": 4.7407,
      "mean_token_accuracy": 0.2256474807858467,
      "num_tokens": 132917003.0,
      "step": 76630
    },
    {
      "entropy": 5.4841399669647215,
      "epoch": 5.962458665629255,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002101157364289315,
      "loss": 4.817,
      "mean_token_accuracy": 0.2278971180319786,
      "num_tokens": 132925008.0,
      "step": 76635
    },
    {
      "entropy": 5.472033834457397,
      "epoch": 5.962847695000972,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021008919875554105,
      "loss": 4.7369,
      "mean_token_accuracy": 0.23309367001056672,
      "num_tokens": 132933238.0,
      "step": 76640
    },
    {
      "entropy": 5.326199388504028,
      "epoch": 5.96323672437269,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002100626620670687,
      "loss": 4.5771,
      "mean_token_accuracy": 0.24518004059791565,
      "num_tokens": 132941469.0,
      "step": 76645
    },
    {
      "entropy": 5.437912368774414,
      "epoch": 5.963625753744408,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002100361263639171,
      "loss": 4.7642,
      "mean_token_accuracy": 0.22345100790262223,
      "num_tokens": 132950888.0,
      "step": 76650
    },
    {
      "entropy": 5.448306655883789,
      "epoch": 5.964014783116125,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021000959164648892,
      "loss": 4.7976,
      "mean_token_accuracy": 0.2331244707107544,
      "num_tokens": 132959602.0,
      "step": 76655
    },
    {
      "entropy": 5.408717012405395,
      "epoch": 5.964403812487843,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020998305791518668,
      "loss": 4.7054,
      "mean_token_accuracy": 0.23554112911224365,
      "num_tokens": 132967606.0,
      "step": 76660
    },
    {
      "entropy": 5.433862924575806,
      "epoch": 5.964792841859561,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002099565251704131,
      "loss": 4.7418,
      "mean_token_accuracy": 0.2312801092863083,
      "num_tokens": 132976574.0,
      "step": 76665
    },
    {
      "entropy": 5.303404092788696,
      "epoch": 5.965181871231278,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020992999341257069,
      "loss": 4.5951,
      "mean_token_accuracy": 0.23995437622070312,
      "num_tokens": 132985778.0,
      "step": 76670
    },
    {
      "entropy": 5.390028429031372,
      "epoch": 5.965570900602995,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020990346264206206,
      "loss": 4.6381,
      "mean_token_accuracy": 0.23171315640211104,
      "num_tokens": 132994755.0,
      "step": 76675
    },
    {
      "entropy": 5.383834266662598,
      "epoch": 5.965959929974713,
      "grad_norm": 1.25,
      "learning_rate": 0.00020987693285928977,
      "loss": 4.642,
      "mean_token_accuracy": 0.23269011378288268,
      "num_tokens": 133003212.0,
      "step": 76680
    },
    {
      "entropy": 5.349404191970825,
      "epoch": 5.966348959346431,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00020985040406465644,
      "loss": 4.6395,
      "mean_token_accuracy": 0.2379076063632965,
      "num_tokens": 133012536.0,
      "step": 76685
    },
    {
      "entropy": 5.353804445266723,
      "epoch": 5.966737988718148,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00020982387625856447,
      "loss": 4.6609,
      "mean_token_accuracy": 0.24110918194055558,
      "num_tokens": 133020960.0,
      "step": 76690
    },
    {
      "entropy": 5.353933238983155,
      "epoch": 5.967127018089866,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002097973494414165,
      "loss": 4.6516,
      "mean_token_accuracy": 0.24323512464761735,
      "num_tokens": 133029251.0,
      "step": 76695
    },
    {
      "entropy": 5.418486070632935,
      "epoch": 5.967516047461583,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020977082361361488,
      "loss": 4.7402,
      "mean_token_accuracy": 0.23197925835847855,
      "num_tokens": 133037971.0,
      "step": 76700
    },
    {
      "entropy": 5.397760486602783,
      "epoch": 5.967905076833301,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00020974429877556228,
      "loss": 4.7048,
      "mean_token_accuracy": 0.2319283053278923,
      "num_tokens": 133047716.0,
      "step": 76705
    },
    {
      "entropy": 5.548833322525025,
      "epoch": 5.968294106205018,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020971777492766102,
      "loss": 4.8258,
      "mean_token_accuracy": 0.22473503202199935,
      "num_tokens": 133056072.0,
      "step": 76710
    },
    {
      "entropy": 5.403546905517578,
      "epoch": 5.968683135576736,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020969125207031365,
      "loss": 4.6829,
      "mean_token_accuracy": 0.24093794971704482,
      "num_tokens": 133064776.0,
      "step": 76715
    },
    {
      "entropy": 5.45290322303772,
      "epoch": 5.969072164948454,
      "grad_norm": 1.21875,
      "learning_rate": 0.00020966473020392256,
      "loss": 4.7314,
      "mean_token_accuracy": 0.23198643177747727,
      "num_tokens": 133073881.0,
      "step": 76720
    },
    {
      "entropy": 5.450905275344849,
      "epoch": 5.969461194320171,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002096382093288901,
      "loss": 4.7755,
      "mean_token_accuracy": 0.22857358455657958,
      "num_tokens": 133083852.0,
      "step": 76725
    },
    {
      "entropy": 5.415627956390381,
      "epoch": 5.969850223691889,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020961168944561888,
      "loss": 4.7533,
      "mean_token_accuracy": 0.23025662302970887,
      "num_tokens": 133093020.0,
      "step": 76730
    },
    {
      "entropy": 5.422198104858398,
      "epoch": 5.970239253063606,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00020958517055451125,
      "loss": 4.8085,
      "mean_token_accuracy": 0.23157608062028884,
      "num_tokens": 133102027.0,
      "step": 76735
    },
    {
      "entropy": 5.442161512374878,
      "epoch": 5.970628282435324,
      "grad_norm": 1.234375,
      "learning_rate": 0.00020955865265596946,
      "loss": 4.7047,
      "mean_token_accuracy": 0.22445060759782792,
      "num_tokens": 133111248.0,
      "step": 76740
    },
    {
      "entropy": 5.48015513420105,
      "epoch": 5.971017311807041,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000209532135750396,
      "loss": 4.7423,
      "mean_token_accuracy": 0.233413402736187,
      "num_tokens": 133120057.0,
      "step": 76745
    },
    {
      "entropy": 5.457960176467895,
      "epoch": 5.971406341178759,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020950561983819316,
      "loss": 4.7205,
      "mean_token_accuracy": 0.23223910182714463,
      "num_tokens": 133128444.0,
      "step": 76750
    },
    {
      "entropy": 5.355844259262085,
      "epoch": 5.971795370550477,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002094791049197633,
      "loss": 4.6649,
      "mean_token_accuracy": 0.23633837699890137,
      "num_tokens": 133136280.0,
      "step": 76755
    },
    {
      "entropy": 5.328870534896851,
      "epoch": 5.972184399922194,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020945259099550866,
      "loss": 4.6786,
      "mean_token_accuracy": 0.23170171231031417,
      "num_tokens": 133145190.0,
      "step": 76760
    },
    {
      "entropy": 5.341886806488037,
      "epoch": 5.972573429293911,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020942607806583174,
      "loss": 4.6255,
      "mean_token_accuracy": 0.23914496451616288,
      "num_tokens": 133153728.0,
      "step": 76765
    },
    {
      "entropy": 5.425568771362305,
      "epoch": 5.972962458665629,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020939956613113465,
      "loss": 4.7195,
      "mean_token_accuracy": 0.22827187329530715,
      "num_tokens": 133161661.0,
      "step": 76770
    },
    {
      "entropy": 5.382685041427612,
      "epoch": 5.973351488037347,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020937305519181976,
      "loss": 4.6334,
      "mean_token_accuracy": 0.23986938446760178,
      "num_tokens": 133169591.0,
      "step": 76775
    },
    {
      "entropy": 5.308913469314575,
      "epoch": 5.973740517409064,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020934654524828927,
      "loss": 4.5732,
      "mean_token_accuracy": 0.2455882653594017,
      "num_tokens": 133178385.0,
      "step": 76780
    },
    {
      "entropy": 5.412352895736694,
      "epoch": 5.974129546780782,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002093200363009455,
      "loss": 4.6682,
      "mean_token_accuracy": 0.2339445799589157,
      "num_tokens": 133186425.0,
      "step": 76785
    },
    {
      "entropy": 5.467498111724853,
      "epoch": 5.9745185761525,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020929352835019055,
      "loss": 4.8113,
      "mean_token_accuracy": 0.2229521855711937,
      "num_tokens": 133194378.0,
      "step": 76790
    },
    {
      "entropy": 5.425801753997803,
      "epoch": 5.974907605524217,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020926702139642677,
      "loss": 4.674,
      "mean_token_accuracy": 0.23077545017004014,
      "num_tokens": 133203113.0,
      "step": 76795
    },
    {
      "entropy": 5.3758197784423825,
      "epoch": 5.975296634895935,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020924051544005636,
      "loss": 4.6824,
      "mean_token_accuracy": 0.23808724731206893,
      "num_tokens": 133212311.0,
      "step": 76800
    },
    {
      "entropy": 5.378590393066406,
      "epoch": 5.975685664267652,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020921401048148137,
      "loss": 4.7026,
      "mean_token_accuracy": 0.2287675067782402,
      "num_tokens": 133220630.0,
      "step": 76805
    },
    {
      "entropy": 5.353485870361328,
      "epoch": 5.97607469363937,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020918750652110414,
      "loss": 4.6469,
      "mean_token_accuracy": 0.2384280100464821,
      "num_tokens": 133229095.0,
      "step": 76810
    },
    {
      "entropy": 5.375460052490235,
      "epoch": 5.976463723011087,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00020916100355932677,
      "loss": 4.7337,
      "mean_token_accuracy": 0.2304761603474617,
      "num_tokens": 133238303.0,
      "step": 76815
    },
    {
      "entropy": 5.512848043441773,
      "epoch": 5.976852752382805,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020913450159655133,
      "loss": 4.793,
      "mean_token_accuracy": 0.22664903104305267,
      "num_tokens": 133247380.0,
      "step": 76820
    },
    {
      "entropy": 5.492810344696045,
      "epoch": 5.977241781754523,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020910800063318004,
      "loss": 4.7434,
      "mean_token_accuracy": 0.2311605766415596,
      "num_tokens": 133256180.0,
      "step": 76825
    },
    {
      "entropy": 5.372041320800781,
      "epoch": 5.97763081112624,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002090815006696149,
      "loss": 4.6271,
      "mean_token_accuracy": 0.24168868213891984,
      "num_tokens": 133263776.0,
      "step": 76830
    },
    {
      "entropy": 5.396262741088867,
      "epoch": 5.978019840497957,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020905500170625808,
      "loss": 4.7167,
      "mean_token_accuracy": 0.23224983513355255,
      "num_tokens": 133272570.0,
      "step": 76835
    },
    {
      "entropy": 5.346427392959595,
      "epoch": 5.978408869869675,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00020902850374351168,
      "loss": 4.7114,
      "mean_token_accuracy": 0.23214575201272963,
      "num_tokens": 133282215.0,
      "step": 76840
    },
    {
      "entropy": 5.357039880752564,
      "epoch": 5.978797899241393,
      "grad_norm": 1.234375,
      "learning_rate": 0.00020900200678177766,
      "loss": 4.681,
      "mean_token_accuracy": 0.23055184185504912,
      "num_tokens": 133290723.0,
      "step": 76845
    },
    {
      "entropy": 5.491289854049683,
      "epoch": 5.97918692861311,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002089755108214582,
      "loss": 4.7916,
      "mean_token_accuracy": 0.2280765637755394,
      "num_tokens": 133300508.0,
      "step": 76850
    },
    {
      "entropy": 5.3928163051605225,
      "epoch": 5.979575957984828,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020894901586295529,
      "loss": 4.6299,
      "mean_token_accuracy": 0.23911733776330948,
      "num_tokens": 133309101.0,
      "step": 76855
    },
    {
      "entropy": 5.396538162231446,
      "epoch": 5.979964987356546,
      "grad_norm": 1.25,
      "learning_rate": 0.00020892252190667088,
      "loss": 4.7025,
      "mean_token_accuracy": 0.22850079834461212,
      "num_tokens": 133317586.0,
      "step": 76860
    },
    {
      "entropy": 5.365128374099731,
      "epoch": 5.980354016728263,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000208896028953007,
      "loss": 4.6528,
      "mean_token_accuracy": 0.24026117473840714,
      "num_tokens": 133326335.0,
      "step": 76865
    },
    {
      "entropy": 5.373180675506592,
      "epoch": 5.98074304609998,
      "grad_norm": 1.203125,
      "learning_rate": 0.00020886953700236576,
      "loss": 4.6607,
      "mean_token_accuracy": 0.2394168734550476,
      "num_tokens": 133335219.0,
      "step": 76870
    },
    {
      "entropy": 5.454305171966553,
      "epoch": 5.981132075471698,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020884304605514899,
      "loss": 4.7711,
      "mean_token_accuracy": 0.23274015337228776,
      "num_tokens": 133343620.0,
      "step": 76875
    },
    {
      "entropy": 5.360507965087891,
      "epoch": 5.981521104843416,
      "grad_norm": 1.25,
      "learning_rate": 0.00020881655611175865,
      "loss": 4.6647,
      "mean_token_accuracy": 0.23662790805101394,
      "num_tokens": 133351919.0,
      "step": 76880
    },
    {
      "entropy": 5.408464860916138,
      "epoch": 5.981910134215133,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002087900671725968,
      "loss": 4.6627,
      "mean_token_accuracy": 0.23448052108287812,
      "num_tokens": 133360437.0,
      "step": 76885
    },
    {
      "entropy": 5.355269956588745,
      "epoch": 5.982299163586851,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020876357923806527,
      "loss": 4.6311,
      "mean_token_accuracy": 0.2424807220697403,
      "num_tokens": 133368835.0,
      "step": 76890
    },
    {
      "entropy": 5.40399956703186,
      "epoch": 5.982688192958569,
      "grad_norm": 1.234375,
      "learning_rate": 0.00020873709230856608,
      "loss": 4.6983,
      "mean_token_accuracy": 0.23902825117111207,
      "num_tokens": 133377677.0,
      "step": 76895
    },
    {
      "entropy": 5.438564300537109,
      "epoch": 5.9830772223302855,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020871060638450101,
      "loss": 4.7355,
      "mean_token_accuracy": 0.2246285006403923,
      "num_tokens": 133385694.0,
      "step": 76900
    },
    {
      "entropy": 5.371127033233643,
      "epoch": 5.983466251702003,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002086841214662719,
      "loss": 4.6203,
      "mean_token_accuracy": 0.24992749840021133,
      "num_tokens": 133393917.0,
      "step": 76905
    },
    {
      "entropy": 5.502904844284058,
      "epoch": 5.983855281073721,
      "grad_norm": 1.25,
      "learning_rate": 0.00020865763755428075,
      "loss": 4.8552,
      "mean_token_accuracy": 0.22348951995372773,
      "num_tokens": 133402784.0,
      "step": 76910
    },
    {
      "entropy": 5.4192009449005125,
      "epoch": 5.984244310445439,
      "grad_norm": 1.25,
      "learning_rate": 0.00020863115464892935,
      "loss": 4.7198,
      "mean_token_accuracy": 0.2285832241177559,
      "num_tokens": 133411693.0,
      "step": 76915
    },
    {
      "entropy": 5.508400535583496,
      "epoch": 5.984633339817156,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00020860467275061952,
      "loss": 4.8303,
      "mean_token_accuracy": 0.22432257235050201,
      "num_tokens": 133420230.0,
      "step": 76920
    },
    {
      "entropy": 5.454340267181396,
      "epoch": 5.985022369188874,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020857819185975308,
      "loss": 4.681,
      "mean_token_accuracy": 0.24011431336402894,
      "num_tokens": 133429069.0,
      "step": 76925
    },
    {
      "entropy": 5.382234907150268,
      "epoch": 5.985411398560592,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020855171197673185,
      "loss": 4.6833,
      "mean_token_accuracy": 0.2321868807077408,
      "num_tokens": 133437133.0,
      "step": 76930
    },
    {
      "entropy": 5.494383811950684,
      "epoch": 5.985800427932309,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002085252331019577,
      "loss": 4.879,
      "mean_token_accuracy": 0.22386016994714736,
      "num_tokens": 133445575.0,
      "step": 76935
    },
    {
      "entropy": 5.500679588317871,
      "epoch": 5.986189457304026,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020849875523583234,
      "loss": 4.7436,
      "mean_token_accuracy": 0.23273541629314423,
      "num_tokens": 133454735.0,
      "step": 76940
    },
    {
      "entropy": 5.5073107242584225,
      "epoch": 5.986578486675744,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020847227837875743,
      "loss": 4.8172,
      "mean_token_accuracy": 0.22638939172029496,
      "num_tokens": 133463537.0,
      "step": 76945
    },
    {
      "entropy": 5.382311773300171,
      "epoch": 5.986967516047462,
      "grad_norm": 1.234375,
      "learning_rate": 0.00020844580253113486,
      "loss": 4.6491,
      "mean_token_accuracy": 0.23310198038816451,
      "num_tokens": 133472680.0,
      "step": 76950
    },
    {
      "entropy": 5.404653120040893,
      "epoch": 5.987356545419179,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002084193276933663,
      "loss": 4.7091,
      "mean_token_accuracy": 0.232947938144207,
      "num_tokens": 133480838.0,
      "step": 76955
    },
    {
      "entropy": 5.394693088531494,
      "epoch": 5.987745574790897,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002083928538658535,
      "loss": 4.6678,
      "mean_token_accuracy": 0.23421524912118913,
      "num_tokens": 133488942.0,
      "step": 76960
    },
    {
      "entropy": 5.40740556716919,
      "epoch": 5.988134604162614,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020836638104899808,
      "loss": 4.748,
      "mean_token_accuracy": 0.22634749561548234,
      "num_tokens": 133497852.0,
      "step": 76965
    },
    {
      "entropy": 5.516544771194458,
      "epoch": 5.9885236335343315,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002083399092432018,
      "loss": 4.8144,
      "mean_token_accuracy": 0.22548949271440505,
      "num_tokens": 133506383.0,
      "step": 76970
    },
    {
      "entropy": 5.504531812667847,
      "epoch": 5.988912662906049,
      "grad_norm": 1.234375,
      "learning_rate": 0.00020831343844886625,
      "loss": 4.8104,
      "mean_token_accuracy": 0.22740807831287385,
      "num_tokens": 133516045.0,
      "step": 76975
    },
    {
      "entropy": 5.350978183746338,
      "epoch": 5.989301692277767,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00020828696866639324,
      "loss": 4.6243,
      "mean_token_accuracy": 0.23697672486305238,
      "num_tokens": 133525076.0,
      "step": 76980
    },
    {
      "entropy": 5.423917293548584,
      "epoch": 5.989690721649485,
      "grad_norm": 1.171875,
      "learning_rate": 0.00020826049989618422,
      "loss": 4.7456,
      "mean_token_accuracy": 0.23019199669361115,
      "num_tokens": 133533986.0,
      "step": 76985
    },
    {
      "entropy": 5.4026953220367435,
      "epoch": 5.990079751021202,
      "grad_norm": 1.25,
      "learning_rate": 0.00020823403213864077,
      "loss": 4.6798,
      "mean_token_accuracy": 0.23516948521137238,
      "num_tokens": 133542344.0,
      "step": 76990
    },
    {
      "entropy": 5.410585975646972,
      "epoch": 5.99046878039292,
      "grad_norm": 1.234375,
      "learning_rate": 0.00020820756539416475,
      "loss": 4.6983,
      "mean_token_accuracy": 0.2353983297944069,
      "num_tokens": 133552045.0,
      "step": 76995
    },
    {
      "entropy": 5.432126855850219,
      "epoch": 5.990857809764638,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020818109966315757,
      "loss": 4.6737,
      "mean_token_accuracy": 0.23556697219610215,
      "num_tokens": 133560854.0,
      "step": 77000
    },
    {
      "entropy": 5.487018680572509,
      "epoch": 5.9912468391363545,
      "grad_norm": 1.21875,
      "learning_rate": 0.00020815463494602095,
      "loss": 4.8386,
      "mean_token_accuracy": 0.22504654675722122,
      "num_tokens": 133569920.0,
      "step": 77005
    },
    {
      "entropy": 5.378811597824097,
      "epoch": 5.991635868508072,
      "grad_norm": 1.25,
      "learning_rate": 0.00020812817124315626,
      "loss": 4.6416,
      "mean_token_accuracy": 0.24096276462078095,
      "num_tokens": 133578769.0,
      "step": 77010
    },
    {
      "entropy": 5.447092962265015,
      "epoch": 5.99202489787979,
      "grad_norm": 1.21875,
      "learning_rate": 0.00020810170855496514,
      "loss": 4.7812,
      "mean_token_accuracy": 0.23128073364496232,
      "num_tokens": 133586700.0,
      "step": 77015
    },
    {
      "entropy": 5.43983302116394,
      "epoch": 5.992413927251508,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020807524688184914,
      "loss": 4.7312,
      "mean_token_accuracy": 0.22870711982250214,
      "num_tokens": 133596137.0,
      "step": 77020
    },
    {
      "entropy": 5.428747177124023,
      "epoch": 5.992802956623225,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002080487862242097,
      "loss": 4.7532,
      "mean_token_accuracy": 0.22952639907598496,
      "num_tokens": 133604964.0,
      "step": 77025
    },
    {
      "entropy": 5.461842823028564,
      "epoch": 5.993191985994943,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002080223265824484,
      "loss": 4.7926,
      "mean_token_accuracy": 0.22888732254505156,
      "num_tokens": 133614277.0,
      "step": 77030
    },
    {
      "entropy": 5.36994161605835,
      "epoch": 5.99358101536666,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020799586795696668,
      "loss": 4.6176,
      "mean_token_accuracy": 0.24013999253511428,
      "num_tokens": 133623467.0,
      "step": 77035
    },
    {
      "entropy": 5.393704652786255,
      "epoch": 5.9939700447383775,
      "grad_norm": 1.234375,
      "learning_rate": 0.00020796941034816602,
      "loss": 4.6542,
      "mean_token_accuracy": 0.23727721124887466,
      "num_tokens": 133632695.0,
      "step": 77040
    },
    {
      "entropy": 5.4600080966949465,
      "epoch": 5.994359074110095,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002079429537564479,
      "loss": 4.6937,
      "mean_token_accuracy": 0.23909743130207062,
      "num_tokens": 133641158.0,
      "step": 77045
    },
    {
      "entropy": 5.350060558319091,
      "epoch": 5.994748103481813,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020791649818221365,
      "loss": 4.6488,
      "mean_token_accuracy": 0.23353536128997804,
      "num_tokens": 133650143.0,
      "step": 77050
    },
    {
      "entropy": 5.369654560089112,
      "epoch": 5.995137132853531,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020789004362586477,
      "loss": 4.6546,
      "mean_token_accuracy": 0.2333401322364807,
      "num_tokens": 133658483.0,
      "step": 77055
    },
    {
      "entropy": 5.417748689651489,
      "epoch": 5.995526162225248,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002078635900878027,
      "loss": 4.7246,
      "mean_token_accuracy": 0.23428456038236617,
      "num_tokens": 133667125.0,
      "step": 77060
    },
    {
      "entropy": 5.4392753601074215,
      "epoch": 5.995915191596966,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020783713756842876,
      "loss": 4.7031,
      "mean_token_accuracy": 0.23410880863666533,
      "num_tokens": 133675445.0,
      "step": 77065
    },
    {
      "entropy": 5.351079654693604,
      "epoch": 5.996304220968683,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002078106860681444,
      "loss": 4.698,
      "mean_token_accuracy": 0.23442161828279495,
      "num_tokens": 133684153.0,
      "step": 77070
    },
    {
      "entropy": 5.4541308879852295,
      "epoch": 5.9966932503404005,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020778423558735087,
      "loss": 4.7435,
      "mean_token_accuracy": 0.23942455947399138,
      "num_tokens": 133692635.0,
      "step": 77075
    },
    {
      "entropy": 5.410097265243531,
      "epoch": 5.997082279712118,
      "grad_norm": 1.25,
      "learning_rate": 0.0002077577861264497,
      "loss": 4.6472,
      "mean_token_accuracy": 0.23771531879901886,
      "num_tokens": 133700900.0,
      "step": 77080
    },
    {
      "entropy": 5.369327259063721,
      "epoch": 5.997471309083836,
      "grad_norm": 1.234375,
      "learning_rate": 0.00020773133768584208,
      "loss": 4.6812,
      "mean_token_accuracy": 0.24013512432575226,
      "num_tokens": 133710020.0,
      "step": 77085
    },
    {
      "entropy": 5.383687973022461,
      "epoch": 5.9978603384555536,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020770489026592927,
      "loss": 4.7596,
      "mean_token_accuracy": 0.23252297937870026,
      "num_tokens": 133719195.0,
      "step": 77090
    },
    {
      "entropy": 5.423875761032105,
      "epoch": 5.998249367827271,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020767844386711265,
      "loss": 4.7006,
      "mean_token_accuracy": 0.23653915971517564,
      "num_tokens": 133728378.0,
      "step": 77095
    },
    {
      "entropy": 5.449291133880616,
      "epoch": 5.998638397198988,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002076519984897935,
      "loss": 4.7074,
      "mean_token_accuracy": 0.23889687806367874,
      "num_tokens": 133736788.0,
      "step": 77100
    },
    {
      "entropy": 5.466154766082764,
      "epoch": 5.999027426570706,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020762555413437306,
      "loss": 4.7934,
      "mean_token_accuracy": 0.22288607954978942,
      "num_tokens": 133745236.0,
      "step": 77105
    },
    {
      "entropy": 5.414954757690429,
      "epoch": 5.9994164559424235,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002075991108012526,
      "loss": 4.6903,
      "mean_token_accuracy": 0.23577385991811753,
      "num_tokens": 133753464.0,
      "step": 77110
    },
    {
      "entropy": 5.495847940444946,
      "epoch": 5.999805485314141,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002075726684908334,
      "loss": 4.7714,
      "mean_token_accuracy": 0.2259894922375679,
      "num_tokens": 133762280.0,
      "step": 77115
    },
    {
      "entropy": 5.534921222262913,
      "epoch": 6.000155611748687,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020754622720351663,
      "loss": 4.7676,
      "mean_token_accuracy": 0.2317834115690655,
      "num_tokens": 133769307.0,
      "step": 77120
    },
    {
      "entropy": 5.438278436660767,
      "epoch": 6.000544641120404,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020751978693970335,
      "loss": 4.6478,
      "mean_token_accuracy": 0.248997662961483,
      "num_tokens": 133778219.0,
      "step": 77125
    },
    {
      "entropy": 5.331284666061402,
      "epoch": 6.000933670492122,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020749334769979506,
      "loss": 4.5485,
      "mean_token_accuracy": 0.24521005749702454,
      "num_tokens": 133787213.0,
      "step": 77130
    },
    {
      "entropy": 5.431422853469849,
      "epoch": 6.00132269986384,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020746690948419272,
      "loss": 4.6674,
      "mean_token_accuracy": 0.2399682432413101,
      "num_tokens": 133795403.0,
      "step": 77135
    },
    {
      "entropy": 5.370991468429565,
      "epoch": 6.0017117292355575,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00020744047229329755,
      "loss": 4.6549,
      "mean_token_accuracy": 0.23694311380386351,
      "num_tokens": 133804656.0,
      "step": 77140
    },
    {
      "entropy": 5.34413161277771,
      "epoch": 6.002100758607275,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002074140361275107,
      "loss": 4.5194,
      "mean_token_accuracy": 0.24938977807760238,
      "num_tokens": 133812172.0,
      "step": 77145
    },
    {
      "entropy": 5.311563873291016,
      "epoch": 6.002489787978992,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020738760098723324,
      "loss": 4.6054,
      "mean_token_accuracy": 0.2485032483935356,
      "num_tokens": 133820635.0,
      "step": 77150
    },
    {
      "entropy": 5.428660440444946,
      "epoch": 6.00287881735071,
      "grad_norm": 1.1875,
      "learning_rate": 0.00020736116687286634,
      "loss": 4.6787,
      "mean_token_accuracy": 0.236958110332489,
      "num_tokens": 133830056.0,
      "step": 77155
    },
    {
      "entropy": 5.3535058975219725,
      "epoch": 6.003267846722427,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020733473378481112,
      "loss": 4.5657,
      "mean_token_accuracy": 0.24297693073749543,
      "num_tokens": 133838688.0,
      "step": 77160
    },
    {
      "entropy": 5.362388658523559,
      "epoch": 6.003656876094145,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002073083017234685,
      "loss": 4.5706,
      "mean_token_accuracy": 0.2524039775133133,
      "num_tokens": 133847525.0,
      "step": 77165
    },
    {
      "entropy": 5.343952178955078,
      "epoch": 6.004045905465863,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020728187068923975,
      "loss": 4.666,
      "mean_token_accuracy": 0.23959900587797164,
      "num_tokens": 133856186.0,
      "step": 77170
    },
    {
      "entropy": 5.302262639999389,
      "epoch": 6.0044349348375805,
      "grad_norm": 1.25,
      "learning_rate": 0.00020725544068252584,
      "loss": 4.5156,
      "mean_token_accuracy": 0.25107404589653015,
      "num_tokens": 133864509.0,
      "step": 77175
    },
    {
      "entropy": 5.379871511459351,
      "epoch": 6.004823964209298,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002072290117037277,
      "loss": 4.6181,
      "mean_token_accuracy": 0.24093781560659408,
      "num_tokens": 133873285.0,
      "step": 77180
    },
    {
      "entropy": 5.455144166946411,
      "epoch": 6.005212993581015,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00020720258375324653,
      "loss": 4.7144,
      "mean_token_accuracy": 0.2286084219813347,
      "num_tokens": 133882053.0,
      "step": 77185
    },
    {
      "entropy": 5.429717397689819,
      "epoch": 6.005602022952733,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002071761568314831,
      "loss": 4.7057,
      "mean_token_accuracy": 0.23580383956432344,
      "num_tokens": 133891048.0,
      "step": 77190
    },
    {
      "entropy": 5.367566728591919,
      "epoch": 6.00599105232445,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002071497309388387,
      "loss": 4.6492,
      "mean_token_accuracy": 0.24195114076137542,
      "num_tokens": 133899689.0,
      "step": 77195
    },
    {
      "entropy": 5.487112283706665,
      "epoch": 6.006380081696168,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020712330607571407,
      "loss": 4.6804,
      "mean_token_accuracy": 0.23912966698408128,
      "num_tokens": 133907874.0,
      "step": 77200
    },
    {
      "entropy": 5.347855567932129,
      "epoch": 6.006769111067886,
      "grad_norm": 1.125,
      "learning_rate": 0.00020709688224251029,
      "loss": 4.5642,
      "mean_token_accuracy": 0.24595350474119188,
      "num_tokens": 133917124.0,
      "step": 77205
    },
    {
      "entropy": 5.467001867294312,
      "epoch": 6.0071581404396035,
      "grad_norm": 1.5,
      "learning_rate": 0.00020707045943962817,
      "loss": 4.7333,
      "mean_token_accuracy": 0.23261107504367828,
      "num_tokens": 133926381.0,
      "step": 77210
    },
    {
      "entropy": 5.482918214797974,
      "epoch": 6.007547169811321,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002070440376674687,
      "loss": 4.7332,
      "mean_token_accuracy": 0.23646429032087327,
      "num_tokens": 133934459.0,
      "step": 77215
    },
    {
      "entropy": 5.429720306396485,
      "epoch": 6.007936199183038,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020701761692643283,
      "loss": 4.6425,
      "mean_token_accuracy": 0.24333780407905578,
      "num_tokens": 133943216.0,
      "step": 77220
    },
    {
      "entropy": 5.342742586135865,
      "epoch": 6.008325228554756,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020699119721692138,
      "loss": 4.5096,
      "mean_token_accuracy": 0.25699692517518996,
      "num_tokens": 133951762.0,
      "step": 77225
    },
    {
      "entropy": 5.3234082698822025,
      "epoch": 6.008714257926473,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00020696477853933531,
      "loss": 4.579,
      "mean_token_accuracy": 0.24454302787780763,
      "num_tokens": 133961107.0,
      "step": 77230
    },
    {
      "entropy": 5.440021276473999,
      "epoch": 6.009103287298191,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002069383608940754,
      "loss": 4.7446,
      "mean_token_accuracy": 0.22733221352100372,
      "num_tokens": 133969485.0,
      "step": 77235
    },
    {
      "entropy": 5.376610469818115,
      "epoch": 6.009492316669909,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020691194428154258,
      "loss": 4.6283,
      "mean_token_accuracy": 0.24553134739398957,
      "num_tokens": 133977528.0,
      "step": 77240
    },
    {
      "entropy": 5.453963327407837,
      "epoch": 6.0098813460416265,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020688552870213762,
      "loss": 4.7329,
      "mean_token_accuracy": 0.2372319296002388,
      "num_tokens": 133986205.0,
      "step": 77245
    },
    {
      "entropy": 5.397297334671021,
      "epoch": 6.010270375413343,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002068591141562614,
      "loss": 4.609,
      "mean_token_accuracy": 0.24004323929548263,
      "num_tokens": 133994843.0,
      "step": 77250
    },
    {
      "entropy": 5.4881233215332035,
      "epoch": 6.010659404785061,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020683270064431454,
      "loss": 4.8094,
      "mean_token_accuracy": 0.2298088014125824,
      "num_tokens": 134004005.0,
      "step": 77255
    },
    {
      "entropy": 5.3949610710144045,
      "epoch": 6.011048434156779,
      "grad_norm": 1.296875,
      "learning_rate": 0.000206806288166698,
      "loss": 4.6679,
      "mean_token_accuracy": 0.23687061965465545,
      "num_tokens": 134012765.0,
      "step": 77260
    },
    {
      "entropy": 5.455898714065552,
      "epoch": 6.011437463528496,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020677987672381248,
      "loss": 4.6581,
      "mean_token_accuracy": 0.23155538886785507,
      "num_tokens": 134021397.0,
      "step": 77265
    },
    {
      "entropy": 5.454668283462524,
      "epoch": 6.011826492900214,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002067534663160588,
      "loss": 4.6687,
      "mean_token_accuracy": 0.23130126148462296,
      "num_tokens": 134029803.0,
      "step": 77270
    },
    {
      "entropy": 5.365855598449707,
      "epoch": 6.012215522271932,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020672705694383759,
      "loss": 4.5896,
      "mean_token_accuracy": 0.25146892219781875,
      "num_tokens": 134038804.0,
      "step": 77275
    },
    {
      "entropy": 5.3507527828216555,
      "epoch": 6.0126045516436495,
      "grad_norm": 1.234375,
      "learning_rate": 0.00020670064860754973,
      "loss": 4.5786,
      "mean_token_accuracy": 0.24565448462963105,
      "num_tokens": 134047642.0,
      "step": 77280
    },
    {
      "entropy": 5.3595164775848385,
      "epoch": 6.012993581015366,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002066742413075957,
      "loss": 4.598,
      "mean_token_accuracy": 0.2372659847140312,
      "num_tokens": 134055968.0,
      "step": 77285
    },
    {
      "entropy": 5.366675615310669,
      "epoch": 6.013382610387084,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020664783504437635,
      "loss": 4.6129,
      "mean_token_accuracy": 0.2429476037621498,
      "num_tokens": 134064236.0,
      "step": 77290
    },
    {
      "entropy": 5.411853790283203,
      "epoch": 6.013771639758802,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002066214298182923,
      "loss": 4.7298,
      "mean_token_accuracy": 0.2253555178642273,
      "num_tokens": 134072713.0,
      "step": 77295
    },
    {
      "entropy": 5.330868101119995,
      "epoch": 6.014160669130519,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020659502562974424,
      "loss": 4.5553,
      "mean_token_accuracy": 0.2555545374751091,
      "num_tokens": 134081889.0,
      "step": 77300
    },
    {
      "entropy": 5.375437927246094,
      "epoch": 6.014549698502237,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020656862247913277,
      "loss": 4.5999,
      "mean_token_accuracy": 0.24149583727121354,
      "num_tokens": 134090163.0,
      "step": 77305
    },
    {
      "entropy": 5.399028539657593,
      "epoch": 6.014938727873955,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020654222036685854,
      "loss": 4.6409,
      "mean_token_accuracy": 0.24319502860307693,
      "num_tokens": 134099421.0,
      "step": 77310
    },
    {
      "entropy": 5.453107690811157,
      "epoch": 6.0153277572456725,
      "grad_norm": 1.171875,
      "learning_rate": 0.00020651581929332215,
      "loss": 4.7006,
      "mean_token_accuracy": 0.23674021512269974,
      "num_tokens": 134108132.0,
      "step": 77315
    },
    {
      "entropy": 5.3733922958374025,
      "epoch": 6.015716786617389,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020648941925892418,
      "loss": 4.6844,
      "mean_token_accuracy": 0.23622374534606932,
      "num_tokens": 134116292.0,
      "step": 77320
    },
    {
      "entropy": 5.420813083648682,
      "epoch": 6.016105815989107,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020646302026406516,
      "loss": 4.6797,
      "mean_token_accuracy": 0.23304443061351776,
      "num_tokens": 134124261.0,
      "step": 77325
    },
    {
      "entropy": 5.35989465713501,
      "epoch": 6.016494845360825,
      "grad_norm": 1.375,
      "learning_rate": 0.00020643662230914584,
      "loss": 4.554,
      "mean_token_accuracy": 0.24283139258623124,
      "num_tokens": 134131853.0,
      "step": 77330
    },
    {
      "entropy": 5.364595031738281,
      "epoch": 6.016883874732542,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020641022539456661,
      "loss": 4.6474,
      "mean_token_accuracy": 0.24322579652071,
      "num_tokens": 134141164.0,
      "step": 77335
    },
    {
      "entropy": 5.2948705673217775,
      "epoch": 6.01727290410426,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020638382952072805,
      "loss": 4.5691,
      "mean_token_accuracy": 0.24413906633853913,
      "num_tokens": 134150145.0,
      "step": 77340
    },
    {
      "entropy": 5.4313396453857425,
      "epoch": 6.017661933475978,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020635743468803064,
      "loss": 4.6716,
      "mean_token_accuracy": 0.23640408366918564,
      "num_tokens": 134158705.0,
      "step": 77345
    },
    {
      "entropy": 5.415476369857788,
      "epoch": 6.018050962847695,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002063310408968749,
      "loss": 4.6572,
      "mean_token_accuracy": 0.2484483614563942,
      "num_tokens": 134166852.0,
      "step": 77350
    },
    {
      "entropy": 5.355245161056518,
      "epoch": 6.018439992219412,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002063046481476613,
      "loss": 4.603,
      "mean_token_accuracy": 0.24671286940574647,
      "num_tokens": 134175607.0,
      "step": 77355
    },
    {
      "entropy": 5.3549168586730955,
      "epoch": 6.01882902159113,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002062782564407904,
      "loss": 4.5825,
      "mean_token_accuracy": 0.25098644345998766,
      "num_tokens": 134184958.0,
      "step": 77360
    },
    {
      "entropy": 5.369450473785401,
      "epoch": 6.019218050962848,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020625186577666256,
      "loss": 4.5672,
      "mean_token_accuracy": 0.2531695455312729,
      "num_tokens": 134193250.0,
      "step": 77365
    },
    {
      "entropy": 5.434607410430909,
      "epoch": 6.019607080334565,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00020622547615567816,
      "loss": 4.734,
      "mean_token_accuracy": 0.23089329600334169,
      "num_tokens": 134202459.0,
      "step": 77370
    },
    {
      "entropy": 5.3298104763031,
      "epoch": 6.019996109706283,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002061990875782378,
      "loss": 4.6206,
      "mean_token_accuracy": 0.2451135203242302,
      "num_tokens": 134211073.0,
      "step": 77375
    },
    {
      "entropy": 5.42009916305542,
      "epoch": 6.020385139078001,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020617270004474165,
      "loss": 4.6818,
      "mean_token_accuracy": 0.23192584961652757,
      "num_tokens": 134220633.0,
      "step": 77380
    },
    {
      "entropy": 5.432529067993164,
      "epoch": 6.020774168449718,
      "grad_norm": 1.1875,
      "learning_rate": 0.00020614631355559033,
      "loss": 4.6239,
      "mean_token_accuracy": 0.24781686812639236,
      "num_tokens": 134229574.0,
      "step": 77385
    },
    {
      "entropy": 5.340813302993775,
      "epoch": 6.021163197821435,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000206119928111184,
      "loss": 4.5804,
      "mean_token_accuracy": 0.234741273522377,
      "num_tokens": 134238927.0,
      "step": 77390
    },
    {
      "entropy": 5.405009460449219,
      "epoch": 6.021552227193153,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020609354371192324,
      "loss": 4.6577,
      "mean_token_accuracy": 0.24255121052265166,
      "num_tokens": 134247635.0,
      "step": 77395
    },
    {
      "entropy": 5.429850912094116,
      "epoch": 6.021941256564871,
      "grad_norm": 1.21875,
      "learning_rate": 0.00020606716035820823,
      "loss": 4.6017,
      "mean_token_accuracy": 0.24273055791854858,
      "num_tokens": 134255973.0,
      "step": 77400
    },
    {
      "entropy": 5.338968992233276,
      "epoch": 6.022330285936588,
      "grad_norm": 1.171875,
      "learning_rate": 0.00020604077805043932,
      "loss": 4.6162,
      "mean_token_accuracy": 0.24828810691833497,
      "num_tokens": 134264482.0,
      "step": 77405
    },
    {
      "entropy": 5.4235467433929445,
      "epoch": 6.022719315308306,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002060143967890169,
      "loss": 4.7943,
      "mean_token_accuracy": 0.23333940505981446,
      "num_tokens": 134273849.0,
      "step": 77410
    },
    {
      "entropy": 5.391621017456055,
      "epoch": 6.023108344680024,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020598801657434118,
      "loss": 4.6676,
      "mean_token_accuracy": 0.23804791569709777,
      "num_tokens": 134282421.0,
      "step": 77415
    },
    {
      "entropy": 5.348129606246948,
      "epoch": 6.023497374051741,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020596163740681245,
      "loss": 4.5869,
      "mean_token_accuracy": 0.24492659866809846,
      "num_tokens": 134290832.0,
      "step": 77420
    },
    {
      "entropy": 5.415919637680053,
      "epoch": 6.023886403423458,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000205935259286831,
      "loss": 4.7072,
      "mean_token_accuracy": 0.23694663792848586,
      "num_tokens": 134298762.0,
      "step": 77425
    },
    {
      "entropy": 5.461903095245361,
      "epoch": 6.024275432795176,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002059088822147971,
      "loss": 4.7066,
      "mean_token_accuracy": 0.23211790174245833,
      "num_tokens": 134307058.0,
      "step": 77430
    },
    {
      "entropy": 5.432510805130005,
      "epoch": 6.024664462166894,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020588250619111087,
      "loss": 4.6324,
      "mean_token_accuracy": 0.24352308511734008,
      "num_tokens": 134316081.0,
      "step": 77435
    },
    {
      "entropy": 5.409109973907471,
      "epoch": 6.025053491538611,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002058561312161727,
      "loss": 4.6453,
      "mean_token_accuracy": 0.23787467628717424,
      "num_tokens": 134325033.0,
      "step": 77440
    },
    {
      "entropy": 5.53907356262207,
      "epoch": 6.025442520910329,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020582975729038262,
      "loss": 4.7793,
      "mean_token_accuracy": 0.23341430723667145,
      "num_tokens": 134333954.0,
      "step": 77445
    },
    {
      "entropy": 5.482991743087768,
      "epoch": 6.025831550282046,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020580338441414088,
      "loss": 4.7381,
      "mean_token_accuracy": 0.23181652128696442,
      "num_tokens": 134343686.0,
      "step": 77450
    },
    {
      "entropy": 5.496955823898316,
      "epoch": 6.026220579653764,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020577701258784754,
      "loss": 4.6876,
      "mean_token_accuracy": 0.23586950004100798,
      "num_tokens": 134351977.0,
      "step": 77455
    },
    {
      "entropy": 5.388966083526611,
      "epoch": 6.026609609025481,
      "grad_norm": 1.1875,
      "learning_rate": 0.00020575064181190295,
      "loss": 4.6554,
      "mean_token_accuracy": 0.23672526925802231,
      "num_tokens": 134360809.0,
      "step": 77460
    },
    {
      "entropy": 5.338500118255615,
      "epoch": 6.026998638397199,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020572427208670713,
      "loss": 4.5703,
      "mean_token_accuracy": 0.23825763314962387,
      "num_tokens": 134370027.0,
      "step": 77465
    },
    {
      "entropy": 5.3764149188995365,
      "epoch": 6.027387667768917,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020569790341266027,
      "loss": 4.6011,
      "mean_token_accuracy": 0.24512943774461746,
      "num_tokens": 134378081.0,
      "step": 77470
    },
    {
      "entropy": 5.493074035644531,
      "epoch": 6.027776697140634,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002056715357901624,
      "loss": 4.8286,
      "mean_token_accuracy": 0.22392419576644898,
      "num_tokens": 134386529.0,
      "step": 77475
    },
    {
      "entropy": 5.41727499961853,
      "epoch": 6.028165726512352,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020564516921961362,
      "loss": 4.6641,
      "mean_token_accuracy": 0.23065652698278427,
      "num_tokens": 134394899.0,
      "step": 77480
    },
    {
      "entropy": 5.457625246047973,
      "epoch": 6.028554755884069,
      "grad_norm": 1.328125,
      "learning_rate": 0.000205618803701414,
      "loss": 4.639,
      "mean_token_accuracy": 0.24349479377269745,
      "num_tokens": 134403724.0,
      "step": 77485
    },
    {
      "entropy": 5.444463872909546,
      "epoch": 6.028943785255787,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002055924392359636,
      "loss": 4.7205,
      "mean_token_accuracy": 0.2358221247792244,
      "num_tokens": 134412491.0,
      "step": 77490
    },
    {
      "entropy": 5.359148979187012,
      "epoch": 6.029332814627504,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002055660758236625,
      "loss": 4.6576,
      "mean_token_accuracy": 0.23717488944530488,
      "num_tokens": 134421232.0,
      "step": 77495
    },
    {
      "entropy": 5.312301158905029,
      "epoch": 6.029721843999222,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020553971346491063,
      "loss": 4.5496,
      "mean_token_accuracy": 0.24213007241487502,
      "num_tokens": 134429976.0,
      "step": 77500
    },
    {
      "entropy": 5.413352108001709,
      "epoch": 6.03011087337094,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020551335216010804,
      "loss": 4.6575,
      "mean_token_accuracy": 0.23858368694782256,
      "num_tokens": 134439214.0,
      "step": 77505
    },
    {
      "entropy": 5.436894369125366,
      "epoch": 6.030499902742657,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002054869919096548,
      "loss": 4.6705,
      "mean_token_accuracy": 0.23456440716981888,
      "num_tokens": 134447392.0,
      "step": 77510
    },
    {
      "entropy": 5.358858728408814,
      "epoch": 6.030888932114375,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020546063271395076,
      "loss": 4.522,
      "mean_token_accuracy": 0.2514137804508209,
      "num_tokens": 134454825.0,
      "step": 77515
    },
    {
      "entropy": 5.434662294387818,
      "epoch": 6.031277961486092,
      "grad_norm": 1.25,
      "learning_rate": 0.00020543427457339592,
      "loss": 4.7025,
      "mean_token_accuracy": 0.23259417414665223,
      "num_tokens": 134463195.0,
      "step": 77520
    },
    {
      "entropy": 5.344886398315429,
      "epoch": 6.0316669908578096,
      "grad_norm": 1.1875,
      "learning_rate": 0.00020540791748839028,
      "loss": 4.5801,
      "mean_token_accuracy": 0.24360698610544204,
      "num_tokens": 134471714.0,
      "step": 77525
    },
    {
      "entropy": 5.457853746414185,
      "epoch": 6.032056020229527,
      "grad_norm": 1.25,
      "learning_rate": 0.00020538156145933373,
      "loss": 4.6745,
      "mean_token_accuracy": 0.23651942759752273,
      "num_tokens": 134479952.0,
      "step": 77530
    },
    {
      "entropy": 5.393327617645264,
      "epoch": 6.032445049601245,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020535520648662614,
      "loss": 4.6552,
      "mean_token_accuracy": 0.24701514542102815,
      "num_tokens": 134488776.0,
      "step": 77535
    },
    {
      "entropy": 5.324007892608643,
      "epoch": 6.032834078972963,
      "grad_norm": 1.375,
      "learning_rate": 0.00020532885257066748,
      "loss": 4.6066,
      "mean_token_accuracy": 0.23661684691905976,
      "num_tokens": 134497493.0,
      "step": 77540
    },
    {
      "entropy": 5.431989336013794,
      "epoch": 6.03322310834468,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020530249971185754,
      "loss": 4.7552,
      "mean_token_accuracy": 0.23398895412683487,
      "num_tokens": 134507491.0,
      "step": 77545
    },
    {
      "entropy": 5.5316530704498295,
      "epoch": 6.033612137716397,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020527614791059635,
      "loss": 4.7337,
      "mean_token_accuracy": 0.2320709139108658,
      "num_tokens": 134515561.0,
      "step": 77550
    },
    {
      "entropy": 5.343206167221069,
      "epoch": 6.034001167088115,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020524979716728353,
      "loss": 4.5463,
      "mean_token_accuracy": 0.25084415823221207,
      "num_tokens": 134524160.0,
      "step": 77555
    },
    {
      "entropy": 5.403464031219483,
      "epoch": 6.0343901964598325,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020522344748231903,
      "loss": 4.7009,
      "mean_token_accuracy": 0.24279077649116515,
      "num_tokens": 134533045.0,
      "step": 77560
    },
    {
      "entropy": 5.406049585342407,
      "epoch": 6.03477922583155,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020519709885610267,
      "loss": 4.658,
      "mean_token_accuracy": 0.23879704028367996,
      "num_tokens": 134541254.0,
      "step": 77565
    },
    {
      "entropy": 5.349804544448853,
      "epoch": 6.035168255203268,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020517075128903423,
      "loss": 4.5813,
      "mean_token_accuracy": 0.2480908066034317,
      "num_tokens": 134549873.0,
      "step": 77570
    },
    {
      "entropy": 5.496742010116577,
      "epoch": 6.035557284574986,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020514440478151352,
      "loss": 4.7539,
      "mean_token_accuracy": 0.23577855676412582,
      "num_tokens": 134558971.0,
      "step": 77575
    },
    {
      "entropy": 5.442083406448364,
      "epoch": 6.035946313946703,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020511805933394022,
      "loss": 4.7215,
      "mean_token_accuracy": 0.23782805353403091,
      "num_tokens": 134567675.0,
      "step": 77580
    },
    {
      "entropy": 5.440418338775634,
      "epoch": 6.03633534331842,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020509171494671418,
      "loss": 4.7577,
      "mean_token_accuracy": 0.23647967278957366,
      "num_tokens": 134575737.0,
      "step": 77585
    },
    {
      "entropy": 5.444126033782959,
      "epoch": 6.036724372690138,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020506537162023503,
      "loss": 4.6956,
      "mean_token_accuracy": 0.23379986733198166,
      "num_tokens": 134585098.0,
      "step": 77590
    },
    {
      "entropy": 5.310971879959107,
      "epoch": 6.0371134020618555,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020503902935490266,
      "loss": 4.5183,
      "mean_token_accuracy": 0.2511457070708275,
      "num_tokens": 134594125.0,
      "step": 77595
    },
    {
      "entropy": 5.334113883972168,
      "epoch": 6.037502431433573,
      "grad_norm": 1.171875,
      "learning_rate": 0.00020501268815111657,
      "loss": 4.6063,
      "mean_token_accuracy": 0.24488144963979722,
      "num_tokens": 134602993.0,
      "step": 77600
    },
    {
      "entropy": 5.346925544738769,
      "epoch": 6.037891460805291,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020498634800927663,
      "loss": 4.5693,
      "mean_token_accuracy": 0.2447124019265175,
      "num_tokens": 134611363.0,
      "step": 77605
    },
    {
      "entropy": 5.386383152008056,
      "epoch": 6.038280490177009,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020496000892978233,
      "loss": 4.7031,
      "mean_token_accuracy": 0.23290566653013228,
      "num_tokens": 134619785.0,
      "step": 77610
    },
    {
      "entropy": 5.399725341796875,
      "epoch": 6.038669519548726,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002049336709130335,
      "loss": 4.637,
      "mean_token_accuracy": 0.2401324227452278,
      "num_tokens": 134627753.0,
      "step": 77615
    },
    {
      "entropy": 5.406328916549683,
      "epoch": 6.039058548920443,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020490733395942969,
      "loss": 4.6806,
      "mean_token_accuracy": 0.24063858091831208,
      "num_tokens": 134636302.0,
      "step": 77620
    },
    {
      "entropy": 5.286217546463012,
      "epoch": 6.039447578292161,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020488099806937043,
      "loss": 4.513,
      "mean_token_accuracy": 0.255272401869297,
      "num_tokens": 134645398.0,
      "step": 77625
    },
    {
      "entropy": 5.417007493972778,
      "epoch": 6.0398366076638785,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002048546632432555,
      "loss": 4.5983,
      "mean_token_accuracy": 0.24469622522592543,
      "num_tokens": 134653751.0,
      "step": 77630
    },
    {
      "entropy": 5.344275712966919,
      "epoch": 6.040225637035596,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020482832948148444,
      "loss": 4.6186,
      "mean_token_accuracy": 0.23836632519960405,
      "num_tokens": 134662226.0,
      "step": 77635
    },
    {
      "entropy": 5.275543260574341,
      "epoch": 6.040614666407314,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002048019967844567,
      "loss": 4.5993,
      "mean_token_accuracy": 0.24343917965888978,
      "num_tokens": 134670108.0,
      "step": 77640
    },
    {
      "entropy": 5.412452697753906,
      "epoch": 6.041003695779032,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020477566515257197,
      "loss": 4.6759,
      "mean_token_accuracy": 0.23912333250045775,
      "num_tokens": 134678521.0,
      "step": 77645
    },
    {
      "entropy": 5.42821192741394,
      "epoch": 6.041392725150749,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020474933458622974,
      "loss": 4.6726,
      "mean_token_accuracy": 0.23464707136154175,
      "num_tokens": 134687504.0,
      "step": 77650
    },
    {
      "entropy": 5.369917297363282,
      "epoch": 6.041781754522466,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020472300508582946,
      "loss": 4.5893,
      "mean_token_accuracy": 0.2439635291695595,
      "num_tokens": 134696119.0,
      "step": 77655
    },
    {
      "entropy": 5.381813001632691,
      "epoch": 6.042170783894184,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020469667665177084,
      "loss": 4.6071,
      "mean_token_accuracy": 0.2444898694753647,
      "num_tokens": 134704497.0,
      "step": 77660
    },
    {
      "entropy": 5.36807861328125,
      "epoch": 6.0425598132659015,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020467034928445323,
      "loss": 4.6621,
      "mean_token_accuracy": 0.23884664922952653,
      "num_tokens": 134713108.0,
      "step": 77665
    },
    {
      "entropy": 5.4359273433685305,
      "epoch": 6.042948842637619,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020464402298427608,
      "loss": 4.7173,
      "mean_token_accuracy": 0.23718032687902452,
      "num_tokens": 134721747.0,
      "step": 77670
    },
    {
      "entropy": 5.426615810394287,
      "epoch": 6.043337872009337,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020461769775163897,
      "loss": 4.6956,
      "mean_token_accuracy": 0.2366696611046791,
      "num_tokens": 134730595.0,
      "step": 77675
    },
    {
      "entropy": 5.364318895339966,
      "epoch": 6.043726901381055,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020459137358694126,
      "loss": 4.5714,
      "mean_token_accuracy": 0.24899971038103103,
      "num_tokens": 134738359.0,
      "step": 77680
    },
    {
      "entropy": 5.365635013580322,
      "epoch": 6.044115930752771,
      "grad_norm": 1.234375,
      "learning_rate": 0.00020456505049058234,
      "loss": 4.6476,
      "mean_token_accuracy": 0.24341872334480286,
      "num_tokens": 134747427.0,
      "step": 77685
    },
    {
      "entropy": 5.373874282836914,
      "epoch": 6.044504960124489,
      "grad_norm": 1.375,
      "learning_rate": 0.00020453872846296168,
      "loss": 4.6803,
      "mean_token_accuracy": 0.23554707914590836,
      "num_tokens": 134756177.0,
      "step": 77690
    },
    {
      "entropy": 5.430577182769776,
      "epoch": 6.044893989496207,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020451240750447868,
      "loss": 4.6787,
      "mean_token_accuracy": 0.24205006659030914,
      "num_tokens": 134764715.0,
      "step": 77695
    },
    {
      "entropy": 5.313101577758789,
      "epoch": 6.0452830188679245,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020448608761553273,
      "loss": 4.5327,
      "mean_token_accuracy": 0.24447813630104065,
      "num_tokens": 134773934.0,
      "step": 77700
    },
    {
      "entropy": 5.414363622665405,
      "epoch": 6.045672048239642,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020445976879652312,
      "loss": 4.7215,
      "mean_token_accuracy": 0.23208819329738617,
      "num_tokens": 134781957.0,
      "step": 77705
    },
    {
      "entropy": 5.377001047134399,
      "epoch": 6.04606107761136,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020443345104784934,
      "loss": 4.5883,
      "mean_token_accuracy": 0.2485720098018646,
      "num_tokens": 134790228.0,
      "step": 77710
    },
    {
      "entropy": 5.4520289421081545,
      "epoch": 6.046450106983078,
      "grad_norm": 1.5,
      "learning_rate": 0.00020440713436991055,
      "loss": 4.7041,
      "mean_token_accuracy": 0.2431233271956444,
      "num_tokens": 134798298.0,
      "step": 77715
    },
    {
      "entropy": 5.379271411895752,
      "epoch": 6.046839136354794,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000204380818763106,
      "loss": 4.5882,
      "mean_token_accuracy": 0.24223340451717376,
      "num_tokens": 134807623.0,
      "step": 77720
    },
    {
      "entropy": 5.389578723907471,
      "epoch": 6.047228165726512,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020435450422783525,
      "loss": 4.6399,
      "mean_token_accuracy": 0.24712451696395873,
      "num_tokens": 134816004.0,
      "step": 77725
    },
    {
      "entropy": 5.349229621887207,
      "epoch": 6.04761719509823,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020432819076449745,
      "loss": 4.6097,
      "mean_token_accuracy": 0.24354492872953415,
      "num_tokens": 134825059.0,
      "step": 77730
    },
    {
      "entropy": 5.385253524780273,
      "epoch": 6.0480062244699475,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020430187837349186,
      "loss": 4.6294,
      "mean_token_accuracy": 0.2420099124312401,
      "num_tokens": 134833526.0,
      "step": 77735
    },
    {
      "entropy": 5.327546548843384,
      "epoch": 6.048395253841665,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020427556705521777,
      "loss": 4.5617,
      "mean_token_accuracy": 0.24564434736967086,
      "num_tokens": 134841575.0,
      "step": 77740
    },
    {
      "entropy": 5.442416191101074,
      "epoch": 6.048784283213383,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002042492568100744,
      "loss": 4.6746,
      "mean_token_accuracy": 0.24136011600494384,
      "num_tokens": 134850506.0,
      "step": 77745
    },
    {
      "entropy": 5.369341468811035,
      "epoch": 6.049173312585101,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002042229476384609,
      "loss": 4.6804,
      "mean_token_accuracy": 0.23446427285671234,
      "num_tokens": 134858954.0,
      "step": 77750
    },
    {
      "entropy": 5.358863973617554,
      "epoch": 6.049562341956817,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00020419663954077656,
      "loss": 4.6,
      "mean_token_accuracy": 0.24840070456266403,
      "num_tokens": 134867390.0,
      "step": 77755
    },
    {
      "entropy": 5.309463167190552,
      "epoch": 6.049951371328535,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020417033251742046,
      "loss": 4.5522,
      "mean_token_accuracy": 0.24821835160255432,
      "num_tokens": 134876499.0,
      "step": 77760
    },
    {
      "entropy": 5.3706114292144775,
      "epoch": 6.050340400700253,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002041440265687919,
      "loss": 4.6251,
      "mean_token_accuracy": 0.24118445813655853,
      "num_tokens": 134885215.0,
      "step": 77765
    },
    {
      "entropy": 5.315656518936157,
      "epoch": 6.0507294300719705,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020411772169528987,
      "loss": 4.5089,
      "mean_token_accuracy": 0.2516279101371765,
      "num_tokens": 134893834.0,
      "step": 77770
    },
    {
      "entropy": 5.456667947769165,
      "epoch": 6.051118459443688,
      "grad_norm": 1.25,
      "learning_rate": 0.00020409141789731362,
      "loss": 4.7455,
      "mean_token_accuracy": 0.2319573163986206,
      "num_tokens": 134902376.0,
      "step": 77775
    },
    {
      "entropy": 5.3662622451782225,
      "epoch": 6.051507488815406,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020406511517526228,
      "loss": 4.6237,
      "mean_token_accuracy": 0.24100393205881118,
      "num_tokens": 134911066.0,
      "step": 77780
    },
    {
      "entropy": 5.418172073364258,
      "epoch": 6.051896518187123,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020403881352953486,
      "loss": 4.721,
      "mean_token_accuracy": 0.2305980369448662,
      "num_tokens": 134919259.0,
      "step": 77785
    },
    {
      "entropy": 5.367647409439087,
      "epoch": 6.05228554755884,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020401251296053047,
      "loss": 4.6212,
      "mean_token_accuracy": 0.23806293904781342,
      "num_tokens": 134928052.0,
      "step": 77790
    },
    {
      "entropy": 5.492748212814331,
      "epoch": 6.052674576930558,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020398621346864827,
      "loss": 4.8058,
      "mean_token_accuracy": 0.22472223043441772,
      "num_tokens": 134936570.0,
      "step": 77795
    },
    {
      "entropy": 5.425838851928711,
      "epoch": 6.053063606302276,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020395991505428723,
      "loss": 4.6482,
      "mean_token_accuracy": 0.2324936121702194,
      "num_tokens": 134945376.0,
      "step": 77800
    },
    {
      "entropy": 5.508853197097778,
      "epoch": 6.0534526356739935,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020393361771784647,
      "loss": 4.7286,
      "mean_token_accuracy": 0.22682449519634246,
      "num_tokens": 134954748.0,
      "step": 77805
    },
    {
      "entropy": 5.410128593444824,
      "epoch": 6.053841665045711,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002039073214597249,
      "loss": 4.6579,
      "mean_token_accuracy": 0.23687517493963242,
      "num_tokens": 134963039.0,
      "step": 77810
    },
    {
      "entropy": 5.364406728744507,
      "epoch": 6.054230694417429,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020388102628032156,
      "loss": 4.6177,
      "mean_token_accuracy": 0.23688792437314987,
      "num_tokens": 134971721.0,
      "step": 77815
    },
    {
      "entropy": 5.437514066696167,
      "epoch": 6.054619723789146,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020385473218003546,
      "loss": 4.7142,
      "mean_token_accuracy": 0.2358041912317276,
      "num_tokens": 134980287.0,
      "step": 77820
    },
    {
      "entropy": 5.332333183288574,
      "epoch": 6.055008753160863,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020382843915926557,
      "loss": 4.5685,
      "mean_token_accuracy": 0.2465401753783226,
      "num_tokens": 134989128.0,
      "step": 77825
    },
    {
      "entropy": 5.363819408416748,
      "epoch": 6.055397782532581,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020380214721841083,
      "loss": 4.6604,
      "mean_token_accuracy": 0.2352599561214447,
      "num_tokens": 134998182.0,
      "step": 77830
    },
    {
      "entropy": 5.411900424957276,
      "epoch": 6.055786811904299,
      "grad_norm": 1.234375,
      "learning_rate": 0.00020377585635787015,
      "loss": 4.7158,
      "mean_token_accuracy": 0.24174374341964722,
      "num_tokens": 135007581.0,
      "step": 77835
    },
    {
      "entropy": 5.443863439559936,
      "epoch": 6.0561758412760165,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020374956657804254,
      "loss": 4.6952,
      "mean_token_accuracy": 0.23185891211032866,
      "num_tokens": 135016207.0,
      "step": 77840
    },
    {
      "entropy": 5.383582067489624,
      "epoch": 6.056564870647734,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002037232778793268,
      "loss": 4.555,
      "mean_token_accuracy": 0.24811094105243683,
      "num_tokens": 135024516.0,
      "step": 77845
    },
    {
      "entropy": 5.4438573837280275,
      "epoch": 6.056953900019452,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002036969902621219,
      "loss": 4.7496,
      "mean_token_accuracy": 0.23171769976615905,
      "num_tokens": 135032653.0,
      "step": 77850
    },
    {
      "entropy": 5.426685523986817,
      "epoch": 6.057342929391169,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002036707037268266,
      "loss": 4.6668,
      "mean_token_accuracy": 0.2369655728340149,
      "num_tokens": 135041341.0,
      "step": 77855
    },
    {
      "entropy": 5.390874719619751,
      "epoch": 6.057731958762886,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020364441827383994,
      "loss": 4.6252,
      "mean_token_accuracy": 0.25102262794971464,
      "num_tokens": 135049297.0,
      "step": 77860
    },
    {
      "entropy": 5.300956153869629,
      "epoch": 6.058120988134604,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020361813390356066,
      "loss": 4.5314,
      "mean_token_accuracy": 0.24673874527215958,
      "num_tokens": 135057919.0,
      "step": 77865
    },
    {
      "entropy": 5.386100149154663,
      "epoch": 6.058510017506322,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020359185061638753,
      "loss": 4.6181,
      "mean_token_accuracy": 0.24043917506933213,
      "num_tokens": 135066452.0,
      "step": 77870
    },
    {
      "entropy": 5.473205661773681,
      "epoch": 6.0588990468780395,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002035655684127194,
      "loss": 4.7348,
      "mean_token_accuracy": 0.2284441828727722,
      "num_tokens": 135074558.0,
      "step": 77875
    },
    {
      "entropy": 5.2776771068573,
      "epoch": 6.059288076249757,
      "grad_norm": 1.25,
      "learning_rate": 0.0002035392872929551,
      "loss": 4.5298,
      "mean_token_accuracy": 0.24325364977121353,
      "num_tokens": 135083118.0,
      "step": 77880
    },
    {
      "entropy": 5.399184608459473,
      "epoch": 6.059677105621474,
      "grad_norm": 1.234375,
      "learning_rate": 0.00020351300725749338,
      "loss": 4.6052,
      "mean_token_accuracy": 0.23999981135129927,
      "num_tokens": 135091832.0,
      "step": 77885
    },
    {
      "entropy": 5.443254613876343,
      "epoch": 6.060066134993192,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000203486728306733,
      "loss": 4.7572,
      "mean_token_accuracy": 0.23332567363977433,
      "num_tokens": 135100025.0,
      "step": 77890
    },
    {
      "entropy": 5.333177757263184,
      "epoch": 6.060455164364909,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020346045044107269,
      "loss": 4.5889,
      "mean_token_accuracy": 0.2422758847475052,
      "num_tokens": 135108414.0,
      "step": 77895
    },
    {
      "entropy": 5.355976200103759,
      "epoch": 6.060844193736627,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002034341736609112,
      "loss": 4.5853,
      "mean_token_accuracy": 0.24422654062509536,
      "num_tokens": 135117409.0,
      "step": 77900
    },
    {
      "entropy": 5.353270673751831,
      "epoch": 6.061233223108345,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020340789796664715,
      "loss": 4.6139,
      "mean_token_accuracy": 0.24613844454288483,
      "num_tokens": 135125757.0,
      "step": 77905
    },
    {
      "entropy": 5.344522857666016,
      "epoch": 6.0616222524800625,
      "grad_norm": 1.25,
      "learning_rate": 0.00020338162335867938,
      "loss": 4.6505,
      "mean_token_accuracy": 0.23396904468536378,
      "num_tokens": 135134334.0,
      "step": 77910
    },
    {
      "entropy": 5.452578067779541,
      "epoch": 6.06201128185178,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00020335534983740645,
      "loss": 4.6934,
      "mean_token_accuracy": 0.23344435542821884,
      "num_tokens": 135142992.0,
      "step": 77915
    },
    {
      "entropy": 5.332722949981689,
      "epoch": 6.062400311223497,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020332907740322693,
      "loss": 4.5864,
      "mean_token_accuracy": 0.24226741194725038,
      "num_tokens": 135151401.0,
      "step": 77920
    },
    {
      "entropy": 5.470785665512085,
      "epoch": 6.062789340595215,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020330280605653967,
      "loss": 4.7259,
      "mean_token_accuracy": 0.2371002107858658,
      "num_tokens": 135160120.0,
      "step": 77925
    },
    {
      "entropy": 5.431211996078491,
      "epoch": 6.063178369966932,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020327653579774331,
      "loss": 4.6778,
      "mean_token_accuracy": 0.23626890927553176,
      "num_tokens": 135169012.0,
      "step": 77930
    },
    {
      "entropy": 5.336164236068726,
      "epoch": 6.06356739933865,
      "grad_norm": 1.234375,
      "learning_rate": 0.00020325026662723622,
      "loss": 4.57,
      "mean_token_accuracy": 0.2500456467270851,
      "num_tokens": 135178001.0,
      "step": 77935
    },
    {
      "entropy": 5.380569219589233,
      "epoch": 6.063956428710368,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020322399854541724,
      "loss": 4.6569,
      "mean_token_accuracy": 0.2396145209670067,
      "num_tokens": 135186881.0,
      "step": 77940
    },
    {
      "entropy": 5.32777156829834,
      "epoch": 6.0643454580820855,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002031977315526848,
      "loss": 4.54,
      "mean_token_accuracy": 0.24685431271791458,
      "num_tokens": 135195591.0,
      "step": 77945
    },
    {
      "entropy": 5.4235669612884525,
      "epoch": 6.064734487453803,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020317146564943745,
      "loss": 4.6907,
      "mean_token_accuracy": 0.23657889515161515,
      "num_tokens": 135204327.0,
      "step": 77950
    },
    {
      "entropy": 5.338595771789551,
      "epoch": 6.06512351682552,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002031452008360738,
      "loss": 4.5892,
      "mean_token_accuracy": 0.2426821455359459,
      "num_tokens": 135213233.0,
      "step": 77955
    },
    {
      "entropy": 5.340740156173706,
      "epoch": 6.065512546197238,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002031189371129924,
      "loss": 4.6201,
      "mean_token_accuracy": 0.24165795147418975,
      "num_tokens": 135221977.0,
      "step": 77960
    },
    {
      "entropy": 5.301126003265381,
      "epoch": 6.065901575568955,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002030926744805917,
      "loss": 4.5509,
      "mean_token_accuracy": 0.24297299981117249,
      "num_tokens": 135231234.0,
      "step": 77965
    },
    {
      "entropy": 5.3734029769897464,
      "epoch": 6.066290604940673,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002030664129392702,
      "loss": 4.6207,
      "mean_token_accuracy": 0.2408313736319542,
      "num_tokens": 135240110.0,
      "step": 77970
    },
    {
      "entropy": 5.45770902633667,
      "epoch": 6.066679634312391,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020304015248942642,
      "loss": 4.6509,
      "mean_token_accuracy": 0.24261791557073592,
      "num_tokens": 135248280.0,
      "step": 77975
    },
    {
      "entropy": 5.353919982910156,
      "epoch": 6.0670686636841085,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020301389313145886,
      "loss": 4.6067,
      "mean_token_accuracy": 0.23615747392177583,
      "num_tokens": 135256842.0,
      "step": 77980
    },
    {
      "entropy": 5.4359081268310545,
      "epoch": 6.067457693055825,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020298763486576566,
      "loss": 4.6891,
      "mean_token_accuracy": 0.23562852293252945,
      "num_tokens": 135265088.0,
      "step": 77985
    },
    {
      "entropy": 5.4074122428894045,
      "epoch": 6.067846722427543,
      "grad_norm": 1.25,
      "learning_rate": 0.00020296137769274563,
      "loss": 4.664,
      "mean_token_accuracy": 0.24390268325805664,
      "num_tokens": 135273544.0,
      "step": 77990
    },
    {
      "entropy": 5.3657269954681395,
      "epoch": 6.068235751799261,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020293512161279704,
      "loss": 4.6597,
      "mean_token_accuracy": 0.2351357415318489,
      "num_tokens": 135282059.0,
      "step": 77995
    },
    {
      "entropy": 5.4972363948822025,
      "epoch": 6.068624781170978,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020290886662631825,
      "loss": 4.7335,
      "mean_token_accuracy": 0.23032846450805664,
      "num_tokens": 135290112.0,
      "step": 78000
    },
    {
      "epoch": 6.068624781170978,
      "eval_entropy": 5.146211392978158,
      "eval_loss": 4.877217769622803,
      "eval_mean_token_accuracy": 0.23330182733876528,
      "eval_num_tokens": 135290112.0,
      "eval_runtime": 28.9243,
      "eval_samples_per_second": 1436.786,
      "eval_steps_per_second": 179.607,
      "step": 78000
    },
    {
      "entropy": 5.4060132026672365,
      "epoch": 6.069013810542696,
      "grad_norm": 1.25,
      "learning_rate": 0.00020288261273370767,
      "loss": 4.6182,
      "mean_token_accuracy": 0.2393279105424881,
      "num_tokens": 135298314.0,
      "step": 78005
    },
    {
      "entropy": 5.308161449432373,
      "epoch": 6.069402839914414,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020285635993536372,
      "loss": 4.5937,
      "mean_token_accuracy": 0.24137768000364304,
      "num_tokens": 135306557.0,
      "step": 78010
    },
    {
      "entropy": 5.325861072540283,
      "epoch": 6.0697918692861315,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020283010823168464,
      "loss": 4.6518,
      "mean_token_accuracy": 0.23814892321825026,
      "num_tokens": 135315100.0,
      "step": 78015
    },
    {
      "entropy": 5.42292857170105,
      "epoch": 6.070180898657848,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020280385762306874,
      "loss": 4.622,
      "mean_token_accuracy": 0.2399055615067482,
      "num_tokens": 135323426.0,
      "step": 78020
    },
    {
      "entropy": 5.332621717453003,
      "epoch": 6.070569928029566,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020277760810991446,
      "loss": 4.5428,
      "mean_token_accuracy": 0.25161524713039396,
      "num_tokens": 135331706.0,
      "step": 78025
    },
    {
      "entropy": 5.416768360137939,
      "epoch": 6.070958957401284,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020275135969262,
      "loss": 4.6018,
      "mean_token_accuracy": 0.23867940455675124,
      "num_tokens": 135340552.0,
      "step": 78030
    },
    {
      "entropy": 5.38787727355957,
      "epoch": 6.071347986773001,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020272511237158364,
      "loss": 4.6064,
      "mean_token_accuracy": 0.23204150348901748,
      "num_tokens": 135349100.0,
      "step": 78035
    },
    {
      "entropy": 5.358548974990844,
      "epoch": 6.071737016144719,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020269886614720367,
      "loss": 4.6085,
      "mean_token_accuracy": 0.24614261388778685,
      "num_tokens": 135357447.0,
      "step": 78040
    },
    {
      "entropy": 5.379813718795776,
      "epoch": 6.072126045516437,
      "grad_norm": 1.25,
      "learning_rate": 0.00020267262101987833,
      "loss": 4.6054,
      "mean_token_accuracy": 0.242340749502182,
      "num_tokens": 135366132.0,
      "step": 78045
    },
    {
      "entropy": 5.35446949005127,
      "epoch": 6.0725150748881545,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020264637699000587,
      "loss": 4.5612,
      "mean_token_accuracy": 0.2479987159371376,
      "num_tokens": 135374436.0,
      "step": 78050
    },
    {
      "entropy": 5.370483303070069,
      "epoch": 6.072904104259871,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002026201340579844,
      "loss": 4.559,
      "mean_token_accuracy": 0.244822359085083,
      "num_tokens": 135382951.0,
      "step": 78055
    },
    {
      "entropy": 5.452025461196899,
      "epoch": 6.073293133631589,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020259389222421225,
      "loss": 4.7917,
      "mean_token_accuracy": 0.22936633080244065,
      "num_tokens": 135392549.0,
      "step": 78060
    },
    {
      "entropy": 5.342155456542969,
      "epoch": 6.073682163003307,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020256765148908752,
      "loss": 4.5841,
      "mean_token_accuracy": 0.2426443174481392,
      "num_tokens": 135400548.0,
      "step": 78065
    },
    {
      "entropy": 5.391656303405762,
      "epoch": 6.074071192375024,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002025414118530084,
      "loss": 4.5835,
      "mean_token_accuracy": 0.24599498957395555,
      "num_tokens": 135409501.0,
      "step": 78070
    },
    {
      "entropy": 5.393645572662353,
      "epoch": 6.074460221746742,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020251517331637303,
      "loss": 4.6538,
      "mean_token_accuracy": 0.2417784109711647,
      "num_tokens": 135417753.0,
      "step": 78075
    },
    {
      "entropy": 5.3836315155029295,
      "epoch": 6.07484925111846,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020248893587957955,
      "loss": 4.6482,
      "mean_token_accuracy": 0.24358249753713607,
      "num_tokens": 135426485.0,
      "step": 78080
    },
    {
      "entropy": 5.373294544219971,
      "epoch": 6.0752382804901774,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020246269954302603,
      "loss": 4.614,
      "mean_token_accuracy": 0.24455467760562896,
      "num_tokens": 135435292.0,
      "step": 78085
    },
    {
      "entropy": 5.324078798294067,
      "epoch": 6.075627309861894,
      "grad_norm": 1.1875,
      "learning_rate": 0.00020243646430711065,
      "loss": 4.5683,
      "mean_token_accuracy": 0.25234498232603075,
      "num_tokens": 135443990.0,
      "step": 78090
    },
    {
      "entropy": 5.343260288238525,
      "epoch": 6.076016339233612,
      "grad_norm": 1.234375,
      "learning_rate": 0.00020241023017223138,
      "loss": 4.6236,
      "mean_token_accuracy": 0.24609242081642152,
      "num_tokens": 135452798.0,
      "step": 78095
    },
    {
      "entropy": 5.383029747009277,
      "epoch": 6.07640536860533,
      "grad_norm": 1.234375,
      "learning_rate": 0.00020238399713878635,
      "loss": 4.6219,
      "mean_token_accuracy": 0.23520573377609252,
      "num_tokens": 135461158.0,
      "step": 78100
    },
    {
      "entropy": 5.419666481018067,
      "epoch": 6.076794397977047,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020235776520717352,
      "loss": 4.6741,
      "mean_token_accuracy": 0.23241727203130721,
      "num_tokens": 135469890.0,
      "step": 78105
    },
    {
      "entropy": 5.446431970596313,
      "epoch": 6.077183427348765,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000202331534377791,
      "loss": 4.7075,
      "mean_token_accuracy": 0.23330069184303284,
      "num_tokens": 135478042.0,
      "step": 78110
    },
    {
      "entropy": 5.404907655715943,
      "epoch": 6.077572456720483,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020230530465103686,
      "loss": 4.6755,
      "mean_token_accuracy": 0.23925051242113113,
      "num_tokens": 135486595.0,
      "step": 78115
    },
    {
      "entropy": 5.329097652435303,
      "epoch": 6.0779614860921996,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020227907602730888,
      "loss": 4.5735,
      "mean_token_accuracy": 0.24599978178739548,
      "num_tokens": 135495093.0,
      "step": 78120
    },
    {
      "entropy": 5.334168291091919,
      "epoch": 6.078350515463917,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020225284850700526,
      "loss": 4.5984,
      "mean_token_accuracy": 0.24583181291818618,
      "num_tokens": 135503917.0,
      "step": 78125
    },
    {
      "entropy": 5.393647193908691,
      "epoch": 6.078739544835635,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020222662209052388,
      "loss": 4.6158,
      "mean_token_accuracy": 0.23712670356035231,
      "num_tokens": 135512430.0,
      "step": 78130
    },
    {
      "entropy": 5.4796082973480225,
      "epoch": 6.079128574207353,
      "grad_norm": 1.25,
      "learning_rate": 0.00020220039677826263,
      "loss": 4.7248,
      "mean_token_accuracy": 0.2333470866084099,
      "num_tokens": 135521116.0,
      "step": 78135
    },
    {
      "entropy": 5.403421640396118,
      "epoch": 6.07951760357907,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020217417257061953,
      "loss": 4.5055,
      "mean_token_accuracy": 0.25195791125297545,
      "num_tokens": 135529044.0,
      "step": 78140
    },
    {
      "entropy": 5.373752117156982,
      "epoch": 6.079906632950788,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020214794946799247,
      "loss": 4.7334,
      "mean_token_accuracy": 0.22973351925611496,
      "num_tokens": 135537172.0,
      "step": 78145
    },
    {
      "entropy": 5.4183669090271,
      "epoch": 6.080295662322506,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002021217274707793,
      "loss": 4.6809,
      "mean_token_accuracy": 0.23833460062742234,
      "num_tokens": 135545613.0,
      "step": 78150
    },
    {
      "entropy": 5.4724328994750975,
      "epoch": 6.0806846916942225,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020209550657937787,
      "loss": 4.7936,
      "mean_token_accuracy": 0.23817306011915207,
      "num_tokens": 135554418.0,
      "step": 78155
    },
    {
      "entropy": 5.439923095703125,
      "epoch": 6.08107372106594,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020206928679418612,
      "loss": 4.6746,
      "mean_token_accuracy": 0.2423946738243103,
      "num_tokens": 135563667.0,
      "step": 78160
    },
    {
      "entropy": 5.332416439056397,
      "epoch": 6.081462750437658,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020204306811560187,
      "loss": 4.53,
      "mean_token_accuracy": 0.24317148476839065,
      "num_tokens": 135572156.0,
      "step": 78165
    },
    {
      "entropy": 5.347363996505737,
      "epoch": 6.081851779809376,
      "grad_norm": 1.375,
      "learning_rate": 0.00020201685054402303,
      "loss": 4.5526,
      "mean_token_accuracy": 0.2427965745329857,
      "num_tokens": 135580397.0,
      "step": 78170
    },
    {
      "entropy": 5.32891812324524,
      "epoch": 6.082240809181093,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002019906340798472,
      "loss": 4.6706,
      "mean_token_accuracy": 0.24101684838533402,
      "num_tokens": 135588649.0,
      "step": 78175
    },
    {
      "entropy": 5.333479738235473,
      "epoch": 6.082629838552811,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020196441872347232,
      "loss": 4.6121,
      "mean_token_accuracy": 0.23681754767894744,
      "num_tokens": 135597548.0,
      "step": 78180
    },
    {
      "entropy": 5.350278234481811,
      "epoch": 6.083018867924529,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020193820447529605,
      "loss": 4.5703,
      "mean_token_accuracy": 0.25180537402629855,
      "num_tokens": 135606121.0,
      "step": 78185
    },
    {
      "entropy": 5.509008741378784,
      "epoch": 6.0834078972962455,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002019119913357163,
      "loss": 4.7829,
      "mean_token_accuracy": 0.23049566596746446,
      "num_tokens": 135615092.0,
      "step": 78190
    },
    {
      "entropy": 5.409129476547241,
      "epoch": 6.083796926667963,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020188577930513074,
      "loss": 4.6077,
      "mean_token_accuracy": 0.23699478209018707,
      "num_tokens": 135622916.0,
      "step": 78195
    },
    {
      "entropy": 5.338917636871338,
      "epoch": 6.084185956039681,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002018595683839372,
      "loss": 4.6751,
      "mean_token_accuracy": 0.24006969928741456,
      "num_tokens": 135631816.0,
      "step": 78200
    },
    {
      "entropy": 5.417550563812256,
      "epoch": 6.084574985411399,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020183335857253327,
      "loss": 4.6291,
      "mean_token_accuracy": 0.23627813011407853,
      "num_tokens": 135640501.0,
      "step": 78205
    },
    {
      "entropy": 5.419645833969116,
      "epoch": 6.084964014783116,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020180714987131658,
      "loss": 4.6692,
      "mean_token_accuracy": 0.23422762900590896,
      "num_tokens": 135649978.0,
      "step": 78210
    },
    {
      "entropy": 5.361103916168213,
      "epoch": 6.085353044154834,
      "grad_norm": 1.3125,
      "learning_rate": 0.000201780942280685,
      "loss": 4.6258,
      "mean_token_accuracy": 0.2394057795405388,
      "num_tokens": 135658552.0,
      "step": 78215
    },
    {
      "entropy": 5.359128618240357,
      "epoch": 6.085742073526551,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000201754735801036,
      "loss": 4.7068,
      "mean_token_accuracy": 0.2343481585383415,
      "num_tokens": 135667756.0,
      "step": 78220
    },
    {
      "entropy": 5.3251978874206545,
      "epoch": 6.0861311028982685,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020172853043276735,
      "loss": 4.5869,
      "mean_token_accuracy": 0.24228523671627045,
      "num_tokens": 135675763.0,
      "step": 78225
    },
    {
      "entropy": 5.321824169158935,
      "epoch": 6.086520132269986,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002017023261762766,
      "loss": 4.583,
      "mean_token_accuracy": 0.24340703189373017,
      "num_tokens": 135684211.0,
      "step": 78230
    },
    {
      "entropy": 5.409508895874024,
      "epoch": 6.086909161641704,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002016761230319613,
      "loss": 4.7598,
      "mean_token_accuracy": 0.23273080587387085,
      "num_tokens": 135692792.0,
      "step": 78235
    },
    {
      "entropy": 5.500107336044311,
      "epoch": 6.087298191013422,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020164992100021924,
      "loss": 4.7555,
      "mean_token_accuracy": 0.22770076990127563,
      "num_tokens": 135701178.0,
      "step": 78240
    },
    {
      "entropy": 5.414925289154053,
      "epoch": 6.087687220385139,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020162372008144792,
      "loss": 4.6757,
      "mean_token_accuracy": 0.23936578780412673,
      "num_tokens": 135710453.0,
      "step": 78245
    },
    {
      "entropy": 5.448999547958374,
      "epoch": 6.088076249756857,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020159752027604477,
      "loss": 4.6837,
      "mean_token_accuracy": 0.23721446543931962,
      "num_tokens": 135718996.0,
      "step": 78250
    },
    {
      "entropy": 5.288850879669189,
      "epoch": 6.088465279128574,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020157132158440732,
      "loss": 4.4845,
      "mean_token_accuracy": 0.25437273383140563,
      "num_tokens": 135727587.0,
      "step": 78255
    },
    {
      "entropy": 5.454306983947754,
      "epoch": 6.0888543085002915,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020154512400693338,
      "loss": 4.8231,
      "mean_token_accuracy": 0.22893631011247634,
      "num_tokens": 135737228.0,
      "step": 78260
    },
    {
      "entropy": 5.345508670806884,
      "epoch": 6.089243337872009,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002015189275440202,
      "loss": 4.5916,
      "mean_token_accuracy": 0.24814069271087646,
      "num_tokens": 135745753.0,
      "step": 78265
    },
    {
      "entropy": 5.430965375900269,
      "epoch": 6.089632367243727,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020149273219606528,
      "loss": 4.5841,
      "mean_token_accuracy": 0.24944130033254625,
      "num_tokens": 135753930.0,
      "step": 78270
    },
    {
      "entropy": 5.4203132629394535,
      "epoch": 6.090021396615445,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002014665379634662,
      "loss": 4.7058,
      "mean_token_accuracy": 0.233579421043396,
      "num_tokens": 135762762.0,
      "step": 78275
    },
    {
      "entropy": 5.342609834671021,
      "epoch": 6.090410425987162,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020144034484662044,
      "loss": 4.6148,
      "mean_token_accuracy": 0.24460069239139556,
      "num_tokens": 135770887.0,
      "step": 78280
    },
    {
      "entropy": 5.386258125305176,
      "epoch": 6.09079945535888,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020141415284592528,
      "loss": 4.6376,
      "mean_token_accuracy": 0.23933984190225602,
      "num_tokens": 135779545.0,
      "step": 78285
    },
    {
      "entropy": 5.336141872406006,
      "epoch": 6.091188484730597,
      "grad_norm": 1.25,
      "learning_rate": 0.00020138796196177827,
      "loss": 4.5384,
      "mean_token_accuracy": 0.2491322413086891,
      "num_tokens": 135787886.0,
      "step": 78290
    },
    {
      "entropy": 5.364180088043213,
      "epoch": 6.0915775141023145,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020136177219457675,
      "loss": 4.5716,
      "mean_token_accuracy": 0.249770849943161,
      "num_tokens": 135796883.0,
      "step": 78295
    },
    {
      "entropy": 5.392657136917114,
      "epoch": 6.091966543474032,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002013355835447181,
      "loss": 4.6926,
      "mean_token_accuracy": 0.22681605964899063,
      "num_tokens": 135805347.0,
      "step": 78300
    },
    {
      "entropy": 5.4189722537994385,
      "epoch": 6.09235557284575,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020130939601259974,
      "loss": 4.7437,
      "mean_token_accuracy": 0.2345716953277588,
      "num_tokens": 135813791.0,
      "step": 78305
    },
    {
      "entropy": 5.46801061630249,
      "epoch": 6.092744602217468,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020128320959861902,
      "loss": 4.6749,
      "mean_token_accuracy": 0.23406531512737275,
      "num_tokens": 135822523.0,
      "step": 78310
    },
    {
      "entropy": 5.367745494842529,
      "epoch": 6.093133631589185,
      "grad_norm": 1.203125,
      "learning_rate": 0.00020125702430317323,
      "loss": 4.5641,
      "mean_token_accuracy": 0.24882566183805466,
      "num_tokens": 135831155.0,
      "step": 78315
    },
    {
      "entropy": 5.415117597579956,
      "epoch": 6.093522660960902,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002012308401266597,
      "loss": 4.6171,
      "mean_token_accuracy": 0.2403356358408928,
      "num_tokens": 135838928.0,
      "step": 78320
    },
    {
      "entropy": 5.3508056640625,
      "epoch": 6.09391169033262,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020120465706947584,
      "loss": 4.6363,
      "mean_token_accuracy": 0.2362497106194496,
      "num_tokens": 135847774.0,
      "step": 78325
    },
    {
      "entropy": 5.26229681968689,
      "epoch": 6.0943007197043375,
      "grad_norm": 1.25,
      "learning_rate": 0.00020117847513201876,
      "loss": 4.4997,
      "mean_token_accuracy": 0.2513694614171982,
      "num_tokens": 135856279.0,
      "step": 78330
    },
    {
      "entropy": 5.2496318340301515,
      "epoch": 6.094689749076055,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002011522943146859,
      "loss": 4.475,
      "mean_token_accuracy": 0.25967146903276445,
      "num_tokens": 135864989.0,
      "step": 78335
    },
    {
      "entropy": 5.346390581130981,
      "epoch": 6.095078778447773,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002011261146178744,
      "loss": 4.6781,
      "mean_token_accuracy": 0.23574930876493455,
      "num_tokens": 135873621.0,
      "step": 78340
    },
    {
      "entropy": 5.435528945922852,
      "epoch": 6.095467807819491,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020109993604198152,
      "loss": 4.6918,
      "mean_token_accuracy": 0.2408699259161949,
      "num_tokens": 135881648.0,
      "step": 78345
    },
    {
      "entropy": 5.4342247486114506,
      "epoch": 6.095856837191208,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002010737585874045,
      "loss": 4.6707,
      "mean_token_accuracy": 0.2374473676085472,
      "num_tokens": 135890291.0,
      "step": 78350
    },
    {
      "entropy": 5.405874490737915,
      "epoch": 6.096245866562925,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002010475822545405,
      "loss": 4.5995,
      "mean_token_accuracy": 0.24467612504959108,
      "num_tokens": 135898546.0,
      "step": 78355
    },
    {
      "entropy": 5.427912473678589,
      "epoch": 6.096634895934643,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002010214070437868,
      "loss": 4.7628,
      "mean_token_accuracy": 0.22852298021316528,
      "num_tokens": 135907836.0,
      "step": 78360
    },
    {
      "entropy": 5.451780509948731,
      "epoch": 6.0970239253063605,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020099523295554045,
      "loss": 4.6572,
      "mean_token_accuracy": 0.24187387377023697,
      "num_tokens": 135916268.0,
      "step": 78365
    },
    {
      "entropy": 5.437407350540161,
      "epoch": 6.097412954678078,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020096905999019871,
      "loss": 4.6954,
      "mean_token_accuracy": 0.24311623275279998,
      "num_tokens": 135924637.0,
      "step": 78370
    },
    {
      "entropy": 5.416843032836914,
      "epoch": 6.097801984049796,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020094288814815858,
      "loss": 4.7034,
      "mean_token_accuracy": 0.2385851949453354,
      "num_tokens": 135934109.0,
      "step": 78375
    },
    {
      "entropy": 5.410534143447876,
      "epoch": 6.098191013421514,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020091671742981725,
      "loss": 4.6272,
      "mean_token_accuracy": 0.2447964444756508,
      "num_tokens": 135942630.0,
      "step": 78380
    },
    {
      "entropy": 5.439756202697754,
      "epoch": 6.098580042793231,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020089054783557176,
      "loss": 4.6184,
      "mean_token_accuracy": 0.2420434072613716,
      "num_tokens": 135950857.0,
      "step": 78385
    },
    {
      "entropy": 5.383828401565552,
      "epoch": 6.098969072164948,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020086437936581936,
      "loss": 4.6651,
      "mean_token_accuracy": 0.2389864668250084,
      "num_tokens": 135958991.0,
      "step": 78390
    },
    {
      "entropy": 5.403201246261597,
      "epoch": 6.099358101536666,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020083821202095697,
      "loss": 4.6034,
      "mean_token_accuracy": 0.2437028169631958,
      "num_tokens": 135966793.0,
      "step": 78395
    },
    {
      "entropy": 5.380572986602783,
      "epoch": 6.0997471309083835,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020081204580138162,
      "loss": 4.6182,
      "mean_token_accuracy": 0.23512735813856125,
      "num_tokens": 135975878.0,
      "step": 78400
    },
    {
      "entropy": 5.318931150436401,
      "epoch": 6.100136160280101,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020078588070749048,
      "loss": 4.5191,
      "mean_token_accuracy": 0.2505130484700203,
      "num_tokens": 135984107.0,
      "step": 78405
    },
    {
      "entropy": 5.388053894042969,
      "epoch": 6.100525189651819,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020075971673968042,
      "loss": 4.7015,
      "mean_token_accuracy": 0.2288355365395546,
      "num_tokens": 135992724.0,
      "step": 78410
    },
    {
      "entropy": 5.368625926971435,
      "epoch": 6.100914219023537,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020073355389834852,
      "loss": 4.6744,
      "mean_token_accuracy": 0.23834931403398513,
      "num_tokens": 136001712.0,
      "step": 78415
    },
    {
      "entropy": 5.413221836090088,
      "epoch": 6.101303248395254,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020070739218389173,
      "loss": 4.6484,
      "mean_token_accuracy": 0.24068213552236556,
      "num_tokens": 136010756.0,
      "step": 78420
    },
    {
      "entropy": 5.41457314491272,
      "epoch": 6.101692277766971,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020068123159670698,
      "loss": 4.6137,
      "mean_token_accuracy": 0.24685857594013214,
      "num_tokens": 136019150.0,
      "step": 78425
    },
    {
      "entropy": 5.4194879055023195,
      "epoch": 6.102081307138689,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020065507213719132,
      "loss": 4.6654,
      "mean_token_accuracy": 0.2342357352375984,
      "num_tokens": 136028889.0,
      "step": 78430
    },
    {
      "entropy": 5.476228857040406,
      "epoch": 6.1024703365104065,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020062891380574156,
      "loss": 4.6804,
      "mean_token_accuracy": 0.2393524870276451,
      "num_tokens": 136037833.0,
      "step": 78435
    },
    {
      "entropy": 5.401272964477539,
      "epoch": 6.102859365882124,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020060275660275468,
      "loss": 4.6526,
      "mean_token_accuracy": 0.2363521307706833,
      "num_tokens": 136046479.0,
      "step": 78440
    },
    {
      "entropy": 5.368519830703735,
      "epoch": 6.103248395253842,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020057660052862753,
      "loss": 4.6704,
      "mean_token_accuracy": 0.2414141997694969,
      "num_tokens": 136054985.0,
      "step": 78445
    },
    {
      "entropy": 5.393288660049438,
      "epoch": 6.10363742462556,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020055044558375695,
      "loss": 4.6369,
      "mean_token_accuracy": 0.2390091061592102,
      "num_tokens": 136063539.0,
      "step": 78450
    },
    {
      "entropy": 5.4311028003692625,
      "epoch": 6.104026453997276,
      "grad_norm": 1.25,
      "learning_rate": 0.00020052429176853993,
      "loss": 4.6273,
      "mean_token_accuracy": 0.24230020642280578,
      "num_tokens": 136072476.0,
      "step": 78455
    },
    {
      "entropy": 5.383445119857788,
      "epoch": 6.104415483368994,
      "grad_norm": 1.125,
      "learning_rate": 0.00020049813908337323,
      "loss": 4.6434,
      "mean_token_accuracy": 0.2392372101545334,
      "num_tokens": 136081153.0,
      "step": 78460
    },
    {
      "entropy": 5.445254468917847,
      "epoch": 6.104804512740712,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002004719875286537,
      "loss": 4.69,
      "mean_token_accuracy": 0.23070433139801025,
      "num_tokens": 136089875.0,
      "step": 78465
    },
    {
      "entropy": 5.440861797332763,
      "epoch": 6.1051935421124295,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002004458371047782,
      "loss": 4.6872,
      "mean_token_accuracy": 0.23757457882165908,
      "num_tokens": 136098115.0,
      "step": 78470
    },
    {
      "entropy": 5.311525106430054,
      "epoch": 6.105582571484147,
      "grad_norm": 1.25,
      "learning_rate": 0.00020041968781214336,
      "loss": 4.5822,
      "mean_token_accuracy": 0.24353455752134323,
      "num_tokens": 136106591.0,
      "step": 78475
    },
    {
      "entropy": 5.42152247428894,
      "epoch": 6.105971600855865,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002003935396511461,
      "loss": 4.7476,
      "mean_token_accuracy": 0.2334782987833023,
      "num_tokens": 136116953.0,
      "step": 78480
    },
    {
      "entropy": 5.388136386871338,
      "epoch": 6.106360630227583,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020036739262218313,
      "loss": 4.6125,
      "mean_token_accuracy": 0.2413247123360634,
      "num_tokens": 136126220.0,
      "step": 78485
    },
    {
      "entropy": 5.403662967681885,
      "epoch": 6.106749659599299,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020034124672565112,
      "loss": 4.6159,
      "mean_token_accuracy": 0.24158646017313004,
      "num_tokens": 136134902.0,
      "step": 78490
    },
    {
      "entropy": 5.372243118286133,
      "epoch": 6.107138688971017,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002003151019619469,
      "loss": 4.5953,
      "mean_token_accuracy": 0.24275404661893846,
      "num_tokens": 136143199.0,
      "step": 78495
    },
    {
      "entropy": 5.403549480438232,
      "epoch": 6.107527718342735,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020028895833146715,
      "loss": 4.688,
      "mean_token_accuracy": 0.24496107399463654,
      "num_tokens": 136151444.0,
      "step": 78500
    },
    {
      "entropy": 5.47201018333435,
      "epoch": 6.1079167477144525,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020026281583460854,
      "loss": 4.7863,
      "mean_token_accuracy": 0.22825094163417817,
      "num_tokens": 136160316.0,
      "step": 78505
    },
    {
      "entropy": 5.416810178756714,
      "epoch": 6.10830577708617,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020023667447176776,
      "loss": 4.6178,
      "mean_token_accuracy": 0.2448534846305847,
      "num_tokens": 136168537.0,
      "step": 78510
    },
    {
      "entropy": 5.45249309539795,
      "epoch": 6.108694806457888,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020021053424334142,
      "loss": 4.6548,
      "mean_token_accuracy": 0.24075135439634324,
      "num_tokens": 136176974.0,
      "step": 78515
    },
    {
      "entropy": 5.304251480102539,
      "epoch": 6.109083835829606,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020018439514972615,
      "loss": 4.5876,
      "mean_token_accuracy": 0.2412341922521591,
      "num_tokens": 136185164.0,
      "step": 78520
    },
    {
      "entropy": 5.38652868270874,
      "epoch": 6.109472865201322,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002001582571913186,
      "loss": 4.6588,
      "mean_token_accuracy": 0.24100582003593446,
      "num_tokens": 136193526.0,
      "step": 78525
    },
    {
      "entropy": 5.331762218475342,
      "epoch": 6.10986189457304,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020013212036851542,
      "loss": 4.5815,
      "mean_token_accuracy": 0.25378274768590925,
      "num_tokens": 136202873.0,
      "step": 78530
    },
    {
      "entropy": 5.372622680664063,
      "epoch": 6.110250923944758,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020010598468171316,
      "loss": 4.6153,
      "mean_token_accuracy": 0.23556251525878907,
      "num_tokens": 136211870.0,
      "step": 78535
    },
    {
      "entropy": 5.393870496749878,
      "epoch": 6.1106399533164755,
      "grad_norm": 1.25,
      "learning_rate": 0.0002000798501313083,
      "loss": 4.6494,
      "mean_token_accuracy": 0.24751973003149033,
      "num_tokens": 136220244.0,
      "step": 78540
    },
    {
      "entropy": 5.4217931747436525,
      "epoch": 6.111028982688193,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020005371671769756,
      "loss": 4.6978,
      "mean_token_accuracy": 0.2300085261464119,
      "num_tokens": 136229184.0,
      "step": 78545
    },
    {
      "entropy": 5.4494706153869625,
      "epoch": 6.111418012059911,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020002758444127738,
      "loss": 4.6749,
      "mean_token_accuracy": 0.23971299082040787,
      "num_tokens": 136238049.0,
      "step": 78550
    },
    {
      "entropy": 5.544940662384033,
      "epoch": 6.111807041431628,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020000145330244424,
      "loss": 4.8538,
      "mean_token_accuracy": 0.22665495574474334,
      "num_tokens": 136246656.0,
      "step": 78555
    },
    {
      "entropy": 5.394204092025757,
      "epoch": 6.112196070803345,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001999753233015947,
      "loss": 4.6366,
      "mean_token_accuracy": 0.23951656967401505,
      "num_tokens": 136255067.0,
      "step": 78560
    },
    {
      "entropy": 5.3613770484924315,
      "epoch": 6.112585100175063,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019994919443912514,
      "loss": 4.6067,
      "mean_token_accuracy": 0.24936345219612122,
      "num_tokens": 136264089.0,
      "step": 78565
    },
    {
      "entropy": 5.400396347045898,
      "epoch": 6.112974129546781,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019992306671543213,
      "loss": 4.6203,
      "mean_token_accuracy": 0.23989343494176865,
      "num_tokens": 136272115.0,
      "step": 78570
    },
    {
      "entropy": 5.407944011688232,
      "epoch": 6.1133631589184985,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019989694013091214,
      "loss": 4.5682,
      "mean_token_accuracy": 0.2481945797801018,
      "num_tokens": 136280468.0,
      "step": 78575
    },
    {
      "entropy": 5.375766801834106,
      "epoch": 6.113752188290216,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00019987081468596146,
      "loss": 4.6496,
      "mean_token_accuracy": 0.23621022701263428,
      "num_tokens": 136288834.0,
      "step": 78580
    },
    {
      "entropy": 5.251183986663818,
      "epoch": 6.114141217661934,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019984469038097658,
      "loss": 4.4715,
      "mean_token_accuracy": 0.25766897201538086,
      "num_tokens": 136297487.0,
      "step": 78585
    },
    {
      "entropy": 5.378200960159302,
      "epoch": 6.114530247033651,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000199818567216354,
      "loss": 4.5597,
      "mean_token_accuracy": 0.24707991927862166,
      "num_tokens": 136305606.0,
      "step": 78590
    },
    {
      "entropy": 5.409097480773926,
      "epoch": 6.114919276405368,
      "grad_norm": 1.28125,
      "learning_rate": 0.00019979244519248996,
      "loss": 4.6916,
      "mean_token_accuracy": 0.24260441362857818,
      "num_tokens": 136314306.0,
      "step": 78595
    },
    {
      "entropy": 5.314045667648315,
      "epoch": 6.115308305777086,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001997663243097808,
      "loss": 4.5545,
      "mean_token_accuracy": 0.2475971356034279,
      "num_tokens": 136323285.0,
      "step": 78600
    },
    {
      "entropy": 5.398833703994751,
      "epoch": 6.115697335148804,
      "grad_norm": 1.375,
      "learning_rate": 0.00019974020456862295,
      "loss": 4.6802,
      "mean_token_accuracy": 0.2368849530816078,
      "num_tokens": 136331885.0,
      "step": 78605
    },
    {
      "entropy": 5.429479455947876,
      "epoch": 6.1160863645205215,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019971408596941283,
      "loss": 4.6957,
      "mean_token_accuracy": 0.23618034273386002,
      "num_tokens": 136340356.0,
      "step": 78610
    },
    {
      "entropy": 5.465973234176635,
      "epoch": 6.116475393892239,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001996879685125465,
      "loss": 4.7638,
      "mean_token_accuracy": 0.23170444667339324,
      "num_tokens": 136348780.0,
      "step": 78615
    },
    {
      "entropy": 5.3300282001495365,
      "epoch": 6.116864423263957,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019966185219842038,
      "loss": 4.5135,
      "mean_token_accuracy": 0.2544509008526802,
      "num_tokens": 136358142.0,
      "step": 78620
    },
    {
      "entropy": 5.3742622375488285,
      "epoch": 6.117253452635674,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019963573702743077,
      "loss": 4.6213,
      "mean_token_accuracy": 0.24851785153150557,
      "num_tokens": 136367105.0,
      "step": 78625
    },
    {
      "entropy": 5.384827327728272,
      "epoch": 6.117642482007391,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019960962299997394,
      "loss": 4.6863,
      "mean_token_accuracy": 0.2396633118391037,
      "num_tokens": 136376469.0,
      "step": 78630
    },
    {
      "entropy": 5.42590012550354,
      "epoch": 6.118031511379109,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019958351011644605,
      "loss": 4.6558,
      "mean_token_accuracy": 0.23362908661365508,
      "num_tokens": 136384755.0,
      "step": 78635
    },
    {
      "entropy": 5.390380239486694,
      "epoch": 6.118420540750827,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001995573983772433,
      "loss": 4.6073,
      "mean_token_accuracy": 0.2402822345495224,
      "num_tokens": 136393920.0,
      "step": 78640
    },
    {
      "entropy": 5.4698608875274655,
      "epoch": 6.1188095701225445,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019953128778276197,
      "loss": 4.6866,
      "mean_token_accuracy": 0.22820471078157425,
      "num_tokens": 136402053.0,
      "step": 78645
    },
    {
      "entropy": 5.36732931137085,
      "epoch": 6.119198599494262,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001995051783333982,
      "loss": 4.6305,
      "mean_token_accuracy": 0.23818558752536773,
      "num_tokens": 136411447.0,
      "step": 78650
    },
    {
      "entropy": 5.376296949386597,
      "epoch": 6.119587628865979,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00019947907002954818,
      "loss": 4.5818,
      "mean_token_accuracy": 0.24653354585170745,
      "num_tokens": 136420588.0,
      "step": 78655
    },
    {
      "entropy": 5.2859306812286375,
      "epoch": 6.119976658237697,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019945296287160818,
      "loss": 4.5772,
      "mean_token_accuracy": 0.24933183342218398,
      "num_tokens": 136428935.0,
      "step": 78660
    },
    {
      "entropy": 5.432226228713989,
      "epoch": 6.120365687609414,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001994268568599742,
      "loss": 4.6884,
      "mean_token_accuracy": 0.23573703914880753,
      "num_tokens": 136437990.0,
      "step": 78665
    },
    {
      "entropy": 5.401709747314453,
      "epoch": 6.120754716981132,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019940075199504232,
      "loss": 4.6278,
      "mean_token_accuracy": 0.24394212514162064,
      "num_tokens": 136445819.0,
      "step": 78670
    },
    {
      "entropy": 5.400432348251343,
      "epoch": 6.12114374635285,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019937464827720874,
      "loss": 4.6936,
      "mean_token_accuracy": 0.2342011660337448,
      "num_tokens": 136454902.0,
      "step": 78675
    },
    {
      "entropy": 5.455522346496582,
      "epoch": 6.121532775724567,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019934854570686944,
      "loss": 4.6604,
      "mean_token_accuracy": 0.23763135075569153,
      "num_tokens": 136463314.0,
      "step": 78680
    },
    {
      "entropy": 5.425014162063599,
      "epoch": 6.121921805096285,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019932244428442065,
      "loss": 4.6955,
      "mean_token_accuracy": 0.23292257636785507,
      "num_tokens": 136471903.0,
      "step": 78685
    },
    {
      "entropy": 5.463228178024292,
      "epoch": 6.122310834468002,
      "grad_norm": 1.234375,
      "learning_rate": 0.00019929634401025825,
      "loss": 4.7576,
      "mean_token_accuracy": 0.22872737050056458,
      "num_tokens": 136480955.0,
      "step": 78690
    },
    {
      "entropy": 5.424966955184937,
      "epoch": 6.12269986383972,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001992702448847783,
      "loss": 4.6785,
      "mean_token_accuracy": 0.23830610662698745,
      "num_tokens": 136489974.0,
      "step": 78695
    },
    {
      "entropy": 5.346819877624512,
      "epoch": 6.123088893211437,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019924414690837688,
      "loss": 4.6012,
      "mean_token_accuracy": 0.24029412716627122,
      "num_tokens": 136499111.0,
      "step": 78700
    },
    {
      "entropy": 5.333548498153687,
      "epoch": 6.123477922583155,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019921805008145,
      "loss": 4.612,
      "mean_token_accuracy": 0.24185976535081863,
      "num_tokens": 136508077.0,
      "step": 78705
    },
    {
      "entropy": 5.411976385116577,
      "epoch": 6.123866951954873,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019919195440439358,
      "loss": 4.6457,
      "mean_token_accuracy": 0.23717589378356935,
      "num_tokens": 136517138.0,
      "step": 78710
    },
    {
      "entropy": 5.437945556640625,
      "epoch": 6.12425598132659,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019916585987760356,
      "loss": 4.6711,
      "mean_token_accuracy": 0.23220251202583314,
      "num_tokens": 136526112.0,
      "step": 78715
    },
    {
      "entropy": 5.467208909988403,
      "epoch": 6.124645010698308,
      "grad_norm": 1.375,
      "learning_rate": 0.00019913976650147585,
      "loss": 4.7084,
      "mean_token_accuracy": 0.2441462278366089,
      "num_tokens": 136535138.0,
      "step": 78720
    },
    {
      "entropy": 5.376082277297973,
      "epoch": 6.125034040070025,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019911367427640647,
      "loss": 4.5849,
      "mean_token_accuracy": 0.2531227692961693,
      "num_tokens": 136543550.0,
      "step": 78725
    },
    {
      "entropy": 5.432673120498658,
      "epoch": 6.125423069441743,
      "grad_norm": 1.265625,
      "learning_rate": 0.00019908758320279134,
      "loss": 4.74,
      "mean_token_accuracy": 0.22677767872810364,
      "num_tokens": 136551892.0,
      "step": 78730
    },
    {
      "entropy": 5.448094463348388,
      "epoch": 6.12581209881346,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019906149328102635,
      "loss": 4.7245,
      "mean_token_accuracy": 0.23276785910129547,
      "num_tokens": 136560575.0,
      "step": 78735
    },
    {
      "entropy": 5.386369752883911,
      "epoch": 6.126201128185178,
      "grad_norm": 1.234375,
      "learning_rate": 0.00019903540451150732,
      "loss": 4.6916,
      "mean_token_accuracy": 0.24483714401721954,
      "num_tokens": 136569388.0,
      "step": 78740
    },
    {
      "entropy": 5.364436054229737,
      "epoch": 6.126590157556896,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019900931689463008,
      "loss": 4.6121,
      "mean_token_accuracy": 0.2412711039185524,
      "num_tokens": 136576780.0,
      "step": 78745
    },
    {
      "entropy": 5.439519548416138,
      "epoch": 6.126979186928613,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019898323043079054,
      "loss": 4.7164,
      "mean_token_accuracy": 0.2306882008910179,
      "num_tokens": 136584966.0,
      "step": 78750
    },
    {
      "entropy": 5.347755432128906,
      "epoch": 6.127368216300331,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019895714512038436,
      "loss": 4.5882,
      "mean_token_accuracy": 0.24828485995531083,
      "num_tokens": 136592783.0,
      "step": 78755
    },
    {
      "entropy": 5.474159049987793,
      "epoch": 6.127757245672048,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019893106096380763,
      "loss": 4.741,
      "mean_token_accuracy": 0.2391098454594612,
      "num_tokens": 136602247.0,
      "step": 78760
    },
    {
      "entropy": 5.417570304870606,
      "epoch": 6.128146275043766,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019890497796145584,
      "loss": 4.6281,
      "mean_token_accuracy": 0.2407189592719078,
      "num_tokens": 136611333.0,
      "step": 78765
    },
    {
      "entropy": 5.397199440002441,
      "epoch": 6.128535304415483,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019887889611372502,
      "loss": 4.6675,
      "mean_token_accuracy": 0.2364713355898857,
      "num_tokens": 136620568.0,
      "step": 78770
    },
    {
      "entropy": 5.357023477554321,
      "epoch": 6.128924333787201,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001988528154210107,
      "loss": 4.6811,
      "mean_token_accuracy": 0.23577784299850463,
      "num_tokens": 136629814.0,
      "step": 78775
    },
    {
      "entropy": 5.414559745788575,
      "epoch": 6.129313363158919,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001988267358837087,
      "loss": 4.6257,
      "mean_token_accuracy": 0.2457575261592865,
      "num_tokens": 136637677.0,
      "step": 78780
    },
    {
      "entropy": 5.371972990036011,
      "epoch": 6.129702392530636,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001988006575022147,
      "loss": 4.5924,
      "mean_token_accuracy": 0.24243347942829133,
      "num_tokens": 136646089.0,
      "step": 78785
    },
    {
      "entropy": 5.364004421234131,
      "epoch": 6.130091421902353,
      "grad_norm": 1.234375,
      "learning_rate": 0.00019877458027692458,
      "loss": 4.5982,
      "mean_token_accuracy": 0.24705359488725662,
      "num_tokens": 136654593.0,
      "step": 78790
    },
    {
      "entropy": 5.430990219116211,
      "epoch": 6.130480451274071,
      "grad_norm": 1.375,
      "learning_rate": 0.00019874850420823376,
      "loss": 4.6963,
      "mean_token_accuracy": 0.23520189821720122,
      "num_tokens": 136662487.0,
      "step": 78795
    },
    {
      "entropy": 5.305890941619873,
      "epoch": 6.130869480645789,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019872242929653812,
      "loss": 4.5721,
      "mean_token_accuracy": 0.24916601777076722,
      "num_tokens": 136671008.0,
      "step": 78800
    },
    {
      "entropy": 5.403901863098144,
      "epoch": 6.131258510017506,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019869635554223313,
      "loss": 4.6991,
      "mean_token_accuracy": 0.2359052374958992,
      "num_tokens": 136680627.0,
      "step": 78805
    },
    {
      "entropy": 5.377779388427735,
      "epoch": 6.131647539389224,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019867028294571454,
      "loss": 4.5645,
      "mean_token_accuracy": 0.2528917476534843,
      "num_tokens": 136688777.0,
      "step": 78810
    },
    {
      "entropy": 5.367270040512085,
      "epoch": 6.132036568760942,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019864421150737787,
      "loss": 4.6066,
      "mean_token_accuracy": 0.2446463868021965,
      "num_tokens": 136698245.0,
      "step": 78815
    },
    {
      "entropy": 5.424227476119995,
      "epoch": 6.132425598132659,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019861814122761883,
      "loss": 4.6571,
      "mean_token_accuracy": 0.24348833858966829,
      "num_tokens": 136707422.0,
      "step": 78820
    },
    {
      "entropy": 5.313541030883789,
      "epoch": 6.132814627504376,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001985920721068329,
      "loss": 4.5704,
      "mean_token_accuracy": 0.2517102718353271,
      "num_tokens": 136716271.0,
      "step": 78825
    },
    {
      "entropy": 5.461903238296509,
      "epoch": 6.133203656876094,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019856600414541566,
      "loss": 4.7452,
      "mean_token_accuracy": 0.22699992805719377,
      "num_tokens": 136724671.0,
      "step": 78830
    },
    {
      "entropy": 5.344994926452637,
      "epoch": 6.133592686247812,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019853993734376263,
      "loss": 4.5999,
      "mean_token_accuracy": 0.24475709050893785,
      "num_tokens": 136733421.0,
      "step": 78835
    },
    {
      "entropy": 5.317690134048462,
      "epoch": 6.133981715619529,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019851387170226936,
      "loss": 4.5444,
      "mean_token_accuracy": 0.2592227280139923,
      "num_tokens": 136741906.0,
      "step": 78840
    },
    {
      "entropy": 5.369586324691772,
      "epoch": 6.134370744991247,
      "grad_norm": 1.375,
      "learning_rate": 0.0001984878072213313,
      "loss": 4.6155,
      "mean_token_accuracy": 0.24382208436727523,
      "num_tokens": 136750519.0,
      "step": 78845
    },
    {
      "entropy": 5.467478036880493,
      "epoch": 6.134759774362965,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019846174390134396,
      "loss": 4.6685,
      "mean_token_accuracy": 0.2351030796766281,
      "num_tokens": 136758654.0,
      "step": 78850
    },
    {
      "entropy": 5.396537113189697,
      "epoch": 6.1351488037346815,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019843568174270293,
      "loss": 4.6888,
      "mean_token_accuracy": 0.24944429397583007,
      "num_tokens": 136767423.0,
      "step": 78855
    },
    {
      "entropy": 5.362281370162964,
      "epoch": 6.135537833106399,
      "grad_norm": 1.5,
      "learning_rate": 0.00019840962074580348,
      "loss": 4.677,
      "mean_token_accuracy": 0.24165502786636353,
      "num_tokens": 136776333.0,
      "step": 78860
    },
    {
      "entropy": 5.401207113265992,
      "epoch": 6.135926862478117,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001983835609110412,
      "loss": 4.6538,
      "mean_token_accuracy": 0.24165072590112685,
      "num_tokens": 136785006.0,
      "step": 78865
    },
    {
      "entropy": 5.326067495346069,
      "epoch": 6.136315891849835,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019835750223881144,
      "loss": 4.5693,
      "mean_token_accuracy": 0.24751589447259903,
      "num_tokens": 136793773.0,
      "step": 78870
    },
    {
      "entropy": 5.346798372268677,
      "epoch": 6.136704921221552,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019833144472950953,
      "loss": 4.6725,
      "mean_token_accuracy": 0.24423033744096756,
      "num_tokens": 136802485.0,
      "step": 78875
    },
    {
      "entropy": 5.432577657699585,
      "epoch": 6.13709395059327,
      "grad_norm": 1.421875,
      "learning_rate": 0.000198305388383531,
      "loss": 4.6888,
      "mean_token_accuracy": 0.23250516206026078,
      "num_tokens": 136810873.0,
      "step": 78880
    },
    {
      "entropy": 5.441373205184936,
      "epoch": 6.137482979964988,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019827933320127106,
      "loss": 4.7249,
      "mean_token_accuracy": 0.22696203142404556,
      "num_tokens": 136818717.0,
      "step": 78885
    },
    {
      "entropy": 5.448186302185059,
      "epoch": 6.1378720093367045,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019825327918312515,
      "loss": 4.7247,
      "mean_token_accuracy": 0.22789814025163652,
      "num_tokens": 136827462.0,
      "step": 78890
    },
    {
      "entropy": 5.485416555404663,
      "epoch": 6.138261038708422,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019822722632948858,
      "loss": 4.7347,
      "mean_token_accuracy": 0.2327155962586403,
      "num_tokens": 136836256.0,
      "step": 78895
    },
    {
      "entropy": 5.474920082092285,
      "epoch": 6.13865006808014,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001982011746407567,
      "loss": 4.6989,
      "mean_token_accuracy": 0.2365957260131836,
      "num_tokens": 136844124.0,
      "step": 78900
    },
    {
      "entropy": 5.430308103561401,
      "epoch": 6.139039097451858,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001981751241173247,
      "loss": 4.6869,
      "mean_token_accuracy": 0.24781902730464936,
      "num_tokens": 136853240.0,
      "step": 78905
    },
    {
      "entropy": 5.438841056823731,
      "epoch": 6.139428126823575,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019814907475958793,
      "loss": 4.6477,
      "mean_token_accuracy": 0.23496609330177307,
      "num_tokens": 136861806.0,
      "step": 78910
    },
    {
      "entropy": 5.448329448699951,
      "epoch": 6.139817156195293,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001981230265679415,
      "loss": 4.6894,
      "mean_token_accuracy": 0.22877499610185623,
      "num_tokens": 136869557.0,
      "step": 78915
    },
    {
      "entropy": 5.449795150756836,
      "epoch": 6.140206185567011,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019809697954278088,
      "loss": 4.7933,
      "mean_token_accuracy": 0.2255861148238182,
      "num_tokens": 136879419.0,
      "step": 78920
    },
    {
      "entropy": 5.4592588424682615,
      "epoch": 6.1405952149387275,
      "grad_norm": 1.25,
      "learning_rate": 0.0001980709336845012,
      "loss": 4.6714,
      "mean_token_accuracy": 0.23592642545700074,
      "num_tokens": 136888290.0,
      "step": 78925
    },
    {
      "entropy": 5.429002857208252,
      "epoch": 6.140984244310445,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019804488899349772,
      "loss": 4.7318,
      "mean_token_accuracy": 0.2355785682797432,
      "num_tokens": 136897034.0,
      "step": 78930
    },
    {
      "entropy": 5.425620126724243,
      "epoch": 6.141373273682163,
      "grad_norm": 1.28125,
      "learning_rate": 0.00019801884547016553,
      "loss": 4.6765,
      "mean_token_accuracy": 0.2403073400259018,
      "num_tokens": 136906155.0,
      "step": 78935
    },
    {
      "entropy": 5.522509622573852,
      "epoch": 6.141762303053881,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001979928031148998,
      "loss": 4.6987,
      "mean_token_accuracy": 0.23565604388713837,
      "num_tokens": 136914641.0,
      "step": 78940
    },
    {
      "entropy": 5.450054502487182,
      "epoch": 6.142151332425598,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001979667619280957,
      "loss": 4.7438,
      "mean_token_accuracy": 0.23120893388986588,
      "num_tokens": 136922819.0,
      "step": 78945
    },
    {
      "entropy": 5.465826463699341,
      "epoch": 6.142540361797316,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019794072191014838,
      "loss": 4.8755,
      "mean_token_accuracy": 0.22128087133169175,
      "num_tokens": 136931981.0,
      "step": 78950
    },
    {
      "entropy": 5.4688468933105465,
      "epoch": 6.142929391169034,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019791468306145298,
      "loss": 4.7517,
      "mean_token_accuracy": 0.22968698740005494,
      "num_tokens": 136940377.0,
      "step": 78955
    },
    {
      "entropy": 5.518280458450318,
      "epoch": 6.1433184205407505,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001978886453824046,
      "loss": 4.6772,
      "mean_token_accuracy": 0.24540621489286424,
      "num_tokens": 136949042.0,
      "step": 78960
    },
    {
      "entropy": 5.4317841053009035,
      "epoch": 6.143707449912468,
      "grad_norm": 1.25,
      "learning_rate": 0.0001978626088733982,
      "loss": 4.6359,
      "mean_token_accuracy": 0.23654134422540665,
      "num_tokens": 136957625.0,
      "step": 78965
    },
    {
      "entropy": 5.372042608261109,
      "epoch": 6.144096479284186,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019783657353482904,
      "loss": 4.5659,
      "mean_token_accuracy": 0.2506356716156006,
      "num_tokens": 136966915.0,
      "step": 78970
    },
    {
      "entropy": 5.350477886199951,
      "epoch": 6.144485508655904,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019781053936709198,
      "loss": 4.5745,
      "mean_token_accuracy": 0.24752390533685684,
      "num_tokens": 136975850.0,
      "step": 78975
    },
    {
      "entropy": 5.343332958221436,
      "epoch": 6.144874538027621,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019778450637058216,
      "loss": 4.63,
      "mean_token_accuracy": 0.23954952359199524,
      "num_tokens": 136984526.0,
      "step": 78980
    },
    {
      "entropy": 5.396504640579224,
      "epoch": 6.145263567399339,
      "grad_norm": 1.28125,
      "learning_rate": 0.00019775847454569446,
      "loss": 4.5996,
      "mean_token_accuracy": 0.24154503345489503,
      "num_tokens": 136992947.0,
      "step": 78985
    },
    {
      "entropy": 5.363021564483643,
      "epoch": 6.145652596771056,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019773244389282408,
      "loss": 4.6315,
      "mean_token_accuracy": 0.23812874257564545,
      "num_tokens": 137002186.0,
      "step": 78990
    },
    {
      "entropy": 5.3873930931091305,
      "epoch": 6.1460416261427735,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001977064144123658,
      "loss": 4.6354,
      "mean_token_accuracy": 0.2386369675397873,
      "num_tokens": 137010587.0,
      "step": 78995
    },
    {
      "entropy": 5.419131278991699,
      "epoch": 6.146430655514491,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019768038610471473,
      "loss": 4.7393,
      "mean_token_accuracy": 0.2293958529829979,
      "num_tokens": 137018806.0,
      "step": 79000
    },
    {
      "entropy": 5.470091104507446,
      "epoch": 6.146819684886209,
      "grad_norm": 1.28125,
      "learning_rate": 0.00019765435897026575,
      "loss": 4.7306,
      "mean_token_accuracy": 0.2350328490138054,
      "num_tokens": 137027958.0,
      "step": 79005
    },
    {
      "entropy": 5.43677282333374,
      "epoch": 6.147208714257927,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019762833300941373,
      "loss": 4.6542,
      "mean_token_accuracy": 0.24548936188220977,
      "num_tokens": 137036257.0,
      "step": 79010
    },
    {
      "entropy": 5.397232294082642,
      "epoch": 6.147597743629644,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019760230822255358,
      "loss": 4.6355,
      "mean_token_accuracy": 0.24341279566287993,
      "num_tokens": 137045098.0,
      "step": 79015
    },
    {
      "entropy": 5.417088460922241,
      "epoch": 6.147986773001362,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019757628461008025,
      "loss": 4.6177,
      "mean_token_accuracy": 0.24664041250944138,
      "num_tokens": 137053997.0,
      "step": 79020
    },
    {
      "entropy": 5.36372594833374,
      "epoch": 6.148375802373079,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001975502621723886,
      "loss": 4.624,
      "mean_token_accuracy": 0.2483044907450676,
      "num_tokens": 137062953.0,
      "step": 79025
    },
    {
      "entropy": 5.29890809059143,
      "epoch": 6.1487648317447965,
      "grad_norm": 1.28125,
      "learning_rate": 0.00019752424090987338,
      "loss": 4.4874,
      "mean_token_accuracy": 0.2565107852220535,
      "num_tokens": 137071563.0,
      "step": 79030
    },
    {
      "entropy": 5.346020936965942,
      "epoch": 6.149153861116514,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019749822082292947,
      "loss": 4.5628,
      "mean_token_accuracy": 0.24831537753343583,
      "num_tokens": 137079372.0,
      "step": 79035
    },
    {
      "entropy": 5.435690402984619,
      "epoch": 6.149542890488232,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019747220191195175,
      "loss": 4.7064,
      "mean_token_accuracy": 0.24443809688091278,
      "num_tokens": 137087646.0,
      "step": 79040
    },
    {
      "entropy": 5.326570749282837,
      "epoch": 6.14993191985995,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019744618417733494,
      "loss": 4.612,
      "mean_token_accuracy": 0.2442912682890892,
      "num_tokens": 137095899.0,
      "step": 79045
    },
    {
      "entropy": 5.439183235168457,
      "epoch": 6.150320949231667,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001974201676194738,
      "loss": 4.6825,
      "mean_token_accuracy": 0.23681214898824693,
      "num_tokens": 137104479.0,
      "step": 79050
    },
    {
      "entropy": 5.402071285247803,
      "epoch": 6.150709978603385,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019739415223876322,
      "loss": 4.7507,
      "mean_token_accuracy": 0.22732164114713668,
      "num_tokens": 137112950.0,
      "step": 79055
    },
    {
      "entropy": 5.326495170593262,
      "epoch": 6.151099007975102,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019736813803559783,
      "loss": 4.5343,
      "mean_token_accuracy": 0.24398083239793777,
      "num_tokens": 137122407.0,
      "step": 79060
    },
    {
      "entropy": 5.443326663970947,
      "epoch": 6.1514880373468195,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001973421250103724,
      "loss": 4.7241,
      "mean_token_accuracy": 0.22959715723991395,
      "num_tokens": 137131172.0,
      "step": 79065
    },
    {
      "entropy": 5.295599126815796,
      "epoch": 6.151877066718537,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019731611316348158,
      "loss": 4.5872,
      "mean_token_accuracy": 0.2469363957643509,
      "num_tokens": 137139393.0,
      "step": 79070
    },
    {
      "entropy": 5.333982372283936,
      "epoch": 6.152266096090255,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019729010249532015,
      "loss": 4.6813,
      "mean_token_accuracy": 0.2387621209025383,
      "num_tokens": 137149154.0,
      "step": 79075
    },
    {
      "entropy": 5.380866670608521,
      "epoch": 6.152655125461973,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019726409300628266,
      "loss": 4.631,
      "mean_token_accuracy": 0.2356287121772766,
      "num_tokens": 137157500.0,
      "step": 79080
    },
    {
      "entropy": 5.3818169116973875,
      "epoch": 6.15304415483369,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019723808469676384,
      "loss": 4.5801,
      "mean_token_accuracy": 0.24681150317192077,
      "num_tokens": 137166388.0,
      "step": 79085
    },
    {
      "entropy": 5.4126551151275635,
      "epoch": 6.153433184205407,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019721207756715842,
      "loss": 4.5855,
      "mean_token_accuracy": 0.24178925305604934,
      "num_tokens": 137175174.0,
      "step": 79090
    },
    {
      "entropy": 5.45515718460083,
      "epoch": 6.153822213577125,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019718607161786083,
      "loss": 4.7633,
      "mean_token_accuracy": 0.23033933937549592,
      "num_tokens": 137184334.0,
      "step": 79095
    },
    {
      "entropy": 5.459223699569702,
      "epoch": 6.1542112429488425,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001971600668492658,
      "loss": 4.7423,
      "mean_token_accuracy": 0.2361503705382347,
      "num_tokens": 137192543.0,
      "step": 79100
    },
    {
      "entropy": 5.4387311935424805,
      "epoch": 6.15460027232056,
      "grad_norm": 1.21875,
      "learning_rate": 0.00019713406326176775,
      "loss": 4.6493,
      "mean_token_accuracy": 0.2353041484951973,
      "num_tokens": 137200992.0,
      "step": 79105
    },
    {
      "entropy": 5.405796337127685,
      "epoch": 6.154989301692278,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00019710806085576144,
      "loss": 4.6241,
      "mean_token_accuracy": 0.24453434199094773,
      "num_tokens": 137209093.0,
      "step": 79110
    },
    {
      "entropy": 5.420788288116455,
      "epoch": 6.155378331063996,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001970820596316412,
      "loss": 4.6774,
      "mean_token_accuracy": 0.23129548728466034,
      "num_tokens": 137217500.0,
      "step": 79115
    },
    {
      "entropy": 5.503490686416626,
      "epoch": 6.155767360435713,
      "grad_norm": 1.25,
      "learning_rate": 0.00019705605958980182,
      "loss": 4.79,
      "mean_token_accuracy": 0.2300945922732353,
      "num_tokens": 137226642.0,
      "step": 79120
    },
    {
      "entropy": 5.516183948516845,
      "epoch": 6.15615638980743,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019703006073063767,
      "loss": 4.7022,
      "mean_token_accuracy": 0.241116301715374,
      "num_tokens": 137234893.0,
      "step": 79125
    },
    {
      "entropy": 5.334193992614746,
      "epoch": 6.156545419179148,
      "grad_norm": 1.28125,
      "learning_rate": 0.00019700406305454322,
      "loss": 4.56,
      "mean_token_accuracy": 0.24646274149417877,
      "num_tokens": 137243524.0,
      "step": 79130
    },
    {
      "entropy": 5.424464130401612,
      "epoch": 6.1569344485508655,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019697806656191296,
      "loss": 4.7139,
      "mean_token_accuracy": 0.23481153100728988,
      "num_tokens": 137251277.0,
      "step": 79135
    },
    {
      "entropy": 5.38928713798523,
      "epoch": 6.157323477922583,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001969520712531414,
      "loss": 4.5558,
      "mean_token_accuracy": 0.25526061207056044,
      "num_tokens": 137260032.0,
      "step": 79140
    },
    {
      "entropy": 5.343547201156616,
      "epoch": 6.157712507294301,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019692607712862287,
      "loss": 4.6324,
      "mean_token_accuracy": 0.2342696711421013,
      "num_tokens": 137268506.0,
      "step": 79145
    },
    {
      "entropy": 5.305695056915283,
      "epoch": 6.158101536666019,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019690008418875188,
      "loss": 4.5899,
      "mean_token_accuracy": 0.2441267728805542,
      "num_tokens": 137277054.0,
      "step": 79150
    },
    {
      "entropy": 5.444936466217041,
      "epoch": 6.158490566037736,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019687409243392278,
      "loss": 4.7466,
      "mean_token_accuracy": 0.2337896093726158,
      "num_tokens": 137285293.0,
      "step": 79155
    },
    {
      "entropy": 5.371608448028565,
      "epoch": 6.158879595409453,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019684810186453,
      "loss": 4.6208,
      "mean_token_accuracy": 0.24750885963439942,
      "num_tokens": 137294367.0,
      "step": 79160
    },
    {
      "entropy": 5.50307297706604,
      "epoch": 6.159268624781171,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019682211248096793,
      "loss": 4.7344,
      "mean_token_accuracy": 0.23041231632232667,
      "num_tokens": 137303050.0,
      "step": 79165
    },
    {
      "entropy": 5.405861282348633,
      "epoch": 6.1596576541528885,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001967961242836308,
      "loss": 4.6182,
      "mean_token_accuracy": 0.24997203797101974,
      "num_tokens": 137311225.0,
      "step": 79170
    },
    {
      "entropy": 5.42248420715332,
      "epoch": 6.160046683524606,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001967701372729131,
      "loss": 4.6814,
      "mean_token_accuracy": 0.23063574731349945,
      "num_tokens": 137320629.0,
      "step": 79175
    },
    {
      "entropy": 5.466156530380249,
      "epoch": 6.160435712896324,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019674415144920894,
      "loss": 4.7085,
      "mean_token_accuracy": 0.2359488919377327,
      "num_tokens": 137329661.0,
      "step": 79180
    },
    {
      "entropy": 5.442575740814209,
      "epoch": 6.1608247422680416,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001967181668129127,
      "loss": 4.6845,
      "mean_token_accuracy": 0.23231832236051558,
      "num_tokens": 137337655.0,
      "step": 79185
    },
    {
      "entropy": 5.417653560638428,
      "epoch": 6.161213771639758,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019669218336441875,
      "loss": 4.686,
      "mean_token_accuracy": 0.23088178038597107,
      "num_tokens": 137346407.0,
      "step": 79190
    },
    {
      "entropy": 5.300498676300049,
      "epoch": 6.161602801011476,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019666620110412125,
      "loss": 4.533,
      "mean_token_accuracy": 0.24398079067468642,
      "num_tokens": 137354983.0,
      "step": 79195
    },
    {
      "entropy": 5.431863737106323,
      "epoch": 6.161991830383194,
      "grad_norm": 1.234375,
      "learning_rate": 0.00019664022003241456,
      "loss": 4.7401,
      "mean_token_accuracy": 0.23567648082971573,
      "num_tokens": 137364568.0,
      "step": 79200
    },
    {
      "entropy": 5.386041784286499,
      "epoch": 6.1623808597549115,
      "grad_norm": 1.28125,
      "learning_rate": 0.00019661424014969276,
      "loss": 4.6233,
      "mean_token_accuracy": 0.23776525557041167,
      "num_tokens": 137373132.0,
      "step": 79205
    },
    {
      "entropy": 5.389340162277222,
      "epoch": 6.162769889126629,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019658826145635007,
      "loss": 4.646,
      "mean_token_accuracy": 0.23617079555988313,
      "num_tokens": 137382051.0,
      "step": 79210
    },
    {
      "entropy": 5.338533067703247,
      "epoch": 6.163158918498347,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019656228395278076,
      "loss": 4.6032,
      "mean_token_accuracy": 0.24431201815605164,
      "num_tokens": 137390144.0,
      "step": 79215
    },
    {
      "entropy": 5.401076936721802,
      "epoch": 6.1635479478700645,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001965363076393789,
      "loss": 4.6879,
      "mean_token_accuracy": 0.23491137623786926,
      "num_tokens": 137398313.0,
      "step": 79220
    },
    {
      "entropy": 5.346828031539917,
      "epoch": 6.163936977241781,
      "grad_norm": 1.28125,
      "learning_rate": 0.00019651033251653866,
      "loss": 4.6427,
      "mean_token_accuracy": 0.24152909070253373,
      "num_tokens": 137407081.0,
      "step": 79225
    },
    {
      "entropy": 5.355836200714111,
      "epoch": 6.164326006613499,
      "grad_norm": 1.375,
      "learning_rate": 0.00019648435858465424,
      "loss": 4.6818,
      "mean_token_accuracy": 0.23410223722457885,
      "num_tokens": 137415448.0,
      "step": 79230
    },
    {
      "entropy": 5.441721105575562,
      "epoch": 6.164715035985217,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019645838584411967,
      "loss": 4.7636,
      "mean_token_accuracy": 0.23450849950313568,
      "num_tokens": 137424710.0,
      "step": 79235
    },
    {
      "entropy": 5.409655046463013,
      "epoch": 6.1651040653569344,
      "grad_norm": 1.25,
      "learning_rate": 0.00019643241429532916,
      "loss": 4.6934,
      "mean_token_accuracy": 0.2325771078467369,
      "num_tokens": 137434472.0,
      "step": 79240
    },
    {
      "entropy": 5.349524974822998,
      "epoch": 6.165493094728652,
      "grad_norm": 1.1875,
      "learning_rate": 0.00019640644393867663,
      "loss": 4.5941,
      "mean_token_accuracy": 0.24110008776187897,
      "num_tokens": 137442452.0,
      "step": 79245
    },
    {
      "entropy": 5.337251424789429,
      "epoch": 6.16588212410037,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019638047477455616,
      "loss": 4.6419,
      "mean_token_accuracy": 0.2397637441754341,
      "num_tokens": 137451758.0,
      "step": 79250
    },
    {
      "entropy": 5.408027505874633,
      "epoch": 6.1662711534720875,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019635450680336198,
      "loss": 4.733,
      "mean_token_accuracy": 0.23420953303575515,
      "num_tokens": 137460335.0,
      "step": 79255
    },
    {
      "entropy": 5.383549737930298,
      "epoch": 6.166660182843804,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019632854002548796,
      "loss": 4.6157,
      "mean_token_accuracy": 0.23964510709047318,
      "num_tokens": 137468113.0,
      "step": 79260
    },
    {
      "entropy": 5.439716815948486,
      "epoch": 6.167049212215522,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019630257444132803,
      "loss": 4.722,
      "mean_token_accuracy": 0.23583550453186036,
      "num_tokens": 137477277.0,
      "step": 79265
    },
    {
      "entropy": 5.372001218795776,
      "epoch": 6.16743824158724,
      "grad_norm": 1.375,
      "learning_rate": 0.00019627661005127634,
      "loss": 4.7296,
      "mean_token_accuracy": 0.23639536052942275,
      "num_tokens": 137486033.0,
      "step": 79270
    },
    {
      "entropy": 5.428027868270874,
      "epoch": 6.167827270958957,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001962506468557268,
      "loss": 4.7002,
      "mean_token_accuracy": 0.24111622273921968,
      "num_tokens": 137494414.0,
      "step": 79275
    },
    {
      "entropy": 5.322297477722168,
      "epoch": 6.168216300330675,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001962246848550734,
      "loss": 4.5601,
      "mean_token_accuracy": 0.24543844014406205,
      "num_tokens": 137502173.0,
      "step": 79280
    },
    {
      "entropy": 5.404531812667846,
      "epoch": 6.168605329702393,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019619872404971002,
      "loss": 4.6391,
      "mean_token_accuracy": 0.2438809171319008,
      "num_tokens": 137511028.0,
      "step": 79285
    },
    {
      "entropy": 5.363803958892822,
      "epoch": 6.1689943590741105,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019617276444003053,
      "loss": 4.6261,
      "mean_token_accuracy": 0.23028571903705597,
      "num_tokens": 137519415.0,
      "step": 79290
    },
    {
      "entropy": 5.371300411224365,
      "epoch": 6.169383388445827,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00019614680602642886,
      "loss": 4.6002,
      "mean_token_accuracy": 0.24473311603069306,
      "num_tokens": 137527957.0,
      "step": 79295
    },
    {
      "entropy": 5.362251901626587,
      "epoch": 6.169772417817545,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019612084880929888,
      "loss": 4.5941,
      "mean_token_accuracy": 0.24410882443189622,
      "num_tokens": 137535965.0,
      "step": 79300
    },
    {
      "entropy": 5.397421169281006,
      "epoch": 6.170161447189263,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019609489278903452,
      "loss": 4.6896,
      "mean_token_accuracy": 0.23279963433742523,
      "num_tokens": 137545051.0,
      "step": 79305
    },
    {
      "entropy": 5.401891231536865,
      "epoch": 6.17055047656098,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019606893796602953,
      "loss": 4.6485,
      "mean_token_accuracy": 0.23451224565505982,
      "num_tokens": 137553451.0,
      "step": 79310
    },
    {
      "entropy": 5.426145553588867,
      "epoch": 6.170939505932698,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001960429843406778,
      "loss": 4.7048,
      "mean_token_accuracy": 0.2308053269982338,
      "num_tokens": 137562424.0,
      "step": 79315
    },
    {
      "entropy": 5.3988243579864506,
      "epoch": 6.171328535304416,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001960170319133731,
      "loss": 4.6822,
      "mean_token_accuracy": 0.2299165576696396,
      "num_tokens": 137570722.0,
      "step": 79320
    },
    {
      "entropy": 5.399321556091309,
      "epoch": 6.171717564676133,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019599108068450922,
      "loss": 4.6463,
      "mean_token_accuracy": 0.2440319299697876,
      "num_tokens": 137579544.0,
      "step": 79325
    },
    {
      "entropy": 5.401704025268555,
      "epoch": 6.17210659404785,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019596513065447995,
      "loss": 4.6344,
      "mean_token_accuracy": 0.24342752397060394,
      "num_tokens": 137587993.0,
      "step": 79330
    },
    {
      "entropy": 5.421851873397827,
      "epoch": 6.172495623419568,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019593918182367903,
      "loss": 4.6799,
      "mean_token_accuracy": 0.24177489280700684,
      "num_tokens": 137596665.0,
      "step": 79335
    },
    {
      "entropy": 5.355125522613525,
      "epoch": 6.172884652791286,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019591323419250017,
      "loss": 4.5972,
      "mean_token_accuracy": 0.24165622591972352,
      "num_tokens": 137605265.0,
      "step": 79340
    },
    {
      "entropy": 5.339280080795288,
      "epoch": 6.173273682163003,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019588728776133714,
      "loss": 4.6183,
      "mean_token_accuracy": 0.23938302099704742,
      "num_tokens": 137613950.0,
      "step": 79345
    },
    {
      "entropy": 5.369203329086304,
      "epoch": 6.173662711534721,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019586134253058357,
      "loss": 4.6597,
      "mean_token_accuracy": 0.23759396970272065,
      "num_tokens": 137622753.0,
      "step": 79350
    },
    {
      "entropy": 5.345684814453125,
      "epoch": 6.174051740906439,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019583539850063315,
      "loss": 4.6101,
      "mean_token_accuracy": 0.24625101685523987,
      "num_tokens": 137631401.0,
      "step": 79355
    },
    {
      "entropy": 5.351523685455322,
      "epoch": 6.174440770278156,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019580945567187956,
      "loss": 4.5662,
      "mean_token_accuracy": 0.24811996072530745,
      "num_tokens": 137640368.0,
      "step": 79360
    },
    {
      "entropy": 5.386639738082886,
      "epoch": 6.174829799649873,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001957835140447165,
      "loss": 4.6927,
      "mean_token_accuracy": 0.23124420195817946,
      "num_tokens": 137649327.0,
      "step": 79365
    },
    {
      "entropy": 5.408903884887695,
      "epoch": 6.175218829021591,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001957575736195375,
      "loss": 4.6785,
      "mean_token_accuracy": 0.24356093555688857,
      "num_tokens": 137657225.0,
      "step": 79370
    },
    {
      "entropy": 5.46279935836792,
      "epoch": 6.175607858393309,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001957316343967362,
      "loss": 4.7073,
      "mean_token_accuracy": 0.23467136025428773,
      "num_tokens": 137666552.0,
      "step": 79375
    },
    {
      "entropy": 5.334253311157227,
      "epoch": 6.175996887765026,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019570569637670604,
      "loss": 4.5983,
      "mean_token_accuracy": 0.24107507169246672,
      "num_tokens": 137675776.0,
      "step": 79380
    },
    {
      "entropy": 5.364078187942505,
      "epoch": 6.176385917136744,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019567975955984079,
      "loss": 4.6365,
      "mean_token_accuracy": 0.24801819175481796,
      "num_tokens": 137684242.0,
      "step": 79385
    },
    {
      "entropy": 5.356059312820435,
      "epoch": 6.176774946508462,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019565382394653403,
      "loss": 4.677,
      "mean_token_accuracy": 0.239322529733181,
      "num_tokens": 137692724.0,
      "step": 79390
    },
    {
      "entropy": 5.413623094558716,
      "epoch": 6.177163975880179,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001956278895371792,
      "loss": 4.6469,
      "mean_token_accuracy": 0.24042658805847167,
      "num_tokens": 137701373.0,
      "step": 79395
    },
    {
      "entropy": 5.28484616279602,
      "epoch": 6.177553005251896,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019560195633216976,
      "loss": 4.5653,
      "mean_token_accuracy": 0.24396198838949204,
      "num_tokens": 137709751.0,
      "step": 79400
    },
    {
      "entropy": 5.38649091720581,
      "epoch": 6.177942034623614,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019557602433189925,
      "loss": 4.7061,
      "mean_token_accuracy": 0.23197105377912522,
      "num_tokens": 137718928.0,
      "step": 79405
    },
    {
      "entropy": 5.374450445175171,
      "epoch": 6.178331063995332,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019555009353676116,
      "loss": 4.6428,
      "mean_token_accuracy": 0.23928750306367874,
      "num_tokens": 137727157.0,
      "step": 79410
    },
    {
      "entropy": 5.429249811172485,
      "epoch": 6.178720093367049,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019552416394714894,
      "loss": 4.715,
      "mean_token_accuracy": 0.22579919993877412,
      "num_tokens": 137735954.0,
      "step": 79415
    },
    {
      "entropy": 5.44999132156372,
      "epoch": 6.179109122738767,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019549823556345603,
      "loss": 4.7136,
      "mean_token_accuracy": 0.2429208368062973,
      "num_tokens": 137743992.0,
      "step": 79420
    },
    {
      "entropy": 5.473904228210449,
      "epoch": 6.179498152110484,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019547230838607593,
      "loss": 4.7325,
      "mean_token_accuracy": 0.23508021980524063,
      "num_tokens": 137752070.0,
      "step": 79425
    },
    {
      "entropy": 5.379948329925537,
      "epoch": 6.179887181482202,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019544638241540186,
      "loss": 4.6747,
      "mean_token_accuracy": 0.2350342482328415,
      "num_tokens": 137760675.0,
      "step": 79430
    },
    {
      "entropy": 5.424725151062011,
      "epoch": 6.180276210853919,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019542045765182737,
      "loss": 4.7651,
      "mean_token_accuracy": 0.2324303299188614,
      "num_tokens": 137769724.0,
      "step": 79435
    },
    {
      "entropy": 5.462090969085693,
      "epoch": 6.180665240225637,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019539453409574576,
      "loss": 4.7334,
      "mean_token_accuracy": 0.23942847698926925,
      "num_tokens": 137777972.0,
      "step": 79440
    },
    {
      "entropy": 5.376684093475342,
      "epoch": 6.181054269597355,
      "grad_norm": 1.28125,
      "learning_rate": 0.00019536861174755045,
      "loss": 4.5413,
      "mean_token_accuracy": 0.2480020761489868,
      "num_tokens": 137786599.0,
      "step": 79445
    },
    {
      "entropy": 5.45131402015686,
      "epoch": 6.181443298969072,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019534269060763454,
      "loss": 4.632,
      "mean_token_accuracy": 0.23995295017957688,
      "num_tokens": 137794300.0,
      "step": 79450
    },
    {
      "entropy": 5.3206517696380615,
      "epoch": 6.18183232834079,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019531677067639164,
      "loss": 4.55,
      "mean_token_accuracy": 0.24876846373081207,
      "num_tokens": 137802487.0,
      "step": 79455
    },
    {
      "entropy": 5.319809627532959,
      "epoch": 6.182221357712507,
      "grad_norm": 1.234375,
      "learning_rate": 0.00019529085195421492,
      "loss": 4.6964,
      "mean_token_accuracy": 0.23383576422929764,
      "num_tokens": 137811626.0,
      "step": 79460
    },
    {
      "entropy": 5.366305255889893,
      "epoch": 6.182610387084225,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019526493444149763,
      "loss": 4.6429,
      "mean_token_accuracy": 0.24318985342979432,
      "num_tokens": 137820105.0,
      "step": 79465
    },
    {
      "entropy": 5.395595645904541,
      "epoch": 6.182999416455942,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019523901813863303,
      "loss": 4.7296,
      "mean_token_accuracy": 0.22988881319761276,
      "num_tokens": 137829491.0,
      "step": 79470
    },
    {
      "entropy": 5.360602378845215,
      "epoch": 6.18338844582766,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019521310304601447,
      "loss": 4.6283,
      "mean_token_accuracy": 0.2401544541120529,
      "num_tokens": 137837787.0,
      "step": 79475
    },
    {
      "entropy": 5.450046873092651,
      "epoch": 6.183777475199378,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000195187189164035,
      "loss": 4.7226,
      "mean_token_accuracy": 0.230457603931427,
      "num_tokens": 137846272.0,
      "step": 79480
    },
    {
      "entropy": 5.375442361831665,
      "epoch": 6.184166504571095,
      "grad_norm": 1.25,
      "learning_rate": 0.0001951612764930879,
      "loss": 4.6476,
      "mean_token_accuracy": 0.23887607753276824,
      "num_tokens": 137854434.0,
      "step": 79485
    },
    {
      "entropy": 5.2854009628295895,
      "epoch": 6.184555533942813,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00019513536503356637,
      "loss": 4.5122,
      "mean_token_accuracy": 0.24911149889230727,
      "num_tokens": 137863871.0,
      "step": 79490
    },
    {
      "entropy": 5.3375180721282955,
      "epoch": 6.18494456331453,
      "grad_norm": 1.234375,
      "learning_rate": 0.00019510945478586351,
      "loss": 4.6047,
      "mean_token_accuracy": 0.24016215652227402,
      "num_tokens": 137873028.0,
      "step": 79495
    },
    {
      "entropy": 5.385112619400024,
      "epoch": 6.185333592686248,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019508354575037258,
      "loss": 4.693,
      "mean_token_accuracy": 0.23691432178020477,
      "num_tokens": 137881467.0,
      "step": 79500
    },
    {
      "entropy": 5.349452543258667,
      "epoch": 6.185722622057965,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001950576379274866,
      "loss": 4.6223,
      "mean_token_accuracy": 0.23969205021858214,
      "num_tokens": 137890368.0,
      "step": 79505
    },
    {
      "entropy": 5.383006286621094,
      "epoch": 6.186111651429683,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001950317313175987,
      "loss": 4.633,
      "mean_token_accuracy": 0.24486603438854218,
      "num_tokens": 137898830.0,
      "step": 79510
    },
    {
      "entropy": 5.445719051361084,
      "epoch": 6.186500680801401,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019500582592110207,
      "loss": 4.6727,
      "mean_token_accuracy": 0.2389020875096321,
      "num_tokens": 137907402.0,
      "step": 79515
    },
    {
      "entropy": 5.437014627456665,
      "epoch": 6.186889710173118,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019497992173838963,
      "loss": 4.74,
      "mean_token_accuracy": 0.2357453003525734,
      "num_tokens": 137916206.0,
      "step": 79520
    },
    {
      "entropy": 5.462190914154053,
      "epoch": 6.187278739544835,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019495401876985457,
      "loss": 4.6837,
      "mean_token_accuracy": 0.23994932174682618,
      "num_tokens": 137925558.0,
      "step": 79525
    },
    {
      "entropy": 5.411999368667603,
      "epoch": 6.187667768916553,
      "grad_norm": 1.25,
      "learning_rate": 0.00019492811701588982,
      "loss": 4.6518,
      "mean_token_accuracy": 0.24090273827314376,
      "num_tokens": 137934317.0,
      "step": 79530
    },
    {
      "entropy": 5.401230001449585,
      "epoch": 6.188056798288271,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001949022164768885,
      "loss": 4.7002,
      "mean_token_accuracy": 0.2349329560995102,
      "num_tokens": 137942325.0,
      "step": 79535
    },
    {
      "entropy": 5.328482007980346,
      "epoch": 6.188445827659988,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001948763171532435,
      "loss": 4.652,
      "mean_token_accuracy": 0.23770998418331146,
      "num_tokens": 137951076.0,
      "step": 79540
    },
    {
      "entropy": 5.37667932510376,
      "epoch": 6.188834857031706,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001948504190453479,
      "loss": 4.6463,
      "mean_token_accuracy": 0.2392538845539093,
      "num_tokens": 137960060.0,
      "step": 79545
    },
    {
      "entropy": 5.386412954330444,
      "epoch": 6.189223886403424,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019482452215359463,
      "loss": 4.6402,
      "mean_token_accuracy": 0.2414134532213211,
      "num_tokens": 137968780.0,
      "step": 79550
    },
    {
      "entropy": 5.295425319671631,
      "epoch": 6.189612915775141,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00019479862647837665,
      "loss": 4.5269,
      "mean_token_accuracy": 0.2573930099606514,
      "num_tokens": 137977805.0,
      "step": 79555
    },
    {
      "entropy": 5.261436986923218,
      "epoch": 6.190001945146858,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019477273202008678,
      "loss": 4.5922,
      "mean_token_accuracy": 0.2438929334282875,
      "num_tokens": 137986210.0,
      "step": 79560
    },
    {
      "entropy": 5.400560760498047,
      "epoch": 6.190390974518576,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019474683877911808,
      "loss": 4.7467,
      "mean_token_accuracy": 0.23514029681682586,
      "num_tokens": 137994590.0,
      "step": 79565
    },
    {
      "entropy": 5.447697162628174,
      "epoch": 6.190780003890294,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019472094675586333,
      "loss": 4.6207,
      "mean_token_accuracy": 0.24900805503129958,
      "num_tokens": 138002578.0,
      "step": 79570
    },
    {
      "entropy": 5.494809675216675,
      "epoch": 6.191169033262011,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001946950559507154,
      "loss": 4.7143,
      "mean_token_accuracy": 0.23388391137123107,
      "num_tokens": 138011186.0,
      "step": 79575
    },
    {
      "entropy": 5.359992742538452,
      "epoch": 6.191558062633729,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019466916636406718,
      "loss": 4.6051,
      "mean_token_accuracy": 0.24019882678985596,
      "num_tokens": 138020000.0,
      "step": 79580
    },
    {
      "entropy": 5.406331396102905,
      "epoch": 6.191947092005447,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019464327799631153,
      "loss": 4.6009,
      "mean_token_accuracy": 0.2418159916996956,
      "num_tokens": 138028873.0,
      "step": 79585
    },
    {
      "entropy": 5.404008674621582,
      "epoch": 6.192336121377164,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019461739084784124,
      "loss": 4.7031,
      "mean_token_accuracy": 0.2306203305721283,
      "num_tokens": 138037377.0,
      "step": 79590
    },
    {
      "entropy": 5.297249698638916,
      "epoch": 6.192725150748881,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001945915049190491,
      "loss": 4.5167,
      "mean_token_accuracy": 0.254434609413147,
      "num_tokens": 138045701.0,
      "step": 79595
    },
    {
      "entropy": 5.379953384399414,
      "epoch": 6.193114180120599,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019456562021032787,
      "loss": 4.663,
      "mean_token_accuracy": 0.24161863476037979,
      "num_tokens": 138054051.0,
      "step": 79600
    },
    {
      "entropy": 5.417987108230591,
      "epoch": 6.193503209492317,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019453973672207037,
      "loss": 4.7277,
      "mean_token_accuracy": 0.23365566581487657,
      "num_tokens": 138063383.0,
      "step": 79605
    },
    {
      "entropy": 5.439590167999268,
      "epoch": 6.193892238864034,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001945138544546693,
      "loss": 4.6472,
      "mean_token_accuracy": 0.2409509927034378,
      "num_tokens": 138072212.0,
      "step": 79610
    },
    {
      "entropy": 5.348165225982666,
      "epoch": 6.194281268235752,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001944879734085173,
      "loss": 4.517,
      "mean_token_accuracy": 0.25133979320526123,
      "num_tokens": 138080971.0,
      "step": 79615
    },
    {
      "entropy": 5.416686201095581,
      "epoch": 6.19467029760747,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019446209358400723,
      "loss": 4.704,
      "mean_token_accuracy": 0.2385067492723465,
      "num_tokens": 138089581.0,
      "step": 79620
    },
    {
      "entropy": 5.434207582473755,
      "epoch": 6.195059326979187,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001944362149815317,
      "loss": 4.7589,
      "mean_token_accuracy": 0.23387827724218369,
      "num_tokens": 138097910.0,
      "step": 79625
    },
    {
      "entropy": 5.381994533538818,
      "epoch": 6.195448356350904,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019441033760148337,
      "loss": 4.6136,
      "mean_token_accuracy": 0.24343781769275666,
      "num_tokens": 138106884.0,
      "step": 79630
    },
    {
      "entropy": 5.3677996635437015,
      "epoch": 6.195837385722622,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019438446144425493,
      "loss": 4.6326,
      "mean_token_accuracy": 0.2453823834657669,
      "num_tokens": 138115350.0,
      "step": 79635
    },
    {
      "entropy": 5.408381366729737,
      "epoch": 6.19622641509434,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00019435858651023892,
      "loss": 4.7098,
      "mean_token_accuracy": 0.24581923633813857,
      "num_tokens": 138124315.0,
      "step": 79640
    },
    {
      "entropy": 5.36551342010498,
      "epoch": 6.196615444466057,
      "grad_norm": 1.375,
      "learning_rate": 0.00019433271279982795,
      "loss": 4.5971,
      "mean_token_accuracy": 0.24363201707601548,
      "num_tokens": 138132534.0,
      "step": 79645
    },
    {
      "entropy": 5.252813053131104,
      "epoch": 6.197004473837775,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001943068403134147,
      "loss": 4.5615,
      "mean_token_accuracy": 0.24849214106798173,
      "num_tokens": 138141040.0,
      "step": 79650
    },
    {
      "entropy": 5.315002870559693,
      "epoch": 6.197393503209493,
      "grad_norm": 1.265625,
      "learning_rate": 0.00019428096905139175,
      "loss": 4.6091,
      "mean_token_accuracy": 0.2416517272591591,
      "num_tokens": 138149934.0,
      "step": 79655
    },
    {
      "entropy": 5.450355482101441,
      "epoch": 6.1977825325812095,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019425509901415162,
      "loss": 4.7696,
      "mean_token_accuracy": 0.2297012209892273,
      "num_tokens": 138158455.0,
      "step": 79660
    },
    {
      "entropy": 5.407970333099366,
      "epoch": 6.198171561952927,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019422923020208688,
      "loss": 4.6009,
      "mean_token_accuracy": 0.24301806688308716,
      "num_tokens": 138167121.0,
      "step": 79665
    },
    {
      "entropy": 5.4395670890808105,
      "epoch": 6.198560591324645,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019420336261558995,
      "loss": 4.7238,
      "mean_token_accuracy": 0.23100040405988692,
      "num_tokens": 138176545.0,
      "step": 79670
    },
    {
      "entropy": 5.395820093154907,
      "epoch": 6.198949620696363,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019417749625505343,
      "loss": 4.6299,
      "mean_token_accuracy": 0.24387394040822982,
      "num_tokens": 138184320.0,
      "step": 79675
    },
    {
      "entropy": 5.43826813697815,
      "epoch": 6.19933865006808,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019415163112086975,
      "loss": 4.7461,
      "mean_token_accuracy": 0.23201263099908828,
      "num_tokens": 138193374.0,
      "step": 79680
    },
    {
      "entropy": 5.426840925216675,
      "epoch": 6.199727679439798,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019412576721343134,
      "loss": 4.6503,
      "mean_token_accuracy": 0.2419420674443245,
      "num_tokens": 138201652.0,
      "step": 79685
    },
    {
      "entropy": 5.349707221984863,
      "epoch": 6.200116708811516,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019409990453313076,
      "loss": 4.5937,
      "mean_token_accuracy": 0.2412685662508011,
      "num_tokens": 138211520.0,
      "step": 79690
    },
    {
      "entropy": 5.464655447006225,
      "epoch": 6.2005057381832325,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001940740430803603,
      "loss": 4.7451,
      "mean_token_accuracy": 0.23529671132564545,
      "num_tokens": 138219964.0,
      "step": 79695
    },
    {
      "entropy": 5.324163436889648,
      "epoch": 6.20089476755495,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019404818285551252,
      "loss": 4.6781,
      "mean_token_accuracy": 0.2403939202427864,
      "num_tokens": 138228636.0,
      "step": 79700
    },
    {
      "entropy": 5.337571430206299,
      "epoch": 6.201283796926668,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001940223238589796,
      "loss": 4.661,
      "mean_token_accuracy": 0.2322801023721695,
      "num_tokens": 138237557.0,
      "step": 79705
    },
    {
      "entropy": 5.388698101043701,
      "epoch": 6.201672826298386,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019399646609115408,
      "loss": 4.6627,
      "mean_token_accuracy": 0.23783407360315323,
      "num_tokens": 138245790.0,
      "step": 79710
    },
    {
      "entropy": 5.502856063842773,
      "epoch": 6.202061855670103,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019397060955242824,
      "loss": 4.7281,
      "mean_token_accuracy": 0.23926471322774887,
      "num_tokens": 138253926.0,
      "step": 79715
    },
    {
      "entropy": 5.366575860977173,
      "epoch": 6.202450885041821,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019394475424319442,
      "loss": 4.681,
      "mean_token_accuracy": 0.23168591409921646,
      "num_tokens": 138262267.0,
      "step": 79720
    },
    {
      "entropy": 5.3291308879852295,
      "epoch": 6.202839914413538,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019391890016384495,
      "loss": 4.6295,
      "mean_token_accuracy": 0.23911878764629363,
      "num_tokens": 138271667.0,
      "step": 79725
    },
    {
      "entropy": 5.377816820144654,
      "epoch": 6.2032289437852555,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00019389304731477213,
      "loss": 4.5984,
      "mean_token_accuracy": 0.2513256996870041,
      "num_tokens": 138280548.0,
      "step": 79730
    },
    {
      "entropy": 5.479285144805909,
      "epoch": 6.203617973156973,
      "grad_norm": 1.1875,
      "learning_rate": 0.00019386719569636823,
      "loss": 4.6855,
      "mean_token_accuracy": 0.2388111963868141,
      "num_tokens": 138288950.0,
      "step": 79735
    },
    {
      "entropy": 5.445222854614258,
      "epoch": 6.204007002528691,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00019384134530902553,
      "loss": 4.8147,
      "mean_token_accuracy": 0.22782792448997496,
      "num_tokens": 138297998.0,
      "step": 79740
    },
    {
      "entropy": 5.412259531021118,
      "epoch": 6.204396031900409,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001938154961531362,
      "loss": 4.6403,
      "mean_token_accuracy": 0.2391327440738678,
      "num_tokens": 138306518.0,
      "step": 79745
    },
    {
      "entropy": 5.422231960296631,
      "epoch": 6.204785061272126,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001937896482290925,
      "loss": 4.6543,
      "mean_token_accuracy": 0.23523647487163543,
      "num_tokens": 138315783.0,
      "step": 79750
    },
    {
      "entropy": 5.4698422908782955,
      "epoch": 6.205174090643844,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019376380153728667,
      "loss": 4.7056,
      "mean_token_accuracy": 0.23454420119524003,
      "num_tokens": 138323993.0,
      "step": 79755
    },
    {
      "entropy": 5.441685533523559,
      "epoch": 6.205563120015561,
      "grad_norm": 1.28125,
      "learning_rate": 0.00019373795607811078,
      "loss": 4.7067,
      "mean_token_accuracy": 0.2368060380220413,
      "num_tokens": 138333158.0,
      "step": 79760
    },
    {
      "entropy": 5.351660966873169,
      "epoch": 6.2059521493872785,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019371211185195708,
      "loss": 4.6309,
      "mean_token_accuracy": 0.2397200807929039,
      "num_tokens": 138341430.0,
      "step": 79765
    },
    {
      "entropy": 5.399773263931275,
      "epoch": 6.206341178758996,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001936862688592177,
      "loss": 4.7283,
      "mean_token_accuracy": 0.23391542136669158,
      "num_tokens": 138350299.0,
      "step": 79770
    },
    {
      "entropy": 5.405533313751221,
      "epoch": 6.206730208130714,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001936604271002848,
      "loss": 4.6147,
      "mean_token_accuracy": 0.24709745049476622,
      "num_tokens": 138359500.0,
      "step": 79775
    },
    {
      "entropy": 5.380837249755859,
      "epoch": 6.2071192375024316,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019363458657555038,
      "loss": 4.5734,
      "mean_token_accuracy": 0.24708219468593598,
      "num_tokens": 138367919.0,
      "step": 79780
    },
    {
      "entropy": 5.337766790390015,
      "epoch": 6.207508266874149,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019360874728540668,
      "loss": 4.5418,
      "mean_token_accuracy": 0.24348196834325792,
      "num_tokens": 138376196.0,
      "step": 79785
    },
    {
      "entropy": 5.297143173217774,
      "epoch": 6.207897296245867,
      "grad_norm": 1.234375,
      "learning_rate": 0.00019358290923024562,
      "loss": 4.6134,
      "mean_token_accuracy": 0.23904112875461578,
      "num_tokens": 138385719.0,
      "step": 79790
    },
    {
      "entropy": 5.453077602386474,
      "epoch": 6.208286325617584,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001935570724104594,
      "loss": 4.7058,
      "mean_token_accuracy": 0.2348807469010353,
      "num_tokens": 138394035.0,
      "step": 79795
    },
    {
      "entropy": 5.371989297866821,
      "epoch": 6.2086753549893015,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019353123682643998,
      "loss": 4.5695,
      "mean_token_accuracy": 0.24460267275571823,
      "num_tokens": 138401838.0,
      "step": 79800
    },
    {
      "entropy": 5.36424708366394,
      "epoch": 6.209064384361019,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019350540247857928,
      "loss": 4.6086,
      "mean_token_accuracy": 0.2410960853099823,
      "num_tokens": 138410191.0,
      "step": 79805
    },
    {
      "entropy": 5.355542898178101,
      "epoch": 6.209453413732737,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019347956936726946,
      "loss": 4.6172,
      "mean_token_accuracy": 0.2405377671122551,
      "num_tokens": 138418426.0,
      "step": 79810
    },
    {
      "entropy": 5.351362657546997,
      "epoch": 6.2098424431044545,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001934537374929024,
      "loss": 4.624,
      "mean_token_accuracy": 0.2394194096326828,
      "num_tokens": 138426208.0,
      "step": 79815
    },
    {
      "entropy": 5.265998506546021,
      "epoch": 6.210231472476172,
      "grad_norm": 1.28125,
      "learning_rate": 0.00019342790685587008,
      "loss": 4.5232,
      "mean_token_accuracy": 0.2580756083130836,
      "num_tokens": 138435187.0,
      "step": 79820
    },
    {
      "entropy": 5.340147161483765,
      "epoch": 6.21062050184789,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001934020774565645,
      "loss": 4.6525,
      "mean_token_accuracy": 0.24407271295785904,
      "num_tokens": 138443721.0,
      "step": 79825
    },
    {
      "entropy": 5.463840627670288,
      "epoch": 6.211009531219607,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001933762492953775,
      "loss": 4.7334,
      "mean_token_accuracy": 0.23506601452827453,
      "num_tokens": 138451813.0,
      "step": 79830
    },
    {
      "entropy": 5.35977463722229,
      "epoch": 6.2113985605913244,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019335042237270095,
      "loss": 4.6287,
      "mean_token_accuracy": 0.23949581682682036,
      "num_tokens": 138460235.0,
      "step": 79835
    },
    {
      "entropy": 5.327545833587647,
      "epoch": 6.211787589963042,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019332459668892676,
      "loss": 4.567,
      "mean_token_accuracy": 0.24725427031517028,
      "num_tokens": 138468796.0,
      "step": 79840
    },
    {
      "entropy": 5.351621961593628,
      "epoch": 6.21217661933476,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019329877224444675,
      "loss": 4.6142,
      "mean_token_accuracy": 0.23658262193202972,
      "num_tokens": 138477343.0,
      "step": 79845
    },
    {
      "entropy": 5.427374315261841,
      "epoch": 6.2125656487064775,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001932729490396529,
      "loss": 4.7385,
      "mean_token_accuracy": 0.2314409926533699,
      "num_tokens": 138486471.0,
      "step": 79850
    },
    {
      "entropy": 5.425639820098877,
      "epoch": 6.212954678078195,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019324712707493707,
      "loss": 4.6545,
      "mean_token_accuracy": 0.2411415159702301,
      "num_tokens": 138495164.0,
      "step": 79855
    },
    {
      "entropy": 5.467065763473511,
      "epoch": 6.213343707449912,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001932213063506909,
      "loss": 4.716,
      "mean_token_accuracy": 0.24003915786743163,
      "num_tokens": 138503339.0,
      "step": 79860
    },
    {
      "entropy": 5.417418193817139,
      "epoch": 6.21373273682163,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019319548686730625,
      "loss": 4.6534,
      "mean_token_accuracy": 0.24243364185094834,
      "num_tokens": 138512115.0,
      "step": 79865
    },
    {
      "entropy": 5.360242557525635,
      "epoch": 6.214121766193347,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019316966862517482,
      "loss": 4.5965,
      "mean_token_accuracy": 0.24542619436979293,
      "num_tokens": 138520710.0,
      "step": 79870
    },
    {
      "entropy": 5.3844828605651855,
      "epoch": 6.214510795565065,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001931438516246885,
      "loss": 4.6699,
      "mean_token_accuracy": 0.24177227318286895,
      "num_tokens": 138528989.0,
      "step": 79875
    },
    {
      "entropy": 5.376661109924316,
      "epoch": 6.214899824936783,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001931180358662389,
      "loss": 4.6739,
      "mean_token_accuracy": 0.2335123211145401,
      "num_tokens": 138537920.0,
      "step": 79880
    },
    {
      "entropy": 5.43037223815918,
      "epoch": 6.2152888543085005,
      "grad_norm": 1.21875,
      "learning_rate": 0.00019309222135021776,
      "loss": 4.6376,
      "mean_token_accuracy": 0.2388914242386818,
      "num_tokens": 138546516.0,
      "step": 79885
    },
    {
      "entropy": 5.431724500656128,
      "epoch": 6.215677883680218,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001930664080770168,
      "loss": 4.7276,
      "mean_token_accuracy": 0.2335727334022522,
      "num_tokens": 138555262.0,
      "step": 79890
    },
    {
      "entropy": 5.431961393356323,
      "epoch": 6.216066913051935,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019304059604702772,
      "loss": 4.6901,
      "mean_token_accuracy": 0.2328224927186966,
      "num_tokens": 138563773.0,
      "step": 79895
    },
    {
      "entropy": 5.4424591064453125,
      "epoch": 6.216455942423653,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019301478526064213,
      "loss": 4.7198,
      "mean_token_accuracy": 0.23592216670513153,
      "num_tokens": 138572298.0,
      "step": 79900
    },
    {
      "entropy": 5.420798063278198,
      "epoch": 6.21684497179537,
      "grad_norm": 1.375,
      "learning_rate": 0.00019298897571825175,
      "loss": 4.7522,
      "mean_token_accuracy": 0.22739098072052003,
      "num_tokens": 138580951.0,
      "step": 79905
    },
    {
      "entropy": 5.304519462585449,
      "epoch": 6.217234001167088,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019296316742024806,
      "loss": 4.5423,
      "mean_token_accuracy": 0.2525974214076996,
      "num_tokens": 138589069.0,
      "step": 79910
    },
    {
      "entropy": 5.404778671264649,
      "epoch": 6.217623030538806,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019293736036702259,
      "loss": 4.6902,
      "mean_token_accuracy": 0.2397646948695183,
      "num_tokens": 138597955.0,
      "step": 79915
    },
    {
      "entropy": 5.350574588775634,
      "epoch": 6.2180120599105235,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001929115545589672,
      "loss": 4.5852,
      "mean_token_accuracy": 0.24022107124328612,
      "num_tokens": 138606168.0,
      "step": 79920
    },
    {
      "entropy": 5.365299034118652,
      "epoch": 6.218401089282241,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001928857499964733,
      "loss": 4.6614,
      "mean_token_accuracy": 0.23673480600118638,
      "num_tokens": 138615004.0,
      "step": 79925
    },
    {
      "entropy": 5.365364360809326,
      "epoch": 6.218790118653958,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019285994667993245,
      "loss": 4.6523,
      "mean_token_accuracy": 0.24387177973985671,
      "num_tokens": 138623994.0,
      "step": 79930
    },
    {
      "entropy": 5.426820755004883,
      "epoch": 6.219179148025676,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019283414460973626,
      "loss": 4.6674,
      "mean_token_accuracy": 0.23529060035943986,
      "num_tokens": 138632387.0,
      "step": 79935
    },
    {
      "entropy": 5.462334251403808,
      "epoch": 6.219568177397393,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019280834378627605,
      "loss": 4.6766,
      "mean_token_accuracy": 0.23831190168857574,
      "num_tokens": 138640826.0,
      "step": 79940
    },
    {
      "entropy": 5.404705047607422,
      "epoch": 6.219957206769111,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001927825442099434,
      "loss": 4.6469,
      "mean_token_accuracy": 0.24509579986333846,
      "num_tokens": 138649993.0,
      "step": 79945
    },
    {
      "entropy": 5.278166151046753,
      "epoch": 6.220346236140829,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001927567458811298,
      "loss": 4.5293,
      "mean_token_accuracy": 0.2563353955745697,
      "num_tokens": 138659305.0,
      "step": 79950
    },
    {
      "entropy": 5.3453442573547365,
      "epoch": 6.2207352655125465,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019273094880022662,
      "loss": 4.6343,
      "mean_token_accuracy": 0.2367136910557747,
      "num_tokens": 138667889.0,
      "step": 79955
    },
    {
      "entropy": 5.426275682449341,
      "epoch": 6.221124294884264,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019270515296762543,
      "loss": 4.6887,
      "mean_token_accuracy": 0.23379596173763276,
      "num_tokens": 138675819.0,
      "step": 79960
    },
    {
      "entropy": 5.467107772827148,
      "epoch": 6.221513324255981,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019267935838371752,
      "loss": 4.6757,
      "mean_token_accuracy": 0.23296771049499512,
      "num_tokens": 138684049.0,
      "step": 79965
    },
    {
      "entropy": 5.390197372436523,
      "epoch": 6.221902353627699,
      "grad_norm": 1.375,
      "learning_rate": 0.0001926535650488943,
      "loss": 4.6354,
      "mean_token_accuracy": 0.23491803854703902,
      "num_tokens": 138692778.0,
      "step": 79970
    },
    {
      "entropy": 5.372357749938965,
      "epoch": 6.222291382999416,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001926277729635472,
      "loss": 4.6492,
      "mean_token_accuracy": 0.2371863067150116,
      "num_tokens": 138701440.0,
      "step": 79975
    },
    {
      "entropy": 5.4828802108764645,
      "epoch": 6.222680412371134,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019260198212806747,
      "loss": 4.6819,
      "mean_token_accuracy": 0.23703991770744323,
      "num_tokens": 138710145.0,
      "step": 79980
    },
    {
      "entropy": 5.425070428848267,
      "epoch": 6.223069441742852,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001925761925428466,
      "loss": 4.7237,
      "mean_token_accuracy": 0.24031783193349837,
      "num_tokens": 138719644.0,
      "step": 79985
    },
    {
      "entropy": 5.445164155960083,
      "epoch": 6.2234584711145695,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019255040420827576,
      "loss": 4.6877,
      "mean_token_accuracy": 0.23490922152996063,
      "num_tokens": 138728089.0,
      "step": 79990
    },
    {
      "entropy": 5.422434520721436,
      "epoch": 6.223847500486286,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019252461712474629,
      "loss": 4.6771,
      "mean_token_accuracy": 0.23843155801296234,
      "num_tokens": 138736948.0,
      "step": 79995
    },
    {
      "entropy": 5.410606813430786,
      "epoch": 6.224236529858004,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001924988312926495,
      "loss": 4.7362,
      "mean_token_accuracy": 0.2325963333249092,
      "num_tokens": 138745897.0,
      "step": 80000
    },
    {
      "entropy": 5.452065467834473,
      "epoch": 6.224625559229722,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019247304671237664,
      "loss": 4.767,
      "mean_token_accuracy": 0.22549825310707092,
      "num_tokens": 138755814.0,
      "step": 80005
    },
    {
      "entropy": 5.431494569778442,
      "epoch": 6.225014588601439,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019244726338431896,
      "loss": 4.6776,
      "mean_token_accuracy": 0.2380765274167061,
      "num_tokens": 138764910.0,
      "step": 80010
    },
    {
      "entropy": 5.36698579788208,
      "epoch": 6.225403617973157,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019242148130886773,
      "loss": 4.5878,
      "mean_token_accuracy": 0.24124503284692764,
      "num_tokens": 138772901.0,
      "step": 80015
    },
    {
      "entropy": 5.358956527709961,
      "epoch": 6.225792647344875,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019239570048641398,
      "loss": 4.6123,
      "mean_token_accuracy": 0.2508848637342453,
      "num_tokens": 138781219.0,
      "step": 80020
    },
    {
      "entropy": 5.392642545700073,
      "epoch": 6.2261816767165925,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000192369920917349,
      "loss": 4.7079,
      "mean_token_accuracy": 0.23904809802770616,
      "num_tokens": 138789359.0,
      "step": 80025
    },
    {
      "entropy": 5.389347696304322,
      "epoch": 6.226570706088309,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000192344142602064,
      "loss": 4.6831,
      "mean_token_accuracy": 0.23707407265901564,
      "num_tokens": 138798375.0,
      "step": 80030
    },
    {
      "entropy": 5.396453809738159,
      "epoch": 6.226959735460027,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019231836554095,
      "loss": 4.66,
      "mean_token_accuracy": 0.24134524315595626,
      "num_tokens": 138807358.0,
      "step": 80035
    },
    {
      "entropy": 5.3784623622894285,
      "epoch": 6.227348764831745,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019229258973439827,
      "loss": 4.6511,
      "mean_token_accuracy": 0.23737898766994475,
      "num_tokens": 138816109.0,
      "step": 80040
    },
    {
      "entropy": 5.380625915527344,
      "epoch": 6.227737794203462,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019226681518279975,
      "loss": 4.6622,
      "mean_token_accuracy": 0.22804000079631806,
      "num_tokens": 138824613.0,
      "step": 80045
    },
    {
      "entropy": 5.414347457885742,
      "epoch": 6.22812682357518,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019224104188654572,
      "loss": 4.6284,
      "mean_token_accuracy": 0.23702718764543534,
      "num_tokens": 138832516.0,
      "step": 80050
    },
    {
      "entropy": 5.402330684661865,
      "epoch": 6.228515852946898,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019221526984602717,
      "loss": 4.6244,
      "mean_token_accuracy": 0.24230293482542037,
      "num_tokens": 138841690.0,
      "step": 80055
    },
    {
      "entropy": 5.361441802978516,
      "epoch": 6.228904882318615,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001921894990616351,
      "loss": 4.612,
      "mean_token_accuracy": 0.23755445778369905,
      "num_tokens": 138850045.0,
      "step": 80060
    },
    {
      "entropy": 5.352422094345092,
      "epoch": 6.229293911690332,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019216372953376055,
      "loss": 4.6746,
      "mean_token_accuracy": 0.23036280125379563,
      "num_tokens": 138858878.0,
      "step": 80065
    },
    {
      "entropy": 5.36416711807251,
      "epoch": 6.22968294106205,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001921379612627946,
      "loss": 4.6478,
      "mean_token_accuracy": 0.2341684579849243,
      "num_tokens": 138866956.0,
      "step": 80070
    },
    {
      "entropy": 5.438803577423096,
      "epoch": 6.230071970433768,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019211219424912818,
      "loss": 4.746,
      "mean_token_accuracy": 0.23682404905557633,
      "num_tokens": 138876097.0,
      "step": 80075
    },
    {
      "entropy": 5.387078952789307,
      "epoch": 6.230460999805485,
      "grad_norm": 1.25,
      "learning_rate": 0.0001920864284931523,
      "loss": 4.6457,
      "mean_token_accuracy": 0.24215710759162903,
      "num_tokens": 138884721.0,
      "step": 80080
    },
    {
      "entropy": 5.467601776123047,
      "epoch": 6.230850029177203,
      "grad_norm": 1.375,
      "learning_rate": 0.00019206066399525784,
      "loss": 4.7297,
      "mean_token_accuracy": 0.2294338807463646,
      "num_tokens": 138893243.0,
      "step": 80085
    },
    {
      "entropy": 5.428174734115601,
      "epoch": 6.231239058548921,
      "grad_norm": 1.265625,
      "learning_rate": 0.00019203490075583586,
      "loss": 4.6676,
      "mean_token_accuracy": 0.23627914637327194,
      "num_tokens": 138901621.0,
      "step": 80090
    },
    {
      "entropy": 5.441163158416748,
      "epoch": 6.231628087920638,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001920091387752772,
      "loss": 4.6927,
      "mean_token_accuracy": 0.23248979896306993,
      "num_tokens": 138910807.0,
      "step": 80095
    },
    {
      "entropy": 5.4242078304290775,
      "epoch": 6.232017117292355,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019198337805397282,
      "loss": 4.6373,
      "mean_token_accuracy": 0.23331674486398696,
      "num_tokens": 138919202.0,
      "step": 80100
    },
    {
      "entropy": 5.4012044906616214,
      "epoch": 6.232406146664073,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00019195761859231342,
      "loss": 4.7971,
      "mean_token_accuracy": 0.2282021388411522,
      "num_tokens": 138928711.0,
      "step": 80105
    },
    {
      "entropy": 5.334251880645752,
      "epoch": 6.232795176035791,
      "grad_norm": 1.1875,
      "learning_rate": 0.00019193186039068993,
      "loss": 4.5859,
      "mean_token_accuracy": 0.242653751373291,
      "num_tokens": 138938023.0,
      "step": 80110
    },
    {
      "entropy": 5.415680170059204,
      "epoch": 6.233184205407508,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019190610344949337,
      "loss": 4.6906,
      "mean_token_accuracy": 0.24405065774917603,
      "num_tokens": 138946216.0,
      "step": 80115
    },
    {
      "entropy": 5.405815649032593,
      "epoch": 6.233573234779226,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019188034776911438,
      "loss": 4.6735,
      "mean_token_accuracy": 0.24194093495607377,
      "num_tokens": 138954749.0,
      "step": 80120
    },
    {
      "entropy": 5.415266990661621,
      "epoch": 6.233962264150944,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019185459334994387,
      "loss": 4.6752,
      "mean_token_accuracy": 0.23676856607198715,
      "num_tokens": 138964064.0,
      "step": 80125
    },
    {
      "entropy": 5.377191543579102,
      "epoch": 6.234351293522661,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019182884019237258,
      "loss": 4.6957,
      "mean_token_accuracy": 0.24328095763921737,
      "num_tokens": 138973156.0,
      "step": 80130
    },
    {
      "entropy": 5.381244087219239,
      "epoch": 6.234740322894378,
      "grad_norm": 1.375,
      "learning_rate": 0.0001918030882967912,
      "loss": 4.6344,
      "mean_token_accuracy": 0.24285353273153304,
      "num_tokens": 138981307.0,
      "step": 80135
    },
    {
      "entropy": 5.347476863861084,
      "epoch": 6.235129352266096,
      "grad_norm": 1.25,
      "learning_rate": 0.00019177733766359057,
      "loss": 4.5991,
      "mean_token_accuracy": 0.24503027498722077,
      "num_tokens": 138990537.0,
      "step": 80140
    },
    {
      "entropy": 5.397099256515503,
      "epoch": 6.235518381637814,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019175158829316135,
      "loss": 4.6059,
      "mean_token_accuracy": 0.24933990836143494,
      "num_tokens": 138999740.0,
      "step": 80145
    },
    {
      "entropy": 5.268996715545654,
      "epoch": 6.235907411009531,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019172584018589429,
      "loss": 4.587,
      "mean_token_accuracy": 0.23901741057634354,
      "num_tokens": 139008257.0,
      "step": 80150
    },
    {
      "entropy": 5.342515373229981,
      "epoch": 6.236296440381249,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019170009334218004,
      "loss": 4.6657,
      "mean_token_accuracy": 0.2384684920310974,
      "num_tokens": 139017386.0,
      "step": 80155
    },
    {
      "entropy": 5.385322189331054,
      "epoch": 6.236685469752967,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019167434776240933,
      "loss": 4.645,
      "mean_token_accuracy": 0.24339460581541061,
      "num_tokens": 139025579.0,
      "step": 80160
    },
    {
      "entropy": 5.32856330871582,
      "epoch": 6.237074499124684,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019164860344697272,
      "loss": 4.5721,
      "mean_token_accuracy": 0.24736848026514052,
      "num_tokens": 139033753.0,
      "step": 80165
    },
    {
      "entropy": 5.384770154953003,
      "epoch": 6.237463528496401,
      "grad_norm": 1.28125,
      "learning_rate": 0.00019162286039626093,
      "loss": 4.6895,
      "mean_token_accuracy": 0.24098296761512755,
      "num_tokens": 139043217.0,
      "step": 80170
    },
    {
      "entropy": 5.432442903518677,
      "epoch": 6.237852557868119,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019159711861066453,
      "loss": 4.7237,
      "mean_token_accuracy": 0.2292504593729973,
      "num_tokens": 139051872.0,
      "step": 80175
    },
    {
      "entropy": 5.47056884765625,
      "epoch": 6.238241587239837,
      "grad_norm": 1.359375,
      "learning_rate": 0.000191571378090574,
      "loss": 4.7545,
      "mean_token_accuracy": 0.22625083923339845,
      "num_tokens": 139060402.0,
      "step": 80180
    },
    {
      "entropy": 5.4540447235107425,
      "epoch": 6.238630616611554,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001915456388363801,
      "loss": 4.7254,
      "mean_token_accuracy": 0.236455500125885,
      "num_tokens": 139069189.0,
      "step": 80185
    },
    {
      "entropy": 5.374243831634521,
      "epoch": 6.239019645983272,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001915199008484733,
      "loss": 4.6412,
      "mean_token_accuracy": 0.23824235200881957,
      "num_tokens": 139078596.0,
      "step": 80190
    },
    {
      "entropy": 5.384951114654541,
      "epoch": 6.239408675354989,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019149416412724414,
      "loss": 4.6453,
      "mean_token_accuracy": 0.23983217477798463,
      "num_tokens": 139087108.0,
      "step": 80195
    },
    {
      "entropy": 5.435272169113159,
      "epoch": 6.239797704726707,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019146842867308306,
      "loss": 4.7552,
      "mean_token_accuracy": 0.23350372165441513,
      "num_tokens": 139095484.0,
      "step": 80200
    },
    {
      "entropy": 5.505098867416382,
      "epoch": 6.240186734098424,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019144269448638073,
      "loss": 4.7535,
      "mean_token_accuracy": 0.23539184778928757,
      "num_tokens": 139104261.0,
      "step": 80205
    },
    {
      "entropy": 5.365941381454467,
      "epoch": 6.240575763470142,
      "grad_norm": 1.5,
      "learning_rate": 0.00019141696156752748,
      "loss": 4.6413,
      "mean_token_accuracy": 0.2406003639101982,
      "num_tokens": 139112466.0,
      "step": 80210
    },
    {
      "entropy": 5.343252038955688,
      "epoch": 6.24096479284186,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00019139122991691378,
      "loss": 4.6096,
      "mean_token_accuracy": 0.2436447873711586,
      "num_tokens": 139121077.0,
      "step": 80215
    },
    {
      "entropy": 5.386625671386719,
      "epoch": 6.241353822213577,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001913654995349301,
      "loss": 4.6233,
      "mean_token_accuracy": 0.24164265394210815,
      "num_tokens": 139129224.0,
      "step": 80220
    },
    {
      "entropy": 5.56432695388794,
      "epoch": 6.241742851585295,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019133977042196682,
      "loss": 4.7861,
      "mean_token_accuracy": 0.22455255687236786,
      "num_tokens": 139137418.0,
      "step": 80225
    },
    {
      "entropy": 5.368259572982788,
      "epoch": 6.242131880957012,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001913140425784144,
      "loss": 4.5903,
      "mean_token_accuracy": 0.24392179250717164,
      "num_tokens": 139146286.0,
      "step": 80230
    },
    {
      "entropy": 5.314634990692139,
      "epoch": 6.24252091032873,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019128831600466313,
      "loss": 4.5464,
      "mean_token_accuracy": 0.2463858738541603,
      "num_tokens": 139154745.0,
      "step": 80235
    },
    {
      "entropy": 5.353270196914673,
      "epoch": 6.242909939700447,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019126259070110347,
      "loss": 4.754,
      "mean_token_accuracy": 0.23059714287519456,
      "num_tokens": 139163563.0,
      "step": 80240
    },
    {
      "entropy": 5.390217256546021,
      "epoch": 6.243298969072165,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00019123686666812562,
      "loss": 4.6684,
      "mean_token_accuracy": 0.24012877196073532,
      "num_tokens": 139171964.0,
      "step": 80245
    },
    {
      "entropy": 5.35038275718689,
      "epoch": 6.243687998443883,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001912111439061201,
      "loss": 4.5338,
      "mean_token_accuracy": 0.24900649338960648,
      "num_tokens": 139180673.0,
      "step": 80250
    },
    {
      "entropy": 5.408588123321533,
      "epoch": 6.2440770278156,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001911854224154771,
      "loss": 4.6378,
      "mean_token_accuracy": 0.23947800993919371,
      "num_tokens": 139189815.0,
      "step": 80255
    },
    {
      "entropy": 5.403171682357788,
      "epoch": 6.244466057187318,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001911597021965869,
      "loss": 4.6539,
      "mean_token_accuracy": 0.2397583693265915,
      "num_tokens": 139197852.0,
      "step": 80260
    },
    {
      "entropy": 5.398620080947876,
      "epoch": 6.244855086559035,
      "grad_norm": 1.375,
      "learning_rate": 0.00019113398324983972,
      "loss": 4.6274,
      "mean_token_accuracy": 0.24101149737834932,
      "num_tokens": 139206752.0,
      "step": 80265
    },
    {
      "entropy": 5.386550951004028,
      "epoch": 6.245244115930753,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019110826557562594,
      "loss": 4.6185,
      "mean_token_accuracy": 0.23750204890966414,
      "num_tokens": 139215686.0,
      "step": 80270
    },
    {
      "entropy": 5.33972134590149,
      "epoch": 6.24563314530247,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019108254917433564,
      "loss": 4.5515,
      "mean_token_accuracy": 0.23720190525054932,
      "num_tokens": 139224680.0,
      "step": 80275
    },
    {
      "entropy": 5.357876348495483,
      "epoch": 6.246022174674188,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019105683404635915,
      "loss": 4.5749,
      "mean_token_accuracy": 0.24470401406288148,
      "num_tokens": 139233263.0,
      "step": 80280
    },
    {
      "entropy": 5.375097608566284,
      "epoch": 6.246411204045906,
      "grad_norm": 1.734375,
      "learning_rate": 0.00019103112019208662,
      "loss": 4.677,
      "mean_token_accuracy": 0.2349293276667595,
      "num_tokens": 139241970.0,
      "step": 80285
    },
    {
      "entropy": 5.407288694381714,
      "epoch": 6.246800233417623,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019100540761190815,
      "loss": 4.6583,
      "mean_token_accuracy": 0.22442522794008254,
      "num_tokens": 139250634.0,
      "step": 80290
    },
    {
      "entropy": 5.454122114181518,
      "epoch": 6.24718926278934,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019097969630621386,
      "loss": 4.7141,
      "mean_token_accuracy": 0.23517605364322663,
      "num_tokens": 139259473.0,
      "step": 80295
    },
    {
      "entropy": 5.367725801467896,
      "epoch": 6.247578292161058,
      "grad_norm": 1.3125,
      "learning_rate": 0.000190953986275394,
      "loss": 4.5974,
      "mean_token_accuracy": 0.24284541755914688,
      "num_tokens": 139268199.0,
      "step": 80300
    },
    {
      "entropy": 5.393930864334107,
      "epoch": 6.247967321532776,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019092827751983866,
      "loss": 4.7253,
      "mean_token_accuracy": 0.23260626941919327,
      "num_tokens": 139277130.0,
      "step": 80305
    },
    {
      "entropy": 5.400803279876709,
      "epoch": 6.248356350904493,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001909025700399377,
      "loss": 4.7084,
      "mean_token_accuracy": 0.23934050947427749,
      "num_tokens": 139285776.0,
      "step": 80310
    },
    {
      "entropy": 5.458319902420044,
      "epoch": 6.248745380276211,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001908768638360816,
      "loss": 4.7803,
      "mean_token_accuracy": 0.2337967962026596,
      "num_tokens": 139294796.0,
      "step": 80315
    },
    {
      "entropy": 5.504234457015992,
      "epoch": 6.249134409647929,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019085115890866018,
      "loss": 4.7262,
      "mean_token_accuracy": 0.23483271300792694,
      "num_tokens": 139303770.0,
      "step": 80320
    },
    {
      "entropy": 5.415128087997436,
      "epoch": 6.249523439019646,
      "grad_norm": 1.375,
      "learning_rate": 0.00019082545525806344,
      "loss": 4.5478,
      "mean_token_accuracy": 0.24559057056903838,
      "num_tokens": 139311298.0,
      "step": 80325
    },
    {
      "entropy": 5.414852952957153,
      "epoch": 6.249912468391363,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019079975288468142,
      "loss": 4.6416,
      "mean_token_accuracy": 0.24122584760189056,
      "num_tokens": 139319572.0,
      "step": 80330
    },
    {
      "entropy": 5.377610445022583,
      "epoch": 6.250301497763081,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019077405178890417,
      "loss": 4.6439,
      "mean_token_accuracy": 0.2359106123447418,
      "num_tokens": 139327296.0,
      "step": 80335
    },
    {
      "entropy": 5.379631280899048,
      "epoch": 6.250690527134799,
      "grad_norm": 1.28125,
      "learning_rate": 0.00019074835197112162,
      "loss": 4.6763,
      "mean_token_accuracy": 0.23707904815673828,
      "num_tokens": 139336212.0,
      "step": 80340
    },
    {
      "entropy": 5.417950820922852,
      "epoch": 6.251079556506516,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019072265343172368,
      "loss": 4.6473,
      "mean_token_accuracy": 0.2485795646905899,
      "num_tokens": 139344538.0,
      "step": 80345
    },
    {
      "entropy": 5.397574329376221,
      "epoch": 6.251468585878234,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019069695617110038,
      "loss": 4.5975,
      "mean_token_accuracy": 0.24355843663215637,
      "num_tokens": 139353205.0,
      "step": 80350
    },
    {
      "entropy": 5.296426105499267,
      "epoch": 6.251857615249952,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001906712601896416,
      "loss": 4.5262,
      "mean_token_accuracy": 0.2529544472694397,
      "num_tokens": 139362205.0,
      "step": 80355
    },
    {
      "entropy": 5.495827531814575,
      "epoch": 6.252246644621669,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001906455654877372,
      "loss": 4.7365,
      "mean_token_accuracy": 0.22831109911203384,
      "num_tokens": 139369977.0,
      "step": 80360
    },
    {
      "entropy": 5.365428400039673,
      "epoch": 6.252635673993386,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019061987206577712,
      "loss": 4.6598,
      "mean_token_accuracy": 0.2404581278562546,
      "num_tokens": 139378401.0,
      "step": 80365
    },
    {
      "entropy": 5.399304056167603,
      "epoch": 6.253024703365104,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001905941799241511,
      "loss": 4.6244,
      "mean_token_accuracy": 0.2493083357810974,
      "num_tokens": 139387651.0,
      "step": 80370
    },
    {
      "entropy": 5.439533758163452,
      "epoch": 6.2534137327368216,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019056848906324914,
      "loss": 4.7235,
      "mean_token_accuracy": 0.23571151047945021,
      "num_tokens": 139396428.0,
      "step": 80375
    },
    {
      "entropy": 5.405953121185303,
      "epoch": 6.253802762108539,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001905427994834608,
      "loss": 4.6203,
      "mean_token_accuracy": 0.24121590852737426,
      "num_tokens": 139405861.0,
      "step": 80380
    },
    {
      "entropy": 5.356570672988892,
      "epoch": 6.254191791480257,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019051711118517618,
      "loss": 4.6711,
      "mean_token_accuracy": 0.2392266035079956,
      "num_tokens": 139415403.0,
      "step": 80385
    },
    {
      "entropy": 5.386609888076782,
      "epoch": 6.254580820851975,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019049142416878496,
      "loss": 4.6361,
      "mean_token_accuracy": 0.2461826503276825,
      "num_tokens": 139424360.0,
      "step": 80390
    },
    {
      "entropy": 5.356417465209961,
      "epoch": 6.2549698502236915,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019046573843467684,
      "loss": 4.5616,
      "mean_token_accuracy": 0.24548180550336837,
      "num_tokens": 139432833.0,
      "step": 80395
    },
    {
      "entropy": 5.394099235534668,
      "epoch": 6.255358879595409,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001904400539832416,
      "loss": 4.6715,
      "mean_token_accuracy": 0.2481598824262619,
      "num_tokens": 139442715.0,
      "step": 80400
    },
    {
      "entropy": 5.470827245712281,
      "epoch": 6.255747908967127,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019041437081486896,
      "loss": 4.665,
      "mean_token_accuracy": 0.24230485856533052,
      "num_tokens": 139451179.0,
      "step": 80405
    },
    {
      "entropy": 5.369745540618896,
      "epoch": 6.2561369383388445,
      "grad_norm": 1.25,
      "learning_rate": 0.00019038868892994854,
      "loss": 4.6888,
      "mean_token_accuracy": 0.23095116317272185,
      "num_tokens": 139460286.0,
      "step": 80410
    },
    {
      "entropy": 5.431776809692383,
      "epoch": 6.256525967710562,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019036300832887015,
      "loss": 4.7018,
      "mean_token_accuracy": 0.23521775305271148,
      "num_tokens": 139468888.0,
      "step": 80415
    },
    {
      "entropy": 5.453109979629517,
      "epoch": 6.25691499708228,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019033732901202339,
      "loss": 4.6602,
      "mean_token_accuracy": 0.23661701679229735,
      "num_tokens": 139477123.0,
      "step": 80420
    },
    {
      "entropy": 5.423855257034302,
      "epoch": 6.257304026453998,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001903116509797979,
      "loss": 4.6417,
      "mean_token_accuracy": 0.23728996366262436,
      "num_tokens": 139486211.0,
      "step": 80425
    },
    {
      "entropy": 5.383970117568969,
      "epoch": 6.2576930558257144,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001902859742325833,
      "loss": 4.5875,
      "mean_token_accuracy": 0.25242227166891096,
      "num_tokens": 139495389.0,
      "step": 80430
    },
    {
      "entropy": 5.3530810356140135,
      "epoch": 6.258082085197432,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019026029877076923,
      "loss": 4.5874,
      "mean_token_accuracy": 0.24760037213563918,
      "num_tokens": 139503649.0,
      "step": 80435
    },
    {
      "entropy": 5.389848279953003,
      "epoch": 6.25847111456915,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001902346245947452,
      "loss": 4.7814,
      "mean_token_accuracy": 0.2215313196182251,
      "num_tokens": 139511679.0,
      "step": 80440
    },
    {
      "entropy": 5.384269189834595,
      "epoch": 6.2588601439408675,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019020895170490082,
      "loss": 4.6092,
      "mean_token_accuracy": 0.24144149720668792,
      "num_tokens": 139520095.0,
      "step": 80445
    },
    {
      "entropy": 5.399913263320923,
      "epoch": 6.259249173312585,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019018328010162567,
      "loss": 4.6015,
      "mean_token_accuracy": 0.24492089748382567,
      "num_tokens": 139528955.0,
      "step": 80450
    },
    {
      "entropy": 5.394786834716797,
      "epoch": 6.259638202684303,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019015760978530926,
      "loss": 4.6633,
      "mean_token_accuracy": 0.24236784726381302,
      "num_tokens": 139536907.0,
      "step": 80455
    },
    {
      "entropy": 5.353962421417236,
      "epoch": 6.260027232056021,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019013194075634106,
      "loss": 4.6374,
      "mean_token_accuracy": 0.2408549204468727,
      "num_tokens": 139545500.0,
      "step": 80460
    },
    {
      "entropy": 5.438731908798218,
      "epoch": 6.260416261427737,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019010627301511057,
      "loss": 4.67,
      "mean_token_accuracy": 0.23069639950990678,
      "num_tokens": 139553877.0,
      "step": 80465
    },
    {
      "entropy": 5.386192655563354,
      "epoch": 6.260805290799455,
      "grad_norm": 1.265625,
      "learning_rate": 0.00019008060656200732,
      "loss": 4.5683,
      "mean_token_accuracy": 0.2467222183942795,
      "num_tokens": 139562504.0,
      "step": 80470
    },
    {
      "entropy": 5.435913944244385,
      "epoch": 6.261194320171173,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019005494139742068,
      "loss": 4.7214,
      "mean_token_accuracy": 0.23638402074575424,
      "num_tokens": 139570969.0,
      "step": 80475
    },
    {
      "entropy": 5.397916555404663,
      "epoch": 6.2615833495428905,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001900292775217401,
      "loss": 4.6522,
      "mean_token_accuracy": 0.2320972502231598,
      "num_tokens": 139579615.0,
      "step": 80480
    },
    {
      "entropy": 5.388457012176514,
      "epoch": 6.261972378914608,
      "grad_norm": 1.40625,
      "learning_rate": 0.000190003614935355,
      "loss": 4.6128,
      "mean_token_accuracy": 0.24789093881845475,
      "num_tokens": 139587660.0,
      "step": 80485
    },
    {
      "entropy": 5.364661359786988,
      "epoch": 6.262361408286326,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018997795363865473,
      "loss": 4.6784,
      "mean_token_accuracy": 0.23412997275590897,
      "num_tokens": 139596822.0,
      "step": 80490
    },
    {
      "entropy": 5.465862226486206,
      "epoch": 6.262750437658044,
      "grad_norm": 1.25,
      "learning_rate": 0.0001899522936320287,
      "loss": 4.7171,
      "mean_token_accuracy": 0.23117563277482986,
      "num_tokens": 139605639.0,
      "step": 80495
    },
    {
      "entropy": 5.379311656951904,
      "epoch": 6.26313946702976,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001899266349158662,
      "loss": 4.5879,
      "mean_token_accuracy": 0.24366039633750916,
      "num_tokens": 139614511.0,
      "step": 80500
    },
    {
      "entropy": 5.417331266403198,
      "epoch": 6.263528496401478,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018990097749055669,
      "loss": 4.6977,
      "mean_token_accuracy": 0.2411031022667885,
      "num_tokens": 139622856.0,
      "step": 80505
    },
    {
      "entropy": 5.33144474029541,
      "epoch": 6.263917525773196,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018987532135648926,
      "loss": 4.5647,
      "mean_token_accuracy": 0.2460334524512291,
      "num_tokens": 139631718.0,
      "step": 80510
    },
    {
      "entropy": 5.315123701095581,
      "epoch": 6.2643065551449135,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001898496665140535,
      "loss": 4.564,
      "mean_token_accuracy": 0.24318933337926865,
      "num_tokens": 139640252.0,
      "step": 80515
    },
    {
      "entropy": 5.313232278823852,
      "epoch": 6.264695584516631,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018982401296363844,
      "loss": 4.63,
      "mean_token_accuracy": 0.23560360372066497,
      "num_tokens": 139649013.0,
      "step": 80520
    },
    {
      "entropy": 5.408234739303589,
      "epoch": 6.265084613888349,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018979836070563339,
      "loss": 4.6446,
      "mean_token_accuracy": 0.2374725192785263,
      "num_tokens": 139658113.0,
      "step": 80525
    },
    {
      "entropy": 5.362346363067627,
      "epoch": 6.265473643260066,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018977270974042764,
      "loss": 4.6271,
      "mean_token_accuracy": 0.24053695797920227,
      "num_tokens": 139666430.0,
      "step": 80530
    },
    {
      "entropy": 5.463942384719848,
      "epoch": 6.265862672631783,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018974706006841035,
      "loss": 4.68,
      "mean_token_accuracy": 0.24543213993310928,
      "num_tokens": 139675283.0,
      "step": 80535
    },
    {
      "entropy": 5.3325052738189695,
      "epoch": 6.266251702003501,
      "grad_norm": 1.421875,
      "learning_rate": 0.00018972141168997075,
      "loss": 4.5359,
      "mean_token_accuracy": 0.24635443687438965,
      "num_tokens": 139683487.0,
      "step": 80540
    },
    {
      "entropy": 5.376246356964112,
      "epoch": 6.266640731375219,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00018969576460549798,
      "loss": 4.661,
      "mean_token_accuracy": 0.2364721938967705,
      "num_tokens": 139692376.0,
      "step": 80545
    },
    {
      "entropy": 5.381367254257202,
      "epoch": 6.2670297607469365,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018967011881538122,
      "loss": 4.6077,
      "mean_token_accuracy": 0.24235492646694184,
      "num_tokens": 139700855.0,
      "step": 80550
    },
    {
      "entropy": 5.388552904129028,
      "epoch": 6.267418790118654,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018964447432000964,
      "loss": 4.6846,
      "mean_token_accuracy": 0.23735942244529723,
      "num_tokens": 139709689.0,
      "step": 80555
    },
    {
      "entropy": 5.446264600753784,
      "epoch": 6.267807819490372,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018961883111977225,
      "loss": 4.6971,
      "mean_token_accuracy": 0.23705108761787413,
      "num_tokens": 139718016.0,
      "step": 80560
    },
    {
      "entropy": 5.373779392242431,
      "epoch": 6.268196848862089,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001895931892150582,
      "loss": 4.6269,
      "mean_token_accuracy": 0.2398022949695587,
      "num_tokens": 139726464.0,
      "step": 80565
    },
    {
      "entropy": 5.417043256759643,
      "epoch": 6.268585878233806,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018956754860625653,
      "loss": 4.6176,
      "mean_token_accuracy": 0.23868566751480103,
      "num_tokens": 139735074.0,
      "step": 80570
    },
    {
      "entropy": 5.384158992767334,
      "epoch": 6.268974907605524,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018954190929375625,
      "loss": 4.613,
      "mean_token_accuracy": 0.2355734646320343,
      "num_tokens": 139743915.0,
      "step": 80575
    },
    {
      "entropy": 5.446339464187622,
      "epoch": 6.269363936977242,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018951627127794656,
      "loss": 4.6959,
      "mean_token_accuracy": 0.23905368596315385,
      "num_tokens": 139751839.0,
      "step": 80580
    },
    {
      "entropy": 5.3889909267425535,
      "epoch": 6.2697529663489595,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001894906345592165,
      "loss": 4.6365,
      "mean_token_accuracy": 0.23967001140117644,
      "num_tokens": 139761155.0,
      "step": 80585
    },
    {
      "entropy": 5.35652527809143,
      "epoch": 6.270141995720677,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001894649991379549,
      "loss": 4.5839,
      "mean_token_accuracy": 0.24570319056510925,
      "num_tokens": 139768916.0,
      "step": 80590
    },
    {
      "entropy": 5.40945725440979,
      "epoch": 6.270531025092394,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018943936501455077,
      "loss": 4.6615,
      "mean_token_accuracy": 0.23720156252384186,
      "num_tokens": 139777065.0,
      "step": 80595
    },
    {
      "entropy": 5.422574424743653,
      "epoch": 6.270920054464112,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018941373218939306,
      "loss": 4.6438,
      "mean_token_accuracy": 0.23594748824834824,
      "num_tokens": 139785633.0,
      "step": 80600
    },
    {
      "entropy": 5.349172353744507,
      "epoch": 6.271309083835829,
      "grad_norm": 1.171875,
      "learning_rate": 0.00018938810066287078,
      "loss": 4.5972,
      "mean_token_accuracy": 0.23926064372062683,
      "num_tokens": 139794601.0,
      "step": 80605
    },
    {
      "entropy": 5.419375467300415,
      "epoch": 6.271698113207547,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001893624704353728,
      "loss": 4.7497,
      "mean_token_accuracy": 0.23352694511413574,
      "num_tokens": 139804169.0,
      "step": 80610
    },
    {
      "entropy": 5.360839557647705,
      "epoch": 6.272087142579265,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018933684150728802,
      "loss": 4.6138,
      "mean_token_accuracy": 0.2471909612417221,
      "num_tokens": 139813003.0,
      "step": 80615
    },
    {
      "entropy": 5.351366138458252,
      "epoch": 6.2724761719509825,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018931121387900536,
      "loss": 4.5867,
      "mean_token_accuracy": 0.24640681147575377,
      "num_tokens": 139821941.0,
      "step": 80620
    },
    {
      "entropy": 5.385867738723755,
      "epoch": 6.2728652013227,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018928558755091358,
      "loss": 4.7004,
      "mean_token_accuracy": 0.23446635156869888,
      "num_tokens": 139830080.0,
      "step": 80625
    },
    {
      "entropy": 5.4055643558502195,
      "epoch": 6.273254230694418,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018925996252340167,
      "loss": 4.6936,
      "mean_token_accuracy": 0.2436074882745743,
      "num_tokens": 139838660.0,
      "step": 80630
    },
    {
      "entropy": 5.346486616134643,
      "epoch": 6.273643260066135,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001892343387968583,
      "loss": 4.6234,
      "mean_token_accuracy": 0.2451077178120613,
      "num_tokens": 139846915.0,
      "step": 80635
    },
    {
      "entropy": 5.349494314193725,
      "epoch": 6.274032289437852,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018920871637167236,
      "loss": 4.6108,
      "mean_token_accuracy": 0.24402038007974625,
      "num_tokens": 139855507.0,
      "step": 80640
    },
    {
      "entropy": 5.459961223602295,
      "epoch": 6.27442131880957,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018918309524823247,
      "loss": 4.6567,
      "mean_token_accuracy": 0.24426183104515076,
      "num_tokens": 139863895.0,
      "step": 80645
    },
    {
      "entropy": 5.446266508102417,
      "epoch": 6.274810348181288,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00018915747542692758,
      "loss": 4.7131,
      "mean_token_accuracy": 0.23641157597303392,
      "num_tokens": 139872636.0,
      "step": 80650
    },
    {
      "entropy": 5.454562044143676,
      "epoch": 6.2751993775530055,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018913185690814637,
      "loss": 4.7336,
      "mean_token_accuracy": 0.23571633845567702,
      "num_tokens": 139881147.0,
      "step": 80655
    },
    {
      "entropy": 5.329338407516479,
      "epoch": 6.275588406924723,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018910623969227764,
      "loss": 4.5654,
      "mean_token_accuracy": 0.24510546028614044,
      "num_tokens": 139890009.0,
      "step": 80660
    },
    {
      "entropy": 5.387473726272583,
      "epoch": 6.27597743629644,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018908062377971,
      "loss": 4.6465,
      "mean_token_accuracy": 0.24790654927492142,
      "num_tokens": 139899255.0,
      "step": 80665
    },
    {
      "entropy": 5.345875930786133,
      "epoch": 6.276366465668158,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018905500917083213,
      "loss": 4.6637,
      "mean_token_accuracy": 0.2340237095952034,
      "num_tokens": 139908119.0,
      "step": 80670
    },
    {
      "entropy": 5.45190920829773,
      "epoch": 6.276755495039875,
      "grad_norm": 1.421875,
      "learning_rate": 0.00018902939586603262,
      "loss": 4.6488,
      "mean_token_accuracy": 0.2409891664981842,
      "num_tokens": 139916819.0,
      "step": 80675
    },
    {
      "entropy": 5.430678796768189,
      "epoch": 6.277144524411593,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018900378386570027,
      "loss": 4.7093,
      "mean_token_accuracy": 0.2340683728456497,
      "num_tokens": 139925334.0,
      "step": 80680
    },
    {
      "entropy": 5.42761082649231,
      "epoch": 6.277533553783311,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001889781731702236,
      "loss": 4.668,
      "mean_token_accuracy": 0.24220959544181825,
      "num_tokens": 139934458.0,
      "step": 80685
    },
    {
      "entropy": 5.417144155502319,
      "epoch": 6.2779225831550285,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018895256377999121,
      "loss": 4.6779,
      "mean_token_accuracy": 0.23350462764501573,
      "num_tokens": 139943227.0,
      "step": 80690
    },
    {
      "entropy": 5.327381944656372,
      "epoch": 6.278311612526746,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001889269556953917,
      "loss": 4.547,
      "mean_token_accuracy": 0.25325956642627717,
      "num_tokens": 139952580.0,
      "step": 80695
    },
    {
      "entropy": 5.425938749313355,
      "epoch": 6.278700641898463,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001889013489168137,
      "loss": 4.6854,
      "mean_token_accuracy": 0.2421893984079361,
      "num_tokens": 139961239.0,
      "step": 80700
    },
    {
      "entropy": 5.377617454528808,
      "epoch": 6.279089671270181,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018887574344464563,
      "loss": 4.6381,
      "mean_token_accuracy": 0.23616444319486618,
      "num_tokens": 139969646.0,
      "step": 80705
    },
    {
      "entropy": 5.451131868362427,
      "epoch": 6.279478700641898,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018885013927927608,
      "loss": 4.7057,
      "mean_token_accuracy": 0.23788418024778366,
      "num_tokens": 139977674.0,
      "step": 80710
    },
    {
      "entropy": 5.388899230957032,
      "epoch": 6.279867730013616,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018882453642109353,
      "loss": 4.6776,
      "mean_token_accuracy": 0.23725140541791917,
      "num_tokens": 139986957.0,
      "step": 80715
    },
    {
      "entropy": 5.445253849029541,
      "epoch": 6.280256759385334,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001887989348704865,
      "loss": 4.7516,
      "mean_token_accuracy": 0.23832908421754836,
      "num_tokens": 139996532.0,
      "step": 80720
    },
    {
      "entropy": 5.279995965957641,
      "epoch": 6.2806457887570515,
      "grad_norm": 1.25,
      "learning_rate": 0.00018877333462784346,
      "loss": 4.5009,
      "mean_token_accuracy": 0.2641653805971146,
      "num_tokens": 140005859.0,
      "step": 80725
    },
    {
      "entropy": 5.473435115814209,
      "epoch": 6.281034818128768,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001887477356935528,
      "loss": 4.7472,
      "mean_token_accuracy": 0.22952835857868195,
      "num_tokens": 140015113.0,
      "step": 80730
    },
    {
      "entropy": 5.369877815246582,
      "epoch": 6.281423847500486,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018872213806800297,
      "loss": 4.6231,
      "mean_token_accuracy": 0.23888950496912004,
      "num_tokens": 140023944.0,
      "step": 80735
    },
    {
      "entropy": 5.428376626968384,
      "epoch": 6.281812876872204,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018869654175158234,
      "loss": 4.6476,
      "mean_token_accuracy": 0.24705903530120848,
      "num_tokens": 140032287.0,
      "step": 80740
    },
    {
      "entropy": 5.44474515914917,
      "epoch": 6.282201906243921,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001886709467446794,
      "loss": 4.7759,
      "mean_token_accuracy": 0.23433725386857987,
      "num_tokens": 140040344.0,
      "step": 80745
    },
    {
      "entropy": 5.379383087158203,
      "epoch": 6.282590935615639,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018864535304768238,
      "loss": 4.6911,
      "mean_token_accuracy": 0.23669513016939164,
      "num_tokens": 140048661.0,
      "step": 80750
    },
    {
      "entropy": 5.332699823379516,
      "epoch": 6.282979964987357,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018861976066097968,
      "loss": 4.6223,
      "mean_token_accuracy": 0.2412927195429802,
      "num_tokens": 140057573.0,
      "step": 80755
    },
    {
      "entropy": 5.435874700546265,
      "epoch": 6.2833689943590745,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018859416958495962,
      "loss": 4.7095,
      "mean_token_accuracy": 0.2424481675028801,
      "num_tokens": 140066850.0,
      "step": 80760
    },
    {
      "entropy": 5.420366001129151,
      "epoch": 6.283758023730791,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00018856857982001051,
      "loss": 4.7108,
      "mean_token_accuracy": 0.23222759366035461,
      "num_tokens": 140076067.0,
      "step": 80765
    },
    {
      "entropy": 5.369050788879394,
      "epoch": 6.284147053102509,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001885429913665207,
      "loss": 4.6313,
      "mean_token_accuracy": 0.24103206098079683,
      "num_tokens": 140085417.0,
      "step": 80770
    },
    {
      "entropy": 5.374463367462158,
      "epoch": 6.284536082474227,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018851740422487829,
      "loss": 4.5843,
      "mean_token_accuracy": 0.24336903989315034,
      "num_tokens": 140094569.0,
      "step": 80775
    },
    {
      "entropy": 5.366031885147095,
      "epoch": 6.284925111845944,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00018849181839547164,
      "loss": 4.6207,
      "mean_token_accuracy": 0.2436074748635292,
      "num_tokens": 140103518.0,
      "step": 80780
    },
    {
      "entropy": 5.44331316947937,
      "epoch": 6.285314141217662,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018846623387868894,
      "loss": 4.6867,
      "mean_token_accuracy": 0.22773874402046204,
      "num_tokens": 140111352.0,
      "step": 80785
    },
    {
      "entropy": 5.3641032695770265,
      "epoch": 6.28570317058938,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018844065067491844,
      "loss": 4.5772,
      "mean_token_accuracy": 0.24944186508655547,
      "num_tokens": 140119417.0,
      "step": 80790
    },
    {
      "entropy": 5.383664989471436,
      "epoch": 6.2860921999610975,
      "grad_norm": 1.25,
      "learning_rate": 0.00018841506878454828,
      "loss": 4.6643,
      "mean_token_accuracy": 0.23693789690732955,
      "num_tokens": 140128171.0,
      "step": 80795
    },
    {
      "entropy": 5.302205371856689,
      "epoch": 6.286481229332814,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018838948820796664,
      "loss": 4.6041,
      "mean_token_accuracy": 0.24092597812414168,
      "num_tokens": 140137594.0,
      "step": 80800
    },
    {
      "entropy": 5.418511819839478,
      "epoch": 6.286870258704532,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00018836390894556165,
      "loss": 4.6908,
      "mean_token_accuracy": 0.2334694132208824,
      "num_tokens": 140146415.0,
      "step": 80805
    },
    {
      "entropy": 5.528481960296631,
      "epoch": 6.28725928807625,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00018833833099772147,
      "loss": 4.8212,
      "mean_token_accuracy": 0.22140052169561386,
      "num_tokens": 140154414.0,
      "step": 80810
    },
    {
      "entropy": 5.345439815521241,
      "epoch": 6.287648317447967,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018831275436483418,
      "loss": 4.6023,
      "mean_token_accuracy": 0.24361783862113953,
      "num_tokens": 140162947.0,
      "step": 80815
    },
    {
      "entropy": 5.459061241149902,
      "epoch": 6.288037346819685,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018828717904728788,
      "loss": 4.7638,
      "mean_token_accuracy": 0.23411581963300704,
      "num_tokens": 140171853.0,
      "step": 80820
    },
    {
      "entropy": 5.490664100646972,
      "epoch": 6.288426376191403,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018826160504547063,
      "loss": 4.7824,
      "mean_token_accuracy": 0.23064191341400148,
      "num_tokens": 140181090.0,
      "step": 80825
    },
    {
      "entropy": 5.416301727294922,
      "epoch": 6.2888154055631205,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018823603235977043,
      "loss": 4.6359,
      "mean_token_accuracy": 0.24007037580013274,
      "num_tokens": 140189725.0,
      "step": 80830
    },
    {
      "entropy": 5.359653043746948,
      "epoch": 6.289204434934837,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018821046099057532,
      "loss": 4.5873,
      "mean_token_accuracy": 0.24688362032175065,
      "num_tokens": 140197791.0,
      "step": 80835
    },
    {
      "entropy": 5.324598598480224,
      "epoch": 6.289593464306555,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018818489093827334,
      "loss": 4.5822,
      "mean_token_accuracy": 0.25362949818372726,
      "num_tokens": 140207029.0,
      "step": 80840
    },
    {
      "entropy": 5.395852041244507,
      "epoch": 6.289982493678273,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018815932220325239,
      "loss": 4.6426,
      "mean_token_accuracy": 0.2395107626914978,
      "num_tokens": 140216295.0,
      "step": 80845
    },
    {
      "entropy": 5.371996450424194,
      "epoch": 6.29037152304999,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001881337547859006,
      "loss": 4.6489,
      "mean_token_accuracy": 0.23724377304315566,
      "num_tokens": 140225219.0,
      "step": 80850
    },
    {
      "entropy": 5.425933599472046,
      "epoch": 6.290760552421708,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018810818868660585,
      "loss": 4.6497,
      "mean_token_accuracy": 0.24207971841096879,
      "num_tokens": 140233437.0,
      "step": 80855
    },
    {
      "entropy": 5.397924089431763,
      "epoch": 6.291149581793426,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018808262390575598,
      "loss": 4.6797,
      "mean_token_accuracy": 0.23178303837776185,
      "num_tokens": 140242084.0,
      "step": 80860
    },
    {
      "entropy": 5.375138950347901,
      "epoch": 6.291538611165143,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018805706044373893,
      "loss": 4.6235,
      "mean_token_accuracy": 0.23665915578603744,
      "num_tokens": 140250889.0,
      "step": 80865
    },
    {
      "entropy": 5.322999477386475,
      "epoch": 6.29192764053686,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018803149830094268,
      "loss": 4.6268,
      "mean_token_accuracy": 0.2407200291752815,
      "num_tokens": 140259149.0,
      "step": 80870
    },
    {
      "entropy": 5.4123610496521,
      "epoch": 6.292316669908578,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00018800593747775492,
      "loss": 4.702,
      "mean_token_accuracy": 0.22775417268276216,
      "num_tokens": 140267905.0,
      "step": 80875
    },
    {
      "entropy": 5.387538766860962,
      "epoch": 6.292705699280296,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018798037797456368,
      "loss": 4.5653,
      "mean_token_accuracy": 0.24655814170837403,
      "num_tokens": 140276990.0,
      "step": 80880
    },
    {
      "entropy": 5.432442617416382,
      "epoch": 6.293094728652013,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001879548197917566,
      "loss": 4.6861,
      "mean_token_accuracy": 0.23746777921915055,
      "num_tokens": 140284897.0,
      "step": 80885
    },
    {
      "entropy": 5.387422800064087,
      "epoch": 6.293483758023731,
      "grad_norm": 1.375,
      "learning_rate": 0.00018792926292972168,
      "loss": 4.6508,
      "mean_token_accuracy": 0.2403615042567253,
      "num_tokens": 140293737.0,
      "step": 80890
    },
    {
      "entropy": 5.437067365646362,
      "epoch": 6.293872787395449,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018790370738884653,
      "loss": 4.738,
      "mean_token_accuracy": 0.2402784824371338,
      "num_tokens": 140302544.0,
      "step": 80895
    },
    {
      "entropy": 5.376826810836792,
      "epoch": 6.294261816767166,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018787815316951905,
      "loss": 4.632,
      "mean_token_accuracy": 0.24814800173044205,
      "num_tokens": 140312052.0,
      "step": 80900
    },
    {
      "entropy": 5.396762323379517,
      "epoch": 6.294650846138883,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018785260027212692,
      "loss": 4.7373,
      "mean_token_accuracy": 0.23124824166297914,
      "num_tokens": 140321429.0,
      "step": 80905
    },
    {
      "entropy": 5.443631219863891,
      "epoch": 6.295039875510601,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001878270486970578,
      "loss": 4.667,
      "mean_token_accuracy": 0.24035699516534806,
      "num_tokens": 140329503.0,
      "step": 80910
    },
    {
      "entropy": 5.361793088912964,
      "epoch": 6.295428904882319,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018780149844469958,
      "loss": 4.5446,
      "mean_token_accuracy": 0.244250525534153,
      "num_tokens": 140338355.0,
      "step": 80915
    },
    {
      "entropy": 5.32082233428955,
      "epoch": 6.295817934254036,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018777594951543974,
      "loss": 4.5653,
      "mean_token_accuracy": 0.24929204136133193,
      "num_tokens": 140347292.0,
      "step": 80920
    },
    {
      "entropy": 5.323388719558716,
      "epoch": 6.296206963625754,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018775040190966607,
      "loss": 4.6138,
      "mean_token_accuracy": 0.24406735897064208,
      "num_tokens": 140356137.0,
      "step": 80925
    },
    {
      "entropy": 5.410298919677734,
      "epoch": 6.296595992997471,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001877248556277662,
      "loss": 4.6981,
      "mean_token_accuracy": 0.22988900393247605,
      "num_tokens": 140364352.0,
      "step": 80930
    },
    {
      "entropy": 5.429219484329224,
      "epoch": 6.296985022369189,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018769931067012775,
      "loss": 4.6928,
      "mean_token_accuracy": 0.23361212760210037,
      "num_tokens": 140373075.0,
      "step": 80935
    },
    {
      "entropy": 5.44643349647522,
      "epoch": 6.297374051740906,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018767376703713828,
      "loss": 4.6833,
      "mean_token_accuracy": 0.24257458746433258,
      "num_tokens": 140382463.0,
      "step": 80940
    },
    {
      "entropy": 5.4415696144104,
      "epoch": 6.297763081112624,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018764822472918535,
      "loss": 4.6784,
      "mean_token_accuracy": 0.23956273943185807,
      "num_tokens": 140390747.0,
      "step": 80945
    },
    {
      "entropy": 5.406580972671509,
      "epoch": 6.298152110484342,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018762268374665665,
      "loss": 4.6893,
      "mean_token_accuracy": 0.23705904930830002,
      "num_tokens": 140399656.0,
      "step": 80950
    },
    {
      "entropy": 5.342750597000122,
      "epoch": 6.298541139856059,
      "grad_norm": 1.15625,
      "learning_rate": 0.00018759714408993962,
      "loss": 4.6293,
      "mean_token_accuracy": 0.2334432601928711,
      "num_tokens": 140408483.0,
      "step": 80955
    },
    {
      "entropy": 5.3609397411346436,
      "epoch": 6.298930169227777,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018757160575942183,
      "loss": 4.631,
      "mean_token_accuracy": 0.24056716561317443,
      "num_tokens": 140417186.0,
      "step": 80960
    },
    {
      "entropy": 5.387080192565918,
      "epoch": 6.299319198599494,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018754606875549074,
      "loss": 4.6466,
      "mean_token_accuracy": 0.23672270625829697,
      "num_tokens": 140425302.0,
      "step": 80965
    },
    {
      "entropy": 5.365395021438599,
      "epoch": 6.2997082279712115,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018752053307853384,
      "loss": 4.6618,
      "mean_token_accuracy": 0.23655004501342775,
      "num_tokens": 140433738.0,
      "step": 80970
    },
    {
      "entropy": 5.395830059051514,
      "epoch": 6.300097257342929,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001874949987289386,
      "loss": 4.7404,
      "mean_token_accuracy": 0.2288475066423416,
      "num_tokens": 140442131.0,
      "step": 80975
    },
    {
      "entropy": 5.388530111312866,
      "epoch": 6.300486286714647,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001874694657070925,
      "loss": 4.6064,
      "mean_token_accuracy": 0.24498714357614518,
      "num_tokens": 140449757.0,
      "step": 80980
    },
    {
      "entropy": 5.480423307418823,
      "epoch": 6.300875316086365,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018744393401338294,
      "loss": 4.7635,
      "mean_token_accuracy": 0.22712938636541366,
      "num_tokens": 140458101.0,
      "step": 80985
    },
    {
      "entropy": 5.438992834091186,
      "epoch": 6.301264345458082,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018741840364819735,
      "loss": 4.6687,
      "mean_token_accuracy": 0.24204365015029908,
      "num_tokens": 140466006.0,
      "step": 80990
    },
    {
      "entropy": 5.465781354904175,
      "epoch": 6.3016533748298,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018739287461192307,
      "loss": 4.8018,
      "mean_token_accuracy": 0.2240292027592659,
      "num_tokens": 140474101.0,
      "step": 80995
    },
    {
      "entropy": 5.413680219650269,
      "epoch": 6.302042404201517,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00018736734690494745,
      "loss": 4.6341,
      "mean_token_accuracy": 0.24746754318475722,
      "num_tokens": 140482564.0,
      "step": 81000
    },
    {
      "epoch": 6.302042404201517,
      "eval_entropy": 5.1702743285200246,
      "eval_loss": 4.871466636657715,
      "eval_mean_token_accuracy": 0.23425593627571945,
      "eval_num_tokens": 140482564.0,
      "eval_runtime": 28.9505,
      "eval_samples_per_second": 1435.486,
      "eval_steps_per_second": 179.444,
      "step": 81000
    },
    {
      "entropy": 5.492792844772339,
      "epoch": 6.3024314335732345,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001873418205276579,
      "loss": 4.7178,
      "mean_token_accuracy": 0.2378475472331047,
      "num_tokens": 140492149.0,
      "step": 81005
    },
    {
      "entropy": 5.456882953643799,
      "epoch": 6.302820462944952,
      "grad_norm": 1.25,
      "learning_rate": 0.00018731629548044166,
      "loss": 4.6964,
      "mean_token_accuracy": 0.24167658686637877,
      "num_tokens": 140500869.0,
      "step": 81010
    },
    {
      "entropy": 5.433547782897949,
      "epoch": 6.30320949231667,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018729077176368614,
      "loss": 4.6593,
      "mean_token_accuracy": 0.24496403187513352,
      "num_tokens": 140509531.0,
      "step": 81015
    },
    {
      "entropy": 5.369193744659424,
      "epoch": 6.303598521688388,
      "grad_norm": 1.421875,
      "learning_rate": 0.00018726524937777856,
      "loss": 4.6153,
      "mean_token_accuracy": 0.23776921182870864,
      "num_tokens": 140518368.0,
      "step": 81020
    },
    {
      "entropy": 5.372480344772339,
      "epoch": 6.303987551060105,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001872397283231061,
      "loss": 4.6043,
      "mean_token_accuracy": 0.24504177123308182,
      "num_tokens": 140526062.0,
      "step": 81025
    },
    {
      "entropy": 5.4226398944854735,
      "epoch": 6.304376580431823,
      "grad_norm": 1.25,
      "learning_rate": 0.00018721420860005612,
      "loss": 4.72,
      "mean_token_accuracy": 0.24133848547935485,
      "num_tokens": 140534982.0,
      "step": 81030
    },
    {
      "entropy": 5.437431716918946,
      "epoch": 6.30476560980354,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018718869020901575,
      "loss": 4.7102,
      "mean_token_accuracy": 0.23205984085798265,
      "num_tokens": 140543105.0,
      "step": 81035
    },
    {
      "entropy": 5.3646622657775875,
      "epoch": 6.3051546391752575,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00018716317315037217,
      "loss": 4.6474,
      "mean_token_accuracy": 0.24133993238210677,
      "num_tokens": 140551747.0,
      "step": 81040
    },
    {
      "entropy": 5.322890186309815,
      "epoch": 6.305543668546975,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00018713765742451284,
      "loss": 4.5592,
      "mean_token_accuracy": 0.2465199872851372,
      "num_tokens": 140560594.0,
      "step": 81045
    },
    {
      "entropy": 5.395346832275391,
      "epoch": 6.305932697918693,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018711214303182455,
      "loss": 4.6322,
      "mean_token_accuracy": 0.24815107882022858,
      "num_tokens": 140569955.0,
      "step": 81050
    },
    {
      "entropy": 5.461027002334594,
      "epoch": 6.306321727290411,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018708662997269466,
      "loss": 4.7449,
      "mean_token_accuracy": 0.23678188621997834,
      "num_tokens": 140578296.0,
      "step": 81055
    },
    {
      "entropy": 5.480560207366944,
      "epoch": 6.306710756662128,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018706111824751026,
      "loss": 4.6617,
      "mean_token_accuracy": 0.2437663272023201,
      "num_tokens": 140586456.0,
      "step": 81060
    },
    {
      "entropy": 5.405377292633057,
      "epoch": 6.307099786033845,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001870356078566584,
      "loss": 4.6903,
      "mean_token_accuracy": 0.23729097247123718,
      "num_tokens": 140595512.0,
      "step": 81065
    },
    {
      "entropy": 5.37175235748291,
      "epoch": 6.307488815405563,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018701009880052615,
      "loss": 4.6751,
      "mean_token_accuracy": 0.23304772526025772,
      "num_tokens": 140604383.0,
      "step": 81070
    },
    {
      "entropy": 5.446683597564697,
      "epoch": 6.3078778447772805,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018698459107950062,
      "loss": 4.714,
      "mean_token_accuracy": 0.23509371727705003,
      "num_tokens": 140612800.0,
      "step": 81075
    },
    {
      "entropy": 5.424234771728516,
      "epoch": 6.308266874148998,
      "grad_norm": 1.375,
      "learning_rate": 0.00018695908469396882,
      "loss": 4.7054,
      "mean_token_accuracy": 0.23271385878324508,
      "num_tokens": 140622238.0,
      "step": 81080
    },
    {
      "entropy": 5.448532962799073,
      "epoch": 6.308655903520716,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018693357964431778,
      "loss": 4.713,
      "mean_token_accuracy": 0.23707250207662584,
      "num_tokens": 140631976.0,
      "step": 81085
    },
    {
      "entropy": 5.321987867355347,
      "epoch": 6.309044932892434,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001869080759309345,
      "loss": 4.5722,
      "mean_token_accuracy": 0.24573880881071092,
      "num_tokens": 140640574.0,
      "step": 81090
    },
    {
      "entropy": 5.351877117156983,
      "epoch": 6.309433962264151,
      "grad_norm": 1.28125,
      "learning_rate": 0.000186882573554206,
      "loss": 4.6426,
      "mean_token_accuracy": 0.24525856971740723,
      "num_tokens": 140649387.0,
      "step": 81095
    },
    {
      "entropy": 5.312054014205932,
      "epoch": 6.309822991635868,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018685707251451923,
      "loss": 4.5095,
      "mean_token_accuracy": 0.25375395715236665,
      "num_tokens": 140657745.0,
      "step": 81100
    },
    {
      "entropy": 5.388836526870728,
      "epoch": 6.310212021007586,
      "grad_norm": 1.40625,
      "learning_rate": 0.000186831572812261,
      "loss": 4.6609,
      "mean_token_accuracy": 0.24664531350135804,
      "num_tokens": 140666005.0,
      "step": 81105
    },
    {
      "entropy": 5.373680591583252,
      "epoch": 6.3106010503793035,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018680607444781822,
      "loss": 4.6448,
      "mean_token_accuracy": 0.23789597749710084,
      "num_tokens": 140674754.0,
      "step": 81110
    },
    {
      "entropy": 5.35059700012207,
      "epoch": 6.310990079751021,
      "grad_norm": 1.375,
      "learning_rate": 0.000186780577421578,
      "loss": 4.6961,
      "mean_token_accuracy": 0.23871206790208815,
      "num_tokens": 140683039.0,
      "step": 81115
    },
    {
      "entropy": 5.336535930633545,
      "epoch": 6.311379109122739,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018675508173392715,
      "loss": 4.6639,
      "mean_token_accuracy": 0.24021504074335098,
      "num_tokens": 140691637.0,
      "step": 81120
    },
    {
      "entropy": 5.288360118865967,
      "epoch": 6.311768138494457,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018672958738525242,
      "loss": 4.5585,
      "mean_token_accuracy": 0.24004936963319778,
      "num_tokens": 140699711.0,
      "step": 81125
    },
    {
      "entropy": 5.384496974945068,
      "epoch": 6.312157167866174,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001867040943759407,
      "loss": 4.6188,
      "mean_token_accuracy": 0.23622289150953293,
      "num_tokens": 140708784.0,
      "step": 81130
    },
    {
      "entropy": 5.382076215744019,
      "epoch": 6.312546197237891,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018667860270637883,
      "loss": 4.626,
      "mean_token_accuracy": 0.2470238596200943,
      "num_tokens": 140717302.0,
      "step": 81135
    },
    {
      "entropy": 5.380614423751831,
      "epoch": 6.312935226609609,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00018665311237695358,
      "loss": 4.6772,
      "mean_token_accuracy": 0.23968446105718613,
      "num_tokens": 140726101.0,
      "step": 81140
    },
    {
      "entropy": 5.388350772857666,
      "epoch": 6.3133242559813265,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00018662762338805174,
      "loss": 4.6452,
      "mean_token_accuracy": 0.24135989099740982,
      "num_tokens": 140735350.0,
      "step": 81145
    },
    {
      "entropy": 5.384521484375,
      "epoch": 6.313713285353044,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018660213574006007,
      "loss": 4.6256,
      "mean_token_accuracy": 0.240645532310009,
      "num_tokens": 140744544.0,
      "step": 81150
    },
    {
      "entropy": 5.304063415527343,
      "epoch": 6.314102314724762,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018657664943336528,
      "loss": 4.5217,
      "mean_token_accuracy": 0.24785464406013488,
      "num_tokens": 140753140.0,
      "step": 81155
    },
    {
      "entropy": 5.341644716262818,
      "epoch": 6.31449134409648,
      "grad_norm": 1.515625,
      "learning_rate": 0.00018655116446835408,
      "loss": 4.5604,
      "mean_token_accuracy": 0.25025922656059263,
      "num_tokens": 140761643.0,
      "step": 81160
    },
    {
      "entropy": 5.4166912078857425,
      "epoch": 6.314880373468197,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018652568084541316,
      "loss": 4.6842,
      "mean_token_accuracy": 0.23657781928777694,
      "num_tokens": 140770551.0,
      "step": 81165
    },
    {
      "entropy": 5.334477472305298,
      "epoch": 6.315269402839914,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018650019856492924,
      "loss": 4.5742,
      "mean_token_accuracy": 0.24270375221967697,
      "num_tokens": 140779663.0,
      "step": 81170
    },
    {
      "entropy": 5.320580387115479,
      "epoch": 6.315658432211632,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001864747176272889,
      "loss": 4.5879,
      "mean_token_accuracy": 0.24315948486328126,
      "num_tokens": 140788723.0,
      "step": 81175
    },
    {
      "entropy": 5.339796876907348,
      "epoch": 6.3160474615833495,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018644923803287882,
      "loss": 4.6052,
      "mean_token_accuracy": 0.2427134245634079,
      "num_tokens": 140796963.0,
      "step": 81180
    },
    {
      "entropy": 5.391323184967041,
      "epoch": 6.316436490955067,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018642375978208565,
      "loss": 4.662,
      "mean_token_accuracy": 0.2399229645729065,
      "num_tokens": 140805449.0,
      "step": 81185
    },
    {
      "entropy": 5.391846227645874,
      "epoch": 6.316825520326785,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018639828287529591,
      "loss": 4.6752,
      "mean_token_accuracy": 0.23541017025709152,
      "num_tokens": 140813976.0,
      "step": 81190
    },
    {
      "entropy": 5.3814839839935305,
      "epoch": 6.317214549698503,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018637280731289625,
      "loss": 4.6732,
      "mean_token_accuracy": 0.23460946828126908,
      "num_tokens": 140823371.0,
      "step": 81195
    },
    {
      "entropy": 5.33790602684021,
      "epoch": 6.317603579070219,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001863473330952732,
      "loss": 4.6473,
      "mean_token_accuracy": 0.23416151404380797,
      "num_tokens": 140833015.0,
      "step": 81200
    },
    {
      "entropy": 5.3981811046600345,
      "epoch": 6.317992608441937,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018632186022281315,
      "loss": 4.6395,
      "mean_token_accuracy": 0.2366352215409279,
      "num_tokens": 140840899.0,
      "step": 81205
    },
    {
      "entropy": 5.368739748001099,
      "epoch": 6.318381637813655,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018629638869590288,
      "loss": 4.6562,
      "mean_token_accuracy": 0.24022720009088516,
      "num_tokens": 140849437.0,
      "step": 81210
    },
    {
      "entropy": 5.400656032562256,
      "epoch": 6.3187706671853725,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001862709185149286,
      "loss": 4.689,
      "mean_token_accuracy": 0.23592578768730163,
      "num_tokens": 140858500.0,
      "step": 81215
    },
    {
      "entropy": 5.413471841812134,
      "epoch": 6.31915969655709,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018624544968027696,
      "loss": 4.6866,
      "mean_token_accuracy": 0.2459576576948166,
      "num_tokens": 140867288.0,
      "step": 81220
    },
    {
      "entropy": 5.381202745437622,
      "epoch": 6.319548725928808,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018621998219233438,
      "loss": 4.6401,
      "mean_token_accuracy": 0.23597426265478133,
      "num_tokens": 140876076.0,
      "step": 81225
    },
    {
      "entropy": 5.3892652034759525,
      "epoch": 6.319937755300526,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018619451605148717,
      "loss": 4.695,
      "mean_token_accuracy": 0.22948427200317384,
      "num_tokens": 140884456.0,
      "step": 81230
    },
    {
      "entropy": 5.420752811431885,
      "epoch": 6.320326784672242,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018616905125812188,
      "loss": 4.6692,
      "mean_token_accuracy": 0.24222567528486252,
      "num_tokens": 140892688.0,
      "step": 81235
    },
    {
      "entropy": 5.370156621932983,
      "epoch": 6.32071581404396,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00018614358781262485,
      "loss": 4.5986,
      "mean_token_accuracy": 0.24845696091651917,
      "num_tokens": 140900705.0,
      "step": 81240
    },
    {
      "entropy": 5.470141124725342,
      "epoch": 6.321104843415678,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001861181257153824,
      "loss": 4.6983,
      "mean_token_accuracy": 0.23500063121318818,
      "num_tokens": 140909234.0,
      "step": 81245
    },
    {
      "entropy": 5.342524003982544,
      "epoch": 6.3214938727873955,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018609266496678095,
      "loss": 4.6018,
      "mean_token_accuracy": 0.23946299105882646,
      "num_tokens": 140917410.0,
      "step": 81250
    },
    {
      "entropy": 5.357816171646118,
      "epoch": 6.321882902159113,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018606720556720686,
      "loss": 4.6305,
      "mean_token_accuracy": 0.2347670614719391,
      "num_tokens": 140925930.0,
      "step": 81255
    },
    {
      "entropy": 5.417809200286865,
      "epoch": 6.322271931530831,
      "grad_norm": 1.25,
      "learning_rate": 0.00018604174751704633,
      "loss": 4.6805,
      "mean_token_accuracy": 0.2429317131638527,
      "num_tokens": 140934197.0,
      "step": 81260
    },
    {
      "entropy": 5.527235078811645,
      "epoch": 6.322660960902548,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001860162908166857,
      "loss": 4.7774,
      "mean_token_accuracy": 0.2270856335759163,
      "num_tokens": 140943468.0,
      "step": 81265
    },
    {
      "entropy": 5.4127270698547365,
      "epoch": 6.323049990274265,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00018599083546651119,
      "loss": 4.6735,
      "mean_token_accuracy": 0.23654923141002654,
      "num_tokens": 140952979.0,
      "step": 81270
    },
    {
      "entropy": 5.466534566879273,
      "epoch": 6.323439019645983,
      "grad_norm": 1.25,
      "learning_rate": 0.0001859653814669091,
      "loss": 4.7331,
      "mean_token_accuracy": 0.2366110473871231,
      "num_tokens": 140961488.0,
      "step": 81275
    },
    {
      "entropy": 5.350458145141602,
      "epoch": 6.323828049017701,
      "grad_norm": 1.25,
      "learning_rate": 0.00018593992881826566,
      "loss": 4.5891,
      "mean_token_accuracy": 0.24671894907951356,
      "num_tokens": 140970144.0,
      "step": 81280
    },
    {
      "entropy": 5.454417181015015,
      "epoch": 6.3242170783894185,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018591447752096702,
      "loss": 4.7548,
      "mean_token_accuracy": 0.23188957571983337,
      "num_tokens": 140979533.0,
      "step": 81285
    },
    {
      "entropy": 5.398789739608764,
      "epoch": 6.324606107761136,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018588902757539944,
      "loss": 4.7132,
      "mean_token_accuracy": 0.2318116009235382,
      "num_tokens": 140987860.0,
      "step": 81290
    },
    {
      "entropy": 5.286217117309571,
      "epoch": 6.324995137132854,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000185863578981949,
      "loss": 4.5603,
      "mean_token_accuracy": 0.26059999614953994,
      "num_tokens": 140996697.0,
      "step": 81295
    },
    {
      "entropy": 5.381032466888428,
      "epoch": 6.325384166504571,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018583813174100183,
      "loss": 4.6654,
      "mean_token_accuracy": 0.24351117312908171,
      "num_tokens": 141005287.0,
      "step": 81300
    },
    {
      "entropy": 5.352812814712524,
      "epoch": 6.325773195876288,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00018581268585294414,
      "loss": 4.6373,
      "mean_token_accuracy": 0.2463349297642708,
      "num_tokens": 141013651.0,
      "step": 81305
    },
    {
      "entropy": 5.353053712844849,
      "epoch": 6.326162225248006,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018578724131816183,
      "loss": 4.6424,
      "mean_token_accuracy": 0.23821027874946593,
      "num_tokens": 141022508.0,
      "step": 81310
    },
    {
      "entropy": 5.330224657058716,
      "epoch": 6.326551254619724,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001857617981370413,
      "loss": 4.5831,
      "mean_token_accuracy": 0.2505003586411476,
      "num_tokens": 141031280.0,
      "step": 81315
    },
    {
      "entropy": 5.368175411224366,
      "epoch": 6.3269402839914415,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018573635630996844,
      "loss": 4.5695,
      "mean_token_accuracy": 0.24152146726846696,
      "num_tokens": 141040232.0,
      "step": 81320
    },
    {
      "entropy": 5.357767581939697,
      "epoch": 6.327329313363159,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001857109158373293,
      "loss": 4.6427,
      "mean_token_accuracy": 0.24082940816879272,
      "num_tokens": 141048556.0,
      "step": 81325
    },
    {
      "entropy": 5.346646070480347,
      "epoch": 6.327718342734877,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018568547671950986,
      "loss": 4.6072,
      "mean_token_accuracy": 0.2389387845993042,
      "num_tokens": 141056623.0,
      "step": 81330
    },
    {
      "entropy": 5.302301692962646,
      "epoch": 6.328107372106594,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018566003895689615,
      "loss": 4.544,
      "mean_token_accuracy": 0.24730204194784164,
      "num_tokens": 141064840.0,
      "step": 81335
    },
    {
      "entropy": 5.34109354019165,
      "epoch": 6.328496401478311,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00018563460254987418,
      "loss": 4.6427,
      "mean_token_accuracy": 0.24042963832616807,
      "num_tokens": 141073964.0,
      "step": 81340
    },
    {
      "entropy": 5.422534799575805,
      "epoch": 6.328885430850029,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001856091674988299,
      "loss": 4.7437,
      "mean_token_accuracy": 0.23336666822433472,
      "num_tokens": 141082808.0,
      "step": 81345
    },
    {
      "entropy": 5.395958089828492,
      "epoch": 6.329274460221747,
      "grad_norm": 1.421875,
      "learning_rate": 0.00018558373380414916,
      "loss": 4.6317,
      "mean_token_accuracy": 0.24446526318788528,
      "num_tokens": 141091665.0,
      "step": 81350
    },
    {
      "entropy": 5.37772421836853,
      "epoch": 6.3296634895934645,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018555830146621793,
      "loss": 4.6713,
      "mean_token_accuracy": 0.23446041196584702,
      "num_tokens": 141101090.0,
      "step": 81355
    },
    {
      "entropy": 5.414840316772461,
      "epoch": 6.330052518965182,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018553287048542216,
      "loss": 4.6934,
      "mean_token_accuracy": 0.23614775538444518,
      "num_tokens": 141109950.0,
      "step": 81360
    },
    {
      "entropy": 5.389267158508301,
      "epoch": 6.3304415483369,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018550744086214766,
      "loss": 4.6182,
      "mean_token_accuracy": 0.25309999734163285,
      "num_tokens": 141118653.0,
      "step": 81365
    },
    {
      "entropy": 5.372490930557251,
      "epoch": 6.330830577708617,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001854820125967804,
      "loss": 4.6062,
      "mean_token_accuracy": 0.24692053943872452,
      "num_tokens": 141127293.0,
      "step": 81370
    },
    {
      "entropy": 5.338057899475098,
      "epoch": 6.331219607080334,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018545658568970585,
      "loss": 4.558,
      "mean_token_accuracy": 0.2484857991337776,
      "num_tokens": 141136046.0,
      "step": 81375
    },
    {
      "entropy": 5.261360168457031,
      "epoch": 6.331608636452052,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001854311601413103,
      "loss": 4.5122,
      "mean_token_accuracy": 0.2576581031084061,
      "num_tokens": 141144609.0,
      "step": 81380
    },
    {
      "entropy": 5.400969314575195,
      "epoch": 6.33199766582377,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001854057359519793,
      "loss": 4.6874,
      "mean_token_accuracy": 0.23151611685752868,
      "num_tokens": 141153478.0,
      "step": 81385
    },
    {
      "entropy": 5.384892416000366,
      "epoch": 6.3323866951954875,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018538031312209863,
      "loss": 4.6075,
      "mean_token_accuracy": 0.24390310198068618,
      "num_tokens": 141161615.0,
      "step": 81390
    },
    {
      "entropy": 5.284872579574585,
      "epoch": 6.332775724567205,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001853548916520541,
      "loss": 4.5209,
      "mean_token_accuracy": 0.24744713604450225,
      "num_tokens": 141169816.0,
      "step": 81395
    },
    {
      "entropy": 5.324034070968628,
      "epoch": 6.333164753938922,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001853294715422314,
      "loss": 4.5625,
      "mean_token_accuracy": 0.24780093431472777,
      "num_tokens": 141178534.0,
      "step": 81400
    },
    {
      "entropy": 5.406375741958618,
      "epoch": 6.33355378331064,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001853040527930162,
      "loss": 4.7243,
      "mean_token_accuracy": 0.23195602297782897,
      "num_tokens": 141187645.0,
      "step": 81405
    },
    {
      "entropy": 5.294068384170532,
      "epoch": 6.333942812682357,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00018527863540479427,
      "loss": 4.5718,
      "mean_token_accuracy": 0.2476823151111603,
      "num_tokens": 141196700.0,
      "step": 81410
    },
    {
      "entropy": 5.410242891311645,
      "epoch": 6.334331842054075,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018525321937795127,
      "loss": 4.7193,
      "mean_token_accuracy": 0.2285415455698967,
      "num_tokens": 141205885.0,
      "step": 81415
    },
    {
      "entropy": 5.347273588180542,
      "epoch": 6.334720871425793,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018522780471287277,
      "loss": 4.6131,
      "mean_token_accuracy": 0.25738555043935774,
      "num_tokens": 141214291.0,
      "step": 81420
    },
    {
      "entropy": 5.3703343868255615,
      "epoch": 6.3351099007975105,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018520239140994445,
      "loss": 4.6488,
      "mean_token_accuracy": 0.23806258291006088,
      "num_tokens": 141223367.0,
      "step": 81425
    },
    {
      "entropy": 5.358229303359986,
      "epoch": 6.335498930169228,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001851769794695519,
      "loss": 4.6135,
      "mean_token_accuracy": 0.23853325545787812,
      "num_tokens": 141231510.0,
      "step": 81430
    },
    {
      "entropy": 5.372674942016602,
      "epoch": 6.335887959540945,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018515156889208079,
      "loss": 4.6921,
      "mean_token_accuracy": 0.23693556487560272,
      "num_tokens": 141239731.0,
      "step": 81435
    },
    {
      "entropy": 5.392068147659302,
      "epoch": 6.336276988912663,
      "grad_norm": 1.375,
      "learning_rate": 0.0001851261596779165,
      "loss": 4.6887,
      "mean_token_accuracy": 0.23697367012500764,
      "num_tokens": 141247728.0,
      "step": 81440
    },
    {
      "entropy": 5.32354211807251,
      "epoch": 6.33666601828438,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018510075182744473,
      "loss": 4.5643,
      "mean_token_accuracy": 0.24208179712295533,
      "num_tokens": 141256408.0,
      "step": 81445
    },
    {
      "entropy": 5.364908599853516,
      "epoch": 6.337055047656098,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00018507534534105102,
      "loss": 4.6313,
      "mean_token_accuracy": 0.2366676524281502,
      "num_tokens": 141265115.0,
      "step": 81450
    },
    {
      "entropy": 5.446081924438476,
      "epoch": 6.337444077027816,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018504994021912075,
      "loss": 4.7411,
      "mean_token_accuracy": 0.22627198100090026,
      "num_tokens": 141273850.0,
      "step": 81455
    },
    {
      "entropy": 5.438184404373169,
      "epoch": 6.3378331063995335,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018502453646203947,
      "loss": 4.726,
      "mean_token_accuracy": 0.23495671451091765,
      "num_tokens": 141283150.0,
      "step": 81460
    },
    {
      "entropy": 5.445001173019409,
      "epoch": 6.33822213577125,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001849991340701927,
      "loss": 4.746,
      "mean_token_accuracy": 0.23022539913654327,
      "num_tokens": 141291748.0,
      "step": 81465
    },
    {
      "entropy": 5.406159925460815,
      "epoch": 6.338611165142968,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001849737330439658,
      "loss": 4.6401,
      "mean_token_accuracy": 0.24514369070529937,
      "num_tokens": 141299948.0,
      "step": 81470
    },
    {
      "entropy": 5.3478223323822025,
      "epoch": 6.339000194514686,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018494833338374416,
      "loss": 4.5961,
      "mean_token_accuracy": 0.24438956379890442,
      "num_tokens": 141308108.0,
      "step": 81475
    },
    {
      "entropy": 5.369941663742066,
      "epoch": 6.339389223886403,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001849229350899133,
      "loss": 4.5926,
      "mean_token_accuracy": 0.24449653774499894,
      "num_tokens": 141316776.0,
      "step": 81480
    },
    {
      "entropy": 5.383547163009643,
      "epoch": 6.339778253258121,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001848975381628585,
      "loss": 4.7699,
      "mean_token_accuracy": 0.23061421811580657,
      "num_tokens": 141325580.0,
      "step": 81485
    },
    {
      "entropy": 5.37119836807251,
      "epoch": 6.340167282629839,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018487214260296515,
      "loss": 4.6038,
      "mean_token_accuracy": 0.2491126298904419,
      "num_tokens": 141333895.0,
      "step": 81490
    },
    {
      "entropy": 5.420374774932862,
      "epoch": 6.3405563120015564,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018484674841061854,
      "loss": 4.6663,
      "mean_token_accuracy": 0.2458408921957016,
      "num_tokens": 141342418.0,
      "step": 81495
    },
    {
      "entropy": 5.361188316345215,
      "epoch": 6.340945341373274,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018482135558620406,
      "loss": 4.6848,
      "mean_token_accuracy": 0.237904454767704,
      "num_tokens": 141351001.0,
      "step": 81500
    },
    {
      "entropy": 5.388563632965088,
      "epoch": 6.341334370744991,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001847959641301069,
      "loss": 4.7017,
      "mean_token_accuracy": 0.2412562131881714,
      "num_tokens": 141360296.0,
      "step": 81505
    },
    {
      "entropy": 5.282053899765015,
      "epoch": 6.341723400116709,
      "grad_norm": 1.375,
      "learning_rate": 0.00018477057404271246,
      "loss": 4.5945,
      "mean_token_accuracy": 0.23875509053468705,
      "num_tokens": 141368719.0,
      "step": 81510
    },
    {
      "entropy": 5.497072649002075,
      "epoch": 6.342112429488426,
      "grad_norm": 1.328125,
      "learning_rate": 0.000184745185324406,
      "loss": 4.8174,
      "mean_token_accuracy": 0.2303106278181076,
      "num_tokens": 141377342.0,
      "step": 81515
    },
    {
      "entropy": 5.467520475387573,
      "epoch": 6.342501458860144,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018471979797557265,
      "loss": 4.7381,
      "mean_token_accuracy": 0.232885479927063,
      "num_tokens": 141385276.0,
      "step": 81520
    },
    {
      "entropy": 5.346171188354492,
      "epoch": 6.342890488231862,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001846944119965977,
      "loss": 4.6736,
      "mean_token_accuracy": 0.2336803361773491,
      "num_tokens": 141395070.0,
      "step": 81525
    },
    {
      "entropy": 5.449671983718872,
      "epoch": 6.343279517603579,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001846690273878663,
      "loss": 4.6512,
      "mean_token_accuracy": 0.24038079679012297,
      "num_tokens": 141403263.0,
      "step": 81530
    },
    {
      "entropy": 5.45230770111084,
      "epoch": 6.343668546975296,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018464364414976365,
      "loss": 4.7301,
      "mean_token_accuracy": 0.23384605646133422,
      "num_tokens": 141411666.0,
      "step": 81535
    },
    {
      "entropy": 5.37118558883667,
      "epoch": 6.344057576347014,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001846182622826748,
      "loss": 4.6383,
      "mean_token_accuracy": 0.2437436357140541,
      "num_tokens": 141419668.0,
      "step": 81540
    },
    {
      "entropy": 5.417099952697754,
      "epoch": 6.344446605718732,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018459288178698503,
      "loss": 4.7154,
      "mean_token_accuracy": 0.23912646770477294,
      "num_tokens": 141427413.0,
      "step": 81545
    },
    {
      "entropy": 5.312813568115234,
      "epoch": 6.344835635090449,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001845675026630794,
      "loss": 4.5676,
      "mean_token_accuracy": 0.24752648919820786,
      "num_tokens": 141436079.0,
      "step": 81550
    },
    {
      "entropy": 5.342815017700195,
      "epoch": 6.345224664462167,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018454212491134293,
      "loss": 4.6627,
      "mean_token_accuracy": 0.24328865259885787,
      "num_tokens": 141444734.0,
      "step": 81555
    },
    {
      "entropy": 5.40089979171753,
      "epoch": 6.345613693833885,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018451674853216084,
      "loss": 4.7126,
      "mean_token_accuracy": 0.2367200180888176,
      "num_tokens": 141453346.0,
      "step": 81560
    },
    {
      "entropy": 5.434236335754394,
      "epoch": 6.346002723205602,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018449137352591794,
      "loss": 4.7107,
      "mean_token_accuracy": 0.2329110860824585,
      "num_tokens": 141462029.0,
      "step": 81565
    },
    {
      "entropy": 5.363843297958374,
      "epoch": 6.346391752577319,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00018446599989299946,
      "loss": 4.5728,
      "mean_token_accuracy": 0.24300796389579774,
      "num_tokens": 141471221.0,
      "step": 81570
    },
    {
      "entropy": 5.308146572113037,
      "epoch": 6.346780781949037,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00018444062763379023,
      "loss": 4.5383,
      "mean_token_accuracy": 0.24766848534345626,
      "num_tokens": 141479531.0,
      "step": 81575
    },
    {
      "entropy": 5.3172389507293705,
      "epoch": 6.347169811320755,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018441525674867544,
      "loss": 4.6372,
      "mean_token_accuracy": 0.23629772812128066,
      "num_tokens": 141488423.0,
      "step": 81580
    },
    {
      "entropy": 5.341009902954101,
      "epoch": 6.347558840692472,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001843898872380399,
      "loss": 4.6214,
      "mean_token_accuracy": 0.23789576888084413,
      "num_tokens": 141496406.0,
      "step": 81585
    },
    {
      "entropy": 5.40327091217041,
      "epoch": 6.34794787006419,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018436451910226864,
      "loss": 4.7026,
      "mean_token_accuracy": 0.23814139813184737,
      "num_tokens": 141504670.0,
      "step": 81590
    },
    {
      "entropy": 5.327240324020385,
      "epoch": 6.348336899435908,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018433915234174653,
      "loss": 4.6855,
      "mean_token_accuracy": 0.24208236187696458,
      "num_tokens": 141514095.0,
      "step": 81595
    },
    {
      "entropy": 5.38925313949585,
      "epoch": 6.3487259288076245,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018431378695685845,
      "loss": 4.6524,
      "mean_token_accuracy": 0.24351726323366166,
      "num_tokens": 141522735.0,
      "step": 81600
    },
    {
      "entropy": 5.355852794647217,
      "epoch": 6.349114958179342,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00018428842294798936,
      "loss": 4.5913,
      "mean_token_accuracy": 0.25054378658533094,
      "num_tokens": 141531932.0,
      "step": 81605
    },
    {
      "entropy": 5.3909155368804935,
      "epoch": 6.34950398755106,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018426306031552408,
      "loss": 4.595,
      "mean_token_accuracy": 0.24641761630773545,
      "num_tokens": 141540951.0,
      "step": 81610
    },
    {
      "entropy": 5.315446853637695,
      "epoch": 6.349893016922778,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018423769905984738,
      "loss": 4.5867,
      "mean_token_accuracy": 0.2426667556166649,
      "num_tokens": 141549524.0,
      "step": 81615
    },
    {
      "entropy": 5.342108345031738,
      "epoch": 6.350282046294495,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018421233918134418,
      "loss": 4.6401,
      "mean_token_accuracy": 0.2486641824245453,
      "num_tokens": 141557348.0,
      "step": 81620
    },
    {
      "entropy": 5.360350942611694,
      "epoch": 6.350671075666213,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001841869806803992,
      "loss": 4.6562,
      "mean_token_accuracy": 0.23844075947999954,
      "num_tokens": 141565571.0,
      "step": 81625
    },
    {
      "entropy": 5.3840258598327635,
      "epoch": 6.351060105037931,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018416162355739725,
      "loss": 4.6818,
      "mean_token_accuracy": 0.2327819988131523,
      "num_tokens": 141574231.0,
      "step": 81630
    },
    {
      "entropy": 5.383302068710327,
      "epoch": 6.3514491344096475,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001841362678127231,
      "loss": 4.6206,
      "mean_token_accuracy": 0.24712872505187988,
      "num_tokens": 141582653.0,
      "step": 81635
    },
    {
      "entropy": 5.402147245407105,
      "epoch": 6.351838163781365,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001841109134467614,
      "loss": 4.7056,
      "mean_token_accuracy": 0.23616975247859956,
      "num_tokens": 141591131.0,
      "step": 81640
    },
    {
      "entropy": 5.369479322433472,
      "epoch": 6.352227193153083,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018408556045989693,
      "loss": 4.6496,
      "mean_token_accuracy": 0.24195281118154527,
      "num_tokens": 141600352.0,
      "step": 81645
    },
    {
      "entropy": 5.377424049377441,
      "epoch": 6.352616222524801,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001840602088525144,
      "loss": 4.6687,
      "mean_token_accuracy": 0.24609695076942445,
      "num_tokens": 141608908.0,
      "step": 81650
    },
    {
      "entropy": 5.4525219917297365,
      "epoch": 6.353005251896518,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00018403485862499846,
      "loss": 4.7483,
      "mean_token_accuracy": 0.2348211631178856,
      "num_tokens": 141619012.0,
      "step": 81655
    },
    {
      "entropy": 5.419418668746948,
      "epoch": 6.353394281268236,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018400950977773379,
      "loss": 4.6502,
      "mean_token_accuracy": 0.24188035130500793,
      "num_tokens": 141628922.0,
      "step": 81660
    },
    {
      "entropy": 5.3910266876220705,
      "epoch": 6.353783310639954,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018398416231110492,
      "loss": 4.6726,
      "mean_token_accuracy": 0.24056272357702255,
      "num_tokens": 141637628.0,
      "step": 81665
    },
    {
      "entropy": 5.34161958694458,
      "epoch": 6.3541723400116705,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001839588162254966,
      "loss": 4.6119,
      "mean_token_accuracy": 0.24218100756406785,
      "num_tokens": 141646362.0,
      "step": 81670
    },
    {
      "entropy": 5.290633726119995,
      "epoch": 6.354561369383388,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001839334715212933,
      "loss": 4.612,
      "mean_token_accuracy": 0.24143652319908143,
      "num_tokens": 141656120.0,
      "step": 81675
    },
    {
      "entropy": 5.4283034324646,
      "epoch": 6.354950398755106,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001839081281988796,
      "loss": 4.707,
      "mean_token_accuracy": 0.23435209542512894,
      "num_tokens": 141664833.0,
      "step": 81680
    },
    {
      "entropy": 5.488491344451904,
      "epoch": 6.355339428126824,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018388278625864008,
      "loss": 4.714,
      "mean_token_accuracy": 0.23372249007225038,
      "num_tokens": 141673823.0,
      "step": 81685
    },
    {
      "entropy": 5.325438070297241,
      "epoch": 6.355728457498541,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018385744570095924,
      "loss": 4.5165,
      "mean_token_accuracy": 0.25430536866188047,
      "num_tokens": 141682086.0,
      "step": 81690
    },
    {
      "entropy": 5.383369112014771,
      "epoch": 6.356117486870259,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001838321065262216,
      "loss": 4.666,
      "mean_token_accuracy": 0.24089957624673844,
      "num_tokens": 141691167.0,
      "step": 81695
    },
    {
      "entropy": 5.360728359222412,
      "epoch": 6.356506516241977,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018380676873481168,
      "loss": 4.671,
      "mean_token_accuracy": 0.23873160928487777,
      "num_tokens": 141699781.0,
      "step": 81700
    },
    {
      "entropy": 5.375424194335937,
      "epoch": 6.3568955456136935,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018378143232711375,
      "loss": 4.6201,
      "mean_token_accuracy": 0.24610953778028488,
      "num_tokens": 141708646.0,
      "step": 81705
    },
    {
      "entropy": 5.3503447532653805,
      "epoch": 6.357284574985411,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001837560973035125,
      "loss": 4.5863,
      "mean_token_accuracy": 0.24339659959077836,
      "num_tokens": 141717105.0,
      "step": 81710
    },
    {
      "entropy": 5.3968620777130125,
      "epoch": 6.357673604357129,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018373076366439228,
      "loss": 4.6245,
      "mean_token_accuracy": 0.23972884118556975,
      "num_tokens": 141725712.0,
      "step": 81715
    },
    {
      "entropy": 5.366573333740234,
      "epoch": 6.358062633728847,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001837054314101374,
      "loss": 4.6461,
      "mean_token_accuracy": 0.2380531519651413,
      "num_tokens": 141735142.0,
      "step": 81720
    },
    {
      "entropy": 5.387720060348511,
      "epoch": 6.358451663100564,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018368010054113233,
      "loss": 4.5899,
      "mean_token_accuracy": 0.24297941774129866,
      "num_tokens": 141743732.0,
      "step": 81725
    },
    {
      "entropy": 5.460933637619019,
      "epoch": 6.358840692472282,
      "grad_norm": 1.375,
      "learning_rate": 0.0001836547710577613,
      "loss": 4.666,
      "mean_token_accuracy": 0.23835860341787338,
      "num_tokens": 141752459.0,
      "step": 81730
    },
    {
      "entropy": 5.478852891921997,
      "epoch": 6.359229721843999,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001836294429604088,
      "loss": 4.7983,
      "mean_token_accuracy": 0.23477291762828828,
      "num_tokens": 141761345.0,
      "step": 81735
    },
    {
      "entropy": 5.441839361190796,
      "epoch": 6.3596187512157165,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018360411624945906,
      "loss": 4.7293,
      "mean_token_accuracy": 0.22620386332273484,
      "num_tokens": 141770326.0,
      "step": 81740
    },
    {
      "entropy": 5.475213432312012,
      "epoch": 6.360007780587434,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001835787909252964,
      "loss": 4.7563,
      "mean_token_accuracy": 0.2387345850467682,
      "num_tokens": 141778794.0,
      "step": 81745
    },
    {
      "entropy": 5.472924900054932,
      "epoch": 6.360396809959152,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018355346698830505,
      "loss": 4.7818,
      "mean_token_accuracy": 0.23250727355480194,
      "num_tokens": 141787885.0,
      "step": 81750
    },
    {
      "entropy": 5.402604818344116,
      "epoch": 6.36078583933087,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00018352814443886928,
      "loss": 4.6261,
      "mean_token_accuracy": 0.2402128666639328,
      "num_tokens": 141796974.0,
      "step": 81755
    },
    {
      "entropy": 5.460384559631348,
      "epoch": 6.361174868702587,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00018350282327737338,
      "loss": 4.6876,
      "mean_token_accuracy": 0.24160350412130355,
      "num_tokens": 141805712.0,
      "step": 81760
    },
    {
      "entropy": 5.4405077457427975,
      "epoch": 6.361563898074305,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00018347750350420145,
      "loss": 4.7107,
      "mean_token_accuracy": 0.23920608460903167,
      "num_tokens": 141814597.0,
      "step": 81765
    },
    {
      "entropy": 5.417399883270264,
      "epoch": 6.361952927446022,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018345218511973773,
      "loss": 4.7357,
      "mean_token_accuracy": 0.2342400997877121,
      "num_tokens": 141823668.0,
      "step": 81770
    },
    {
      "entropy": 5.331061840057373,
      "epoch": 6.3623419568177395,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018342686812436633,
      "loss": 4.6442,
      "mean_token_accuracy": 0.2369713693857193,
      "num_tokens": 141832040.0,
      "step": 81775
    },
    {
      "entropy": 5.342257452011109,
      "epoch": 6.362730986189457,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018340155251847157,
      "loss": 4.5861,
      "mean_token_accuracy": 0.24306774884462357,
      "num_tokens": 141840768.0,
      "step": 81780
    },
    {
      "entropy": 5.4087584018707275,
      "epoch": 6.363120015561175,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001833762383024374,
      "loss": 4.7076,
      "mean_token_accuracy": 0.22625284641981125,
      "num_tokens": 141849931.0,
      "step": 81785
    },
    {
      "entropy": 5.349915027618408,
      "epoch": 6.363509044932893,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000183350925476648,
      "loss": 4.6443,
      "mean_token_accuracy": 0.23782870769500733,
      "num_tokens": 141858955.0,
      "step": 81790
    },
    {
      "entropy": 5.2893274307250975,
      "epoch": 6.36389807430461,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018332561404148739,
      "loss": 4.5283,
      "mean_token_accuracy": 0.25127232670783994,
      "num_tokens": 141867675.0,
      "step": 81795
    },
    {
      "entropy": 5.380553722381592,
      "epoch": 6.364287103676327,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001833003039973397,
      "loss": 4.5793,
      "mean_token_accuracy": 0.24327391088008882,
      "num_tokens": 141876264.0,
      "step": 81800
    },
    {
      "entropy": 5.457530879974366,
      "epoch": 6.364676133048045,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018327499534458896,
      "loss": 4.7301,
      "mean_token_accuracy": 0.233053158223629,
      "num_tokens": 141884805.0,
      "step": 81805
    },
    {
      "entropy": 5.417302036285401,
      "epoch": 6.3650651624197625,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018324968808361913,
      "loss": 4.6416,
      "mean_token_accuracy": 0.2443991258740425,
      "num_tokens": 141892825.0,
      "step": 81810
    },
    {
      "entropy": 5.41368293762207,
      "epoch": 6.36545419179148,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001832243822148143,
      "loss": 4.7199,
      "mean_token_accuracy": 0.23828549683094025,
      "num_tokens": 141902011.0,
      "step": 81815
    },
    {
      "entropy": 5.485558271408081,
      "epoch": 6.365843221163198,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018319907773855837,
      "loss": 4.775,
      "mean_token_accuracy": 0.23361821174621583,
      "num_tokens": 141910837.0,
      "step": 81820
    },
    {
      "entropy": 5.388908433914184,
      "epoch": 6.366232250534916,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018317377465523532,
      "loss": 4.6231,
      "mean_token_accuracy": 0.23674268573522567,
      "num_tokens": 141919447.0,
      "step": 81825
    },
    {
      "entropy": 5.417634391784668,
      "epoch": 6.366621279906633,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018314847296522908,
      "loss": 4.6649,
      "mean_token_accuracy": 0.2364397391676903,
      "num_tokens": 141927628.0,
      "step": 81830
    },
    {
      "entropy": 5.418837594985962,
      "epoch": 6.36701030927835,
      "grad_norm": 1.25,
      "learning_rate": 0.00018312317266892358,
      "loss": 4.6513,
      "mean_token_accuracy": 0.23429521471261977,
      "num_tokens": 141936248.0,
      "step": 81835
    },
    {
      "entropy": 5.376932859420776,
      "epoch": 6.367399338650068,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001830978737667026,
      "loss": 4.6234,
      "mean_token_accuracy": 0.234259994328022,
      "num_tokens": 141944821.0,
      "step": 81840
    },
    {
      "entropy": 5.411740684509278,
      "epoch": 6.3677883680217855,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001830725762589502,
      "loss": 4.7029,
      "mean_token_accuracy": 0.23760416656732558,
      "num_tokens": 141953954.0,
      "step": 81845
    },
    {
      "entropy": 5.331614875793457,
      "epoch": 6.368177397393503,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001830472801460501,
      "loss": 4.595,
      "mean_token_accuracy": 0.24298890829086303,
      "num_tokens": 141962969.0,
      "step": 81850
    },
    {
      "entropy": 5.3686851978302,
      "epoch": 6.368566426765221,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018302198542838622,
      "loss": 4.7064,
      "mean_token_accuracy": 0.23919731229543686,
      "num_tokens": 141971008.0,
      "step": 81855
    },
    {
      "entropy": 5.3630022525787355,
      "epoch": 6.368955456136939,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00018299669210634234,
      "loss": 4.6683,
      "mean_token_accuracy": 0.24461256414651872,
      "num_tokens": 141980802.0,
      "step": 81860
    },
    {
      "entropy": 5.378611040115357,
      "epoch": 6.369344485508656,
      "grad_norm": 1.515625,
      "learning_rate": 0.00018297140018030212,
      "loss": 4.5941,
      "mean_token_accuracy": 0.23938099592924117,
      "num_tokens": 141989303.0,
      "step": 81865
    },
    {
      "entropy": 5.46697154045105,
      "epoch": 6.369733514880373,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018294610965064945,
      "loss": 4.7358,
      "mean_token_accuracy": 0.22796095609664918,
      "num_tokens": 141998146.0,
      "step": 81870
    },
    {
      "entropy": 5.367310476303101,
      "epoch": 6.370122544252091,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018292082051776809,
      "loss": 4.6227,
      "mean_token_accuracy": 0.24019922763109208,
      "num_tokens": 142006303.0,
      "step": 81875
    },
    {
      "entropy": 5.2997650623321535,
      "epoch": 6.3705115736238085,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018289553278204174,
      "loss": 4.5498,
      "mean_token_accuracy": 0.240702848136425,
      "num_tokens": 142014540.0,
      "step": 81880
    },
    {
      "entropy": 5.421625804901123,
      "epoch": 6.370900602995526,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018287024644385397,
      "loss": 4.6242,
      "mean_token_accuracy": 0.23887188732624054,
      "num_tokens": 142022910.0,
      "step": 81885
    },
    {
      "entropy": 5.4814472675323485,
      "epoch": 6.371289632367244,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018284496150358865,
      "loss": 4.7576,
      "mean_token_accuracy": 0.23407843559980393,
      "num_tokens": 142032271.0,
      "step": 81890
    },
    {
      "entropy": 5.386371660232544,
      "epoch": 6.371678661738962,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018281967796162935,
      "loss": 4.6156,
      "mean_token_accuracy": 0.24722603261470794,
      "num_tokens": 142040934.0,
      "step": 81895
    },
    {
      "entropy": 5.373214149475098,
      "epoch": 6.372067691110679,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001827943958183597,
      "loss": 4.6629,
      "mean_token_accuracy": 0.23342284113168715,
      "num_tokens": 142049127.0,
      "step": 81900
    },
    {
      "entropy": 5.3290375709533695,
      "epoch": 6.372456720482396,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00018276911507416333,
      "loss": 4.6203,
      "mean_token_accuracy": 0.2354627639055252,
      "num_tokens": 142057535.0,
      "step": 81905
    },
    {
      "entropy": 5.39911470413208,
      "epoch": 6.372845749854114,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001827438357294238,
      "loss": 4.6748,
      "mean_token_accuracy": 0.23726222962141036,
      "num_tokens": 142065962.0,
      "step": 81910
    },
    {
      "entropy": 5.376339054107666,
      "epoch": 6.3732347792258315,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018271855778452485,
      "loss": 4.6465,
      "mean_token_accuracy": 0.2407628744840622,
      "num_tokens": 142074531.0,
      "step": 81915
    },
    {
      "entropy": 5.351188516616821,
      "epoch": 6.373623808597549,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001826932812398498,
      "loss": 4.6111,
      "mean_token_accuracy": 0.24652115851640702,
      "num_tokens": 142083554.0,
      "step": 81920
    },
    {
      "entropy": 5.392584943771363,
      "epoch": 6.374012837969267,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018266800609578233,
      "loss": 4.6315,
      "mean_token_accuracy": 0.24688238203525542,
      "num_tokens": 142092117.0,
      "step": 81925
    },
    {
      "entropy": 5.379878139495849,
      "epoch": 6.374401867340985,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001826427323527059,
      "loss": 4.6954,
      "mean_token_accuracy": 0.23215968906879425,
      "num_tokens": 142101415.0,
      "step": 81930
    },
    {
      "entropy": 5.403899812698365,
      "epoch": 6.374790896712701,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000182617460011004,
      "loss": 4.6614,
      "mean_token_accuracy": 0.23737689703702927,
      "num_tokens": 142110100.0,
      "step": 81935
    },
    {
      "entropy": 5.4512005805969235,
      "epoch": 6.375179926084419,
      "grad_norm": 1.25,
      "learning_rate": 0.00018259218907106013,
      "loss": 4.6159,
      "mean_token_accuracy": 0.24001251459121703,
      "num_tokens": 142118405.0,
      "step": 81940
    },
    {
      "entropy": 5.27708215713501,
      "epoch": 6.375568955456137,
      "grad_norm": 1.203125,
      "learning_rate": 0.00018256691953325767,
      "loss": 4.5305,
      "mean_token_accuracy": 0.24942998290061952,
      "num_tokens": 142127332.0,
      "step": 81945
    },
    {
      "entropy": 5.393292856216431,
      "epoch": 6.3759579848278545,
      "grad_norm": 1.375,
      "learning_rate": 0.00018254165139798006,
      "loss": 4.6208,
      "mean_token_accuracy": 0.24353747814893723,
      "num_tokens": 142136018.0,
      "step": 81950
    },
    {
      "entropy": 5.3412312984466555,
      "epoch": 6.376347014199572,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018251638466561083,
      "loss": 4.7083,
      "mean_token_accuracy": 0.23640664368867875,
      "num_tokens": 142145040.0,
      "step": 81955
    },
    {
      "entropy": 5.379727935791015,
      "epoch": 6.37673604357129,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018249111933653313,
      "loss": 4.5924,
      "mean_token_accuracy": 0.24310518950223922,
      "num_tokens": 142153879.0,
      "step": 81960
    },
    {
      "entropy": 5.367972373962402,
      "epoch": 6.377125072943008,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018246585541113053,
      "loss": 4.6269,
      "mean_token_accuracy": 0.24690053164958953,
      "num_tokens": 142162269.0,
      "step": 81965
    },
    {
      "entropy": 5.398870229721069,
      "epoch": 6.377514102314724,
      "grad_norm": 1.375,
      "learning_rate": 0.0001824405928897862,
      "loss": 4.6896,
      "mean_token_accuracy": 0.2368488445878029,
      "num_tokens": 142170535.0,
      "step": 81970
    },
    {
      "entropy": 5.373088455200195,
      "epoch": 6.377903131686442,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018241533177288367,
      "loss": 4.6473,
      "mean_token_accuracy": 0.2376559481024742,
      "num_tokens": 142180043.0,
      "step": 81975
    },
    {
      "entropy": 5.321472454071045,
      "epoch": 6.37829216105816,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018239007206080605,
      "loss": 4.583,
      "mean_token_accuracy": 0.24278482645750046,
      "num_tokens": 142188322.0,
      "step": 81980
    },
    {
      "entropy": 5.46144289970398,
      "epoch": 6.3786811904298775,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001823648137539367,
      "loss": 4.6917,
      "mean_token_accuracy": 0.24894859939813613,
      "num_tokens": 142196828.0,
      "step": 81985
    },
    {
      "entropy": 5.350796699523926,
      "epoch": 6.379070219801595,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00018233955685265885,
      "loss": 4.551,
      "mean_token_accuracy": 0.2485095053911209,
      "num_tokens": 142205943.0,
      "step": 81990
    },
    {
      "entropy": 5.363320541381836,
      "epoch": 6.379459249173313,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018231430135735573,
      "loss": 4.6011,
      "mean_token_accuracy": 0.25058511942625045,
      "num_tokens": 142214732.0,
      "step": 81995
    },
    {
      "entropy": 5.398632049560547,
      "epoch": 6.379848278545031,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018228904726841062,
      "loss": 4.6933,
      "mean_token_accuracy": 0.23812761306762695,
      "num_tokens": 142224053.0,
      "step": 82000
    },
    {
      "entropy": 5.315751028060913,
      "epoch": 6.380237307916747,
      "grad_norm": 1.25,
      "learning_rate": 0.00018226379458620656,
      "loss": 4.5815,
      "mean_token_accuracy": 0.2454157754778862,
      "num_tokens": 142232715.0,
      "step": 82005
    },
    {
      "entropy": 5.34528341293335,
      "epoch": 6.380626337288465,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018223854331112695,
      "loss": 4.6575,
      "mean_token_accuracy": 0.23794249147176744,
      "num_tokens": 142241277.0,
      "step": 82010
    },
    {
      "entropy": 5.460067367553711,
      "epoch": 6.381015366660183,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018221329344355464,
      "loss": 4.7708,
      "mean_token_accuracy": 0.23725757449865342,
      "num_tokens": 142250646.0,
      "step": 82015
    },
    {
      "entropy": 5.517052078247071,
      "epoch": 6.3814043960319005,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018218804498387308,
      "loss": 4.76,
      "mean_token_accuracy": 0.23118478804826736,
      "num_tokens": 142259983.0,
      "step": 82020
    },
    {
      "entropy": 5.465401268005371,
      "epoch": 6.381793425403618,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001821627979324651,
      "loss": 4.6741,
      "mean_token_accuracy": 0.23786509335041045,
      "num_tokens": 142269029.0,
      "step": 82025
    },
    {
      "entropy": 5.4050312519073485,
      "epoch": 6.382182454775336,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001821375522897139,
      "loss": 4.6153,
      "mean_token_accuracy": 0.24565163254737854,
      "num_tokens": 142277445.0,
      "step": 82030
    },
    {
      "entropy": 5.33351731300354,
      "epoch": 6.3825714841470536,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001821123080560026,
      "loss": 4.5901,
      "mean_token_accuracy": 0.252828748524189,
      "num_tokens": 142286400.0,
      "step": 82035
    },
    {
      "entropy": 5.271478605270386,
      "epoch": 6.38296051351877,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00018208706523171404,
      "loss": 4.5503,
      "mean_token_accuracy": 0.24840150028467178,
      "num_tokens": 142295528.0,
      "step": 82040
    },
    {
      "entropy": 5.391713237762451,
      "epoch": 6.383349542890488,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018206182381723146,
      "loss": 4.6961,
      "mean_token_accuracy": 0.24021245539188385,
      "num_tokens": 142304148.0,
      "step": 82045
    },
    {
      "entropy": 5.402368545532227,
      "epoch": 6.383738572262206,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018203658381293786,
      "loss": 4.6953,
      "mean_token_accuracy": 0.23334311693906784,
      "num_tokens": 142312720.0,
      "step": 82050
    },
    {
      "entropy": 5.466646814346314,
      "epoch": 6.3841276016339235,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00018201134521921602,
      "loss": 4.6889,
      "mean_token_accuracy": 0.2347816750407219,
      "num_tokens": 142322422.0,
      "step": 82055
    },
    {
      "entropy": 5.3408647060394285,
      "epoch": 6.384516631005641,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001819861080364491,
      "loss": 4.5598,
      "mean_token_accuracy": 0.24860088229179383,
      "num_tokens": 142330081.0,
      "step": 82060
    },
    {
      "entropy": 5.331630086898803,
      "epoch": 6.384905660377359,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001819608722650199,
      "loss": 4.5793,
      "mean_token_accuracy": 0.24902829229831697,
      "num_tokens": 142338588.0,
      "step": 82065
    },
    {
      "entropy": 5.375982713699341,
      "epoch": 6.385294689749076,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018193563790531138,
      "loss": 4.7203,
      "mean_token_accuracy": 0.24178221821784973,
      "num_tokens": 142347680.0,
      "step": 82070
    },
    {
      "entropy": 5.399868488311768,
      "epoch": 6.385683719120793,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001819104049577064,
      "loss": 4.6937,
      "mean_token_accuracy": 0.23545539677143096,
      "num_tokens": 142356503.0,
      "step": 82075
    },
    {
      "entropy": 5.401919794082642,
      "epoch": 6.386072748492511,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018188517342258793,
      "loss": 4.6375,
      "mean_token_accuracy": 0.23690478056669234,
      "num_tokens": 142364985.0,
      "step": 82080
    },
    {
      "entropy": 5.478675603866577,
      "epoch": 6.386461777864229,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018185994330033867,
      "loss": 4.7636,
      "mean_token_accuracy": 0.22478411793708802,
      "num_tokens": 142373322.0,
      "step": 82085
    },
    {
      "entropy": 5.444008922576904,
      "epoch": 6.3868508072359464,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018183471459134155,
      "loss": 4.7221,
      "mean_token_accuracy": 0.23277215510606766,
      "num_tokens": 142382424.0,
      "step": 82090
    },
    {
      "entropy": 5.40822696685791,
      "epoch": 6.387239836607664,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018180948729597935,
      "loss": 4.5735,
      "mean_token_accuracy": 0.24074033945798873,
      "num_tokens": 142390727.0,
      "step": 82095
    },
    {
      "entropy": 5.378018093109131,
      "epoch": 6.387628865979382,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00018178426141463489,
      "loss": 4.6573,
      "mean_token_accuracy": 0.23915650844573974,
      "num_tokens": 142399730.0,
      "step": 82100
    },
    {
      "entropy": 5.3933806896209715,
      "epoch": 6.388017895351099,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00018175903694769074,
      "loss": 4.6636,
      "mean_token_accuracy": 0.23441509753465653,
      "num_tokens": 142408839.0,
      "step": 82105
    },
    {
      "entropy": 5.441501760482788,
      "epoch": 6.388406924722816,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00018173381389552994,
      "loss": 4.6273,
      "mean_token_accuracy": 0.244753797352314,
      "num_tokens": 142417752.0,
      "step": 82110
    },
    {
      "entropy": 5.472310733795166,
      "epoch": 6.388795954094534,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00018170859225853504,
      "loss": 4.7082,
      "mean_token_accuracy": 0.235901902616024,
      "num_tokens": 142425593.0,
      "step": 82115
    },
    {
      "entropy": 5.282119464874268,
      "epoch": 6.389184983466252,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001816833720370888,
      "loss": 4.5791,
      "mean_token_accuracy": 0.2447889655828476,
      "num_tokens": 142434572.0,
      "step": 82120
    },
    {
      "entropy": 5.355295658111572,
      "epoch": 6.389574012837969,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018165815323157377,
      "loss": 4.6522,
      "mean_token_accuracy": 0.23986128866672515,
      "num_tokens": 142443388.0,
      "step": 82125
    },
    {
      "entropy": 5.485706949234009,
      "epoch": 6.389963042209687,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001816329358423728,
      "loss": 4.7697,
      "mean_token_accuracy": 0.23387398272752763,
      "num_tokens": 142451763.0,
      "step": 82130
    },
    {
      "entropy": 5.420133876800537,
      "epoch": 6.390352071581404,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018160771986986836,
      "loss": 4.6341,
      "mean_token_accuracy": 0.23864349573850632,
      "num_tokens": 142459949.0,
      "step": 82135
    },
    {
      "entropy": 5.436772441864013,
      "epoch": 6.390741100953122,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018158250531444316,
      "loss": 4.6801,
      "mean_token_accuracy": 0.24552912563085555,
      "num_tokens": 142468915.0,
      "step": 82140
    },
    {
      "entropy": 5.458391809463501,
      "epoch": 6.391130130324839,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001815572921764797,
      "loss": 4.6829,
      "mean_token_accuracy": 0.23844251185655593,
      "num_tokens": 142477040.0,
      "step": 82145
    },
    {
      "entropy": 5.370807123184204,
      "epoch": 6.391519159696557,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001815320804563606,
      "loss": 4.6127,
      "mean_token_accuracy": 0.25233336091041564,
      "num_tokens": 142485953.0,
      "step": 82150
    },
    {
      "entropy": 5.362776136398315,
      "epoch": 6.391908189068275,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018150687015446842,
      "loss": 4.6676,
      "mean_token_accuracy": 0.2354375198483467,
      "num_tokens": 142494074.0,
      "step": 82155
    },
    {
      "entropy": 5.424072217941284,
      "epoch": 6.392297218439992,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001814816612711857,
      "loss": 4.7416,
      "mean_token_accuracy": 0.23425615727901458,
      "num_tokens": 142504241.0,
      "step": 82160
    },
    {
      "entropy": 5.3328159809112545,
      "epoch": 6.39268624781171,
      "grad_norm": 1.375,
      "learning_rate": 0.0001814564538068949,
      "loss": 4.6194,
      "mean_token_accuracy": 0.24161644876003266,
      "num_tokens": 142514154.0,
      "step": 82165
    },
    {
      "entropy": 5.395192432403564,
      "epoch": 6.393075277183427,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018143124776197851,
      "loss": 4.6553,
      "mean_token_accuracy": 0.2379287913441658,
      "num_tokens": 142522958.0,
      "step": 82170
    },
    {
      "entropy": 5.410838079452515,
      "epoch": 6.393464306555145,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018140604313681902,
      "loss": 4.6611,
      "mean_token_accuracy": 0.23812813460826873,
      "num_tokens": 142531878.0,
      "step": 82175
    },
    {
      "entropy": 5.376241302490234,
      "epoch": 6.393853335926862,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001813808399317989,
      "loss": 4.5557,
      "mean_token_accuracy": 0.24799149185419084,
      "num_tokens": 142539974.0,
      "step": 82180
    },
    {
      "entropy": 5.291793394088745,
      "epoch": 6.39424236529858,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00018135563814730056,
      "loss": 4.5501,
      "mean_token_accuracy": 0.24842403531074525,
      "num_tokens": 142548607.0,
      "step": 82185
    },
    {
      "entropy": 5.406942653656006,
      "epoch": 6.394631394670298,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018133043778370623,
      "loss": 4.7543,
      "mean_token_accuracy": 0.234139883518219,
      "num_tokens": 142557975.0,
      "step": 82190
    },
    {
      "entropy": 5.400814437866211,
      "epoch": 6.395020424042015,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001813052388413985,
      "loss": 4.7087,
      "mean_token_accuracy": 0.23189706206321717,
      "num_tokens": 142566202.0,
      "step": 82195
    },
    {
      "entropy": 5.37941346168518,
      "epoch": 6.395409453413733,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018128004132075953,
      "loss": 4.5918,
      "mean_token_accuracy": 0.24848803728818894,
      "num_tokens": 142574318.0,
      "step": 82200
    },
    {
      "entropy": 5.399551868438721,
      "epoch": 6.39579848278545,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018125484522217195,
      "loss": 4.6182,
      "mean_token_accuracy": 0.2469880610704422,
      "num_tokens": 142581910.0,
      "step": 82205
    },
    {
      "entropy": 5.402128553390503,
      "epoch": 6.396187512157168,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001812296505460177,
      "loss": 4.7084,
      "mean_token_accuracy": 0.23227871656417848,
      "num_tokens": 142590306.0,
      "step": 82210
    },
    {
      "entropy": 5.413269424438477,
      "epoch": 6.396576541528885,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018120445729267936,
      "loss": 4.6914,
      "mean_token_accuracy": 0.23969131857156753,
      "num_tokens": 142599305.0,
      "step": 82215
    },
    {
      "entropy": 5.415508031845093,
      "epoch": 6.396965570900603,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018117926546253906,
      "loss": 4.7172,
      "mean_token_accuracy": 0.23408996313810349,
      "num_tokens": 142607852.0,
      "step": 82220
    },
    {
      "entropy": 5.392915105819702,
      "epoch": 6.397354600272321,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018115407505597909,
      "loss": 4.7124,
      "mean_token_accuracy": 0.23461237251758577,
      "num_tokens": 142617276.0,
      "step": 82225
    },
    {
      "entropy": 5.356106996536255,
      "epoch": 6.397743629644038,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00018112888607338162,
      "loss": 4.6024,
      "mean_token_accuracy": 0.23891111761331557,
      "num_tokens": 142625978.0,
      "step": 82230
    },
    {
      "entropy": 5.342175149917603,
      "epoch": 6.398132659015756,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000181103698515129,
      "loss": 4.5585,
      "mean_token_accuracy": 0.24532577097415925,
      "num_tokens": 142634525.0,
      "step": 82235
    },
    {
      "entropy": 5.46988959312439,
      "epoch": 6.398521688387473,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018107851238160312,
      "loss": 4.7467,
      "mean_token_accuracy": 0.23420187830924988,
      "num_tokens": 142643468.0,
      "step": 82240
    },
    {
      "entropy": 5.3753949165344235,
      "epoch": 6.398910717759191,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018105332767318638,
      "loss": 4.6426,
      "mean_token_accuracy": 0.24643555134534836,
      "num_tokens": 142651761.0,
      "step": 82245
    },
    {
      "entropy": 5.384146642684937,
      "epoch": 6.399299747130908,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000181028144390261,
      "loss": 4.6374,
      "mean_token_accuracy": 0.23885766863822938,
      "num_tokens": 142660400.0,
      "step": 82250
    },
    {
      "entropy": 5.3627174377441404,
      "epoch": 6.399688776502626,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001810029625332088,
      "loss": 4.5905,
      "mean_token_accuracy": 0.24726342111825944,
      "num_tokens": 142669100.0,
      "step": 82255
    },
    {
      "entropy": 5.360354471206665,
      "epoch": 6.400077805874344,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001809777821024121,
      "loss": 4.6297,
      "mean_token_accuracy": 0.23754183501005172,
      "num_tokens": 142677996.0,
      "step": 82260
    },
    {
      "entropy": 5.245414733886719,
      "epoch": 6.400466835246061,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00018095260309825296,
      "loss": 4.5027,
      "mean_token_accuracy": 0.26008217632770536,
      "num_tokens": 142686644.0,
      "step": 82265
    },
    {
      "entropy": 5.435843181610108,
      "epoch": 6.400855864617778,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001809274255211133,
      "loss": 4.7166,
      "mean_token_accuracy": 0.234586638212204,
      "num_tokens": 142695420.0,
      "step": 82270
    },
    {
      "entropy": 5.406366062164307,
      "epoch": 6.401244893989496,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018090224937137529,
      "loss": 4.6784,
      "mean_token_accuracy": 0.23823073506355286,
      "num_tokens": 142704128.0,
      "step": 82275
    },
    {
      "entropy": 5.338760280609131,
      "epoch": 6.401633923361214,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018087707464942078,
      "loss": 4.6316,
      "mean_token_accuracy": 0.24388722330331802,
      "num_tokens": 142712247.0,
      "step": 82280
    },
    {
      "entropy": 5.368616962432862,
      "epoch": 6.402022952732931,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018085190135563194,
      "loss": 4.6137,
      "mean_token_accuracy": 0.24260696917772293,
      "num_tokens": 142720847.0,
      "step": 82285
    },
    {
      "entropy": 5.389946460723877,
      "epoch": 6.402411982104649,
      "grad_norm": 1.5,
      "learning_rate": 0.00018082672949039057,
      "loss": 4.6076,
      "mean_token_accuracy": 0.24145932942628862,
      "num_tokens": 142729054.0,
      "step": 82290
    },
    {
      "entropy": 5.325081253051758,
      "epoch": 6.402801011476367,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018080155905407874,
      "loss": 4.5693,
      "mean_token_accuracy": 0.24333367496728897,
      "num_tokens": 142738125.0,
      "step": 82295
    },
    {
      "entropy": 5.30859785079956,
      "epoch": 6.403190040848084,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001807763900470783,
      "loss": 4.5882,
      "mean_token_accuracy": 0.24494532644748687,
      "num_tokens": 142747092.0,
      "step": 82300
    },
    {
      "entropy": 5.412494850158692,
      "epoch": 6.403579070219801,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001807512224697711,
      "loss": 4.7363,
      "mean_token_accuracy": 0.23632404208183289,
      "num_tokens": 142755537.0,
      "step": 82305
    },
    {
      "entropy": 5.4143472671508786,
      "epoch": 6.403968099591519,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001807260563225392,
      "loss": 4.6516,
      "mean_token_accuracy": 0.24171083122491838,
      "num_tokens": 142763493.0,
      "step": 82310
    },
    {
      "entropy": 5.378700828552246,
      "epoch": 6.404357128963237,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00018070089160576424,
      "loss": 4.6077,
      "mean_token_accuracy": 0.25079862028360367,
      "num_tokens": 142772411.0,
      "step": 82315
    },
    {
      "entropy": 5.397850751876831,
      "epoch": 6.404746158334954,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018067572831982826,
      "loss": 4.6303,
      "mean_token_accuracy": 0.2384672537446022,
      "num_tokens": 142780773.0,
      "step": 82320
    },
    {
      "entropy": 5.410871982574463,
      "epoch": 6.405135187706672,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018065056646511285,
      "loss": 4.729,
      "mean_token_accuracy": 0.2367436781525612,
      "num_tokens": 142788849.0,
      "step": 82325
    },
    {
      "entropy": 5.298367691040039,
      "epoch": 6.40552421707839,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018062540604200006,
      "loss": 4.5614,
      "mean_token_accuracy": 0.24917505979537963,
      "num_tokens": 142797263.0,
      "step": 82330
    },
    {
      "entropy": 5.474892854690552,
      "epoch": 6.4059132464501065,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001806002470508714,
      "loss": 4.7482,
      "mean_token_accuracy": 0.23943687975406647,
      "num_tokens": 142805506.0,
      "step": 82335
    },
    {
      "entropy": 5.462739372253418,
      "epoch": 6.406302275821824,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018057508949210888,
      "loss": 4.7714,
      "mean_token_accuracy": 0.22545326948165895,
      "num_tokens": 142814152.0,
      "step": 82340
    },
    {
      "entropy": 5.4240296363830565,
      "epoch": 6.406691305193542,
      "grad_norm": 1.375,
      "learning_rate": 0.00018054993336609398,
      "loss": 4.6622,
      "mean_token_accuracy": 0.2343586951494217,
      "num_tokens": 142822948.0,
      "step": 82345
    },
    {
      "entropy": 5.361614227294922,
      "epoch": 6.40708033456526,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00018052477867320854,
      "loss": 4.6335,
      "mean_token_accuracy": 0.23653846085071564,
      "num_tokens": 142831710.0,
      "step": 82350
    },
    {
      "entropy": 5.392384481430054,
      "epoch": 6.407469363936977,
      "grad_norm": 1.421875,
      "learning_rate": 0.00018049962541383425,
      "loss": 4.6284,
      "mean_token_accuracy": 0.24847085773944855,
      "num_tokens": 142840163.0,
      "step": 82355
    },
    {
      "entropy": 5.40827751159668,
      "epoch": 6.407858393308695,
      "grad_norm": 1.25,
      "learning_rate": 0.0001804744735883526,
      "loss": 4.68,
      "mean_token_accuracy": 0.23137565106153488,
      "num_tokens": 142848454.0,
      "step": 82360
    },
    {
      "entropy": 5.285017013549805,
      "epoch": 6.408247422680413,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018044932319714546,
      "loss": 4.6028,
      "mean_token_accuracy": 0.24156380146741868,
      "num_tokens": 142856956.0,
      "step": 82365
    },
    {
      "entropy": 5.403048706054688,
      "epoch": 6.40863645205213,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001804241742405942,
      "loss": 4.6772,
      "mean_token_accuracy": 0.23438923358917235,
      "num_tokens": 142865804.0,
      "step": 82370
    },
    {
      "entropy": 5.395246696472168,
      "epoch": 6.409025481423847,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018039902671908076,
      "loss": 4.7213,
      "mean_token_accuracy": 0.2313644528388977,
      "num_tokens": 142874382.0,
      "step": 82375
    },
    {
      "entropy": 5.430659437179566,
      "epoch": 6.409414510795565,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018037388063298648,
      "loss": 4.7254,
      "mean_token_accuracy": 0.23973845392465593,
      "num_tokens": 142883489.0,
      "step": 82380
    },
    {
      "entropy": 5.394609117507935,
      "epoch": 6.409803540167283,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001803487359826929,
      "loss": 4.6146,
      "mean_token_accuracy": 0.24025665372610092,
      "num_tokens": 142892279.0,
      "step": 82385
    },
    {
      "entropy": 5.417500734329224,
      "epoch": 6.410192569539,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018032359276858167,
      "loss": 4.6723,
      "mean_token_accuracy": 0.23485136926174163,
      "num_tokens": 142901571.0,
      "step": 82390
    },
    {
      "entropy": 5.398145771026611,
      "epoch": 6.410581598910718,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018029845099103414,
      "loss": 4.6169,
      "mean_token_accuracy": 0.24927266538143159,
      "num_tokens": 142909296.0,
      "step": 82395
    },
    {
      "entropy": 5.30060658454895,
      "epoch": 6.410970628282436,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001802733106504319,
      "loss": 4.5603,
      "mean_token_accuracy": 0.2482319578528404,
      "num_tokens": 142917249.0,
      "step": 82400
    },
    {
      "entropy": 5.450872230529785,
      "epoch": 6.4113596576541525,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001802481717471564,
      "loss": 4.6992,
      "mean_token_accuracy": 0.23405710756778716,
      "num_tokens": 142925910.0,
      "step": 82405
    },
    {
      "entropy": 5.365111398696899,
      "epoch": 6.41174868702587,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018022303428158915,
      "loss": 4.6425,
      "mean_token_accuracy": 0.23882312178611756,
      "num_tokens": 142934468.0,
      "step": 82410
    },
    {
      "entropy": 5.422766447067261,
      "epoch": 6.412137716397588,
      "grad_norm": 1.421875,
      "learning_rate": 0.00018019789825411136,
      "loss": 4.6462,
      "mean_token_accuracy": 0.23503439575433732,
      "num_tokens": 142942854.0,
      "step": 82415
    },
    {
      "entropy": 5.416478204727173,
      "epoch": 6.412526745769306,
      "grad_norm": 1.25,
      "learning_rate": 0.0001801727636651047,
      "loss": 4.6529,
      "mean_token_accuracy": 0.23765443861484528,
      "num_tokens": 142951656.0,
      "step": 82420
    },
    {
      "entropy": 5.437187242507934,
      "epoch": 6.412915775141023,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018014763051495047,
      "loss": 4.716,
      "mean_token_accuracy": 0.2361622154712677,
      "num_tokens": 142959294.0,
      "step": 82425
    },
    {
      "entropy": 5.447124719619751,
      "epoch": 6.413304804512741,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001801224988040298,
      "loss": 4.7486,
      "mean_token_accuracy": 0.22954119741916656,
      "num_tokens": 142967526.0,
      "step": 82430
    },
    {
      "entropy": 5.432141971588135,
      "epoch": 6.413693833884459,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018009736853272424,
      "loss": 4.7576,
      "mean_token_accuracy": 0.23075587898492814,
      "num_tokens": 142976676.0,
      "step": 82435
    },
    {
      "entropy": 5.461958694458008,
      "epoch": 6.4140828632561755,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018007223970141505,
      "loss": 4.7452,
      "mean_token_accuracy": 0.23252155631780624,
      "num_tokens": 142986004.0,
      "step": 82440
    },
    {
      "entropy": 5.386804628372192,
      "epoch": 6.414471892627893,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001800471123104836,
      "loss": 4.6028,
      "mean_token_accuracy": 0.24331078678369522,
      "num_tokens": 142994367.0,
      "step": 82445
    },
    {
      "entropy": 5.381896591186523,
      "epoch": 6.414860921999611,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018002198636031102,
      "loss": 4.5834,
      "mean_token_accuracy": 0.24277009516954423,
      "num_tokens": 143002913.0,
      "step": 82450
    },
    {
      "entropy": 5.39580135345459,
      "epoch": 6.415249951371329,
      "grad_norm": 1.25,
      "learning_rate": 0.00017999686185127871,
      "loss": 4.6866,
      "mean_token_accuracy": 0.23713803440332412,
      "num_tokens": 143012000.0,
      "step": 82455
    },
    {
      "entropy": 5.414807224273682,
      "epoch": 6.415638980743046,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017997173878376782,
      "loss": 4.6672,
      "mean_token_accuracy": 0.24279556125402452,
      "num_tokens": 143020595.0,
      "step": 82460
    },
    {
      "entropy": 5.356094026565552,
      "epoch": 6.416028010114764,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017994661715815946,
      "loss": 4.5955,
      "mean_token_accuracy": 0.25009226351976394,
      "num_tokens": 143029243.0,
      "step": 82465
    },
    {
      "entropy": 5.358830356597901,
      "epoch": 6.416417039486481,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017992149697483494,
      "loss": 4.6411,
      "mean_token_accuracy": 0.24308784753084184,
      "num_tokens": 143038456.0,
      "step": 82470
    },
    {
      "entropy": 5.422231864929199,
      "epoch": 6.4168060688581985,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017989637823417532,
      "loss": 4.7382,
      "mean_token_accuracy": 0.23100993484258653,
      "num_tokens": 143047033.0,
      "step": 82475
    },
    {
      "entropy": 5.377179479598999,
      "epoch": 6.417195098229916,
      "grad_norm": 1.21875,
      "learning_rate": 0.00017987126093656187,
      "loss": 4.6095,
      "mean_token_accuracy": 0.2589604452252388,
      "num_tokens": 143056103.0,
      "step": 82480
    },
    {
      "entropy": 5.395293426513672,
      "epoch": 6.417584127601634,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001798461450823755,
      "loss": 4.6664,
      "mean_token_accuracy": 0.23565146774053575,
      "num_tokens": 143064403.0,
      "step": 82485
    },
    {
      "entropy": 5.396309900283813,
      "epoch": 6.417973156973352,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017982103067199756,
      "loss": 4.6518,
      "mean_token_accuracy": 0.24212147444486617,
      "num_tokens": 143072681.0,
      "step": 82490
    },
    {
      "entropy": 5.374069261550903,
      "epoch": 6.418362186345069,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017979591770580884,
      "loss": 4.6486,
      "mean_token_accuracy": 0.24003251641988754,
      "num_tokens": 143081360.0,
      "step": 82495
    },
    {
      "entropy": 5.444823360443115,
      "epoch": 6.418751215716787,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017977080618419066,
      "loss": 4.7323,
      "mean_token_accuracy": 0.23150442093610762,
      "num_tokens": 143090267.0,
      "step": 82500
    },
    {
      "entropy": 5.444047164916992,
      "epoch": 6.419140245088504,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00017974569610752384,
      "loss": 4.6656,
      "mean_token_accuracy": 0.23649680316448213,
      "num_tokens": 143098884.0,
      "step": 82505
    },
    {
      "entropy": 5.5026328563690186,
      "epoch": 6.4195292744602215,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017972058747618946,
      "loss": 4.8003,
      "mean_token_accuracy": 0.23001947849988938,
      "num_tokens": 143107512.0,
      "step": 82510
    },
    {
      "entropy": 5.428179359436035,
      "epoch": 6.419918303831939,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017969548029056858,
      "loss": 4.6303,
      "mean_token_accuracy": 0.23728398978710175,
      "num_tokens": 143115253.0,
      "step": 82515
    },
    {
      "entropy": 5.315980243682861,
      "epoch": 6.420307333203657,
      "grad_norm": 1.28125,
      "learning_rate": 0.00017967037455104208,
      "loss": 4.6095,
      "mean_token_accuracy": 0.24113020896911622,
      "num_tokens": 143124094.0,
      "step": 82520
    },
    {
      "entropy": 5.356109666824341,
      "epoch": 6.420696362575375,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017964527025799093,
      "loss": 4.6908,
      "mean_token_accuracy": 0.24234326034784318,
      "num_tokens": 143133109.0,
      "step": 82525
    },
    {
      "entropy": 5.294205904006958,
      "epoch": 6.421085391947092,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017962016741179592,
      "loss": 4.6556,
      "mean_token_accuracy": 0.24496525079011916,
      "num_tokens": 143141930.0,
      "step": 82530
    },
    {
      "entropy": 5.32760066986084,
      "epoch": 6.42147442131881,
      "grad_norm": 1.28125,
      "learning_rate": 0.00017959506601283827,
      "loss": 4.576,
      "mean_token_accuracy": 0.24812858402729035,
      "num_tokens": 143150687.0,
      "step": 82535
    },
    {
      "entropy": 5.430657768249512,
      "epoch": 6.421863450690527,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001795699660614985,
      "loss": 4.6857,
      "mean_token_accuracy": 0.23502505123615264,
      "num_tokens": 143159636.0,
      "step": 82540
    },
    {
      "entropy": 5.444018316268921,
      "epoch": 6.4222524800622445,
      "grad_norm": 1.28125,
      "learning_rate": 0.00017954486755815757,
      "loss": 4.6888,
      "mean_token_accuracy": 0.23968026489019395,
      "num_tokens": 143168227.0,
      "step": 82545
    },
    {
      "entropy": 5.311178350448609,
      "epoch": 6.422641509433962,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017951977050319646,
      "loss": 4.611,
      "mean_token_accuracy": 0.2364587441086769,
      "num_tokens": 143177328.0,
      "step": 82550
    },
    {
      "entropy": 5.378188180923462,
      "epoch": 6.42303053880568,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017949467489699572,
      "loss": 4.6586,
      "mean_token_accuracy": 0.2442087411880493,
      "num_tokens": 143185648.0,
      "step": 82555
    },
    {
      "entropy": 5.422831296920776,
      "epoch": 6.423419568177398,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017946958073993642,
      "loss": 4.7126,
      "mean_token_accuracy": 0.22931693345308304,
      "num_tokens": 143194210.0,
      "step": 82560
    },
    {
      "entropy": 5.294196367263794,
      "epoch": 6.423808597549115,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017944448803239904,
      "loss": 4.542,
      "mean_token_accuracy": 0.24147382378578186,
      "num_tokens": 143202771.0,
      "step": 82565
    },
    {
      "entropy": 5.43730411529541,
      "epoch": 6.424197626920833,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001794193967747645,
      "loss": 4.6435,
      "mean_token_accuracy": 0.24095530956983566,
      "num_tokens": 143211618.0,
      "step": 82570
    },
    {
      "entropy": 5.394436883926391,
      "epoch": 6.42458665629255,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017939430696741355,
      "loss": 4.6331,
      "mean_token_accuracy": 0.23825594484806062,
      "num_tokens": 143219828.0,
      "step": 82575
    },
    {
      "entropy": 5.302516603469849,
      "epoch": 6.4249756856642675,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017936921861072674,
      "loss": 4.5366,
      "mean_token_accuracy": 0.24871233105659485,
      "num_tokens": 143228281.0,
      "step": 82580
    },
    {
      "entropy": 5.27856559753418,
      "epoch": 6.425364715035985,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001793441317050849,
      "loss": 4.4542,
      "mean_token_accuracy": 0.25426729321479796,
      "num_tokens": 143236656.0,
      "step": 82585
    },
    {
      "entropy": 5.362694597244262,
      "epoch": 6.425753744407703,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017931904625086853,
      "loss": 4.6201,
      "mean_token_accuracy": 0.24828188568353654,
      "num_tokens": 143245231.0,
      "step": 82590
    },
    {
      "entropy": 5.370662069320678,
      "epoch": 6.426142773779421,
      "grad_norm": 1.375,
      "learning_rate": 0.00017929396224845845,
      "loss": 4.6314,
      "mean_token_accuracy": 0.23676994889974595,
      "num_tokens": 143254115.0,
      "step": 82595
    },
    {
      "entropy": 5.359732532501221,
      "epoch": 6.426531803151138,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00017926887969823503,
      "loss": 4.6368,
      "mean_token_accuracy": 0.23817794471979142,
      "num_tokens": 143262591.0,
      "step": 82600
    },
    {
      "entropy": 5.382681894302368,
      "epoch": 6.426920832522855,
      "grad_norm": 1.5,
      "learning_rate": 0.0001792437986005791,
      "loss": 4.6259,
      "mean_token_accuracy": 0.23873403966426848,
      "num_tokens": 143271343.0,
      "step": 82605
    },
    {
      "entropy": 5.3405029296875,
      "epoch": 6.427309861894573,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000179218718955871,
      "loss": 4.6296,
      "mean_token_accuracy": 0.25048225969076154,
      "num_tokens": 143279344.0,
      "step": 82610
    },
    {
      "entropy": 5.373542213439942,
      "epoch": 6.4276988912662905,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001791936407644915,
      "loss": 4.663,
      "mean_token_accuracy": 0.23606383204460143,
      "num_tokens": 143287896.0,
      "step": 82615
    },
    {
      "entropy": 5.386887645721435,
      "epoch": 6.428087920638008,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017916856402682102,
      "loss": 4.626,
      "mean_token_accuracy": 0.24582363665103912,
      "num_tokens": 143296144.0,
      "step": 82620
    },
    {
      "entropy": 5.3436016082763675,
      "epoch": 6.428476950009726,
      "grad_norm": 1.375,
      "learning_rate": 0.00017914348874323993,
      "loss": 4.5661,
      "mean_token_accuracy": 0.24722502380609512,
      "num_tokens": 143304299.0,
      "step": 82625
    },
    {
      "entropy": 5.337892532348633,
      "epoch": 6.4288659793814436,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017911841491412894,
      "loss": 4.5634,
      "mean_token_accuracy": 0.24097222089767456,
      "num_tokens": 143312478.0,
      "step": 82630
    },
    {
      "entropy": 5.31866946220398,
      "epoch": 6.429255008753161,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017909334253986826,
      "loss": 4.5188,
      "mean_token_accuracy": 0.251820707321167,
      "num_tokens": 143321075.0,
      "step": 82635
    },
    {
      "entropy": 5.324297666549683,
      "epoch": 6.429644038124878,
      "grad_norm": 1.28125,
      "learning_rate": 0.00017906827162083856,
      "loss": 4.6387,
      "mean_token_accuracy": 0.24618198126554489,
      "num_tokens": 143330142.0,
      "step": 82640
    },
    {
      "entropy": 5.297784519195557,
      "epoch": 6.430033067496596,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001790432021574202,
      "loss": 4.6495,
      "mean_token_accuracy": 0.23837926387786865,
      "num_tokens": 143339008.0,
      "step": 82645
    },
    {
      "entropy": 5.34595308303833,
      "epoch": 6.4304220968683135,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00017901813414999336,
      "loss": 4.665,
      "mean_token_accuracy": 0.24033653885126113,
      "num_tokens": 143348841.0,
      "step": 82650
    },
    {
      "entropy": 5.400599622726441,
      "epoch": 6.430811126240031,
      "grad_norm": 1.375,
      "learning_rate": 0.00017899306759893875,
      "loss": 4.6592,
      "mean_token_accuracy": 0.24718987345695495,
      "num_tokens": 143356681.0,
      "step": 82655
    },
    {
      "entropy": 5.4087811470031735,
      "epoch": 6.431200155611749,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017896800250463636,
      "loss": 4.6724,
      "mean_token_accuracy": 0.242009998857975,
      "num_tokens": 143365396.0,
      "step": 82660
    },
    {
      "entropy": 5.332721185684204,
      "epoch": 6.4315891849834665,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00017894293886746676,
      "loss": 4.5469,
      "mean_token_accuracy": 0.24421672374010087,
      "num_tokens": 143374678.0,
      "step": 82665
    },
    {
      "entropy": 5.359928131103516,
      "epoch": 6.431978214355183,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001789178766878101,
      "loss": 4.6075,
      "mean_token_accuracy": 0.24772242605686187,
      "num_tokens": 143383423.0,
      "step": 82670
    },
    {
      "entropy": 5.35410442352295,
      "epoch": 6.432367243726901,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017889281596604684,
      "loss": 4.6413,
      "mean_token_accuracy": 0.24685716181993483,
      "num_tokens": 143392366.0,
      "step": 82675
    },
    {
      "entropy": 5.44408769607544,
      "epoch": 6.432756273098619,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017886775670255707,
      "loss": 4.754,
      "mean_token_accuracy": 0.23950789123773575,
      "num_tokens": 143400815.0,
      "step": 82680
    },
    {
      "entropy": 5.328939771652221,
      "epoch": 6.4331453024703364,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017884269889772116,
      "loss": 4.5276,
      "mean_token_accuracy": 0.247337943315506,
      "num_tokens": 143410319.0,
      "step": 82685
    },
    {
      "entropy": 5.345132064819336,
      "epoch": 6.433534331842054,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017881764255191912,
      "loss": 4.6627,
      "mean_token_accuracy": 0.2379737988114357,
      "num_tokens": 143419177.0,
      "step": 82690
    },
    {
      "entropy": 5.34805121421814,
      "epoch": 6.433923361213772,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00017879258766553136,
      "loss": 4.6714,
      "mean_token_accuracy": 0.2511689379811287,
      "num_tokens": 143428030.0,
      "step": 82695
    },
    {
      "entropy": 5.389418268203736,
      "epoch": 6.4343123905854895,
      "grad_norm": 1.28125,
      "learning_rate": 0.00017876753423893792,
      "loss": 4.6642,
      "mean_token_accuracy": 0.23998350501060486,
      "num_tokens": 143436930.0,
      "step": 82700
    },
    {
      "entropy": 5.406605768203735,
      "epoch": 6.434701419957207,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017874248227251887,
      "loss": 4.7228,
      "mean_token_accuracy": 0.23316849768161774,
      "num_tokens": 143445831.0,
      "step": 82705
    },
    {
      "entropy": 5.399896621704102,
      "epoch": 6.435090449328924,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017871743176665455,
      "loss": 4.6348,
      "mean_token_accuracy": 0.24600838720798493,
      "num_tokens": 143454345.0,
      "step": 82710
    },
    {
      "entropy": 5.526035642623901,
      "epoch": 6.435479478700642,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017869238272172493,
      "loss": 4.7978,
      "mean_token_accuracy": 0.22638158947229386,
      "num_tokens": 143462770.0,
      "step": 82715
    },
    {
      "entropy": 5.455417060852051,
      "epoch": 6.435868508072359,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017866733513811016,
      "loss": 4.7059,
      "mean_token_accuracy": 0.2342896968126297,
      "num_tokens": 143470495.0,
      "step": 82720
    },
    {
      "entropy": 5.287090015411377,
      "epoch": 6.436257537444077,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00017864228901619028,
      "loss": 4.5918,
      "mean_token_accuracy": 0.24382158517837524,
      "num_tokens": 143479479.0,
      "step": 82725
    },
    {
      "entropy": 5.357227468490601,
      "epoch": 6.436646566815795,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001786172443563452,
      "loss": 4.6235,
      "mean_token_accuracy": 0.24275797754526138,
      "num_tokens": 143488823.0,
      "step": 82730
    },
    {
      "entropy": 5.329410600662231,
      "epoch": 6.4370355961875125,
      "grad_norm": 1.375,
      "learning_rate": 0.0001785922011589551,
      "loss": 4.606,
      "mean_token_accuracy": 0.24503783136606216,
      "num_tokens": 143497305.0,
      "step": 82735
    },
    {
      "entropy": 5.41592378616333,
      "epoch": 6.437424625559229,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00017856715942439982,
      "loss": 4.695,
      "mean_token_accuracy": 0.23348598182201385,
      "num_tokens": 143507312.0,
      "step": 82740
    },
    {
      "entropy": 5.4380169868469235,
      "epoch": 6.437813654930947,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00017854211915305945,
      "loss": 4.7373,
      "mean_token_accuracy": 0.23575616627931595,
      "num_tokens": 143515898.0,
      "step": 82745
    },
    {
      "entropy": 5.353181791305542,
      "epoch": 6.438202684302665,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001785170803453139,
      "loss": 4.6052,
      "mean_token_accuracy": 0.24803001284599305,
      "num_tokens": 143523708.0,
      "step": 82750
    },
    {
      "entropy": 5.357293510437012,
      "epoch": 6.438591713674382,
      "grad_norm": 1.25,
      "learning_rate": 0.0001784920430015431,
      "loss": 4.5987,
      "mean_token_accuracy": 0.24204165637493133,
      "num_tokens": 143532507.0,
      "step": 82755
    },
    {
      "entropy": 5.4409888744354244,
      "epoch": 6.4389807430461,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001784670071221269,
      "loss": 4.7485,
      "mean_token_accuracy": 0.22733623087406157,
      "num_tokens": 143541221.0,
      "step": 82760
    },
    {
      "entropy": 5.3728498935699465,
      "epoch": 6.439369772417818,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001784419727074452,
      "loss": 4.7145,
      "mean_token_accuracy": 0.2306716576218605,
      "num_tokens": 143549885.0,
      "step": 82765
    },
    {
      "entropy": 5.331688261032104,
      "epoch": 6.4397588017895355,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00017841693975787791,
      "loss": 4.5989,
      "mean_token_accuracy": 0.24704163521528244,
      "num_tokens": 143558871.0,
      "step": 82770
    },
    {
      "entropy": 5.369926404953003,
      "epoch": 6.440147831161252,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001783919082738048,
      "loss": 4.6112,
      "mean_token_accuracy": 0.23526297211647035,
      "num_tokens": 143567483.0,
      "step": 82775
    },
    {
      "entropy": 5.318182897567749,
      "epoch": 6.44053686053297,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017836687825560578,
      "loss": 4.6087,
      "mean_token_accuracy": 0.239929136633873,
      "num_tokens": 143576919.0,
      "step": 82780
    },
    {
      "entropy": 5.370616054534912,
      "epoch": 6.440925889904688,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001783418497036605,
      "loss": 4.5892,
      "mean_token_accuracy": 0.2424986407160759,
      "num_tokens": 143585575.0,
      "step": 82785
    },
    {
      "entropy": 5.447087526321411,
      "epoch": 6.441314919276405,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001783168226183489,
      "loss": 4.6691,
      "mean_token_accuracy": 0.2354416325688362,
      "num_tokens": 143595050.0,
      "step": 82790
    },
    {
      "entropy": 5.383753728866577,
      "epoch": 6.441703948648123,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017829179700005056,
      "loss": 4.6575,
      "mean_token_accuracy": 0.2362608253955841,
      "num_tokens": 143603145.0,
      "step": 82795
    },
    {
      "entropy": 5.3817009925842285,
      "epoch": 6.442092978019841,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017826677284914533,
      "loss": 4.6469,
      "mean_token_accuracy": 0.24037082940340043,
      "num_tokens": 143611889.0,
      "step": 82800
    },
    {
      "entropy": 5.43207950592041,
      "epoch": 6.442482007391558,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017824175016601285,
      "loss": 4.6936,
      "mean_token_accuracy": 0.23874589949846267,
      "num_tokens": 143619854.0,
      "step": 82805
    },
    {
      "entropy": 5.3997721672058105,
      "epoch": 6.442871036763275,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001782167289510327,
      "loss": 4.6544,
      "mean_token_accuracy": 0.23907834887504578,
      "num_tokens": 143627882.0,
      "step": 82810
    },
    {
      "entropy": 5.445751523971557,
      "epoch": 6.443260066134993,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017819170920458478,
      "loss": 4.7703,
      "mean_token_accuracy": 0.228787662088871,
      "num_tokens": 143636882.0,
      "step": 82815
    },
    {
      "entropy": 5.4082811832427975,
      "epoch": 6.443649095506711,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017816669092704846,
      "loss": 4.6316,
      "mean_token_accuracy": 0.24099887460470198,
      "num_tokens": 143644372.0,
      "step": 82820
    },
    {
      "entropy": 5.374001216888428,
      "epoch": 6.444038124878428,
      "grad_norm": 1.28125,
      "learning_rate": 0.00017814167411880354,
      "loss": 4.5304,
      "mean_token_accuracy": 0.2501221776008606,
      "num_tokens": 143652764.0,
      "step": 82825
    },
    {
      "entropy": 5.457921028137207,
      "epoch": 6.444427154250146,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017811665878022953,
      "loss": 4.7587,
      "mean_token_accuracy": 0.2417881429195404,
      "num_tokens": 143661722.0,
      "step": 82830
    },
    {
      "entropy": 5.390911960601807,
      "epoch": 6.444816183621864,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017809164491170597,
      "loss": 4.5971,
      "mean_token_accuracy": 0.24584528654813767,
      "num_tokens": 143670097.0,
      "step": 82835
    },
    {
      "entropy": 5.387441110610962,
      "epoch": 6.445205212993581,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001780666325136126,
      "loss": 4.6429,
      "mean_token_accuracy": 0.2411521017551422,
      "num_tokens": 143678954.0,
      "step": 82840
    },
    {
      "entropy": 5.357714033126831,
      "epoch": 6.445594242365298,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017804162158632864,
      "loss": 4.642,
      "mean_token_accuracy": 0.2400539055466652,
      "num_tokens": 143687329.0,
      "step": 82845
    },
    {
      "entropy": 5.338862371444702,
      "epoch": 6.445983271737016,
      "grad_norm": 1.234375,
      "learning_rate": 0.00017801661213023384,
      "loss": 4.6206,
      "mean_token_accuracy": 0.24225640147924424,
      "num_tokens": 143697445.0,
      "step": 82850
    },
    {
      "entropy": 5.400271320343018,
      "epoch": 6.446372301108734,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017799160414570754,
      "loss": 4.6967,
      "mean_token_accuracy": 0.24033423513174057,
      "num_tokens": 143706627.0,
      "step": 82855
    },
    {
      "entropy": 5.465586757659912,
      "epoch": 6.446761330480451,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017796659763312934,
      "loss": 4.7074,
      "mean_token_accuracy": 0.23926141113042831,
      "num_tokens": 143715108.0,
      "step": 82860
    },
    {
      "entropy": 5.317722749710083,
      "epoch": 6.447150359852169,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017794159259287845,
      "loss": 4.5346,
      "mean_token_accuracy": 0.24749047607183455,
      "num_tokens": 143723422.0,
      "step": 82865
    },
    {
      "entropy": 5.402229881286621,
      "epoch": 6.447539389223887,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017791658902533447,
      "loss": 4.6624,
      "mean_token_accuracy": 0.24272151440382003,
      "num_tokens": 143731784.0,
      "step": 82870
    },
    {
      "entropy": 5.343760776519775,
      "epoch": 6.447928418595604,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001778915869308767,
      "loss": 4.6736,
      "mean_token_accuracy": 0.23606982678174973,
      "num_tokens": 143740851.0,
      "step": 82875
    },
    {
      "entropy": 5.3634425640106205,
      "epoch": 6.448317447967321,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017786658630988462,
      "loss": 4.61,
      "mean_token_accuracy": 0.24724558144807815,
      "num_tokens": 143749444.0,
      "step": 82880
    },
    {
      "entropy": 5.414980459213257,
      "epoch": 6.448706477339039,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017784158716273746,
      "loss": 4.6934,
      "mean_token_accuracy": 0.24275663495063782,
      "num_tokens": 143758027.0,
      "step": 82885
    },
    {
      "entropy": 5.286357975006103,
      "epoch": 6.449095506710757,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017781658948981449,
      "loss": 4.5018,
      "mean_token_accuracy": 0.26249654591083527,
      "num_tokens": 143766413.0,
      "step": 82890
    },
    {
      "entropy": 5.456971788406372,
      "epoch": 6.449484536082474,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017779159329149515,
      "loss": 4.7674,
      "mean_token_accuracy": 0.22810399085283278,
      "num_tokens": 143775014.0,
      "step": 82895
    },
    {
      "entropy": 5.339426803588867,
      "epoch": 6.449873565454192,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017776659856815852,
      "loss": 4.5752,
      "mean_token_accuracy": 0.24826463013887407,
      "num_tokens": 143783921.0,
      "step": 82900
    },
    {
      "entropy": 5.3621879577636715,
      "epoch": 6.45026259482591,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017774160532018417,
      "loss": 4.613,
      "mean_token_accuracy": 0.25057166814804077,
      "num_tokens": 143792714.0,
      "step": 82905
    },
    {
      "entropy": 5.375015449523926,
      "epoch": 6.450651624197627,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017771661354795104,
      "loss": 4.5844,
      "mean_token_accuracy": 0.24597897082567216,
      "num_tokens": 143801153.0,
      "step": 82910
    },
    {
      "entropy": 5.377874279022217,
      "epoch": 6.451040653569344,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017769162325183857,
      "loss": 4.5852,
      "mean_token_accuracy": 0.24732966125011444,
      "num_tokens": 143810068.0,
      "step": 82915
    },
    {
      "entropy": 5.379004955291748,
      "epoch": 6.451429682941062,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017766663443222585,
      "loss": 4.6237,
      "mean_token_accuracy": 0.23434246480464935,
      "num_tokens": 143817844.0,
      "step": 82920
    },
    {
      "entropy": 5.380799770355225,
      "epoch": 6.45181871231278,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017764164708949194,
      "loss": 4.6596,
      "mean_token_accuracy": 0.23932394087314607,
      "num_tokens": 143826291.0,
      "step": 82925
    },
    {
      "entropy": 5.43578462600708,
      "epoch": 6.452207741684497,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001776166612240161,
      "loss": 4.6954,
      "mean_token_accuracy": 0.23862385153770446,
      "num_tokens": 143834985.0,
      "step": 82930
    },
    {
      "entropy": 5.339512729644776,
      "epoch": 6.452596771056215,
      "grad_norm": 1.234375,
      "learning_rate": 0.00017759167683617738,
      "loss": 4.6465,
      "mean_token_accuracy": 0.24068357050418854,
      "num_tokens": 143844369.0,
      "step": 82935
    },
    {
      "entropy": 5.496247625350952,
      "epoch": 6.452985800427932,
      "grad_norm": 1.390625,
      "learning_rate": 0.000177566693926355,
      "loss": 4.7513,
      "mean_token_accuracy": 0.23614590018987655,
      "num_tokens": 143853000.0,
      "step": 82940
    },
    {
      "entropy": 5.442478942871094,
      "epoch": 6.45337482979965,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001775417124949279,
      "loss": 4.731,
      "mean_token_accuracy": 0.23147838115692138,
      "num_tokens": 143861623.0,
      "step": 82945
    },
    {
      "entropy": 5.344984102249145,
      "epoch": 6.453763859171367,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001775167325422753,
      "loss": 4.5786,
      "mean_token_accuracy": 0.2535018652677536,
      "num_tokens": 143870148.0,
      "step": 82950
    },
    {
      "entropy": 5.369374847412109,
      "epoch": 6.454152888543085,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017749175406877598,
      "loss": 4.6569,
      "mean_token_accuracy": 0.23541801869869233,
      "num_tokens": 143880379.0,
      "step": 82955
    },
    {
      "entropy": 5.376127099990844,
      "epoch": 6.454541917914803,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001774667770748092,
      "loss": 4.6916,
      "mean_token_accuracy": 0.23773810118436814,
      "num_tokens": 143889027.0,
      "step": 82960
    },
    {
      "entropy": 5.3560628414154055,
      "epoch": 6.45493094728652,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017744180156075372,
      "loss": 4.6129,
      "mean_token_accuracy": 0.2472979798913002,
      "num_tokens": 143897411.0,
      "step": 82965
    },
    {
      "entropy": 5.350114631652832,
      "epoch": 6.455319976658238,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017741682752698868,
      "loss": 4.628,
      "mean_token_accuracy": 0.24208532124757767,
      "num_tokens": 143905314.0,
      "step": 82970
    },
    {
      "entropy": 5.424723958969116,
      "epoch": 6.455709006029955,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017739185497389298,
      "loss": 4.6715,
      "mean_token_accuracy": 0.23854233026504518,
      "num_tokens": 143914037.0,
      "step": 82975
    },
    {
      "entropy": 5.407558488845825,
      "epoch": 6.456098035401673,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017736688390184552,
      "loss": 4.6744,
      "mean_token_accuracy": 0.23678355664014816,
      "num_tokens": 143922816.0,
      "step": 82980
    },
    {
      "entropy": 5.352316856384277,
      "epoch": 6.45648706477339,
      "grad_norm": 1.375,
      "learning_rate": 0.0001773419143112252,
      "loss": 4.5705,
      "mean_token_accuracy": 0.24099220633506774,
      "num_tokens": 143930862.0,
      "step": 82985
    },
    {
      "entropy": 5.473547744750976,
      "epoch": 6.456876094145108,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017731694620241084,
      "loss": 4.7749,
      "mean_token_accuracy": 0.2313386917114258,
      "num_tokens": 143939630.0,
      "step": 82990
    },
    {
      "entropy": 5.405443334579468,
      "epoch": 6.457265123516826,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001772919795757814,
      "loss": 4.5958,
      "mean_token_accuracy": 0.24345363676548004,
      "num_tokens": 143948361.0,
      "step": 82995
    },
    {
      "entropy": 5.4167420864105225,
      "epoch": 6.457654152888543,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017726701443171572,
      "loss": 4.6272,
      "mean_token_accuracy": 0.24736629724502562,
      "num_tokens": 143957460.0,
      "step": 83000
    },
    {
      "entropy": 5.3337325096130375,
      "epoch": 6.45804318226026,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001772420507705924,
      "loss": 4.624,
      "mean_token_accuracy": 0.24406273514032364,
      "num_tokens": 143966404.0,
      "step": 83005
    },
    {
      "entropy": 5.262674856185913,
      "epoch": 6.458432211631978,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017721708859279046,
      "loss": 4.5657,
      "mean_token_accuracy": 0.24264351427555084,
      "num_tokens": 143975064.0,
      "step": 83010
    },
    {
      "entropy": 5.317763185501098,
      "epoch": 6.458821241003696,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001771921278986885,
      "loss": 4.5363,
      "mean_token_accuracy": 0.2511401042342186,
      "num_tokens": 143983780.0,
      "step": 83015
    },
    {
      "entropy": 5.367264080047607,
      "epoch": 6.459210270375413,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001771671686886654,
      "loss": 4.6875,
      "mean_token_accuracy": 0.237623693048954,
      "num_tokens": 143993344.0,
      "step": 83020
    },
    {
      "entropy": 5.378525876998902,
      "epoch": 6.459599299747131,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017714221096309967,
      "loss": 4.6444,
      "mean_token_accuracy": 0.23488162457942963,
      "num_tokens": 144002678.0,
      "step": 83025
    },
    {
      "entropy": 5.449579668045044,
      "epoch": 6.459988329118849,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017711725472237028,
      "loss": 4.6412,
      "mean_token_accuracy": 0.24572697281837463,
      "num_tokens": 144010553.0,
      "step": 83030
    },
    {
      "entropy": 5.402384614944458,
      "epoch": 6.460377358490566,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017709229996685567,
      "loss": 4.6678,
      "mean_token_accuracy": 0.23952512443065643,
      "num_tokens": 144019587.0,
      "step": 83035
    },
    {
      "entropy": 5.410198545455932,
      "epoch": 6.460766387862283,
      "grad_norm": 1.5,
      "learning_rate": 0.00017706734669693447,
      "loss": 4.6733,
      "mean_token_accuracy": 0.23823703229427337,
      "num_tokens": 144027889.0,
      "step": 83040
    },
    {
      "entropy": 5.423977661132812,
      "epoch": 6.461155417234001,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017704239491298556,
      "loss": 4.7003,
      "mean_token_accuracy": 0.23039052486419678,
      "num_tokens": 144036618.0,
      "step": 83045
    },
    {
      "entropy": 5.39475383758545,
      "epoch": 6.461544446605719,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017701744461538733,
      "loss": 4.6854,
      "mean_token_accuracy": 0.23890471011400222,
      "num_tokens": 144046171.0,
      "step": 83050
    },
    {
      "entropy": 5.366714811325073,
      "epoch": 6.461933475977436,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017699249580451848,
      "loss": 4.5689,
      "mean_token_accuracy": 0.24190766364336014,
      "num_tokens": 144054564.0,
      "step": 83055
    },
    {
      "entropy": 5.2825024127960205,
      "epoch": 6.462322505349154,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001769675484807574,
      "loss": 4.492,
      "mean_token_accuracy": 0.25035585165023805,
      "num_tokens": 144062808.0,
      "step": 83060
    },
    {
      "entropy": 5.351992130279541,
      "epoch": 6.462711534720872,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017694260264448287,
      "loss": 4.5962,
      "mean_token_accuracy": 0.2456778109073639,
      "num_tokens": 144071542.0,
      "step": 83065
    },
    {
      "entropy": 5.337395238876343,
      "epoch": 6.463100564092589,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001769176582960732,
      "loss": 4.7133,
      "mean_token_accuracy": 0.23347149640321732,
      "num_tokens": 144080878.0,
      "step": 83070
    },
    {
      "entropy": 5.31376953125,
      "epoch": 6.463489593464306,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00017689271543590702,
      "loss": 4.6278,
      "mean_token_accuracy": 0.2330754280090332,
      "num_tokens": 144089817.0,
      "step": 83075
    },
    {
      "entropy": 5.4150779247283936,
      "epoch": 6.463878622836024,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017686777406436267,
      "loss": 4.727,
      "mean_token_accuracy": 0.23348935395479203,
      "num_tokens": 144099105.0,
      "step": 83080
    },
    {
      "entropy": 5.35163631439209,
      "epoch": 6.464267652207742,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00017684283418181862,
      "loss": 4.5938,
      "mean_token_accuracy": 0.24613300412893296,
      "num_tokens": 144107433.0,
      "step": 83085
    },
    {
      "entropy": 5.280055284500122,
      "epoch": 6.464656681579459,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017681789578865342,
      "loss": 4.4464,
      "mean_token_accuracy": 0.2616216540336609,
      "num_tokens": 144116014.0,
      "step": 83090
    },
    {
      "entropy": 5.352335357666016,
      "epoch": 6.465045710951177,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017679295888524526,
      "loss": 4.6164,
      "mean_token_accuracy": 0.2486771896481514,
      "num_tokens": 144124524.0,
      "step": 83095
    },
    {
      "entropy": 5.431673288345337,
      "epoch": 6.465434740322895,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017676802347197268,
      "loss": 4.7013,
      "mean_token_accuracy": 0.23786157816648484,
      "num_tokens": 144134194.0,
      "step": 83100
    },
    {
      "entropy": 5.416635227203369,
      "epoch": 6.465823769694612,
      "grad_norm": 1.375,
      "learning_rate": 0.00017674308954921403,
      "loss": 4.6527,
      "mean_token_accuracy": 0.23166165202856065,
      "num_tokens": 144142975.0,
      "step": 83105
    },
    {
      "entropy": 5.434501981735229,
      "epoch": 6.466212799066329,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001767181571173475,
      "loss": 4.7677,
      "mean_token_accuracy": 0.23253786265850068,
      "num_tokens": 144152160.0,
      "step": 83110
    },
    {
      "entropy": 5.51481032371521,
      "epoch": 6.466601828438047,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001766932261767517,
      "loss": 4.7508,
      "mean_token_accuracy": 0.23270728588104247,
      "num_tokens": 144160275.0,
      "step": 83115
    },
    {
      "entropy": 5.435375070571899,
      "epoch": 6.466990857809765,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017666829672780454,
      "loss": 4.7002,
      "mean_token_accuracy": 0.23971298038959504,
      "num_tokens": 144168830.0,
      "step": 83120
    },
    {
      "entropy": 5.3598182678222654,
      "epoch": 6.467379887181482,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00017664336877088455,
      "loss": 4.6843,
      "mean_token_accuracy": 0.23895858824253083,
      "num_tokens": 144177532.0,
      "step": 83125
    },
    {
      "entropy": 5.381075048446656,
      "epoch": 6.4677689165532,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017661844230636986,
      "loss": 4.6461,
      "mean_token_accuracy": 0.24438221007585526,
      "num_tokens": 144186522.0,
      "step": 83130
    },
    {
      "entropy": 5.369091463088989,
      "epoch": 6.468157945924918,
      "grad_norm": 1.171875,
      "learning_rate": 0.00017659351733463879,
      "loss": 4.6695,
      "mean_token_accuracy": 0.23817451894283295,
      "num_tokens": 144195265.0,
      "step": 83135
    },
    {
      "entropy": 5.2811665534973145,
      "epoch": 6.4685469752966345,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017656859385606938,
      "loss": 4.5577,
      "mean_token_accuracy": 0.24444673508405684,
      "num_tokens": 144204035.0,
      "step": 83140
    },
    {
      "entropy": 5.478275394439697,
      "epoch": 6.468936004668352,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017654367187103993,
      "loss": 4.7658,
      "mean_token_accuracy": 0.23258228451013566,
      "num_tokens": 144213719.0,
      "step": 83145
    },
    {
      "entropy": 5.423763656616211,
      "epoch": 6.46932503404007,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001765187513799285,
      "loss": 4.5578,
      "mean_token_accuracy": 0.2460702881217003,
      "num_tokens": 144222339.0,
      "step": 83150
    },
    {
      "entropy": 5.430379009246826,
      "epoch": 6.469714063411788,
      "grad_norm": 1.375,
      "learning_rate": 0.00017649383238311334,
      "loss": 4.6843,
      "mean_token_accuracy": 0.23618950098752975,
      "num_tokens": 144230538.0,
      "step": 83155
    },
    {
      "entropy": 5.419036054611206,
      "epoch": 6.470103092783505,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017646891488097249,
      "loss": 4.6879,
      "mean_token_accuracy": 0.2372881218791008,
      "num_tokens": 144239515.0,
      "step": 83160
    },
    {
      "entropy": 5.403009223937988,
      "epoch": 6.470492122155223,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017644399887388395,
      "loss": 4.6246,
      "mean_token_accuracy": 0.24060717225074768,
      "num_tokens": 144248415.0,
      "step": 83165
    },
    {
      "entropy": 5.352075433731079,
      "epoch": 6.470881151526941,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017641908436222582,
      "loss": 4.6019,
      "mean_token_accuracy": 0.24918050467967987,
      "num_tokens": 144256911.0,
      "step": 83170
    },
    {
      "entropy": 5.383292198181152,
      "epoch": 6.4712701808986575,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00017639417134637626,
      "loss": 4.7333,
      "mean_token_accuracy": 0.22776828557252884,
      "num_tokens": 144266495.0,
      "step": 83175
    },
    {
      "entropy": 5.3731663703918455,
      "epoch": 6.471659210270375,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001763692598267132,
      "loss": 4.6339,
      "mean_token_accuracy": 0.24725642800331116,
      "num_tokens": 144275036.0,
      "step": 83180
    },
    {
      "entropy": 5.432977151870728,
      "epoch": 6.472048239642093,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001763443498036147,
      "loss": 4.6342,
      "mean_token_accuracy": 0.24487248808145523,
      "num_tokens": 144283063.0,
      "step": 83185
    },
    {
      "entropy": 5.350069665908814,
      "epoch": 6.472437269013811,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017631944127745853,
      "loss": 4.5786,
      "mean_token_accuracy": 0.2431916981935501,
      "num_tokens": 144291918.0,
      "step": 83190
    },
    {
      "entropy": 5.353489923477173,
      "epoch": 6.472826298385528,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001762945342486229,
      "loss": 4.6198,
      "mean_token_accuracy": 0.23948957920074462,
      "num_tokens": 144300950.0,
      "step": 83195
    },
    {
      "entropy": 5.379969072341919,
      "epoch": 6.473215327757246,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017626962871748547,
      "loss": 4.6049,
      "mean_token_accuracy": 0.2385811135172844,
      "num_tokens": 144309726.0,
      "step": 83200
    },
    {
      "entropy": 5.384865379333496,
      "epoch": 6.473604357128964,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017624472468442444,
      "loss": 4.6355,
      "mean_token_accuracy": 0.24071792364120484,
      "num_tokens": 144318172.0,
      "step": 83205
    },
    {
      "entropy": 5.3996976852417,
      "epoch": 6.4739933865006805,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001762198221498174,
      "loss": 4.7347,
      "mean_token_accuracy": 0.23437222689390183,
      "num_tokens": 144327651.0,
      "step": 83210
    },
    {
      "entropy": 5.371520853042602,
      "epoch": 6.474382415872398,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001761949211140424,
      "loss": 4.5978,
      "mean_token_accuracy": 0.2415680393576622,
      "num_tokens": 144336461.0,
      "step": 83215
    },
    {
      "entropy": 5.353822851181031,
      "epoch": 6.474771445244116,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017617002157747713,
      "loss": 4.5992,
      "mean_token_accuracy": 0.24708184599876404,
      "num_tokens": 144344873.0,
      "step": 83220
    },
    {
      "entropy": 5.4636212348937985,
      "epoch": 6.4751604746158336,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017614512354049956,
      "loss": 4.7315,
      "mean_token_accuracy": 0.2276887536048889,
      "num_tokens": 144354009.0,
      "step": 83225
    },
    {
      "entropy": 5.256934785842896,
      "epoch": 6.475549503987551,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017612022700348734,
      "loss": 4.5065,
      "mean_token_accuracy": 0.25481875985860825,
      "num_tokens": 144362835.0,
      "step": 83230
    },
    {
      "entropy": 5.3872823238372805,
      "epoch": 6.475938533359269,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017609533196681827,
      "loss": 4.7105,
      "mean_token_accuracy": 0.23734594881534576,
      "num_tokens": 144371390.0,
      "step": 83235
    },
    {
      "entropy": 5.415806436538697,
      "epoch": 6.476327562730987,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017607043843087024,
      "loss": 4.6786,
      "mean_token_accuracy": 0.2336479753255844,
      "num_tokens": 144379929.0,
      "step": 83240
    },
    {
      "entropy": 5.451193809509277,
      "epoch": 6.4767165921027035,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001760455463960207,
      "loss": 4.8381,
      "mean_token_accuracy": 0.22816236019134523,
      "num_tokens": 144389218.0,
      "step": 83245
    },
    {
      "entropy": 5.4189568042755125,
      "epoch": 6.477105621474421,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001760206558626476,
      "loss": 4.6398,
      "mean_token_accuracy": 0.24005489945411682,
      "num_tokens": 144398026.0,
      "step": 83250
    },
    {
      "entropy": 5.319013166427612,
      "epoch": 6.477494650846139,
      "grad_norm": 1.21875,
      "learning_rate": 0.00017599576683112843,
      "loss": 4.5606,
      "mean_token_accuracy": 0.24223515540361404,
      "num_tokens": 144406951.0,
      "step": 83255
    },
    {
      "entropy": 5.44429759979248,
      "epoch": 6.4778836802178565,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017597087930184096,
      "loss": 4.7072,
      "mean_token_accuracy": 0.2340103343129158,
      "num_tokens": 144416307.0,
      "step": 83260
    },
    {
      "entropy": 5.368282318115234,
      "epoch": 6.478272709589574,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017594599327516275,
      "loss": 4.6499,
      "mean_token_accuracy": 0.24169748723506929,
      "num_tokens": 144424553.0,
      "step": 83265
    },
    {
      "entropy": 5.380005884170532,
      "epoch": 6.478661738961292,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001759211087514715,
      "loss": 4.6563,
      "mean_token_accuracy": 0.23348379582166673,
      "num_tokens": 144432965.0,
      "step": 83270
    },
    {
      "entropy": 5.417811870574951,
      "epoch": 6.479050768333009,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017589622573114473,
      "loss": 4.6618,
      "mean_token_accuracy": 0.23485977053642274,
      "num_tokens": 144442246.0,
      "step": 83275
    },
    {
      "entropy": 5.3904369354248045,
      "epoch": 6.4794397977047264,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017587134421455985,
      "loss": 4.6345,
      "mean_token_accuracy": 0.24311417192220688,
      "num_tokens": 144450769.0,
      "step": 83280
    },
    {
      "entropy": 5.456230640411377,
      "epoch": 6.479828827076444,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017584646420209471,
      "loss": 4.7351,
      "mean_token_accuracy": 0.23157899379730223,
      "num_tokens": 144459316.0,
      "step": 83285
    },
    {
      "entropy": 5.314313220977783,
      "epoch": 6.480217856448162,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017582158569412655,
      "loss": 4.5794,
      "mean_token_accuracy": 0.24161757230758668,
      "num_tokens": 144468129.0,
      "step": 83290
    },
    {
      "entropy": 5.327884340286255,
      "epoch": 6.4806068858198795,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017579670869103298,
      "loss": 4.5809,
      "mean_token_accuracy": 0.24892695546150206,
      "num_tokens": 144477231.0,
      "step": 83295
    },
    {
      "entropy": 5.488379621505738,
      "epoch": 6.480995915191597,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017577183319319146,
      "loss": 4.7505,
      "mean_token_accuracy": 0.23556938171386718,
      "num_tokens": 144485564.0,
      "step": 83300
    },
    {
      "entropy": 5.4518866539001465,
      "epoch": 6.481384944563315,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001757469592009794,
      "loss": 4.7263,
      "mean_token_accuracy": 0.23517630100250245,
      "num_tokens": 144495001.0,
      "step": 83305
    },
    {
      "entropy": 5.400519466400146,
      "epoch": 6.481773973935032,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017572208671477434,
      "loss": 4.6299,
      "mean_token_accuracy": 0.24035293310880662,
      "num_tokens": 144503201.0,
      "step": 83310
    },
    {
      "entropy": 5.375655746459961,
      "epoch": 6.482163003306749,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017569721573495355,
      "loss": 4.7024,
      "mean_token_accuracy": 0.23829648345708848,
      "num_tokens": 144511117.0,
      "step": 83315
    },
    {
      "entropy": 5.3682609558105465,
      "epoch": 6.482552032678467,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001756723462618945,
      "loss": 4.6645,
      "mean_token_accuracy": 0.23648438453674317,
      "num_tokens": 144519697.0,
      "step": 83320
    },
    {
      "entropy": 5.4353930950164795,
      "epoch": 6.482941062050185,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017564747829597438,
      "loss": 4.6199,
      "mean_token_accuracy": 0.2387355834245682,
      "num_tokens": 144528031.0,
      "step": 83325
    },
    {
      "entropy": 5.388996505737305,
      "epoch": 6.4833300914219025,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017562261183757077,
      "loss": 4.6697,
      "mean_token_accuracy": 0.24037218242883682,
      "num_tokens": 144536971.0,
      "step": 83330
    },
    {
      "entropy": 5.329874706268311,
      "epoch": 6.48371912079362,
      "grad_norm": 1.46875,
      "learning_rate": 0.00017559774688706076,
      "loss": 4.5203,
      "mean_token_accuracy": 0.26216619163751603,
      "num_tokens": 144545407.0,
      "step": 83335
    },
    {
      "entropy": 5.401223945617676,
      "epoch": 6.484108150165337,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001755728834448218,
      "loss": 4.628,
      "mean_token_accuracy": 0.24279364794492722,
      "num_tokens": 144554025.0,
      "step": 83340
    },
    {
      "entropy": 5.376631879806519,
      "epoch": 6.484497179537055,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000175548021511231,
      "loss": 4.6482,
      "mean_token_accuracy": 0.24557300060987472,
      "num_tokens": 144562529.0,
      "step": 83345
    },
    {
      "entropy": 5.336154317855835,
      "epoch": 6.484886208908772,
      "grad_norm": 1.21875,
      "learning_rate": 0.00017552316108666584,
      "loss": 4.6549,
      "mean_token_accuracy": 0.23849573284387587,
      "num_tokens": 144572085.0,
      "step": 83350
    },
    {
      "entropy": 5.369937705993652,
      "epoch": 6.48527523828049,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017549830217150327,
      "loss": 4.6377,
      "mean_token_accuracy": 0.24205496907234192,
      "num_tokens": 144581187.0,
      "step": 83355
    },
    {
      "entropy": 5.4559544086456295,
      "epoch": 6.485664267652208,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017547344476612054,
      "loss": 4.7003,
      "mean_token_accuracy": 0.23278297185897828,
      "num_tokens": 144589266.0,
      "step": 83360
    },
    {
      "entropy": 5.400863218307495,
      "epoch": 6.4860532970239255,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017544858887089487,
      "loss": 4.757,
      "mean_token_accuracy": 0.2346418097615242,
      "num_tokens": 144598662.0,
      "step": 83365
    },
    {
      "entropy": 5.322594690322876,
      "epoch": 6.486442326395643,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001754237344862034,
      "loss": 4.6297,
      "mean_token_accuracy": 0.23753974437713624,
      "num_tokens": 144607721.0,
      "step": 83370
    },
    {
      "entropy": 5.421631097793579,
      "epoch": 6.48683135576736,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001753988816124234,
      "loss": 4.7043,
      "mean_token_accuracy": 0.2296262189745903,
      "num_tokens": 144615860.0,
      "step": 83375
    },
    {
      "entropy": 5.393128395080566,
      "epoch": 6.487220385139078,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017537403024993182,
      "loss": 4.6045,
      "mean_token_accuracy": 0.24792427867650985,
      "num_tokens": 144624243.0,
      "step": 83380
    },
    {
      "entropy": 5.413813066482544,
      "epoch": 6.487609414510795,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017534918039910564,
      "loss": 4.6419,
      "mean_token_accuracy": 0.23878444135189056,
      "num_tokens": 144632924.0,
      "step": 83385
    },
    {
      "entropy": 5.389645576477051,
      "epoch": 6.487998443882513,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017532433206032217,
      "loss": 4.693,
      "mean_token_accuracy": 0.2284933552145958,
      "num_tokens": 144641664.0,
      "step": 83390
    },
    {
      "entropy": 5.260879325866699,
      "epoch": 6.488387473254231,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017529948523395817,
      "loss": 4.6019,
      "mean_token_accuracy": 0.24544305801391603,
      "num_tokens": 144650376.0,
      "step": 83395
    },
    {
      "entropy": 5.392789220809936,
      "epoch": 6.4887765026259485,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017527463992039093,
      "loss": 4.7005,
      "mean_token_accuracy": 0.2376922160387039,
      "num_tokens": 144659295.0,
      "step": 83400
    },
    {
      "entropy": 5.3748095512390135,
      "epoch": 6.489165531997666,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001752497961199972,
      "loss": 4.586,
      "mean_token_accuracy": 0.24188725501298905,
      "num_tokens": 144667291.0,
      "step": 83405
    },
    {
      "entropy": 5.450176525115967,
      "epoch": 6.489554561369383,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00017522495383315412,
      "loss": 4.7079,
      "mean_token_accuracy": 0.22790154218673705,
      "num_tokens": 144676178.0,
      "step": 83410
    },
    {
      "entropy": 5.382223892211914,
      "epoch": 6.489943590741101,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017520011306023847,
      "loss": 4.6562,
      "mean_token_accuracy": 0.24365486055612565,
      "num_tokens": 144685099.0,
      "step": 83415
    },
    {
      "entropy": 5.477406454086304,
      "epoch": 6.490332620112818,
      "grad_norm": 1.25,
      "learning_rate": 0.0001751752738016274,
      "loss": 4.7046,
      "mean_token_accuracy": 0.23478867560625077,
      "num_tokens": 144694159.0,
      "step": 83420
    },
    {
      "entropy": 5.400271987915039,
      "epoch": 6.490721649484536,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001751504360576975,
      "loss": 4.6641,
      "mean_token_accuracy": 0.2375238299369812,
      "num_tokens": 144702377.0,
      "step": 83425
    },
    {
      "entropy": 5.41828670501709,
      "epoch": 6.491110678856254,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017512559982882592,
      "loss": 4.7073,
      "mean_token_accuracy": 0.2315768525004387,
      "num_tokens": 144711024.0,
      "step": 83430
    },
    {
      "entropy": 5.34035472869873,
      "epoch": 6.4914997082279715,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001751007651153893,
      "loss": 4.6499,
      "mean_token_accuracy": 0.23761221617460251,
      "num_tokens": 144719948.0,
      "step": 83435
    },
    {
      "entropy": 5.380842065811157,
      "epoch": 6.491888737599689,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017507593191776454,
      "loss": 4.6441,
      "mean_token_accuracy": 0.2363035723567009,
      "num_tokens": 144728303.0,
      "step": 83440
    },
    {
      "entropy": 5.354175472259522,
      "epoch": 6.492277766971406,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017505110023632858,
      "loss": 4.6216,
      "mean_token_accuracy": 0.24454816728830336,
      "num_tokens": 144736614.0,
      "step": 83445
    },
    {
      "entropy": 5.395017528533936,
      "epoch": 6.492666796343124,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017502627007145804,
      "loss": 4.6354,
      "mean_token_accuracy": 0.24412447810173035,
      "num_tokens": 144745225.0,
      "step": 83450
    },
    {
      "entropy": 5.396701049804688,
      "epoch": 6.493055825714841,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017500144142352974,
      "loss": 4.7659,
      "mean_token_accuracy": 0.23419586718082427,
      "num_tokens": 144753691.0,
      "step": 83455
    },
    {
      "entropy": 5.37310152053833,
      "epoch": 6.493444855086559,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001749766142929205,
      "loss": 4.5724,
      "mean_token_accuracy": 0.24771341383457185,
      "num_tokens": 144761623.0,
      "step": 83460
    },
    {
      "entropy": 5.33272066116333,
      "epoch": 6.493833884458277,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017495178868000679,
      "loss": 4.5398,
      "mean_token_accuracy": 0.24511287957429886,
      "num_tokens": 144770119.0,
      "step": 83465
    },
    {
      "entropy": 5.4937763690948485,
      "epoch": 6.4942229138299945,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001749269645851655,
      "loss": 4.7742,
      "mean_token_accuracy": 0.23412951976060867,
      "num_tokens": 144778535.0,
      "step": 83470
    },
    {
      "entropy": 5.4095234870910645,
      "epoch": 6.494611943201711,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017490214200877323,
      "loss": 4.7165,
      "mean_token_accuracy": 0.23663477897644042,
      "num_tokens": 144787377.0,
      "step": 83475
    },
    {
      "entropy": 5.368360137939453,
      "epoch": 6.495000972573429,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017487732095120665,
      "loss": 4.6441,
      "mean_token_accuracy": 0.23645565509796143,
      "num_tokens": 144795627.0,
      "step": 83480
    },
    {
      "entropy": 5.365503311157227,
      "epoch": 6.495390001945147,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001748525014128423,
      "loss": 4.6106,
      "mean_token_accuracy": 0.23636668622493745,
      "num_tokens": 144804566.0,
      "step": 83485
    },
    {
      "entropy": 5.366048860549927,
      "epoch": 6.495779031316864,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001748276833940569,
      "loss": 4.6629,
      "mean_token_accuracy": 0.24267521351575852,
      "num_tokens": 144812947.0,
      "step": 83490
    },
    {
      "entropy": 5.296392440795898,
      "epoch": 6.496168060688582,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017480286689522693,
      "loss": 4.5359,
      "mean_token_accuracy": 0.24629462659358978,
      "num_tokens": 144820845.0,
      "step": 83495
    },
    {
      "entropy": 5.363618612289429,
      "epoch": 6.4965570900603,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017477805191672897,
      "loss": 4.6564,
      "mean_token_accuracy": 0.24040672332048416,
      "num_tokens": 144829796.0,
      "step": 83500
    },
    {
      "entropy": 5.350198602676391,
      "epoch": 6.4969461194320175,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001747532384589396,
      "loss": 4.6378,
      "mean_token_accuracy": 0.2524542972445488,
      "num_tokens": 144838603.0,
      "step": 83505
    },
    {
      "entropy": 5.407485103607177,
      "epoch": 6.497335148803734,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001747284265222352,
      "loss": 4.6566,
      "mean_token_accuracy": 0.23786742091178895,
      "num_tokens": 144846947.0,
      "step": 83510
    },
    {
      "entropy": 5.306180095672607,
      "epoch": 6.497724178175452,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00017470361610699242,
      "loss": 4.5784,
      "mean_token_accuracy": 0.24295345991849898,
      "num_tokens": 144856526.0,
      "step": 83515
    },
    {
      "entropy": 5.308623838424682,
      "epoch": 6.49811320754717,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017467880721358753,
      "loss": 4.5811,
      "mean_token_accuracy": 0.25063242465257646,
      "num_tokens": 144865108.0,
      "step": 83520
    },
    {
      "entropy": 5.372726345062256,
      "epoch": 6.498502236918887,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001746539998423972,
      "loss": 4.6847,
      "mean_token_accuracy": 0.23263390362262726,
      "num_tokens": 144874099.0,
      "step": 83525
    },
    {
      "entropy": 5.399352216720581,
      "epoch": 6.498891266290605,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017462919399379757,
      "loss": 4.6513,
      "mean_token_accuracy": 0.23591966181993484,
      "num_tokens": 144882622.0,
      "step": 83530
    },
    {
      "entropy": 5.397777080535889,
      "epoch": 6.499280295662323,
      "grad_norm": 1.375,
      "learning_rate": 0.00017460438966816525,
      "loss": 4.694,
      "mean_token_accuracy": 0.23270392268896103,
      "num_tokens": 144891143.0,
      "step": 83535
    },
    {
      "entropy": 5.336880302429199,
      "epoch": 6.49966932503404,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00017457958686587654,
      "loss": 4.5252,
      "mean_token_accuracy": 0.24792672246694564,
      "num_tokens": 144900487.0,
      "step": 83540
    },
    {
      "entropy": 5.3179912090301515,
      "epoch": 6.500058354405757,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00017455478558730769,
      "loss": 4.6041,
      "mean_token_accuracy": 0.24024751484394075,
      "num_tokens": 144909116.0,
      "step": 83545
    },
    {
      "entropy": 5.408647918701172,
      "epoch": 6.500447383777475,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00017452998583283513,
      "loss": 4.6972,
      "mean_token_accuracy": 0.23305072635412216,
      "num_tokens": 144918053.0,
      "step": 83550
    },
    {
      "entropy": 5.4168976783752445,
      "epoch": 6.500836413149193,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00017450518760283508,
      "loss": 4.7059,
      "mean_token_accuracy": 0.23590434491634368,
      "num_tokens": 144926801.0,
      "step": 83555
    },
    {
      "entropy": 5.3464583396911625,
      "epoch": 6.50122544252091,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017448039089768393,
      "loss": 4.5756,
      "mean_token_accuracy": 0.24794217944145203,
      "num_tokens": 144935373.0,
      "step": 83560
    },
    {
      "entropy": 5.380363130569458,
      "epoch": 6.501614471892628,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017445559571775766,
      "loss": 4.6318,
      "mean_token_accuracy": 0.2436601012945175,
      "num_tokens": 144943925.0,
      "step": 83565
    },
    {
      "entropy": 5.439556837081909,
      "epoch": 6.502003501264346,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001744308020634329,
      "loss": 4.6761,
      "mean_token_accuracy": 0.23909944891929627,
      "num_tokens": 144952186.0,
      "step": 83570
    },
    {
      "entropy": 5.378495550155639,
      "epoch": 6.5023925306360635,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017440600993508563,
      "loss": 4.58,
      "mean_token_accuracy": 0.24732628464698792,
      "num_tokens": 144961399.0,
      "step": 83575
    },
    {
      "entropy": 5.42419753074646,
      "epoch": 6.50278156000778,
      "grad_norm": 1.28125,
      "learning_rate": 0.00017438121933309198,
      "loss": 4.6952,
      "mean_token_accuracy": 0.2412168115377426,
      "num_tokens": 144969858.0,
      "step": 83580
    },
    {
      "entropy": 5.340136766433716,
      "epoch": 6.503170589379498,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017435643025782822,
      "loss": 4.5822,
      "mean_token_accuracy": 0.24464589953422547,
      "num_tokens": 144978261.0,
      "step": 83585
    },
    {
      "entropy": 5.400950622558594,
      "epoch": 6.503559618751216,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00017433164270967038,
      "loss": 4.6367,
      "mean_token_accuracy": 0.2401440218091011,
      "num_tokens": 144986373.0,
      "step": 83590
    },
    {
      "entropy": 5.408540058135986,
      "epoch": 6.503948648122933,
      "grad_norm": 1.59375,
      "learning_rate": 0.00017430685668899472,
      "loss": 4.6803,
      "mean_token_accuracy": 0.24499937742948533,
      "num_tokens": 144995046.0,
      "step": 83595
    },
    {
      "entropy": 5.379244565963745,
      "epoch": 6.504337677494651,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017428207219617715,
      "loss": 4.7014,
      "mean_token_accuracy": 0.23217418491840364,
      "num_tokens": 145003878.0,
      "step": 83600
    },
    {
      "entropy": 5.379294443130493,
      "epoch": 6.504726706866369,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017425728923159392,
      "loss": 4.6587,
      "mean_token_accuracy": 0.24162754714488982,
      "num_tokens": 145013195.0,
      "step": 83605
    },
    {
      "entropy": 5.327761316299439,
      "epoch": 6.505115736238086,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001742325077956209,
      "loss": 4.6443,
      "mean_token_accuracy": 0.2402042031288147,
      "num_tokens": 145021874.0,
      "step": 83610
    },
    {
      "entropy": 5.390652704238891,
      "epoch": 6.505504765609803,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017420772788863422,
      "loss": 4.6587,
      "mean_token_accuracy": 0.2386966183781624,
      "num_tokens": 145030055.0,
      "step": 83615
    },
    {
      "entropy": 5.366240406036377,
      "epoch": 6.505893794981521,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017418294951100988,
      "loss": 4.6501,
      "mean_token_accuracy": 0.2400822952389717,
      "num_tokens": 145039136.0,
      "step": 83620
    },
    {
      "entropy": 5.356420183181763,
      "epoch": 6.506282824353239,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017415817266312373,
      "loss": 4.6144,
      "mean_token_accuracy": 0.25107798874378207,
      "num_tokens": 145048021.0,
      "step": 83625
    },
    {
      "entropy": 5.350440168380738,
      "epoch": 6.506671853724956,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001741333973453518,
      "loss": 4.6291,
      "mean_token_accuracy": 0.24473216235637665,
      "num_tokens": 145056891.0,
      "step": 83630
    },
    {
      "entropy": 5.45037202835083,
      "epoch": 6.507060883096674,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017410862355807,
      "loss": 4.7281,
      "mean_token_accuracy": 0.23751418441534042,
      "num_tokens": 145064959.0,
      "step": 83635
    },
    {
      "entropy": 5.4057104110717775,
      "epoch": 6.507449912468392,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001740838513016544,
      "loss": 4.6734,
      "mean_token_accuracy": 0.23462863266468048,
      "num_tokens": 145073000.0,
      "step": 83640
    },
    {
      "entropy": 5.368985509872436,
      "epoch": 6.507838941840109,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017405908057648055,
      "loss": 4.5999,
      "mean_token_accuracy": 0.24118918925523758,
      "num_tokens": 145081984.0,
      "step": 83645
    },
    {
      "entropy": 5.403512907028198,
      "epoch": 6.508227971211826,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017403431138292465,
      "loss": 4.6662,
      "mean_token_accuracy": 0.24083335548639298,
      "num_tokens": 145091038.0,
      "step": 83650
    },
    {
      "entropy": 5.391356468200684,
      "epoch": 6.508617000583544,
      "grad_norm": 1.25,
      "learning_rate": 0.00017400954372136235,
      "loss": 4.7057,
      "mean_token_accuracy": 0.24033572673797607,
      "num_tokens": 145099829.0,
      "step": 83655
    },
    {
      "entropy": 5.3741413116455075,
      "epoch": 6.509006029955262,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017398477759216939,
      "loss": 4.5606,
      "mean_token_accuracy": 0.2511690020561218,
      "num_tokens": 145109039.0,
      "step": 83660
    },
    {
      "entropy": 5.413078641891479,
      "epoch": 6.509395059326979,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017396001299572174,
      "loss": 4.7016,
      "mean_token_accuracy": 0.23531747460365296,
      "num_tokens": 145117156.0,
      "step": 83665
    },
    {
      "entropy": 5.4354172706604,
      "epoch": 6.509784088698697,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017393524993239496,
      "loss": 4.705,
      "mean_token_accuracy": 0.2352612242102623,
      "num_tokens": 145124774.0,
      "step": 83670
    },
    {
      "entropy": 5.252783393859863,
      "epoch": 6.510173118070414,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00017391048840256502,
      "loss": 4.5743,
      "mean_token_accuracy": 0.2417900413274765,
      "num_tokens": 145133471.0,
      "step": 83675
    },
    {
      "entropy": 5.412714719772339,
      "epoch": 6.510562147442132,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017388572840660742,
      "loss": 4.7412,
      "mean_token_accuracy": 0.23891898840665818,
      "num_tokens": 145141839.0,
      "step": 83680
    },
    {
      "entropy": 5.414959192276001,
      "epoch": 6.510951176813849,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017386096994489802,
      "loss": 4.6893,
      "mean_token_accuracy": 0.2417949065566063,
      "num_tokens": 145150258.0,
      "step": 83685
    },
    {
      "entropy": 5.501168251037598,
      "epoch": 6.511340206185567,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001738362130178124,
      "loss": 4.8125,
      "mean_token_accuracy": 0.22415529787540436,
      "num_tokens": 145159325.0,
      "step": 83690
    },
    {
      "entropy": 5.4057403087615965,
      "epoch": 6.511729235557285,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017381145762572622,
      "loss": 4.655,
      "mean_token_accuracy": 0.23063469678163528,
      "num_tokens": 145168223.0,
      "step": 83695
    },
    {
      "entropy": 5.403976106643677,
      "epoch": 6.512118264929002,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017378670376901506,
      "loss": 4.6322,
      "mean_token_accuracy": 0.2381923660635948,
      "num_tokens": 145177510.0,
      "step": 83700
    },
    {
      "entropy": 5.295703601837158,
      "epoch": 6.51250729430072,
      "grad_norm": 1.28125,
      "learning_rate": 0.00017376195144805463,
      "loss": 4.4998,
      "mean_token_accuracy": 0.24944690763950347,
      "num_tokens": 145186323.0,
      "step": 83705
    },
    {
      "entropy": 5.3563133716583256,
      "epoch": 6.512896323672438,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017373720066322048,
      "loss": 4.6157,
      "mean_token_accuracy": 0.24090910255908965,
      "num_tokens": 145195363.0,
      "step": 83710
    },
    {
      "entropy": 5.381863451004028,
      "epoch": 6.513285353044155,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017371245141488805,
      "loss": 4.6385,
      "mean_token_accuracy": 0.2459246277809143,
      "num_tokens": 145204179.0,
      "step": 83715
    },
    {
      "entropy": 5.439093589782715,
      "epoch": 6.513674382415872,
      "grad_norm": 1.25,
      "learning_rate": 0.00017368770370343307,
      "loss": 4.7523,
      "mean_token_accuracy": 0.23796629309654235,
      "num_tokens": 145213116.0,
      "step": 83720
    },
    {
      "entropy": 5.428942537307739,
      "epoch": 6.51406341178759,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017366295752923078,
      "loss": 4.654,
      "mean_token_accuracy": 0.23647023290395736,
      "num_tokens": 145221626.0,
      "step": 83725
    },
    {
      "entropy": 5.394447565078735,
      "epoch": 6.514452441159308,
      "grad_norm": 1.234375,
      "learning_rate": 0.000173638212892657,
      "loss": 4.6136,
      "mean_token_accuracy": 0.2380926489830017,
      "num_tokens": 145230243.0,
      "step": 83730
    },
    {
      "entropy": 5.3869202613830565,
      "epoch": 6.514841470531025,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017361346979408694,
      "loss": 4.6477,
      "mean_token_accuracy": 0.23749045580625533,
      "num_tokens": 145238314.0,
      "step": 83735
    },
    {
      "entropy": 5.376548910140992,
      "epoch": 6.515230499902742,
      "grad_norm": 1.25,
      "learning_rate": 0.000173588728233896,
      "loss": 4.6322,
      "mean_token_accuracy": 0.24324368238449096,
      "num_tokens": 145246867.0,
      "step": 83740
    },
    {
      "entropy": 5.363338565826416,
      "epoch": 6.51561952927446,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017356398821245984,
      "loss": 4.6949,
      "mean_token_accuracy": 0.23921376913785936,
      "num_tokens": 145255436.0,
      "step": 83745
    },
    {
      "entropy": 5.367169046401978,
      "epoch": 6.516008558646178,
      "grad_norm": 1.28125,
      "learning_rate": 0.00017353924973015358,
      "loss": 4.6783,
      "mean_token_accuracy": 0.2320227265357971,
      "num_tokens": 145264089.0,
      "step": 83750
    },
    {
      "entropy": 5.394724702835083,
      "epoch": 6.516397588017895,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001735145127873528,
      "loss": 4.6668,
      "mean_token_accuracy": 0.23478420376777648,
      "num_tokens": 145273119.0,
      "step": 83755
    },
    {
      "entropy": 5.321128940582275,
      "epoch": 6.516786617389613,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017348977738443266,
      "loss": 4.5583,
      "mean_token_accuracy": 0.24828683733940124,
      "num_tokens": 145281254.0,
      "step": 83760
    },
    {
      "entropy": 5.455593681335449,
      "epoch": 6.517175646761331,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017346504352176863,
      "loss": 4.7276,
      "mean_token_accuracy": 0.2350144475698471,
      "num_tokens": 145289839.0,
      "step": 83765
    },
    {
      "entropy": 5.493970632553101,
      "epoch": 6.517564676133048,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017344031119973603,
      "loss": 4.7783,
      "mean_token_accuracy": 0.2354137971997261,
      "num_tokens": 145298517.0,
      "step": 83770
    },
    {
      "entropy": 5.404459428787232,
      "epoch": 6.517953705504766,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017341558041870995,
      "loss": 4.692,
      "mean_token_accuracy": 0.23664177656173707,
      "num_tokens": 145307390.0,
      "step": 83775
    },
    {
      "entropy": 5.368070983886719,
      "epoch": 6.518342734876483,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00017339085117906583,
      "loss": 4.6166,
      "mean_token_accuracy": 0.24402675926685333,
      "num_tokens": 145316555.0,
      "step": 83780
    },
    {
      "entropy": 5.359530353546143,
      "epoch": 6.5187317642482006,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017336612348117875,
      "loss": 4.7184,
      "mean_token_accuracy": 0.24068852216005326,
      "num_tokens": 145325337.0,
      "step": 83785
    },
    {
      "entropy": 5.41115026473999,
      "epoch": 6.519120793619918,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017334139732542403,
      "loss": 4.6374,
      "mean_token_accuracy": 0.24156950265169144,
      "num_tokens": 145334048.0,
      "step": 83790
    },
    {
      "entropy": 5.376961708068848,
      "epoch": 6.519509822991636,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017331667271217675,
      "loss": 4.6782,
      "mean_token_accuracy": 0.2332757368683815,
      "num_tokens": 145343200.0,
      "step": 83795
    },
    {
      "entropy": 5.51034369468689,
      "epoch": 6.519898852363354,
      "grad_norm": 1.25,
      "learning_rate": 0.00017329194964181219,
      "loss": 4.7767,
      "mean_token_accuracy": 0.2358574777841568,
      "num_tokens": 145351656.0,
      "step": 83800
    },
    {
      "entropy": 5.375264120101929,
      "epoch": 6.520287881735071,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017326722811470533,
      "loss": 4.6496,
      "mean_token_accuracy": 0.24498433172702788,
      "num_tokens": 145360381.0,
      "step": 83805
    },
    {
      "entropy": 5.392841482162476,
      "epoch": 6.520676911106788,
      "grad_norm": 1.515625,
      "learning_rate": 0.00017324250813123145,
      "loss": 4.6598,
      "mean_token_accuracy": 0.2369341015815735,
      "num_tokens": 145368889.0,
      "step": 83810
    },
    {
      "entropy": 5.302327442169189,
      "epoch": 6.521065940478506,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001732177896917655,
      "loss": 4.6072,
      "mean_token_accuracy": 0.23790445178747177,
      "num_tokens": 145377921.0,
      "step": 83815
    },
    {
      "entropy": 5.331922483444214,
      "epoch": 6.5214549698502235,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017319307279668256,
      "loss": 4.6125,
      "mean_token_accuracy": 0.24864569902420045,
      "num_tokens": 145387162.0,
      "step": 83820
    },
    {
      "entropy": 5.304832887649536,
      "epoch": 6.521843999221941,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017316835744635777,
      "loss": 4.5387,
      "mean_token_accuracy": 0.24892952740192414,
      "num_tokens": 145395665.0,
      "step": 83825
    },
    {
      "entropy": 5.39853687286377,
      "epoch": 6.522233028593659,
      "grad_norm": 1.203125,
      "learning_rate": 0.00017314364364116587,
      "loss": 4.6491,
      "mean_token_accuracy": 0.24374651461839675,
      "num_tokens": 145404577.0,
      "step": 83830
    },
    {
      "entropy": 5.378909730911255,
      "epoch": 6.522622057965377,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017311893138148227,
      "loss": 4.6161,
      "mean_token_accuracy": 0.2442230299115181,
      "num_tokens": 145413042.0,
      "step": 83835
    },
    {
      "entropy": 5.439112901687622,
      "epoch": 6.523011087337094,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001730942206676817,
      "loss": 4.7345,
      "mean_token_accuracy": 0.23397388756275178,
      "num_tokens": 145421261.0,
      "step": 83840
    },
    {
      "entropy": 5.284405136108399,
      "epoch": 6.523400116708811,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017306951150013904,
      "loss": 4.5188,
      "mean_token_accuracy": 0.24985456317663193,
      "num_tokens": 145430120.0,
      "step": 83845
    },
    {
      "entropy": 5.3332709789276125,
      "epoch": 6.523789146080529,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017304480387922945,
      "loss": 4.6284,
      "mean_token_accuracy": 0.24427114874124528,
      "num_tokens": 145439210.0,
      "step": 83850
    },
    {
      "entropy": 5.34494252204895,
      "epoch": 6.5241781754522465,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00017302009780532752,
      "loss": 4.5656,
      "mean_token_accuracy": 0.2410481885075569,
      "num_tokens": 145447604.0,
      "step": 83855
    },
    {
      "entropy": 5.352521753311157,
      "epoch": 6.524567204823964,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001729953932788084,
      "loss": 4.5831,
      "mean_token_accuracy": 0.2410559758543968,
      "num_tokens": 145455405.0,
      "step": 83860
    },
    {
      "entropy": 5.34260745048523,
      "epoch": 6.524956234195682,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017297069030004672,
      "loss": 4.6003,
      "mean_token_accuracy": 0.24615536034107208,
      "num_tokens": 145463812.0,
      "step": 83865
    },
    {
      "entropy": 5.367448091506958,
      "epoch": 6.5253452635674,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017294598886941754,
      "loss": 4.6749,
      "mean_token_accuracy": 0.24046464264392853,
      "num_tokens": 145473022.0,
      "step": 83870
    },
    {
      "entropy": 5.292556571960449,
      "epoch": 6.525734292939116,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017292128898729544,
      "loss": 4.5794,
      "mean_token_accuracy": 0.24306445717811584,
      "num_tokens": 145481730.0,
      "step": 83875
    },
    {
      "entropy": 5.306625318527222,
      "epoch": 6.526123322310834,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017289659065405542,
      "loss": 4.5989,
      "mean_token_accuracy": 0.2454576388001442,
      "num_tokens": 145490865.0,
      "step": 83880
    },
    {
      "entropy": 5.321991109848023,
      "epoch": 6.526512351682552,
      "grad_norm": 1.15625,
      "learning_rate": 0.00017287189387007201,
      "loss": 4.6169,
      "mean_token_accuracy": 0.2442027822136879,
      "num_tokens": 145500449.0,
      "step": 83885
    },
    {
      "entropy": 5.408231210708618,
      "epoch": 6.5269013810542695,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017284719863572013,
      "loss": 4.6203,
      "mean_token_accuracy": 0.24066005945205687,
      "num_tokens": 145508909.0,
      "step": 83890
    },
    {
      "entropy": 5.418175458908081,
      "epoch": 6.527290410425987,
      "grad_norm": 1.234375,
      "learning_rate": 0.00017282250495137444,
      "loss": 4.6675,
      "mean_token_accuracy": 0.23974123001098632,
      "num_tokens": 145517382.0,
      "step": 83895
    },
    {
      "entropy": 5.275529909133911,
      "epoch": 6.527679439797705,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017279781281740944,
      "loss": 4.5843,
      "mean_token_accuracy": 0.25030107200145724,
      "num_tokens": 145527207.0,
      "step": 83900
    },
    {
      "entropy": 5.301412963867188,
      "epoch": 6.528068469169423,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017277312223420012,
      "loss": 4.6054,
      "mean_token_accuracy": 0.23936323523521424,
      "num_tokens": 145535833.0,
      "step": 83905
    },
    {
      "entropy": 5.383844995498658,
      "epoch": 6.52845749854114,
      "grad_norm": 1.28125,
      "learning_rate": 0.00017274843320212087,
      "loss": 4.757,
      "mean_token_accuracy": 0.23228081315755844,
      "num_tokens": 145544168.0,
      "step": 83910
    },
    {
      "entropy": 5.4047401428222654,
      "epoch": 6.528846527912857,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001727237457215465,
      "loss": 4.6647,
      "mean_token_accuracy": 0.236198353767395,
      "num_tokens": 145552382.0,
      "step": 83915
    },
    {
      "entropy": 5.414656019210815,
      "epoch": 6.529235557284575,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00017269905979285148,
      "loss": 4.6896,
      "mean_token_accuracy": 0.23490134179592131,
      "num_tokens": 145561647.0,
      "step": 83920
    },
    {
      "entropy": 5.3888702392578125,
      "epoch": 6.5296245866562925,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017267437541641033,
      "loss": 4.5747,
      "mean_token_accuracy": 0.25093753039836886,
      "num_tokens": 145569687.0,
      "step": 83925
    },
    {
      "entropy": 5.444630861282349,
      "epoch": 6.53001361602801,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017264969259259773,
      "loss": 4.6901,
      "mean_token_accuracy": 0.23507392257452012,
      "num_tokens": 145578056.0,
      "step": 83930
    },
    {
      "entropy": 5.3281152725219725,
      "epoch": 6.530402645399728,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00017262501132178805,
      "loss": 4.606,
      "mean_token_accuracy": 0.24108273088932036,
      "num_tokens": 145586454.0,
      "step": 83935
    },
    {
      "entropy": 5.308131694793701,
      "epoch": 6.530791674771446,
      "grad_norm": 1.28125,
      "learning_rate": 0.00017260033160435595,
      "loss": 4.5357,
      "mean_token_accuracy": 0.2548290640115738,
      "num_tokens": 145594514.0,
      "step": 83940
    },
    {
      "entropy": 5.342011833190918,
      "epoch": 6.531180704143162,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017257565344067576,
      "loss": 4.6439,
      "mean_token_accuracy": 0.24256763309240342,
      "num_tokens": 145602834.0,
      "step": 83945
    },
    {
      "entropy": 5.354339838027954,
      "epoch": 6.53156973351488,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00017255097683112214,
      "loss": 4.6503,
      "mean_token_accuracy": 0.2321965605020523,
      "num_tokens": 145611353.0,
      "step": 83950
    },
    {
      "entropy": 5.443689060211182,
      "epoch": 6.531958762886598,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017252630177606924,
      "loss": 4.6794,
      "mean_token_accuracy": 0.23898309767246245,
      "num_tokens": 145619896.0,
      "step": 83955
    },
    {
      "entropy": 5.363357973098755,
      "epoch": 6.5323477922583155,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017250162827589171,
      "loss": 4.6155,
      "mean_token_accuracy": 0.24339344948530198,
      "num_tokens": 145628214.0,
      "step": 83960
    },
    {
      "entropy": 5.3466917991638185,
      "epoch": 6.532736821630033,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001724769563309637,
      "loss": 4.6032,
      "mean_token_accuracy": 0.23967114388942717,
      "num_tokens": 145637528.0,
      "step": 83965
    },
    {
      "entropy": 5.32107720375061,
      "epoch": 6.533125851001751,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017245228594165978,
      "loss": 4.6097,
      "mean_token_accuracy": 0.23968523889780044,
      "num_tokens": 145646259.0,
      "step": 83970
    },
    {
      "entropy": 5.424398708343506,
      "epoch": 6.533514880373469,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017242761710835425,
      "loss": 4.6838,
      "mean_token_accuracy": 0.24035162031650542,
      "num_tokens": 145655256.0,
      "step": 83975
    },
    {
      "entropy": 5.427449798583984,
      "epoch": 6.533903909745185,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00017240294983142125,
      "loss": 4.6792,
      "mean_token_accuracy": 0.23770765960216522,
      "num_tokens": 145664261.0,
      "step": 83980
    },
    {
      "entropy": 5.4089349746704105,
      "epoch": 6.534292939116903,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001723782841112353,
      "loss": 4.6323,
      "mean_token_accuracy": 0.23705004751682282,
      "num_tokens": 145672036.0,
      "step": 83985
    },
    {
      "entropy": 5.381989479064941,
      "epoch": 6.534681968488621,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017235361994817044,
      "loss": 4.6581,
      "mean_token_accuracy": 0.24577950537204743,
      "num_tokens": 145680953.0,
      "step": 83990
    },
    {
      "entropy": 5.410299491882324,
      "epoch": 6.5350709978603385,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017232895734260113,
      "loss": 4.6641,
      "mean_token_accuracy": 0.24226383119821548,
      "num_tokens": 145690121.0,
      "step": 83995
    },
    {
      "entropy": 5.369603490829467,
      "epoch": 6.535460027232056,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017230429629490153,
      "loss": 4.6229,
      "mean_token_accuracy": 0.23694472908973693,
      "num_tokens": 145698350.0,
      "step": 84000
    },
    {
      "epoch": 6.535460027232056,
      "eval_entropy": 5.121053772826282,
      "eval_loss": 4.862063407897949,
      "eval_mean_token_accuracy": 0.23522837958132797,
      "eval_num_tokens": 145698350.0,
      "eval_runtime": 28.7926,
      "eval_samples_per_second": 1443.358,
      "eval_steps_per_second": 180.428,
      "step": 84000
    },
    {
      "entropy": 5.393476676940918,
      "epoch": 6.535849056603774,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017227963680544562,
      "loss": 4.6556,
      "mean_token_accuracy": 0.24038407504558562,
      "num_tokens": 145706525.0,
      "step": 84005
    },
    {
      "entropy": 5.381467628479004,
      "epoch": 6.536238085975491,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017225497887460778,
      "loss": 4.6988,
      "mean_token_accuracy": 0.2302231878042221,
      "num_tokens": 145715492.0,
      "step": 84010
    },
    {
      "entropy": 5.338558626174927,
      "epoch": 6.536627115347208,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00017223032250276204,
      "loss": 4.5774,
      "mean_token_accuracy": 0.24618176519870758,
      "num_tokens": 145725016.0,
      "step": 84015
    },
    {
      "entropy": 5.390619897842408,
      "epoch": 6.537016144718926,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001722056676902827,
      "loss": 4.6548,
      "mean_token_accuracy": 0.2448733702301979,
      "num_tokens": 145734390.0,
      "step": 84020
    },
    {
      "entropy": 5.417832374572754,
      "epoch": 6.537405174090644,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017218101443754365,
      "loss": 4.6744,
      "mean_token_accuracy": 0.23179246336221696,
      "num_tokens": 145742937.0,
      "step": 84025
    },
    {
      "entropy": 5.413867378234864,
      "epoch": 6.5377942034623615,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017215636274491903,
      "loss": 4.6641,
      "mean_token_accuracy": 0.24020789563655853,
      "num_tokens": 145752324.0,
      "step": 84030
    },
    {
      "entropy": 5.4198668003082275,
      "epoch": 6.538183232834079,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017213171261278297,
      "loss": 4.6586,
      "mean_token_accuracy": 0.23608543276786803,
      "num_tokens": 145760867.0,
      "step": 84035
    },
    {
      "entropy": 5.434342956542968,
      "epoch": 6.538572262205797,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001721070640415094,
      "loss": 4.7072,
      "mean_token_accuracy": 0.2364867314696312,
      "num_tokens": 145770271.0,
      "step": 84040
    },
    {
      "entropy": 5.448533916473389,
      "epoch": 6.538961291577515,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017208241703147248,
      "loss": 4.7627,
      "mean_token_accuracy": 0.22755660861730576,
      "num_tokens": 145778966.0,
      "step": 84045
    },
    {
      "entropy": 5.359823751449585,
      "epoch": 6.539350320949231,
      "grad_norm": 1.28125,
      "learning_rate": 0.00017205777158304592,
      "loss": 4.6081,
      "mean_token_accuracy": 0.24489447027444838,
      "num_tokens": 145787242.0,
      "step": 84050
    },
    {
      "entropy": 5.392895936965942,
      "epoch": 6.539739350320949,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017203312769660395,
      "loss": 4.6567,
      "mean_token_accuracy": 0.23839914351701735,
      "num_tokens": 145795574.0,
      "step": 84055
    },
    {
      "entropy": 5.342375946044922,
      "epoch": 6.540128379692667,
      "grad_norm": 1.375,
      "learning_rate": 0.00017200848537252028,
      "loss": 4.5905,
      "mean_token_accuracy": 0.24331727176904677,
      "num_tokens": 145804478.0,
      "step": 84060
    },
    {
      "entropy": 5.348700952529907,
      "epoch": 6.5405174090643845,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000171983844611169,
      "loss": 4.5941,
      "mean_token_accuracy": 0.24122008234262465,
      "num_tokens": 145813002.0,
      "step": 84065
    },
    {
      "entropy": 5.433027124404907,
      "epoch": 6.540906438436102,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017195920541292382,
      "loss": 4.6644,
      "mean_token_accuracy": 0.24166579991579057,
      "num_tokens": 145822095.0,
      "step": 84070
    },
    {
      "entropy": 5.438534164428711,
      "epoch": 6.541295467807819,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017193456777815877,
      "loss": 4.6358,
      "mean_token_accuracy": 0.24057725667953492,
      "num_tokens": 145831731.0,
      "step": 84075
    },
    {
      "entropy": 5.452589654922486,
      "epoch": 6.541684497179537,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00017190993170724766,
      "loss": 4.7284,
      "mean_token_accuracy": 0.23002278208732604,
      "num_tokens": 145840464.0,
      "step": 84080
    },
    {
      "entropy": 5.417345714569092,
      "epoch": 6.542073526551254,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001718852972005641,
      "loss": 4.6294,
      "mean_token_accuracy": 0.24934646785259246,
      "num_tokens": 145849294.0,
      "step": 84085
    },
    {
      "entropy": 5.422814512252808,
      "epoch": 6.542462555922972,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017186066425848215,
      "loss": 4.6893,
      "mean_token_accuracy": 0.23790296018123627,
      "num_tokens": 145858055.0,
      "step": 84090
    },
    {
      "entropy": 5.304858636856079,
      "epoch": 6.54285158529469,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00017183603288137527,
      "loss": 4.6146,
      "mean_token_accuracy": 0.24345111101865768,
      "num_tokens": 145867576.0,
      "step": 84095
    },
    {
      "entropy": 5.345287847518921,
      "epoch": 6.5432406146664075,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001718114030696176,
      "loss": 4.6097,
      "mean_token_accuracy": 0.24089200645685196,
      "num_tokens": 145876564.0,
      "step": 84100
    },
    {
      "entropy": 5.42977294921875,
      "epoch": 6.543629644038125,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001717867748235825,
      "loss": 4.7495,
      "mean_token_accuracy": 0.22662174105644226,
      "num_tokens": 145885536.0,
      "step": 84105
    },
    {
      "entropy": 5.338153886795044,
      "epoch": 6.544018673409843,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017176214814364394,
      "loss": 4.5238,
      "mean_token_accuracy": 0.2548313319683075,
      "num_tokens": 145893234.0,
      "step": 84110
    },
    {
      "entropy": 5.364932584762573,
      "epoch": 6.54440770278156,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017173752303017542,
      "loss": 4.6572,
      "mean_token_accuracy": 0.24212122857570648,
      "num_tokens": 145901942.0,
      "step": 84115
    },
    {
      "entropy": 5.328374004364013,
      "epoch": 6.544796732153277,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001717128994835506,
      "loss": 4.5475,
      "mean_token_accuracy": 0.2529362618923187,
      "num_tokens": 145910342.0,
      "step": 84120
    },
    {
      "entropy": 5.346514654159546,
      "epoch": 6.545185761524995,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017168827750414317,
      "loss": 4.6363,
      "mean_token_accuracy": 0.24813128411769866,
      "num_tokens": 145919059.0,
      "step": 84125
    },
    {
      "entropy": 5.345714569091797,
      "epoch": 6.545574790896713,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001716636570923266,
      "loss": 4.5656,
      "mean_token_accuracy": 0.2530625581741333,
      "num_tokens": 145927396.0,
      "step": 84130
    },
    {
      "entropy": 5.325767660140992,
      "epoch": 6.5459638202684305,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00017163903824847471,
      "loss": 4.6023,
      "mean_token_accuracy": 0.25125242620706556,
      "num_tokens": 145936262.0,
      "step": 84135
    },
    {
      "entropy": 5.339072895050049,
      "epoch": 6.546352849640148,
      "grad_norm": 1.234375,
      "learning_rate": 0.00017161442097296076,
      "loss": 4.6205,
      "mean_token_accuracy": 0.2428480327129364,
      "num_tokens": 145946328.0,
      "step": 84140
    },
    {
      "entropy": 5.297600889205933,
      "epoch": 6.546741879011865,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017158980526615854,
      "loss": 4.5646,
      "mean_token_accuracy": 0.24329643100500106,
      "num_tokens": 145954537.0,
      "step": 84145
    },
    {
      "entropy": 5.4114377975463865,
      "epoch": 6.547130908383583,
      "grad_norm": 1.375,
      "learning_rate": 0.00017156519112844132,
      "loss": 4.6566,
      "mean_token_accuracy": 0.24016353189945222,
      "num_tokens": 145963379.0,
      "step": 84150
    },
    {
      "entropy": 5.325254917144775,
      "epoch": 6.5475199377553,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00017154057856018275,
      "loss": 4.6352,
      "mean_token_accuracy": 0.2456674799323082,
      "num_tokens": 145971808.0,
      "step": 84155
    },
    {
      "entropy": 5.357063484191895,
      "epoch": 6.547908967127018,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001715159675617563,
      "loss": 4.6767,
      "mean_token_accuracy": 0.2364422485232353,
      "num_tokens": 145981089.0,
      "step": 84160
    },
    {
      "entropy": 5.347199535369873,
      "epoch": 6.548297996498736,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017149135813353508,
      "loss": 4.6427,
      "mean_token_accuracy": 0.23932726979255675,
      "num_tokens": 145989810.0,
      "step": 84165
    },
    {
      "entropy": 5.321401977539063,
      "epoch": 6.5486870258704535,
      "grad_norm": 1.28125,
      "learning_rate": 0.00017146675027589296,
      "loss": 4.5695,
      "mean_token_accuracy": 0.24530381113290786,
      "num_tokens": 145997771.0,
      "step": 84170
    },
    {
      "entropy": 5.336527919769287,
      "epoch": 6.549076055242171,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017144214398920302,
      "loss": 4.5737,
      "mean_token_accuracy": 0.2472361296415329,
      "num_tokens": 146006266.0,
      "step": 84175
    },
    {
      "entropy": 5.359685182571411,
      "epoch": 6.549465084613888,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001714175392738388,
      "loss": 4.6665,
      "mean_token_accuracy": 0.2403781995177269,
      "num_tokens": 146016097.0,
      "step": 84180
    },
    {
      "entropy": 5.318788671493531,
      "epoch": 6.549854113985606,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001713929361301735,
      "loss": 4.5898,
      "mean_token_accuracy": 0.24367111176252365,
      "num_tokens": 146024377.0,
      "step": 84185
    },
    {
      "entropy": 5.412964868545532,
      "epoch": 6.550243143357323,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017136833455858052,
      "loss": 4.7014,
      "mean_token_accuracy": 0.23616087734699248,
      "num_tokens": 146033105.0,
      "step": 84190
    },
    {
      "entropy": 5.3331825733184814,
      "epoch": 6.550632172729041,
      "grad_norm": 1.28125,
      "learning_rate": 0.00017134373455943315,
      "loss": 4.6121,
      "mean_token_accuracy": 0.2447109505534172,
      "num_tokens": 146041973.0,
      "step": 84195
    },
    {
      "entropy": 5.391699552536011,
      "epoch": 6.551021202100759,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001713191361331045,
      "loss": 4.6927,
      "mean_token_accuracy": 0.23643915355205536,
      "num_tokens": 146051288.0,
      "step": 84200
    },
    {
      "entropy": 5.444328641891479,
      "epoch": 6.5514102314724765,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017129453927996804,
      "loss": 4.7213,
      "mean_token_accuracy": 0.23552737534046173,
      "num_tokens": 146060384.0,
      "step": 84205
    },
    {
      "entropy": 5.342244005203247,
      "epoch": 6.551799260844193,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001712699440003968,
      "loss": 4.56,
      "mean_token_accuracy": 0.24780899286270142,
      "num_tokens": 146069308.0,
      "step": 84210
    },
    {
      "entropy": 5.35835075378418,
      "epoch": 6.552188290215911,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001712453502947642,
      "loss": 4.7067,
      "mean_token_accuracy": 0.23354239910840988,
      "num_tokens": 146077828.0,
      "step": 84215
    },
    {
      "entropy": 5.427068614959717,
      "epoch": 6.552577319587629,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017122075816344314,
      "loss": 4.7367,
      "mean_token_accuracy": 0.22425749003887177,
      "num_tokens": 146086835.0,
      "step": 84220
    },
    {
      "entropy": 5.418323230743408,
      "epoch": 6.552966348959346,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017119616760680688,
      "loss": 4.6803,
      "mean_token_accuracy": 0.24538735300302505,
      "num_tokens": 146095405.0,
      "step": 84225
    },
    {
      "entropy": 5.401887702941894,
      "epoch": 6.553355378331064,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001711715786252286,
      "loss": 4.6136,
      "mean_token_accuracy": 0.24039352536201478,
      "num_tokens": 146104656.0,
      "step": 84230
    },
    {
      "entropy": 5.43229398727417,
      "epoch": 6.553744407702782,
      "grad_norm": 1.453125,
      "learning_rate": 0.00017114699121908132,
      "loss": 4.6263,
      "mean_token_accuracy": 0.24385692179203033,
      "num_tokens": 146113128.0,
      "step": 84235
    },
    {
      "entropy": 5.331828165054321,
      "epoch": 6.5541334370744995,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001711224053887382,
      "loss": 4.6006,
      "mean_token_accuracy": 0.2475523218512535,
      "num_tokens": 146122667.0,
      "step": 84240
    },
    {
      "entropy": 5.393920135498047,
      "epoch": 6.554522466446217,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017109782113457217,
      "loss": 4.6398,
      "mean_token_accuracy": 0.23725885152816772,
      "num_tokens": 146130931.0,
      "step": 84245
    },
    {
      "entropy": 5.327795553207397,
      "epoch": 6.554911495817934,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017107323845695643,
      "loss": 4.5939,
      "mean_token_accuracy": 0.24004062861204148,
      "num_tokens": 146140271.0,
      "step": 84250
    },
    {
      "entropy": 5.37441987991333,
      "epoch": 6.555300525189652,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001710486573562638,
      "loss": 4.6341,
      "mean_token_accuracy": 0.2427230194211006,
      "num_tokens": 146149084.0,
      "step": 84255
    },
    {
      "entropy": 5.376166343688965,
      "epoch": 6.555689554561369,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017102407783286738,
      "loss": 4.638,
      "mean_token_accuracy": 0.237137046456337,
      "num_tokens": 146157836.0,
      "step": 84260
    },
    {
      "entropy": 5.3544927597045895,
      "epoch": 6.556078583933087,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017099949988714006,
      "loss": 4.6047,
      "mean_token_accuracy": 0.24785349220037461,
      "num_tokens": 146165922.0,
      "step": 84265
    },
    {
      "entropy": 5.310955810546875,
      "epoch": 6.556467613304805,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017097492351945482,
      "loss": 4.5598,
      "mean_token_accuracy": 0.25023695826530457,
      "num_tokens": 146174191.0,
      "step": 84270
    },
    {
      "entropy": 5.376591157913208,
      "epoch": 6.5568566426765225,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017095034873018458,
      "loss": 4.6828,
      "mean_token_accuracy": 0.25139724314212797,
      "num_tokens": 146182665.0,
      "step": 84275
    },
    {
      "entropy": 5.421943473815918,
      "epoch": 6.557245672048239,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001709257755197021,
      "loss": 4.6626,
      "mean_token_accuracy": 0.24305889159440994,
      "num_tokens": 146190848.0,
      "step": 84280
    },
    {
      "entropy": 5.417236185073852,
      "epoch": 6.557634701419957,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017090120388838043,
      "loss": 4.6724,
      "mean_token_accuracy": 0.24581272900104523,
      "num_tokens": 146199273.0,
      "step": 84285
    },
    {
      "entropy": 5.318183851242066,
      "epoch": 6.558023730791675,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017087663383659216,
      "loss": 4.506,
      "mean_token_accuracy": 0.24811684638261794,
      "num_tokens": 146207549.0,
      "step": 84290
    },
    {
      "entropy": 5.445098781585694,
      "epoch": 6.558412760163392,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017085206536471026,
      "loss": 4.6487,
      "mean_token_accuracy": 0.24554549008607865,
      "num_tokens": 146216643.0,
      "step": 84295
    },
    {
      "entropy": 5.410681056976318,
      "epoch": 6.55880178953511,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017082749847310757,
      "loss": 4.676,
      "mean_token_accuracy": 0.23692795485258103,
      "num_tokens": 146225834.0,
      "step": 84300
    },
    {
      "entropy": 5.309787654876709,
      "epoch": 6.559190818906828,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001708029331621568,
      "loss": 4.5357,
      "mean_token_accuracy": 0.24205396622419356,
      "num_tokens": 146234553.0,
      "step": 84305
    },
    {
      "entropy": 5.3737420558929445,
      "epoch": 6.5595798482785455,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017077836943223064,
      "loss": 4.6947,
      "mean_token_accuracy": 0.23523975610733033,
      "num_tokens": 146243803.0,
      "step": 84310
    },
    {
      "entropy": 5.491117668151856,
      "epoch": 6.559968877650262,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017075380728370182,
      "loss": 4.8376,
      "mean_token_accuracy": 0.22225705683231353,
      "num_tokens": 146252599.0,
      "step": 84315
    },
    {
      "entropy": 5.3212621212005615,
      "epoch": 6.56035790702198,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001707292467169431,
      "loss": 4.5196,
      "mean_token_accuracy": 0.250051286816597,
      "num_tokens": 146260574.0,
      "step": 84320
    },
    {
      "entropy": 5.43336033821106,
      "epoch": 6.560746936393698,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000170704687732327,
      "loss": 4.6941,
      "mean_token_accuracy": 0.239984230697155,
      "num_tokens": 146269457.0,
      "step": 84325
    },
    {
      "entropy": 5.4719713687896725,
      "epoch": 6.561135965765415,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001706801303302264,
      "loss": 4.7688,
      "mean_token_accuracy": 0.22898236215114592,
      "num_tokens": 146279851.0,
      "step": 84330
    },
    {
      "entropy": 5.465330696105957,
      "epoch": 6.561524995137133,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017065557451101366,
      "loss": 4.7279,
      "mean_token_accuracy": 0.23887331038713455,
      "num_tokens": 146288911.0,
      "step": 84335
    },
    {
      "entropy": 5.3912557601928714,
      "epoch": 6.561914024508851,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017063102027506155,
      "loss": 4.6667,
      "mean_token_accuracy": 0.24109784960746766,
      "num_tokens": 146297820.0,
      "step": 84340
    },
    {
      "entropy": 5.427685880661011,
      "epoch": 6.562303053880568,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017060646762274254,
      "loss": 4.6772,
      "mean_token_accuracy": 0.2452961176633835,
      "num_tokens": 146306421.0,
      "step": 84345
    },
    {
      "entropy": 5.339828586578369,
      "epoch": 6.562692083252285,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017058191655442928,
      "loss": 4.5342,
      "mean_token_accuracy": 0.245089989900589,
      "num_tokens": 146313496.0,
      "step": 84350
    },
    {
      "entropy": 5.373449420928955,
      "epoch": 6.563081112624003,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001705573670704943,
      "loss": 4.664,
      "mean_token_accuracy": 0.2332203656435013,
      "num_tokens": 146323080.0,
      "step": 84355
    },
    {
      "entropy": 5.385708427429199,
      "epoch": 6.563470141995721,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017053281917130993,
      "loss": 4.6024,
      "mean_token_accuracy": 0.24575001299381255,
      "num_tokens": 146332205.0,
      "step": 84360
    },
    {
      "entropy": 5.366835832595825,
      "epoch": 6.563859171367438,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017050827285724872,
      "loss": 4.6298,
      "mean_token_accuracy": 0.2418859377503395,
      "num_tokens": 146340894.0,
      "step": 84365
    },
    {
      "entropy": 5.330213308334351,
      "epoch": 6.564248200739156,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017048372812868316,
      "loss": 4.578,
      "mean_token_accuracy": 0.24439503848552704,
      "num_tokens": 146349592.0,
      "step": 84370
    },
    {
      "entropy": 5.333413410186767,
      "epoch": 6.564637230110874,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001704591849859858,
      "loss": 4.5265,
      "mean_token_accuracy": 0.2485651507973671,
      "num_tokens": 146358012.0,
      "step": 84375
    },
    {
      "entropy": 5.378177118301392,
      "epoch": 6.5650262594825906,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017043464342952883,
      "loss": 4.6278,
      "mean_token_accuracy": 0.24400155395269393,
      "num_tokens": 146366427.0,
      "step": 84380
    },
    {
      "entropy": 5.376384973526001,
      "epoch": 6.565415288854308,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017041010345968468,
      "loss": 4.6448,
      "mean_token_accuracy": 0.24679746180772782,
      "num_tokens": 146375356.0,
      "step": 84385
    },
    {
      "entropy": 5.437932538986206,
      "epoch": 6.565804318226026,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017038556507682586,
      "loss": 4.7177,
      "mean_token_accuracy": 0.24367764443159104,
      "num_tokens": 146384486.0,
      "step": 84390
    },
    {
      "entropy": 5.416551113128662,
      "epoch": 6.566193347597744,
      "grad_norm": 1.28125,
      "learning_rate": 0.00017036102828132444,
      "loss": 4.67,
      "mean_token_accuracy": 0.23539976626634598,
      "num_tokens": 146393421.0,
      "step": 84395
    },
    {
      "entropy": 5.385421705245972,
      "epoch": 6.566582376969461,
      "grad_norm": 1.46875,
      "learning_rate": 0.00017033649307355295,
      "loss": 4.6122,
      "mean_token_accuracy": 0.24254804998636245,
      "num_tokens": 146403419.0,
      "step": 84400
    },
    {
      "entropy": 5.404775953292846,
      "epoch": 6.566971406341179,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001703119594538835,
      "loss": 4.6425,
      "mean_token_accuracy": 0.235706102848053,
      "num_tokens": 146412277.0,
      "step": 84405
    },
    {
      "entropy": 5.425418281555176,
      "epoch": 6.567360435712896,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017028742742268856,
      "loss": 4.6947,
      "mean_token_accuracy": 0.23312044739723206,
      "num_tokens": 146420847.0,
      "step": 84410
    },
    {
      "entropy": 5.35340723991394,
      "epoch": 6.5677494650846135,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017026289698034015,
      "loss": 4.6789,
      "mean_token_accuracy": 0.24067778438329696,
      "num_tokens": 146429632.0,
      "step": 84415
    },
    {
      "entropy": 5.393985843658447,
      "epoch": 6.568138494456331,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017023836812721062,
      "loss": 4.6165,
      "mean_token_accuracy": 0.2482227012515068,
      "num_tokens": 146438206.0,
      "step": 84420
    },
    {
      "entropy": 5.441124725341797,
      "epoch": 6.568527523828049,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00017021384086367204,
      "loss": 4.7284,
      "mean_token_accuracy": 0.24350121170282363,
      "num_tokens": 146447403.0,
      "step": 84425
    },
    {
      "entropy": 5.362552690505981,
      "epoch": 6.568916553199767,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017018931519009663,
      "loss": 4.6072,
      "mean_token_accuracy": 0.24852775186300277,
      "num_tokens": 146456192.0,
      "step": 84430
    },
    {
      "entropy": 5.448221826553345,
      "epoch": 6.569305582571484,
      "grad_norm": 1.53125,
      "learning_rate": 0.00017016479110685667,
      "loss": 4.7287,
      "mean_token_accuracy": 0.23243789076805116,
      "num_tokens": 146464763.0,
      "step": 84435
    },
    {
      "entropy": 5.325114488601685,
      "epoch": 6.569694611943202,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017014026861432403,
      "loss": 4.602,
      "mean_token_accuracy": 0.24513457268476485,
      "num_tokens": 146473149.0,
      "step": 84440
    },
    {
      "entropy": 5.350306034088135,
      "epoch": 6.57008364131492,
      "grad_norm": 1.28125,
      "learning_rate": 0.000170115747712871,
      "loss": 4.6138,
      "mean_token_accuracy": 0.24640905261039733,
      "num_tokens": 146481616.0,
      "step": 84445
    },
    {
      "entropy": 5.460289096832275,
      "epoch": 6.5704726706866365,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017009122840286943,
      "loss": 4.7307,
      "mean_token_accuracy": 0.23854437470436096,
      "num_tokens": 146490491.0,
      "step": 84450
    },
    {
      "entropy": 5.444174194335938,
      "epoch": 6.570861700058354,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017006671068469165,
      "loss": 4.6648,
      "mean_token_accuracy": 0.24520158320665358,
      "num_tokens": 146499986.0,
      "step": 84455
    },
    {
      "entropy": 5.367552280426025,
      "epoch": 6.571250729430072,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017004219455870947,
      "loss": 4.6039,
      "mean_token_accuracy": 0.2479190245270729,
      "num_tokens": 146508361.0,
      "step": 84460
    },
    {
      "entropy": 5.470309734344482,
      "epoch": 6.57163975880179,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017001768002529487,
      "loss": 4.7349,
      "mean_token_accuracy": 0.239715813100338,
      "num_tokens": 146517673.0,
      "step": 84465
    },
    {
      "entropy": 5.396499109268189,
      "epoch": 6.572028788173507,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00016999316708481992,
      "loss": 4.6866,
      "mean_token_accuracy": 0.2461523488163948,
      "num_tokens": 146526378.0,
      "step": 84470
    },
    {
      "entropy": 5.3315455436706545,
      "epoch": 6.572417817545225,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001699686557376565,
      "loss": 4.5572,
      "mean_token_accuracy": 0.24175869822502136,
      "num_tokens": 146535521.0,
      "step": 84475
    },
    {
      "entropy": 5.381885147094726,
      "epoch": 6.572806846916942,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016994414598417662,
      "loss": 4.6292,
      "mean_token_accuracy": 0.24499955773353577,
      "num_tokens": 146544567.0,
      "step": 84480
    },
    {
      "entropy": 5.427047157287598,
      "epoch": 6.5731958762886595,
      "grad_norm": 1.4375,
      "learning_rate": 0.000169919637824752,
      "loss": 4.6761,
      "mean_token_accuracy": 0.23921978920698167,
      "num_tokens": 146552838.0,
      "step": 84485
    },
    {
      "entropy": 5.438476324081421,
      "epoch": 6.573584905660377,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001698951312597547,
      "loss": 4.7514,
      "mean_token_accuracy": 0.2375622197985649,
      "num_tokens": 146560883.0,
      "step": 84490
    },
    {
      "entropy": 5.3991138458251955,
      "epoch": 6.573973935032095,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016987062628955646,
      "loss": 4.6314,
      "mean_token_accuracy": 0.24775750786066056,
      "num_tokens": 146569352.0,
      "step": 84495
    },
    {
      "entropy": 5.400540113449097,
      "epoch": 6.574362964403813,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016984612291452909,
      "loss": 4.6629,
      "mean_token_accuracy": 0.24056964367628098,
      "num_tokens": 146577926.0,
      "step": 84500
    },
    {
      "entropy": 5.385214233398438,
      "epoch": 6.57475199377553,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001698216211350445,
      "loss": 4.6687,
      "mean_token_accuracy": 0.23299350440502167,
      "num_tokens": 146587240.0,
      "step": 84505
    },
    {
      "entropy": 5.442490768432617,
      "epoch": 6.575141023147248,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016979712095147438,
      "loss": 4.6961,
      "mean_token_accuracy": 0.23248315304517747,
      "num_tokens": 146595461.0,
      "step": 84510
    },
    {
      "entropy": 5.388933515548706,
      "epoch": 6.575530052518965,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00016977262236419057,
      "loss": 4.6276,
      "mean_token_accuracy": 0.23133996576070787,
      "num_tokens": 146604128.0,
      "step": 84515
    },
    {
      "entropy": 5.42922945022583,
      "epoch": 6.5759190818906825,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016974812537356459,
      "loss": 4.7031,
      "mean_token_accuracy": 0.23677914589643478,
      "num_tokens": 146612187.0,
      "step": 84520
    },
    {
      "entropy": 5.360751104354859,
      "epoch": 6.5763081112624,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016972362997996846,
      "loss": 4.6123,
      "mean_token_accuracy": 0.24134421944618226,
      "num_tokens": 146620206.0,
      "step": 84525
    },
    {
      "entropy": 5.4562609672546385,
      "epoch": 6.576697140634118,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016969913618377353,
      "loss": 4.7393,
      "mean_token_accuracy": 0.22760725766420364,
      "num_tokens": 146629567.0,
      "step": 84530
    },
    {
      "entropy": 5.411977481842041,
      "epoch": 6.577086170005836,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001696746439853517,
      "loss": 4.6721,
      "mean_token_accuracy": 0.2410157486796379,
      "num_tokens": 146638511.0,
      "step": 84535
    },
    {
      "entropy": 5.424694633483886,
      "epoch": 6.577475199377553,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016965015338507453,
      "loss": 4.6889,
      "mean_token_accuracy": 0.24234678149223327,
      "num_tokens": 146647468.0,
      "step": 84540
    },
    {
      "entropy": 5.320657873153687,
      "epoch": 6.57786422874927,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016962566438331352,
      "loss": 4.507,
      "mean_token_accuracy": 0.25057934373617174,
      "num_tokens": 146656400.0,
      "step": 84545
    },
    {
      "entropy": 5.3214905738830565,
      "epoch": 6.578253258120988,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016960117698044036,
      "loss": 4.5988,
      "mean_token_accuracy": 0.24446096867322922,
      "num_tokens": 146664761.0,
      "step": 84550
    },
    {
      "entropy": 5.376437664031982,
      "epoch": 6.5786422874927055,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016957669117682652,
      "loss": 4.6245,
      "mean_token_accuracy": 0.23707259744405745,
      "num_tokens": 146672986.0,
      "step": 84555
    },
    {
      "entropy": 5.385486268997193,
      "epoch": 6.579031316864423,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016955220697284363,
      "loss": 4.6942,
      "mean_token_accuracy": 0.24149303883314133,
      "num_tokens": 146681828.0,
      "step": 84560
    },
    {
      "entropy": 5.346033906936645,
      "epoch": 6.579420346236141,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016952772436886306,
      "loss": 4.6389,
      "mean_token_accuracy": 0.241556316614151,
      "num_tokens": 146690137.0,
      "step": 84565
    },
    {
      "entropy": 5.326116991043091,
      "epoch": 6.579809375607859,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016950324336525657,
      "loss": 4.5592,
      "mean_token_accuracy": 0.24523078352212907,
      "num_tokens": 146699131.0,
      "step": 84570
    },
    {
      "entropy": 5.401447725296021,
      "epoch": 6.580198404979576,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016947876396239544,
      "loss": 4.7022,
      "mean_token_accuracy": 0.23353791236877441,
      "num_tokens": 146707937.0,
      "step": 84575
    },
    {
      "entropy": 5.363683605194092,
      "epoch": 6.580587434351294,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016945428616065104,
      "loss": 4.6443,
      "mean_token_accuracy": 0.24058277010917664,
      "num_tokens": 146716294.0,
      "step": 84580
    },
    {
      "entropy": 5.349463701248169,
      "epoch": 6.580976463723011,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001694298099603949,
      "loss": 4.6696,
      "mean_token_accuracy": 0.24491775929927825,
      "num_tokens": 146724729.0,
      "step": 84585
    },
    {
      "entropy": 5.425011730194091,
      "epoch": 6.5813654930947285,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001694053353619983,
      "loss": 4.7282,
      "mean_token_accuracy": 0.23404111862182617,
      "num_tokens": 146734169.0,
      "step": 84590
    },
    {
      "entropy": 5.357710361480713,
      "epoch": 6.581754522466446,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016938086236583272,
      "loss": 4.6391,
      "mean_token_accuracy": 0.24091438055038453,
      "num_tokens": 146743085.0,
      "step": 84595
    },
    {
      "entropy": 5.421069097518921,
      "epoch": 6.582143551838164,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001693563909722694,
      "loss": 4.6466,
      "mean_token_accuracy": 0.24066193848848344,
      "num_tokens": 146751716.0,
      "step": 84600
    },
    {
      "entropy": 5.348891878128052,
      "epoch": 6.582532581209882,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016933192118167976,
      "loss": 4.5778,
      "mean_token_accuracy": 0.2498205304145813,
      "num_tokens": 146759769.0,
      "step": 84605
    },
    {
      "entropy": 5.357227754592896,
      "epoch": 6.582921610581598,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016930745299443496,
      "loss": 4.7118,
      "mean_token_accuracy": 0.23699511140584945,
      "num_tokens": 146768702.0,
      "step": 84610
    },
    {
      "entropy": 5.355023002624511,
      "epoch": 6.583310639953316,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001692829864109064,
      "loss": 4.6399,
      "mean_token_accuracy": 0.2422180563211441,
      "num_tokens": 146777338.0,
      "step": 84615
    },
    {
      "entropy": 5.383370304107666,
      "epoch": 6.583699669325034,
      "grad_norm": 1.375,
      "learning_rate": 0.00016925852143146526,
      "loss": 4.6003,
      "mean_token_accuracy": 0.24811945259571075,
      "num_tokens": 146786168.0,
      "step": 84620
    },
    {
      "entropy": 5.518668985366821,
      "epoch": 6.5840886986967515,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016923405805648268,
      "loss": 4.8094,
      "mean_token_accuracy": 0.23005206435918807,
      "num_tokens": 146796015.0,
      "step": 84625
    },
    {
      "entropy": 5.406057262420655,
      "epoch": 6.584477728068469,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00016920959628632988,
      "loss": 4.6889,
      "mean_token_accuracy": 0.24273302853107454,
      "num_tokens": 146805637.0,
      "step": 84630
    },
    {
      "entropy": 5.465916585922241,
      "epoch": 6.584866757440187,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016918513612137809,
      "loss": 4.7824,
      "mean_token_accuracy": 0.23126621544361115,
      "num_tokens": 146815095.0,
      "step": 84635
    },
    {
      "entropy": 5.410434532165527,
      "epoch": 6.585255786811905,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001691606775619986,
      "loss": 4.7135,
      "mean_token_accuracy": 0.23429333716630935,
      "num_tokens": 146824437.0,
      "step": 84640
    },
    {
      "entropy": 5.4217798709869385,
      "epoch": 6.585644816183622,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016913622060856217,
      "loss": 4.6848,
      "mean_token_accuracy": 0.24033688008785248,
      "num_tokens": 146833216.0,
      "step": 84645
    },
    {
      "entropy": 5.359645223617553,
      "epoch": 6.586033845555339,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016911176526144024,
      "loss": 4.5601,
      "mean_token_accuracy": 0.2509561315178871,
      "num_tokens": 146841136.0,
      "step": 84650
    },
    {
      "entropy": 5.338476085662842,
      "epoch": 6.586422874927057,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016908731152100374,
      "loss": 4.613,
      "mean_token_accuracy": 0.24283459335565566,
      "num_tokens": 146849592.0,
      "step": 84655
    },
    {
      "entropy": 5.287589836120605,
      "epoch": 6.5868119042987745,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001690628593876236,
      "loss": 4.603,
      "mean_token_accuracy": 0.2446575567126274,
      "num_tokens": 146859010.0,
      "step": 84660
    },
    {
      "entropy": 5.332306432723999,
      "epoch": 6.587200933670492,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016903840886167104,
      "loss": 4.6066,
      "mean_token_accuracy": 0.24302032589912415,
      "num_tokens": 146867221.0,
      "step": 84665
    },
    {
      "entropy": 5.3716552734375,
      "epoch": 6.58758996304221,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016901395994351688,
      "loss": 4.6089,
      "mean_token_accuracy": 0.24078522622585297,
      "num_tokens": 146875948.0,
      "step": 84670
    },
    {
      "entropy": 5.279925107955933,
      "epoch": 6.587978992413928,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016898951263353228,
      "loss": 4.4908,
      "mean_token_accuracy": 0.24956051558256148,
      "num_tokens": 146884376.0,
      "step": 84675
    },
    {
      "entropy": 5.3656940937042235,
      "epoch": 6.588368021785644,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016896506693208802,
      "loss": 4.6513,
      "mean_token_accuracy": 0.24646958112716674,
      "num_tokens": 146893572.0,
      "step": 84680
    },
    {
      "entropy": 5.405946397781372,
      "epoch": 6.588757051157362,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016894062283955515,
      "loss": 4.6657,
      "mean_token_accuracy": 0.23959125727415084,
      "num_tokens": 146901722.0,
      "step": 84685
    },
    {
      "entropy": 5.424219083786011,
      "epoch": 6.58914608052908,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016891618035630446,
      "loss": 4.6961,
      "mean_token_accuracy": 0.24392360150814058,
      "num_tokens": 146910580.0,
      "step": 84690
    },
    {
      "entropy": 5.369238710403442,
      "epoch": 6.5895351099007975,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016889173948270687,
      "loss": 4.5686,
      "mean_token_accuracy": 0.24335096031427383,
      "num_tokens": 146918444.0,
      "step": 84695
    },
    {
      "entropy": 5.340540027618408,
      "epoch": 6.589924139272515,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016886730021913336,
      "loss": 4.564,
      "mean_token_accuracy": 0.2444102570414543,
      "num_tokens": 146927252.0,
      "step": 84700
    },
    {
      "entropy": 5.3123572826385494,
      "epoch": 6.590313168644233,
      "grad_norm": 1.375,
      "learning_rate": 0.00016884286256595456,
      "loss": 4.631,
      "mean_token_accuracy": 0.24328398406505586,
      "num_tokens": 146935517.0,
      "step": 84705
    },
    {
      "entropy": 5.282245254516601,
      "epoch": 6.590702198015951,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016881842652354136,
      "loss": 4.5288,
      "mean_token_accuracy": 0.2472146987915039,
      "num_tokens": 146944549.0,
      "step": 84710
    },
    {
      "entropy": 5.385653066635132,
      "epoch": 6.591091227387667,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001687939920922645,
      "loss": 4.7421,
      "mean_token_accuracy": 0.23724108040332795,
      "num_tokens": 146952418.0,
      "step": 84715
    },
    {
      "entropy": 5.435217571258545,
      "epoch": 6.591480256759385,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001687695592724949,
      "loss": 4.7498,
      "mean_token_accuracy": 0.23667190968990326,
      "num_tokens": 146961248.0,
      "step": 84720
    },
    {
      "entropy": 5.375254774093628,
      "epoch": 6.591869286131103,
      "grad_norm": 1.25,
      "learning_rate": 0.00016874512806460296,
      "loss": 4.6728,
      "mean_token_accuracy": 0.24463984370231628,
      "num_tokens": 146970062.0,
      "step": 84725
    },
    {
      "entropy": 5.505873537063598,
      "epoch": 6.5922583155028205,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016872069846895978,
      "loss": 4.7693,
      "mean_token_accuracy": 0.23443709164857865,
      "num_tokens": 146978487.0,
      "step": 84730
    },
    {
      "entropy": 5.313940620422363,
      "epoch": 6.592647344874538,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016869627048593573,
      "loss": 4.5458,
      "mean_token_accuracy": 0.24701424688100815,
      "num_tokens": 146986715.0,
      "step": 84735
    },
    {
      "entropy": 5.389333581924438,
      "epoch": 6.593036374246256,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016867184411590154,
      "loss": 4.6683,
      "mean_token_accuracy": 0.2395954415202141,
      "num_tokens": 146994968.0,
      "step": 84740
    },
    {
      "entropy": 5.396217393875122,
      "epoch": 6.593425403617973,
      "grad_norm": 1.3125,
      "learning_rate": 0.000168647419359228,
      "loss": 4.6428,
      "mean_token_accuracy": 0.24097407460212708,
      "num_tokens": 147003679.0,
      "step": 84745
    },
    {
      "entropy": 5.48943133354187,
      "epoch": 6.59381443298969,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016862299621628546,
      "loss": 4.7303,
      "mean_token_accuracy": 0.23437334597110748,
      "num_tokens": 147012637.0,
      "step": 84750
    },
    {
      "entropy": 5.435339498519897,
      "epoch": 6.594203462361408,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016859857468744471,
      "loss": 4.7035,
      "mean_token_accuracy": 0.23244636207818986,
      "num_tokens": 147021294.0,
      "step": 84755
    },
    {
      "entropy": 5.461569118499756,
      "epoch": 6.594592491733126,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001685741547730762,
      "loss": 4.7494,
      "mean_token_accuracy": 0.24023213237524033,
      "num_tokens": 147029998.0,
      "step": 84760
    },
    {
      "entropy": 5.494941854476929,
      "epoch": 6.5949815211048435,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016854973647355045,
      "loss": 4.7367,
      "mean_token_accuracy": 0.23438269048929214,
      "num_tokens": 147038790.0,
      "step": 84765
    },
    {
      "entropy": 5.423483180999756,
      "epoch": 6.595370550476561,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016852531978923816,
      "loss": 4.6321,
      "mean_token_accuracy": 0.24188441187143325,
      "num_tokens": 147046906.0,
      "step": 84770
    },
    {
      "entropy": 5.41530933380127,
      "epoch": 6.595759579848279,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016850090472050956,
      "loss": 4.6975,
      "mean_token_accuracy": 0.235201495885849,
      "num_tokens": 147055958.0,
      "step": 84775
    },
    {
      "entropy": 5.367156648635865,
      "epoch": 6.596148609219997,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016847649126773526,
      "loss": 4.6788,
      "mean_token_accuracy": 0.23820467442274093,
      "num_tokens": 147064533.0,
      "step": 84780
    },
    {
      "entropy": 5.37761492729187,
      "epoch": 6.596537638591713,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001684520794312856,
      "loss": 4.6032,
      "mean_token_accuracy": 0.24528128504753113,
      "num_tokens": 147072377.0,
      "step": 84785
    },
    {
      "entropy": 5.430870723724365,
      "epoch": 6.596926667963431,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016842766921153113,
      "loss": 4.6483,
      "mean_token_accuracy": 0.24028197824954986,
      "num_tokens": 147080301.0,
      "step": 84790
    },
    {
      "entropy": 5.423362398147583,
      "epoch": 6.597315697335149,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016840326060884206,
      "loss": 4.7395,
      "mean_token_accuracy": 0.23221496641635894,
      "num_tokens": 147089231.0,
      "step": 84795
    },
    {
      "entropy": 5.375909948348999,
      "epoch": 6.5977047267068665,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016837885362358895,
      "loss": 4.6124,
      "mean_token_accuracy": 0.24140287190675735,
      "num_tokens": 147098174.0,
      "step": 84800
    },
    {
      "entropy": 5.344957160949707,
      "epoch": 6.598093756078584,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016835444825614193,
      "loss": 4.6304,
      "mean_token_accuracy": 0.24658889472484588,
      "num_tokens": 147106178.0,
      "step": 84805
    },
    {
      "entropy": 5.299638366699218,
      "epoch": 6.598482785450302,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001683300445068715,
      "loss": 4.5776,
      "mean_token_accuracy": 0.23834228515625,
      "num_tokens": 147115512.0,
      "step": 84810
    },
    {
      "entropy": 5.294767570495606,
      "epoch": 6.598871814822019,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001683056423761479,
      "loss": 4.5105,
      "mean_token_accuracy": 0.2546794727444649,
      "num_tokens": 147123784.0,
      "step": 84815
    },
    {
      "entropy": 5.474394273757935,
      "epoch": 6.599260844193736,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00016828124186434121,
      "loss": 4.7325,
      "mean_token_accuracy": 0.23575533032417298,
      "num_tokens": 147131559.0,
      "step": 84820
    },
    {
      "entropy": 5.327744770050049,
      "epoch": 6.599649873565454,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016825684297182194,
      "loss": 4.5967,
      "mean_token_accuracy": 0.24534608721733092,
      "num_tokens": 147140253.0,
      "step": 84825
    },
    {
      "entropy": 5.347218704223633,
      "epoch": 6.600038902937172,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016823244569896,
      "loss": 4.6087,
      "mean_token_accuracy": 0.24162197560071946,
      "num_tokens": 147149245.0,
      "step": 84830
    },
    {
      "entropy": 5.46824951171875,
      "epoch": 6.6004279323088895,
      "grad_norm": 1.25,
      "learning_rate": 0.00016820805004612595,
      "loss": 4.7011,
      "mean_token_accuracy": 0.24054159522056578,
      "num_tokens": 147158305.0,
      "step": 84835
    },
    {
      "entropy": 5.353557538986206,
      "epoch": 6.600816961680607,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016818365601368974,
      "loss": 4.6123,
      "mean_token_accuracy": 0.24318811893463135,
      "num_tokens": 147167022.0,
      "step": 84840
    },
    {
      "entropy": 5.422938346862793,
      "epoch": 6.601205991052325,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016815926360202145,
      "loss": 4.6821,
      "mean_token_accuracy": 0.2423985242843628,
      "num_tokens": 147176061.0,
      "step": 84845
    },
    {
      "entropy": 5.356579351425171,
      "epoch": 6.601595020424042,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001681348728114913,
      "loss": 4.6239,
      "mean_token_accuracy": 0.240465883910656,
      "num_tokens": 147184923.0,
      "step": 84850
    },
    {
      "entropy": 5.3260753631591795,
      "epoch": 6.601984049795759,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016811048364246938,
      "loss": 4.5062,
      "mean_token_accuracy": 0.24692298769950866,
      "num_tokens": 147193366.0,
      "step": 84855
    },
    {
      "entropy": 5.313503170013428,
      "epoch": 6.602373079167477,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016808609609532578,
      "loss": 4.5732,
      "mean_token_accuracy": 0.24501850008964537,
      "num_tokens": 147201778.0,
      "step": 84860
    },
    {
      "entropy": 5.4345612049102785,
      "epoch": 6.602762108539195,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001680617101704304,
      "loss": 4.7815,
      "mean_token_accuracy": 0.23034237325191498,
      "num_tokens": 147211999.0,
      "step": 84865
    },
    {
      "entropy": 5.3611139297485355,
      "epoch": 6.6031511379109125,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016803732586815346,
      "loss": 4.6324,
      "mean_token_accuracy": 0.24090876132249833,
      "num_tokens": 147220353.0,
      "step": 84870
    },
    {
      "entropy": 5.348388767242431,
      "epoch": 6.60354016728263,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016801294318886478,
      "loss": 4.5788,
      "mean_token_accuracy": 0.24277048110961913,
      "num_tokens": 147229180.0,
      "step": 84875
    },
    {
      "entropy": 5.362228775024414,
      "epoch": 6.603929196654347,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001679885621329345,
      "loss": 4.559,
      "mean_token_accuracy": 0.24239525347948074,
      "num_tokens": 147237367.0,
      "step": 84880
    },
    {
      "entropy": 5.305191946029663,
      "epoch": 6.604318226026065,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016796418270073233,
      "loss": 4.6019,
      "mean_token_accuracy": 0.24548857808113098,
      "num_tokens": 147246089.0,
      "step": 84885
    },
    {
      "entropy": 5.417315101623535,
      "epoch": 6.604707255397782,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001679398048926284,
      "loss": 4.6356,
      "mean_token_accuracy": 0.24127229899168015,
      "num_tokens": 147254735.0,
      "step": 84890
    },
    {
      "entropy": 5.340967893600464,
      "epoch": 6.6050962847695,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016791542870899252,
      "loss": 4.6506,
      "mean_token_accuracy": 0.241156867146492,
      "num_tokens": 147263070.0,
      "step": 84895
    },
    {
      "entropy": 5.326006937026977,
      "epoch": 6.605485314141218,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016789105415019448,
      "loss": 4.6349,
      "mean_token_accuracy": 0.23975169360637666,
      "num_tokens": 147271891.0,
      "step": 84900
    },
    {
      "entropy": 5.3107250213623045,
      "epoch": 6.6058743435129355,
      "grad_norm": 1.25,
      "learning_rate": 0.00016786668121660433,
      "loss": 4.612,
      "mean_token_accuracy": 0.24616075307130814,
      "num_tokens": 147280868.0,
      "step": 84905
    },
    {
      "entropy": 5.395116853713989,
      "epoch": 6.606263372884653,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016784230990859173,
      "loss": 4.6782,
      "mean_token_accuracy": 0.24300650209188462,
      "num_tokens": 147290394.0,
      "step": 84910
    },
    {
      "entropy": 5.430595064163208,
      "epoch": 6.606652402256371,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016781794022652656,
      "loss": 4.7249,
      "mean_token_accuracy": 0.2295028895139694,
      "num_tokens": 147298671.0,
      "step": 84915
    },
    {
      "entropy": 5.457510757446289,
      "epoch": 6.607041431628088,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016779357217077851,
      "loss": 4.6875,
      "mean_token_accuracy": 0.2369174227118492,
      "num_tokens": 147306877.0,
      "step": 84920
    },
    {
      "entropy": 5.529739809036255,
      "epoch": 6.607430460999805,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016776920574171735,
      "loss": 4.8239,
      "mean_token_accuracy": 0.2308649316430092,
      "num_tokens": 147315543.0,
      "step": 84925
    },
    {
      "entropy": 5.357408618927002,
      "epoch": 6.607819490371523,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016774484093971288,
      "loss": 4.5691,
      "mean_token_accuracy": 0.2458970680832863,
      "num_tokens": 147323522.0,
      "step": 84930
    },
    {
      "entropy": 5.3845543384552,
      "epoch": 6.608208519743241,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016772047776513464,
      "loss": 4.6907,
      "mean_token_accuracy": 0.24160870015621186,
      "num_tokens": 147332403.0,
      "step": 84935
    },
    {
      "entropy": 5.292087411880493,
      "epoch": 6.6085975491149584,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016769611621835252,
      "loss": 4.5882,
      "mean_token_accuracy": 0.2479159951210022,
      "num_tokens": 147340486.0,
      "step": 84940
    },
    {
      "entropy": 5.395690965652466,
      "epoch": 6.608986578486675,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001676717562997359,
      "loss": 4.6284,
      "mean_token_accuracy": 0.24656738340854645,
      "num_tokens": 147348262.0,
      "step": 84945
    },
    {
      "entropy": 5.371829700469971,
      "epoch": 6.609375607858393,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016764739800965468,
      "loss": 4.5725,
      "mean_token_accuracy": 0.25084731727838516,
      "num_tokens": 147356214.0,
      "step": 84950
    },
    {
      "entropy": 5.301826286315918,
      "epoch": 6.609764637230111,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001676230413484782,
      "loss": 4.5534,
      "mean_token_accuracy": 0.2503005161881447,
      "num_tokens": 147364819.0,
      "step": 84955
    },
    {
      "entropy": 5.333249902725219,
      "epoch": 6.610153666601828,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001675986863165762,
      "loss": 4.57,
      "mean_token_accuracy": 0.24635596871376036,
      "num_tokens": 147372955.0,
      "step": 84960
    },
    {
      "entropy": 5.3653754711151125,
      "epoch": 6.610542695973546,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016757433291431822,
      "loss": 4.6395,
      "mean_token_accuracy": 0.235653255879879,
      "num_tokens": 147380958.0,
      "step": 84965
    },
    {
      "entropy": 5.372679710388184,
      "epoch": 6.610931725345264,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001675499811420737,
      "loss": 4.5926,
      "mean_token_accuracy": 0.2409351646900177,
      "num_tokens": 147390552.0,
      "step": 84970
    },
    {
      "entropy": 5.458901596069336,
      "epoch": 6.611320754716981,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016752563100021228,
      "loss": 4.7176,
      "mean_token_accuracy": 0.2386649176478386,
      "num_tokens": 147398952.0,
      "step": 84975
    },
    {
      "entropy": 5.418434381484985,
      "epoch": 6.611709784088699,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016750128248910328,
      "loss": 4.7212,
      "mean_token_accuracy": 0.2442736104130745,
      "num_tokens": 147407608.0,
      "step": 84980
    },
    {
      "entropy": 5.401566457748413,
      "epoch": 6.612098813460416,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00016747693560911626,
      "loss": 4.6954,
      "mean_token_accuracy": 0.24070698916912078,
      "num_tokens": 147416462.0,
      "step": 84985
    },
    {
      "entropy": 5.353321218490601,
      "epoch": 6.612487842832134,
      "grad_norm": 1.25,
      "learning_rate": 0.00016745259036062056,
      "loss": 4.6087,
      "mean_token_accuracy": 0.24540013819932938,
      "num_tokens": 147424685.0,
      "step": 84990
    },
    {
      "entropy": 5.364633321762085,
      "epoch": 6.612876872203851,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001674282467439857,
      "loss": 4.633,
      "mean_token_accuracy": 0.2417902246117592,
      "num_tokens": 147433750.0,
      "step": 84995
    },
    {
      "entropy": 5.303768539428711,
      "epoch": 6.613265901575569,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016740390475958095,
      "loss": 4.542,
      "mean_token_accuracy": 0.24556280076503753,
      "num_tokens": 147441999.0,
      "step": 85000
    },
    {
      "entropy": 5.316144227981567,
      "epoch": 6.613654930947287,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016737956440777564,
      "loss": 4.6246,
      "mean_token_accuracy": 0.2416437089443207,
      "num_tokens": 147450945.0,
      "step": 85005
    },
    {
      "entropy": 5.362026262283325,
      "epoch": 6.614043960319004,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001673552256889392,
      "loss": 4.722,
      "mean_token_accuracy": 0.23605990707874297,
      "num_tokens": 147459541.0,
      "step": 85010
    },
    {
      "entropy": 5.338880586624145,
      "epoch": 6.614432989690721,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001673308886034408,
      "loss": 4.5975,
      "mean_token_accuracy": 0.24209969341754914,
      "num_tokens": 147467904.0,
      "step": 85015
    },
    {
      "entropy": 5.429802274703979,
      "epoch": 6.614822019062439,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016730655315164985,
      "loss": 4.7035,
      "mean_token_accuracy": 0.2403227210044861,
      "num_tokens": 147476679.0,
      "step": 85020
    },
    {
      "entropy": 5.349461841583252,
      "epoch": 6.615211048434157,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001672822193339354,
      "loss": 4.5948,
      "mean_token_accuracy": 0.24181448817253112,
      "num_tokens": 147485081.0,
      "step": 85025
    },
    {
      "entropy": 5.465399742126465,
      "epoch": 6.615600077805874,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016725788715066698,
      "loss": 4.7435,
      "mean_token_accuracy": 0.23700958639383315,
      "num_tokens": 147493808.0,
      "step": 85030
    },
    {
      "entropy": 5.396971082687378,
      "epoch": 6.615989107177592,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016723355660221368,
      "loss": 4.6666,
      "mean_token_accuracy": 0.24058808386325836,
      "num_tokens": 147502696.0,
      "step": 85035
    },
    {
      "entropy": 5.426829671859741,
      "epoch": 6.61637813654931,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016720922768894447,
      "loss": 4.6999,
      "mean_token_accuracy": 0.23768509328365325,
      "num_tokens": 147511034.0,
      "step": 85040
    },
    {
      "entropy": 5.4506837844848635,
      "epoch": 6.616767165921027,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001671849004112288,
      "loss": 4.7435,
      "mean_token_accuracy": 0.23680549710989,
      "num_tokens": 147520506.0,
      "step": 85045
    },
    {
      "entropy": 5.386112451553345,
      "epoch": 6.617156195292744,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016716057476943554,
      "loss": 4.6802,
      "mean_token_accuracy": 0.24078678488731384,
      "num_tokens": 147529398.0,
      "step": 85050
    },
    {
      "entropy": 5.353237152099609,
      "epoch": 6.617545224664462,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00016713625076393403,
      "loss": 4.5929,
      "mean_token_accuracy": 0.24861593395471573,
      "num_tokens": 147538199.0,
      "step": 85055
    },
    {
      "entropy": 5.317676591873169,
      "epoch": 6.61793425403618,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016711192839509315,
      "loss": 4.602,
      "mean_token_accuracy": 0.24197246134281158,
      "num_tokens": 147547460.0,
      "step": 85060
    },
    {
      "entropy": 5.3474658012390135,
      "epoch": 6.618323283407897,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001670876076632821,
      "loss": 4.5924,
      "mean_token_accuracy": 0.24277315735816957,
      "num_tokens": 147555938.0,
      "step": 85065
    },
    {
      "entropy": 5.39335265159607,
      "epoch": 6.618712312779615,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016706328856886976,
      "loss": 4.6297,
      "mean_token_accuracy": 0.24416491836309434,
      "num_tokens": 147564293.0,
      "step": 85070
    },
    {
      "entropy": 5.420904588699341,
      "epoch": 6.619101342151333,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016703897111222532,
      "loss": 4.7129,
      "mean_token_accuracy": 0.23345373719930648,
      "num_tokens": 147572755.0,
      "step": 85075
    },
    {
      "entropy": 5.397638654708862,
      "epoch": 6.6194903715230495,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016701465529371753,
      "loss": 4.7244,
      "mean_token_accuracy": 0.22805369049310684,
      "num_tokens": 147581677.0,
      "step": 85080
    },
    {
      "entropy": 5.429655075073242,
      "epoch": 6.619879400894767,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016699034111371564,
      "loss": 4.7088,
      "mean_token_accuracy": 0.23997960537672042,
      "num_tokens": 147590830.0,
      "step": 85085
    },
    {
      "entropy": 5.385819339752198,
      "epoch": 6.620268430266485,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00016696602857258833,
      "loss": 4.609,
      "mean_token_accuracy": 0.2437571480870247,
      "num_tokens": 147599746.0,
      "step": 85090
    },
    {
      "entropy": 5.3169252395629885,
      "epoch": 6.620657459638203,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016694171767070443,
      "loss": 4.54,
      "mean_token_accuracy": 0.25219019055366515,
      "num_tokens": 147608138.0,
      "step": 85095
    },
    {
      "entropy": 5.3297018051147464,
      "epoch": 6.62104648900992,
      "grad_norm": 1.375,
      "learning_rate": 0.00016691740840843315,
      "loss": 4.5844,
      "mean_token_accuracy": 0.24312763065099716,
      "num_tokens": 147616425.0,
      "step": 85100
    },
    {
      "entropy": 5.3773948669433596,
      "epoch": 6.621435518381638,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016689310078614307,
      "loss": 4.6407,
      "mean_token_accuracy": 0.2373705491423607,
      "num_tokens": 147623758.0,
      "step": 85105
    },
    {
      "entropy": 5.405201625823975,
      "epoch": 6.621824547753356,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001668687948042032,
      "loss": 4.7049,
      "mean_token_accuracy": 0.23456077873706818,
      "num_tokens": 147631516.0,
      "step": 85110
    },
    {
      "entropy": 5.291733121871948,
      "epoch": 6.622213577125073,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016684449046298227,
      "loss": 4.5699,
      "mean_token_accuracy": 0.2517614096403122,
      "num_tokens": 147639346.0,
      "step": 85115
    },
    {
      "entropy": 5.430687475204468,
      "epoch": 6.62260260649679,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016682018776284898,
      "loss": 4.6832,
      "mean_token_accuracy": 0.2405116379261017,
      "num_tokens": 147647850.0,
      "step": 85120
    },
    {
      "entropy": 5.506042623519898,
      "epoch": 6.622991635868508,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001667958867041722,
      "loss": 4.8014,
      "mean_token_accuracy": 0.22759669423103332,
      "num_tokens": 147656101.0,
      "step": 85125
    },
    {
      "entropy": 5.384186458587647,
      "epoch": 6.623380665240226,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016677158728732055,
      "loss": 4.6297,
      "mean_token_accuracy": 0.24604760706424714,
      "num_tokens": 147664416.0,
      "step": 85130
    },
    {
      "entropy": 5.360487699508667,
      "epoch": 6.623769694611943,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001667472895126629,
      "loss": 4.597,
      "mean_token_accuracy": 0.24831484854221345,
      "num_tokens": 147673396.0,
      "step": 85135
    },
    {
      "entropy": 5.296630382537842,
      "epoch": 6.624158723983661,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016672299338056773,
      "loss": 4.5864,
      "mean_token_accuracy": 0.2457493305206299,
      "num_tokens": 147681389.0,
      "step": 85140
    },
    {
      "entropy": 5.3259330749511715,
      "epoch": 6.624547753355379,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016669869889140388,
      "loss": 4.5853,
      "mean_token_accuracy": 0.2446097120642662,
      "num_tokens": 147689486.0,
      "step": 85145
    },
    {
      "entropy": 5.405546712875366,
      "epoch": 6.6249367827270955,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001666744060455398,
      "loss": 4.725,
      "mean_token_accuracy": 0.23502719402313232,
      "num_tokens": 147698301.0,
      "step": 85150
    },
    {
      "entropy": 5.408657217025757,
      "epoch": 6.625325812098813,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001666501148433443,
      "loss": 4.6597,
      "mean_token_accuracy": 0.24079445898532867,
      "num_tokens": 147707539.0,
      "step": 85155
    },
    {
      "entropy": 5.3553760051727295,
      "epoch": 6.625714841470531,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016662582528518578,
      "loss": 4.6468,
      "mean_token_accuracy": 0.2402677431702614,
      "num_tokens": 147716017.0,
      "step": 85160
    },
    {
      "entropy": 5.328882360458374,
      "epoch": 6.626103870842249,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00016660153737143283,
      "loss": 4.621,
      "mean_token_accuracy": 0.2557135999202728,
      "num_tokens": 147724475.0,
      "step": 85165
    },
    {
      "entropy": 5.466624355316162,
      "epoch": 6.626492900213966,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016657725110245412,
      "loss": 4.7412,
      "mean_token_accuracy": 0.2377592593431473,
      "num_tokens": 147733064.0,
      "step": 85170
    },
    {
      "entropy": 5.302232265472412,
      "epoch": 6.626881929585684,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016655296647861796,
      "loss": 4.5393,
      "mean_token_accuracy": 0.2535153031349182,
      "num_tokens": 147742224.0,
      "step": 85175
    },
    {
      "entropy": 5.344184875488281,
      "epoch": 6.627270958957402,
      "grad_norm": 1.21875,
      "learning_rate": 0.000166528683500293,
      "loss": 4.6178,
      "mean_token_accuracy": 0.24734582901000976,
      "num_tokens": 147751160.0,
      "step": 85180
    },
    {
      "entropy": 5.2657266616821286,
      "epoch": 6.6276599883291185,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016650440216784754,
      "loss": 4.5229,
      "mean_token_accuracy": 0.2464326038956642,
      "num_tokens": 147759507.0,
      "step": 85185
    },
    {
      "entropy": 5.270854043960571,
      "epoch": 6.628049017700836,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016648012248165017,
      "loss": 4.5873,
      "mean_token_accuracy": 0.24960466474294662,
      "num_tokens": 147767740.0,
      "step": 85190
    },
    {
      "entropy": 5.269649314880371,
      "epoch": 6.628438047072554,
      "grad_norm": 1.375,
      "learning_rate": 0.0001664558444420692,
      "loss": 4.5597,
      "mean_token_accuracy": 0.25650366991758344,
      "num_tokens": 147776401.0,
      "step": 85195
    },
    {
      "entropy": 5.377918672561646,
      "epoch": 6.628827076444272,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016643156804947297,
      "loss": 4.6772,
      "mean_token_accuracy": 0.2402199164032936,
      "num_tokens": 147784408.0,
      "step": 85200
    },
    {
      "entropy": 5.460059595108032,
      "epoch": 6.629216105815989,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016640729330422992,
      "loss": 4.7143,
      "mean_token_accuracy": 0.23713289499282836,
      "num_tokens": 147793150.0,
      "step": 85205
    },
    {
      "entropy": 5.404957628250122,
      "epoch": 6.629605135187707,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001663830202067083,
      "loss": 4.6067,
      "mean_token_accuracy": 0.24149061292409896,
      "num_tokens": 147801820.0,
      "step": 85210
    },
    {
      "entropy": 5.421248579025269,
      "epoch": 6.629994164559424,
      "grad_norm": 1.25,
      "learning_rate": 0.00016635874875727658,
      "loss": 4.6836,
      "mean_token_accuracy": 0.2415817067027092,
      "num_tokens": 147810637.0,
      "step": 85215
    },
    {
      "entropy": 5.429814052581787,
      "epoch": 6.6303831939311415,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00016633447895630284,
      "loss": 4.7339,
      "mean_token_accuracy": 0.23105303943157196,
      "num_tokens": 147819310.0,
      "step": 85220
    },
    {
      "entropy": 5.401655387878418,
      "epoch": 6.630772223302859,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001663102108041554,
      "loss": 4.6773,
      "mean_token_accuracy": 0.2368190884590149,
      "num_tokens": 147828255.0,
      "step": 85225
    },
    {
      "entropy": 5.426845216751099,
      "epoch": 6.631161252674577,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016628594430120262,
      "loss": 4.7457,
      "mean_token_accuracy": 0.23459971845149993,
      "num_tokens": 147837309.0,
      "step": 85230
    },
    {
      "entropy": 5.361053085327148,
      "epoch": 6.631550282046295,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016626167944781248,
      "loss": 4.5661,
      "mean_token_accuracy": 0.24625769555568694,
      "num_tokens": 147845869.0,
      "step": 85235
    },
    {
      "entropy": 5.425272703170776,
      "epoch": 6.631939311418012,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001662374162443534,
      "loss": 4.694,
      "mean_token_accuracy": 0.24079749584197999,
      "num_tokens": 147854025.0,
      "step": 85240
    },
    {
      "entropy": 5.388816165924072,
      "epoch": 6.63232834078973,
      "grad_norm": 1.375,
      "learning_rate": 0.0001662131546911933,
      "loss": 4.6375,
      "mean_token_accuracy": 0.24436034858226777,
      "num_tokens": 147862219.0,
      "step": 85245
    },
    {
      "entropy": 5.348510313034057,
      "epoch": 6.632717370161447,
      "grad_norm": 1.375,
      "learning_rate": 0.00016618889478870052,
      "loss": 4.6145,
      "mean_token_accuracy": 0.24737160354852678,
      "num_tokens": 147870585.0,
      "step": 85250
    },
    {
      "entropy": 5.37085452079773,
      "epoch": 6.6331063995331645,
      "grad_norm": 1.34375,
      "learning_rate": 0.000166164636537243,
      "loss": 4.6513,
      "mean_token_accuracy": 0.24065429270267485,
      "num_tokens": 147878969.0,
      "step": 85255
    },
    {
      "entropy": 5.387342882156372,
      "epoch": 6.633495428904882,
      "grad_norm": 1.375,
      "learning_rate": 0.00016614037993718895,
      "loss": 4.6252,
      "mean_token_accuracy": 0.23824930787086487,
      "num_tokens": 147887186.0,
      "step": 85260
    },
    {
      "entropy": 5.357514667510986,
      "epoch": 6.6338844582766,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001661161249889063,
      "loss": 4.6304,
      "mean_token_accuracy": 0.24571343660354614,
      "num_tokens": 147896151.0,
      "step": 85265
    },
    {
      "entropy": 5.421031141281128,
      "epoch": 6.634273487648318,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016609187169276324,
      "loss": 4.759,
      "mean_token_accuracy": 0.23388646245002748,
      "num_tokens": 147904707.0,
      "step": 85270
    },
    {
      "entropy": 5.383432388305664,
      "epoch": 6.634662517020035,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016606762004912762,
      "loss": 4.6627,
      "mean_token_accuracy": 0.24441962391138078,
      "num_tokens": 147912871.0,
      "step": 85275
    },
    {
      "entropy": 5.4062706470489506,
      "epoch": 6.635051546391752,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016604337005836752,
      "loss": 4.7174,
      "mean_token_accuracy": 0.24571072161197663,
      "num_tokens": 147921762.0,
      "step": 85280
    },
    {
      "entropy": 5.427821063995362,
      "epoch": 6.63544057576347,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016601912172085082,
      "loss": 4.6764,
      "mean_token_accuracy": 0.23166516274213791,
      "num_tokens": 147930781.0,
      "step": 85285
    },
    {
      "entropy": 5.444191026687622,
      "epoch": 6.6358296051351875,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016599487503694545,
      "loss": 4.6468,
      "mean_token_accuracy": 0.24590632915496827,
      "num_tokens": 147940029.0,
      "step": 85290
    },
    {
      "entropy": 5.423157644271851,
      "epoch": 6.636218634506905,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016597063000701936,
      "loss": 4.6604,
      "mean_token_accuracy": 0.2413393720984459,
      "num_tokens": 147948597.0,
      "step": 85295
    },
    {
      "entropy": 5.392226362228394,
      "epoch": 6.636607663878623,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001659463866314404,
      "loss": 4.6837,
      "mean_token_accuracy": 0.2324478045105934,
      "num_tokens": 147957030.0,
      "step": 85300
    },
    {
      "entropy": 5.336175489425659,
      "epoch": 6.636996693250341,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016592214491057653,
      "loss": 4.6257,
      "mean_token_accuracy": 0.23795036673545839,
      "num_tokens": 147965757.0,
      "step": 85305
    },
    {
      "entropy": 5.338454627990723,
      "epoch": 6.637385722622058,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016589790484479553,
      "loss": 4.6282,
      "mean_token_accuracy": 0.2338057965040207,
      "num_tokens": 147974334.0,
      "step": 85310
    },
    {
      "entropy": 5.358457803726196,
      "epoch": 6.637774751993776,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016587366643446504,
      "loss": 4.6374,
      "mean_token_accuracy": 0.24028546810150148,
      "num_tokens": 147983122.0,
      "step": 85315
    },
    {
      "entropy": 5.385693883895874,
      "epoch": 6.638163781365493,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016584942967995308,
      "loss": 4.6142,
      "mean_token_accuracy": 0.2455917313694954,
      "num_tokens": 147991834.0,
      "step": 85320
    },
    {
      "entropy": 5.396356678009033,
      "epoch": 6.6385528107372105,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001658251945816272,
      "loss": 4.6337,
      "mean_token_accuracy": 0.24554928094148637,
      "num_tokens": 148001534.0,
      "step": 85325
    },
    {
      "entropy": 5.354276561737061,
      "epoch": 6.638941840108928,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016580096113985536,
      "loss": 4.5468,
      "mean_token_accuracy": 0.2516625002026558,
      "num_tokens": 148009787.0,
      "step": 85330
    },
    {
      "entropy": 5.287834119796753,
      "epoch": 6.639330869480646,
      "grad_norm": 1.328125,
      "learning_rate": 0.000165776729355005,
      "loss": 4.5867,
      "mean_token_accuracy": 0.24549387991428376,
      "num_tokens": 148018538.0,
      "step": 85335
    },
    {
      "entropy": 5.375620746612549,
      "epoch": 6.639719898852364,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016575249922744402,
      "loss": 4.7084,
      "mean_token_accuracy": 0.2332543283700943,
      "num_tokens": 148027375.0,
      "step": 85340
    },
    {
      "entropy": 5.407729578018189,
      "epoch": 6.640108928224081,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016572827075753987,
      "loss": 4.6124,
      "mean_token_accuracy": 0.24296474158763887,
      "num_tokens": 148035638.0,
      "step": 85345
    },
    {
      "entropy": 5.324307823181153,
      "epoch": 6.640497957595798,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001657040439456604,
      "loss": 4.4712,
      "mean_token_accuracy": 0.25399318933486936,
      "num_tokens": 148044458.0,
      "step": 85350
    },
    {
      "entropy": 5.339103698730469,
      "epoch": 6.640886986967516,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00016567981879217307,
      "loss": 4.6416,
      "mean_token_accuracy": 0.24069348573684693,
      "num_tokens": 148054046.0,
      "step": 85355
    },
    {
      "entropy": 5.3384088516235355,
      "epoch": 6.6412760163392335,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001656555952974453,
      "loss": 4.6307,
      "mean_token_accuracy": 0.24218022227287292,
      "num_tokens": 148062590.0,
      "step": 85360
    },
    {
      "entropy": 5.323037624359131,
      "epoch": 6.641665045710951,
      "grad_norm": 1.25,
      "learning_rate": 0.00016563137346184504,
      "loss": 4.6383,
      "mean_token_accuracy": 0.2352748543024063,
      "num_tokens": 148071039.0,
      "step": 85365
    },
    {
      "entropy": 5.353954267501831,
      "epoch": 6.642054075082669,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016560715328573953,
      "loss": 4.6459,
      "mean_token_accuracy": 0.23778789937496186,
      "num_tokens": 148079579.0,
      "step": 85370
    },
    {
      "entropy": 5.347445821762085,
      "epoch": 6.642443104454387,
      "grad_norm": 1.5,
      "learning_rate": 0.00016558293476949643,
      "loss": 4.6571,
      "mean_token_accuracy": 0.24544587284326552,
      "num_tokens": 148088389.0,
      "step": 85375
    },
    {
      "entropy": 5.380689477920532,
      "epoch": 6.642832133826104,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016555871791348303,
      "loss": 4.6538,
      "mean_token_accuracy": 0.24249397665262223,
      "num_tokens": 148097558.0,
      "step": 85380
    },
    {
      "entropy": 5.3454516410827635,
      "epoch": 6.643221163197821,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016553450271806702,
      "loss": 4.5319,
      "mean_token_accuracy": 0.24773604720830916,
      "num_tokens": 148105803.0,
      "step": 85385
    },
    {
      "entropy": 5.315246105194092,
      "epoch": 6.643610192569539,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00016551028918361567,
      "loss": 4.5955,
      "mean_token_accuracy": 0.23978111743927003,
      "num_tokens": 148114209.0,
      "step": 85390
    },
    {
      "entropy": 5.38988037109375,
      "epoch": 6.6439992219412565,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016548607731049634,
      "loss": 4.7168,
      "mean_token_accuracy": 0.23563034534454347,
      "num_tokens": 148122461.0,
      "step": 85395
    },
    {
      "entropy": 5.417348861694336,
      "epoch": 6.644388251312974,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00016546186709907656,
      "loss": 4.7061,
      "mean_token_accuracy": 0.23760205507278442,
      "num_tokens": 148131806.0,
      "step": 85400
    },
    {
      "entropy": 5.412226629257202,
      "epoch": 6.644777280684692,
      "grad_norm": 1.4375,
      "learning_rate": 0.00016543765854972354,
      "loss": 4.7454,
      "mean_token_accuracy": 0.23512753695249558,
      "num_tokens": 148140869.0,
      "step": 85405
    },
    {
      "entropy": 5.344978761672974,
      "epoch": 6.64516631005641,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016541345166280467,
      "loss": 4.5716,
      "mean_token_accuracy": 0.2509961470961571,
      "num_tokens": 148149676.0,
      "step": 85410
    },
    {
      "entropy": 5.295153045654297,
      "epoch": 6.645555339428126,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001653892464386872,
      "loss": 4.4918,
      "mean_token_accuracy": 0.250806125998497,
      "num_tokens": 148157926.0,
      "step": 85415
    },
    {
      "entropy": 5.313762950897217,
      "epoch": 6.645944368799844,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016536504287773858,
      "loss": 4.6046,
      "mean_token_accuracy": 0.24608051925897598,
      "num_tokens": 148167151.0,
      "step": 85420
    },
    {
      "entropy": 5.338822269439698,
      "epoch": 6.646333398171562,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001653408409803258,
      "loss": 4.6145,
      "mean_token_accuracy": 0.2462860658764839,
      "num_tokens": 148175557.0,
      "step": 85425
    },
    {
      "entropy": 5.404410696029663,
      "epoch": 6.6467224275432795,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00016531664074681624,
      "loss": 4.6917,
      "mean_token_accuracy": 0.22996897548437117,
      "num_tokens": 148184278.0,
      "step": 85430
    },
    {
      "entropy": 5.353729248046875,
      "epoch": 6.647111456914997,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016529244217757718,
      "loss": 4.5939,
      "mean_token_accuracy": 0.24608834087848663,
      "num_tokens": 148192778.0,
      "step": 85435
    },
    {
      "entropy": 5.356325674057007,
      "epoch": 6.647500486286715,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016526824527297553,
      "loss": 4.6049,
      "mean_token_accuracy": 0.24799669831991195,
      "num_tokens": 148202376.0,
      "step": 85440
    },
    {
      "entropy": 5.3778783798217775,
      "epoch": 6.647889515658433,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016524405003337877,
      "loss": 4.6661,
      "mean_token_accuracy": 0.23516835421323776,
      "num_tokens": 148210266.0,
      "step": 85445
    },
    {
      "entropy": 5.383705472946167,
      "epoch": 6.64827854503015,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001652198564591537,
      "loss": 4.6896,
      "mean_token_accuracy": 0.23982729464769365,
      "num_tokens": 148218820.0,
      "step": 85450
    },
    {
      "entropy": 5.477152252197266,
      "epoch": 6.648667574401867,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016519566455066773,
      "loss": 4.7463,
      "mean_token_accuracy": 0.23034174144268035,
      "num_tokens": 148227737.0,
      "step": 85455
    },
    {
      "entropy": 5.442525339126587,
      "epoch": 6.649056603773585,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016517147430828767,
      "loss": 4.6808,
      "mean_token_accuracy": 0.23237675279378892,
      "num_tokens": 148236963.0,
      "step": 85460
    },
    {
      "entropy": 5.390676212310791,
      "epoch": 6.6494456331453025,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00016514728573238077,
      "loss": 4.6116,
      "mean_token_accuracy": 0.2443060517311096,
      "num_tokens": 148246374.0,
      "step": 85465
    },
    {
      "entropy": 5.358269357681275,
      "epoch": 6.64983466251702,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016512309882331394,
      "loss": 4.6515,
      "mean_token_accuracy": 0.2373913824558258,
      "num_tokens": 148254920.0,
      "step": 85470
    },
    {
      "entropy": 5.357132291793823,
      "epoch": 6.650223691888738,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016509891358145414,
      "loss": 4.644,
      "mean_token_accuracy": 0.23675850927829742,
      "num_tokens": 148263322.0,
      "step": 85475
    },
    {
      "entropy": 5.331697750091553,
      "epoch": 6.6506127212604556,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016507473000716843,
      "loss": 4.6169,
      "mean_token_accuracy": 0.2437201127409935,
      "num_tokens": 148272221.0,
      "step": 85480
    },
    {
      "entropy": 5.364223861694336,
      "epoch": 6.651001750632172,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001650505481008237,
      "loss": 4.6644,
      "mean_token_accuracy": 0.23907559961080552,
      "num_tokens": 148281086.0,
      "step": 85485
    },
    {
      "entropy": 5.355470132827759,
      "epoch": 6.65139078000389,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00016502636786278686,
      "loss": 4.5965,
      "mean_token_accuracy": 0.23912649899721145,
      "num_tokens": 148289257.0,
      "step": 85490
    },
    {
      "entropy": 5.343170785903931,
      "epoch": 6.651779809375608,
      "grad_norm": 1.25,
      "learning_rate": 0.00016500218929342496,
      "loss": 4.6508,
      "mean_token_accuracy": 0.23619042783975602,
      "num_tokens": 148298188.0,
      "step": 85495
    },
    {
      "entropy": 5.3839606761932375,
      "epoch": 6.6521688387473255,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00016497801239310464,
      "loss": 4.7004,
      "mean_token_accuracy": 0.2332440733909607,
      "num_tokens": 148307374.0,
      "step": 85500
    },
    {
      "entropy": 5.399441242218018,
      "epoch": 6.652557868119043,
      "grad_norm": 1.34375,
      "learning_rate": 0.000164953837162193,
      "loss": 4.6554,
      "mean_token_accuracy": 0.23956432640552522,
      "num_tokens": 148316059.0,
      "step": 85505
    },
    {
      "entropy": 5.50105299949646,
      "epoch": 6.652946897490761,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001649296636010566,
      "loss": 4.7939,
      "mean_token_accuracy": 0.22809799313545226,
      "num_tokens": 148325429.0,
      "step": 85510
    },
    {
      "entropy": 5.450500583648681,
      "epoch": 6.6533359268624785,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016490549171006247,
      "loss": 4.6948,
      "mean_token_accuracy": 0.23416401147842408,
      "num_tokens": 148334540.0,
      "step": 85515
    },
    {
      "entropy": 5.413470029830933,
      "epoch": 6.653724956234195,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001648813214895772,
      "loss": 4.6656,
      "mean_token_accuracy": 0.24416756331920625,
      "num_tokens": 148343211.0,
      "step": 85520
    },
    {
      "entropy": 5.305610036849975,
      "epoch": 6.654113985605913,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016485715293996763,
      "loss": 4.5425,
      "mean_token_accuracy": 0.24662694334983826,
      "num_tokens": 148351899.0,
      "step": 85525
    },
    {
      "entropy": 5.343322849273681,
      "epoch": 6.654503014977631,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001648329860616004,
      "loss": 4.6233,
      "mean_token_accuracy": 0.24089870005846023,
      "num_tokens": 148360220.0,
      "step": 85530
    },
    {
      "entropy": 5.306836557388306,
      "epoch": 6.6548920443493484,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016480882085484233,
      "loss": 4.5761,
      "mean_token_accuracy": 0.24908214807510376,
      "num_tokens": 148368867.0,
      "step": 85535
    },
    {
      "entropy": 5.420171546936035,
      "epoch": 6.655281073721066,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016478465732006,
      "loss": 4.7045,
      "mean_token_accuracy": 0.22974404990673064,
      "num_tokens": 148377201.0,
      "step": 85540
    },
    {
      "entropy": 5.432074642181396,
      "epoch": 6.655670103092784,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016476049545762007,
      "loss": 4.6756,
      "mean_token_accuracy": 0.24211029708385468,
      "num_tokens": 148386022.0,
      "step": 85545
    },
    {
      "entropy": 5.4201983451843265,
      "epoch": 6.656059132464501,
      "grad_norm": 1.25,
      "learning_rate": 0.00016473633526788918,
      "loss": 4.6998,
      "mean_token_accuracy": 0.23235575258731841,
      "num_tokens": 148394937.0,
      "step": 85550
    },
    {
      "entropy": 5.502428293228149,
      "epoch": 6.656448161836218,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001647121767512338,
      "loss": 4.7974,
      "mean_token_accuracy": 0.23492847234010697,
      "num_tokens": 148403741.0,
      "step": 85555
    },
    {
      "entropy": 5.439122915267944,
      "epoch": 6.656837191207936,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016468801990802062,
      "loss": 4.6361,
      "mean_token_accuracy": 0.24071057438850402,
      "num_tokens": 148413014.0,
      "step": 85560
    },
    {
      "entropy": 5.423998308181763,
      "epoch": 6.657226220579654,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016466386473861617,
      "loss": 4.697,
      "mean_token_accuracy": 0.24356161803007126,
      "num_tokens": 148421672.0,
      "step": 85565
    },
    {
      "entropy": 5.41186203956604,
      "epoch": 6.657615249951371,
      "grad_norm": 1.453125,
      "learning_rate": 0.000164639711243387,
      "loss": 4.7599,
      "mean_token_accuracy": 0.23111544102430343,
      "num_tokens": 148430810.0,
      "step": 85570
    },
    {
      "entropy": 5.347298574447632,
      "epoch": 6.658004279323089,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001646155594226996,
      "loss": 4.6207,
      "mean_token_accuracy": 0.244733028113842,
      "num_tokens": 148439463.0,
      "step": 85575
    },
    {
      "entropy": 5.417573928833008,
      "epoch": 6.658393308694807,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016459140927692028,
      "loss": 4.6392,
      "mean_token_accuracy": 0.24074405431747437,
      "num_tokens": 148448089.0,
      "step": 85580
    },
    {
      "entropy": 5.341724491119384,
      "epoch": 6.658782338066524,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016456726080641567,
      "loss": 4.5883,
      "mean_token_accuracy": 0.2562005206942558,
      "num_tokens": 148456580.0,
      "step": 85585
    },
    {
      "entropy": 5.322108364105224,
      "epoch": 6.659171367438241,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016454311401155206,
      "loss": 4.6359,
      "mean_token_accuracy": 0.23228022158145906,
      "num_tokens": 148465484.0,
      "step": 85590
    },
    {
      "entropy": 5.451080656051635,
      "epoch": 6.659560396809959,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016451896889269595,
      "loss": 4.6553,
      "mean_token_accuracy": 0.23936405777931213,
      "num_tokens": 148474050.0,
      "step": 85595
    },
    {
      "entropy": 5.345937299728393,
      "epoch": 6.659949426181677,
      "grad_norm": 1.375,
      "learning_rate": 0.00016449482545021353,
      "loss": 4.6233,
      "mean_token_accuracy": 0.23817439675331115,
      "num_tokens": 148482613.0,
      "step": 85600
    },
    {
      "entropy": 5.3842803001403805,
      "epoch": 6.660338455553394,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016447068368447137,
      "loss": 4.6232,
      "mean_token_accuracy": 0.24066583216190338,
      "num_tokens": 148491576.0,
      "step": 85605
    },
    {
      "entropy": 5.357517862319947,
      "epoch": 6.660727484925112,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001644465435958355,
      "loss": 4.6352,
      "mean_token_accuracy": 0.24488113969564437,
      "num_tokens": 148499915.0,
      "step": 85610
    },
    {
      "entropy": 5.345045709609986,
      "epoch": 6.661116514296829,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016442240518467252,
      "loss": 4.6233,
      "mean_token_accuracy": 0.24977737069129943,
      "num_tokens": 148509554.0,
      "step": 85615
    },
    {
      "entropy": 5.327544450759888,
      "epoch": 6.661505543668547,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016439826845134845,
      "loss": 4.5554,
      "mean_token_accuracy": 0.25001082122325896,
      "num_tokens": 148518366.0,
      "step": 85620
    },
    {
      "entropy": 5.370396280288697,
      "epoch": 6.661894573040264,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016437413339622962,
      "loss": 4.6422,
      "mean_token_accuracy": 0.24231666177511216,
      "num_tokens": 148527629.0,
      "step": 85625
    },
    {
      "entropy": 5.392465353012085,
      "epoch": 6.662283602411982,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016435000001968232,
      "loss": 4.6596,
      "mean_token_accuracy": 0.24017878472805024,
      "num_tokens": 148536641.0,
      "step": 85630
    },
    {
      "entropy": 5.3613543033599855,
      "epoch": 6.6626726317837,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00016432586832207253,
      "loss": 4.5991,
      "mean_token_accuracy": 0.24646140336990358,
      "num_tokens": 148545543.0,
      "step": 85635
    },
    {
      "entropy": 5.367738914489746,
      "epoch": 6.663061661155417,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016430173830376665,
      "loss": 4.6532,
      "mean_token_accuracy": 0.24289954602718353,
      "num_tokens": 148554706.0,
      "step": 85640
    },
    {
      "entropy": 5.410859680175781,
      "epoch": 6.663450690527135,
      "grad_norm": 1.375,
      "learning_rate": 0.00016427760996513057,
      "loss": 4.6744,
      "mean_token_accuracy": 0.24617674350738525,
      "num_tokens": 148562691.0,
      "step": 85645
    },
    {
      "entropy": 5.37871561050415,
      "epoch": 6.663839719898853,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016425348330653065,
      "loss": 4.6621,
      "mean_token_accuracy": 0.2414600670337677,
      "num_tokens": 148571040.0,
      "step": 85650
    },
    {
      "entropy": 5.42986536026001,
      "epoch": 6.66422874927057,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016422935832833286,
      "loss": 4.6992,
      "mean_token_accuracy": 0.23579649925231932,
      "num_tokens": 148580335.0,
      "step": 85655
    },
    {
      "entropy": 5.39847240447998,
      "epoch": 6.664617778642287,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001642052350309031,
      "loss": 4.6642,
      "mean_token_accuracy": 0.24066114127635957,
      "num_tokens": 148589634.0,
      "step": 85660
    },
    {
      "entropy": 5.407503700256347,
      "epoch": 6.665006808014005,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016418111341460773,
      "loss": 4.6594,
      "mean_token_accuracy": 0.24357927292585374,
      "num_tokens": 148598395.0,
      "step": 85665
    },
    {
      "entropy": 5.3296877384185795,
      "epoch": 6.665395837385723,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016415699347981243,
      "loss": 4.5856,
      "mean_token_accuracy": 0.24505438655614853,
      "num_tokens": 148606767.0,
      "step": 85670
    },
    {
      "entropy": 5.33421483039856,
      "epoch": 6.66578486675744,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016413287522688342,
      "loss": 4.5606,
      "mean_token_accuracy": 0.2471531003713608,
      "num_tokens": 148614802.0,
      "step": 85675
    },
    {
      "entropy": 5.4209237575531,
      "epoch": 6.666173896129158,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016410875865618645,
      "loss": 4.6837,
      "mean_token_accuracy": 0.23804503828287124,
      "num_tokens": 148623584.0,
      "step": 85680
    },
    {
      "entropy": 5.448467350006103,
      "epoch": 6.666562925500875,
      "grad_norm": 1.25,
      "learning_rate": 0.00016408464376808774,
      "loss": 4.6769,
      "mean_token_accuracy": 0.2375603809952736,
      "num_tokens": 148632110.0,
      "step": 85685
    },
    {
      "entropy": 5.315150165557862,
      "epoch": 6.666951954872593,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016406053056295286,
      "loss": 4.5388,
      "mean_token_accuracy": 0.257371523976326,
      "num_tokens": 148640507.0,
      "step": 85690
    },
    {
      "entropy": 5.367241144180298,
      "epoch": 6.66734098424431,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00016403641904114786,
      "loss": 4.6133,
      "mean_token_accuracy": 0.23839765340089797,
      "num_tokens": 148649200.0,
      "step": 85695
    },
    {
      "entropy": 5.3596227169036865,
      "epoch": 6.667730013616028,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001640123092030387,
      "loss": 4.6839,
      "mean_token_accuracy": 0.2451276734471321,
      "num_tokens": 148657551.0,
      "step": 85700
    },
    {
      "entropy": 5.420835399627686,
      "epoch": 6.668119042987746,
      "grad_norm": 1.21875,
      "learning_rate": 0.000163988201048991,
      "loss": 4.7776,
      "mean_token_accuracy": 0.22760249227285384,
      "num_tokens": 148666424.0,
      "step": 85705
    },
    {
      "entropy": 5.409557914733886,
      "epoch": 6.668508072359463,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016396409457937073,
      "loss": 4.7031,
      "mean_token_accuracy": 0.2373543992638588,
      "num_tokens": 148674971.0,
      "step": 85710
    },
    {
      "entropy": 5.320564842224121,
      "epoch": 6.668897101731181,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001639399897945435,
      "loss": 4.5422,
      "mean_token_accuracy": 0.2542125850915909,
      "num_tokens": 148682857.0,
      "step": 85715
    },
    {
      "entropy": 5.427887678146362,
      "epoch": 6.669286131102898,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016391588669487526,
      "loss": 4.7451,
      "mean_token_accuracy": 0.23499433547258378,
      "num_tokens": 148692406.0,
      "step": 85720
    },
    {
      "entropy": 5.417051124572754,
      "epoch": 6.669675160474616,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016389178528073152,
      "loss": 4.6291,
      "mean_token_accuracy": 0.2396501272916794,
      "num_tokens": 148700862.0,
      "step": 85725
    },
    {
      "entropy": 5.470126628875732,
      "epoch": 6.670064189846333,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00016386768555247822,
      "loss": 4.7626,
      "mean_token_accuracy": 0.23526046127080918,
      "num_tokens": 148708877.0,
      "step": 85730
    },
    {
      "entropy": 5.419955205917359,
      "epoch": 6.670453219218051,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016384358751048085,
      "loss": 4.6513,
      "mean_token_accuracy": 0.24558581858873368,
      "num_tokens": 148717239.0,
      "step": 85735
    },
    {
      "entropy": 5.42053256034851,
      "epoch": 6.670842248589769,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016381949115510503,
      "loss": 4.6264,
      "mean_token_accuracy": 0.24132100343704224,
      "num_tokens": 148726146.0,
      "step": 85740
    },
    {
      "entropy": 5.442404413223267,
      "epoch": 6.671231277961486,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016379539648671659,
      "loss": 4.6653,
      "mean_token_accuracy": 0.23514719009399415,
      "num_tokens": 148734941.0,
      "step": 85745
    },
    {
      "entropy": 5.419620180130005,
      "epoch": 6.671620307333203,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016377130350568093,
      "loss": 4.7596,
      "mean_token_accuracy": 0.2328447699546814,
      "num_tokens": 148743260.0,
      "step": 85750
    },
    {
      "entropy": 5.427387332916259,
      "epoch": 6.672009336704921,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016374721221236376,
      "loss": 4.6744,
      "mean_token_accuracy": 0.24535293132066727,
      "num_tokens": 148752171.0,
      "step": 85755
    },
    {
      "entropy": 5.39915452003479,
      "epoch": 6.672398366076639,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016372312260713043,
      "loss": 4.6804,
      "mean_token_accuracy": 0.23130587190389634,
      "num_tokens": 148760703.0,
      "step": 85760
    },
    {
      "entropy": 5.395666217803955,
      "epoch": 6.672787395448356,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016369903469034674,
      "loss": 4.6381,
      "mean_token_accuracy": 0.23776402026414872,
      "num_tokens": 148769357.0,
      "step": 85765
    },
    {
      "entropy": 5.423109722137451,
      "epoch": 6.673176424820074,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016367494846237805,
      "loss": 4.6875,
      "mean_token_accuracy": 0.23275558948516845,
      "num_tokens": 148777528.0,
      "step": 85770
    },
    {
      "entropy": 5.421376371383667,
      "epoch": 6.673565454191792,
      "grad_norm": 1.4375,
      "learning_rate": 0.00016365086392358975,
      "loss": 4.733,
      "mean_token_accuracy": 0.23971307426691055,
      "num_tokens": 148786106.0,
      "step": 85775
    },
    {
      "entropy": 5.376003265380859,
      "epoch": 6.673954483563509,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016362678107434745,
      "loss": 4.6502,
      "mean_token_accuracy": 0.23705280423164368,
      "num_tokens": 148795263.0,
      "step": 85780
    },
    {
      "entropy": 5.298519659042358,
      "epoch": 6.674343512935227,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016360269991501635,
      "loss": 4.5479,
      "mean_token_accuracy": 0.2490735098719597,
      "num_tokens": 148803861.0,
      "step": 85785
    },
    {
      "entropy": 5.358462190628051,
      "epoch": 6.674732542306944,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016357862044596208,
      "loss": 4.6323,
      "mean_token_accuracy": 0.23823629021644593,
      "num_tokens": 148813023.0,
      "step": 85790
    },
    {
      "entropy": 5.373043489456177,
      "epoch": 6.675121571678662,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001635545426675498,
      "loss": 4.6461,
      "mean_token_accuracy": 0.24289888739585877,
      "num_tokens": 148821873.0,
      "step": 85795
    },
    {
      "entropy": 5.3703758239746096,
      "epoch": 6.675510601050379,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001635304665801451,
      "loss": 4.5889,
      "mean_token_accuracy": 0.2529910862445831,
      "num_tokens": 148831140.0,
      "step": 85800
    },
    {
      "entropy": 5.426456069946289,
      "epoch": 6.675899630422097,
      "grad_norm": 1.375,
      "learning_rate": 0.00016350639218411304,
      "loss": 4.5917,
      "mean_token_accuracy": 0.24542993903160096,
      "num_tokens": 148840217.0,
      "step": 85805
    },
    {
      "entropy": 5.319226551055908,
      "epoch": 6.676288659793815,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016348231947981907,
      "loss": 4.5832,
      "mean_token_accuracy": 0.24326183646917343,
      "num_tokens": 148848837.0,
      "step": 85810
    },
    {
      "entropy": 5.346635580062866,
      "epoch": 6.6766776891655315,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00016345824846762842,
      "loss": 4.6292,
      "mean_token_accuracy": 0.24182516038417817,
      "num_tokens": 148857478.0,
      "step": 85815
    },
    {
      "entropy": 5.368225336074829,
      "epoch": 6.677066718537249,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001634341791479062,
      "loss": 4.6273,
      "mean_token_accuracy": 0.24243911802768708,
      "num_tokens": 148865284.0,
      "step": 85820
    },
    {
      "entropy": 5.432206392288208,
      "epoch": 6.677455747908967,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016341011152101783,
      "loss": 4.7275,
      "mean_token_accuracy": 0.23467491269111634,
      "num_tokens": 148874134.0,
      "step": 85825
    },
    {
      "entropy": 5.385148620605468,
      "epoch": 6.677844777280685,
      "grad_norm": 1.21875,
      "learning_rate": 0.00016338604558732828,
      "loss": 4.6039,
      "mean_token_accuracy": 0.23824471831321717,
      "num_tokens": 148883070.0,
      "step": 85830
    },
    {
      "entropy": 5.346499490737915,
      "epoch": 6.678233806652402,
      "grad_norm": 1.171875,
      "learning_rate": 0.00016336198134720299,
      "loss": 4.6139,
      "mean_token_accuracy": 0.23946797847747803,
      "num_tokens": 148892093.0,
      "step": 85835
    },
    {
      "entropy": 5.411632537841797,
      "epoch": 6.67862283602412,
      "grad_norm": 1.578125,
      "learning_rate": 0.00016333791880100682,
      "loss": 4.7143,
      "mean_token_accuracy": 0.23436293452978135,
      "num_tokens": 148900450.0,
      "step": 85840
    },
    {
      "entropy": 5.387404060363769,
      "epoch": 6.679011865395838,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016331385794910515,
      "loss": 4.6635,
      "mean_token_accuracy": 0.24276360869407654,
      "num_tokens": 148909334.0,
      "step": 85845
    },
    {
      "entropy": 5.418674802780151,
      "epoch": 6.679400894767555,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016328979879186284,
      "loss": 4.7416,
      "mean_token_accuracy": 0.22215415239334108,
      "num_tokens": 148918333.0,
      "step": 85850
    },
    {
      "entropy": 5.437266874313354,
      "epoch": 6.679789924139272,
      "grad_norm": 1.28125,
      "learning_rate": 0.000163265741329645,
      "loss": 4.7139,
      "mean_token_accuracy": 0.2301343098282814,
      "num_tokens": 148928211.0,
      "step": 85855
    },
    {
      "entropy": 5.4409264087677,
      "epoch": 6.68017895351099,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00016324168556281678,
      "loss": 4.662,
      "mean_token_accuracy": 0.2368478611111641,
      "num_tokens": 148936870.0,
      "step": 85860
    },
    {
      "entropy": 5.389806652069092,
      "epoch": 6.680567982882708,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000163217631491743,
      "loss": 4.672,
      "mean_token_accuracy": 0.23888297528028488,
      "num_tokens": 148945171.0,
      "step": 85865
    },
    {
      "entropy": 5.432330226898193,
      "epoch": 6.680957012254425,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016319357911678885,
      "loss": 4.6445,
      "mean_token_accuracy": 0.23784528225660323,
      "num_tokens": 148953607.0,
      "step": 85870
    },
    {
      "entropy": 5.335233736038208,
      "epoch": 6.681346041626143,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000163169528438319,
      "loss": 4.5725,
      "mean_token_accuracy": 0.24724780768156052,
      "num_tokens": 148962611.0,
      "step": 85875
    },
    {
      "entropy": 5.366178369522094,
      "epoch": 6.681735070997861,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016314547945669872,
      "loss": 4.6726,
      "mean_token_accuracy": 0.23614899069070816,
      "num_tokens": 148970687.0,
      "step": 85880
    },
    {
      "entropy": 5.246035051345825,
      "epoch": 6.6821241003695775,
      "grad_norm": 1.375,
      "learning_rate": 0.0001631214321722927,
      "loss": 4.5004,
      "mean_token_accuracy": 0.25127984285354615,
      "num_tokens": 148979026.0,
      "step": 85885
    },
    {
      "entropy": 5.358155345916748,
      "epoch": 6.682513129741295,
      "grad_norm": 1.25,
      "learning_rate": 0.0001630973865854658,
      "loss": 4.6401,
      "mean_token_accuracy": 0.2361677184700966,
      "num_tokens": 148987533.0,
      "step": 85890
    },
    {
      "entropy": 5.3530029296875,
      "epoch": 6.682902159113013,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001630733426965831,
      "loss": 4.638,
      "mean_token_accuracy": 0.24057910591363907,
      "num_tokens": 148996389.0,
      "step": 85895
    },
    {
      "entropy": 5.362461519241333,
      "epoch": 6.683291188484731,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016304930050600915,
      "loss": 4.5832,
      "mean_token_accuracy": 0.24250404834747313,
      "num_tokens": 149004863.0,
      "step": 85900
    },
    {
      "entropy": 5.4327695846557615,
      "epoch": 6.683680217856448,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016302526001410895,
      "loss": 4.7061,
      "mean_token_accuracy": 0.230624458193779,
      "num_tokens": 149013887.0,
      "step": 85905
    },
    {
      "entropy": 5.438835525512696,
      "epoch": 6.684069247228166,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016300122122124712,
      "loss": 4.6904,
      "mean_token_accuracy": 0.23745026290416718,
      "num_tokens": 149022034.0,
      "step": 85910
    },
    {
      "entropy": 5.379136180877685,
      "epoch": 6.684458276599884,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001629771841277886,
      "loss": 4.7018,
      "mean_token_accuracy": 0.2290496975183487,
      "num_tokens": 149030690.0,
      "step": 85915
    },
    {
      "entropy": 5.403381443023681,
      "epoch": 6.6848473059716005,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016295314873409795,
      "loss": 4.6459,
      "mean_token_accuracy": 0.23669716268777846,
      "num_tokens": 149039414.0,
      "step": 85920
    },
    {
      "entropy": 5.38030424118042,
      "epoch": 6.685236335343318,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016292911504053997,
      "loss": 4.6919,
      "mean_token_accuracy": 0.241168549656868,
      "num_tokens": 149048044.0,
      "step": 85925
    },
    {
      "entropy": 5.3759987354278564,
      "epoch": 6.685625364715036,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016290508304747932,
      "loss": 4.5754,
      "mean_token_accuracy": 0.24588843137025834,
      "num_tokens": 149056992.0,
      "step": 85930
    },
    {
      "entropy": 5.3843772411346436,
      "epoch": 6.686014394086754,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001628810527552805,
      "loss": 4.6102,
      "mean_token_accuracy": 0.2502222806215286,
      "num_tokens": 149065070.0,
      "step": 85935
    },
    {
      "entropy": 5.398129892349243,
      "epoch": 6.686403423458471,
      "grad_norm": 1.375,
      "learning_rate": 0.00016285702416430836,
      "loss": 4.5734,
      "mean_token_accuracy": 0.25228031277656554,
      "num_tokens": 149073290.0,
      "step": 85940
    },
    {
      "entropy": 5.398529815673828,
      "epoch": 6.686792452830189,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001628329972749273,
      "loss": 4.664,
      "mean_token_accuracy": 0.23920369744300843,
      "num_tokens": 149081739.0,
      "step": 85945
    },
    {
      "entropy": 5.37217493057251,
      "epoch": 6.687181482201906,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000162808972087502,
      "loss": 4.6423,
      "mean_token_accuracy": 0.24132792204618453,
      "num_tokens": 149090392.0,
      "step": 85950
    },
    {
      "entropy": 5.308342313766479,
      "epoch": 6.6875705115736235,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016278494860239696,
      "loss": 4.5404,
      "mean_token_accuracy": 0.24768907725811004,
      "num_tokens": 149098903.0,
      "step": 85955
    },
    {
      "entropy": 5.370001363754272,
      "epoch": 6.687959540945341,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016276092681997666,
      "loss": 4.7397,
      "mean_token_accuracy": 0.23736969530582427,
      "num_tokens": 149107935.0,
      "step": 85960
    },
    {
      "entropy": 5.313495731353759,
      "epoch": 6.688348570317059,
      "grad_norm": 1.375,
      "learning_rate": 0.00016273690674060583,
      "loss": 4.5459,
      "mean_token_accuracy": 0.2537378206849098,
      "num_tokens": 149116384.0,
      "step": 85965
    },
    {
      "entropy": 5.3542193412780765,
      "epoch": 6.688737599688777,
      "grad_norm": 1.25,
      "learning_rate": 0.0001627128883646486,
      "loss": 4.6013,
      "mean_token_accuracy": 0.24025869518518447,
      "num_tokens": 149124316.0,
      "step": 85970
    },
    {
      "entropy": 5.389502477645874,
      "epoch": 6.689126629060494,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016268887169246964,
      "loss": 4.6304,
      "mean_token_accuracy": 0.23938551545143127,
      "num_tokens": 149133555.0,
      "step": 85975
    },
    {
      "entropy": 5.401685571670532,
      "epoch": 6.689515658432212,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016266485672443323,
      "loss": 4.6523,
      "mean_token_accuracy": 0.237826144695282,
      "num_tokens": 149142067.0,
      "step": 85980
    },
    {
      "entropy": 5.445377922058105,
      "epoch": 6.68990468780393,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001626408434609039,
      "loss": 4.6876,
      "mean_token_accuracy": 0.2407173901796341,
      "num_tokens": 149150208.0,
      "step": 85985
    },
    {
      "entropy": 5.379765510559082,
      "epoch": 6.6902937171756465,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016261683190224585,
      "loss": 4.6337,
      "mean_token_accuracy": 0.24098272919654845,
      "num_tokens": 149159221.0,
      "step": 85990
    },
    {
      "entropy": 5.372735071182251,
      "epoch": 6.690682746547364,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016259282204882358,
      "loss": 4.5571,
      "mean_token_accuracy": 0.250486795604229,
      "num_tokens": 149167180.0,
      "step": 85995
    },
    {
      "entropy": 5.4077043533325195,
      "epoch": 6.691071775919082,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00016256881390100123,
      "loss": 4.6849,
      "mean_token_accuracy": 0.23053427338600158,
      "num_tokens": 149175654.0,
      "step": 86000
    },
    {
      "entropy": 5.312512636184692,
      "epoch": 6.6914608052908,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016254480745914326,
      "loss": 4.6118,
      "mean_token_accuracy": 0.24819298535585405,
      "num_tokens": 149184748.0,
      "step": 86005
    },
    {
      "entropy": 5.354326486587524,
      "epoch": 6.691849834662517,
      "grad_norm": 1.21875,
      "learning_rate": 0.00016252080272361385,
      "loss": 4.5854,
      "mean_token_accuracy": 0.2381861314177513,
      "num_tokens": 149193747.0,
      "step": 86010
    },
    {
      "entropy": 5.507125616073608,
      "epoch": 6.692238864034235,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016249679969477715,
      "loss": 4.7088,
      "mean_token_accuracy": 0.23161526620388032,
      "num_tokens": 149202330.0,
      "step": 86015
    },
    {
      "entropy": 5.38338680267334,
      "epoch": 6.692627893405952,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00016247279837299755,
      "loss": 4.589,
      "mean_token_accuracy": 0.24806948006153107,
      "num_tokens": 149210757.0,
      "step": 86020
    },
    {
      "entropy": 5.3589647769927975,
      "epoch": 6.6930169227776695,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016244879875863894,
      "loss": 4.6163,
      "mean_token_accuracy": 0.2430060788989067,
      "num_tokens": 149218784.0,
      "step": 86025
    },
    {
      "entropy": 5.253366565704345,
      "epoch": 6.693405952149387,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016242480085206583,
      "loss": 4.4951,
      "mean_token_accuracy": 0.2550062254071236,
      "num_tokens": 149226954.0,
      "step": 86030
    },
    {
      "entropy": 5.332148170471191,
      "epoch": 6.693794981521105,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001624008046536422,
      "loss": 4.5945,
      "mean_token_accuracy": 0.24648851603269578,
      "num_tokens": 149235385.0,
      "step": 86035
    },
    {
      "entropy": 5.392679500579834,
      "epoch": 6.694184010892823,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016237681016373205,
      "loss": 4.6719,
      "mean_token_accuracy": 0.24645027369260789,
      "num_tokens": 149244357.0,
      "step": 86040
    },
    {
      "entropy": 5.387827634811401,
      "epoch": 6.69457304026454,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001623528173826996,
      "loss": 4.6405,
      "mean_token_accuracy": 0.24245555698871613,
      "num_tokens": 149253118.0,
      "step": 86045
    },
    {
      "entropy": 5.320360231399536,
      "epoch": 6.694962069636258,
      "grad_norm": 1.203125,
      "learning_rate": 0.00016232882631090877,
      "loss": 4.5018,
      "mean_token_accuracy": 0.2471695899963379,
      "num_tokens": 149262388.0,
      "step": 86050
    },
    {
      "entropy": 5.3258678913116455,
      "epoch": 6.695351099007975,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016230483694872384,
      "loss": 4.5682,
      "mean_token_accuracy": 0.24792979955673217,
      "num_tokens": 149270264.0,
      "step": 86055
    },
    {
      "entropy": 5.285712671279907,
      "epoch": 6.6957401283796925,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016228084929650847,
      "loss": 4.6234,
      "mean_token_accuracy": 0.24342343658208848,
      "num_tokens": 149279331.0,
      "step": 86060
    },
    {
      "entropy": 5.402322435379029,
      "epoch": 6.69612915775141,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001622568633546269,
      "loss": 4.6747,
      "mean_token_accuracy": 0.23903605043888093,
      "num_tokens": 149288391.0,
      "step": 86065
    },
    {
      "entropy": 5.39708948135376,
      "epoch": 6.696518187123128,
      "grad_norm": 1.375,
      "learning_rate": 0.0001622328791234429,
      "loss": 4.6023,
      "mean_token_accuracy": 0.24187932312488555,
      "num_tokens": 149297067.0,
      "step": 86070
    },
    {
      "entropy": 5.382947158813477,
      "epoch": 6.6969072164948455,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00016220889660332054,
      "loss": 4.6348,
      "mean_token_accuracy": 0.24589806497097016,
      "num_tokens": 149306223.0,
      "step": 86075
    },
    {
      "entropy": 5.402849960327148,
      "epoch": 6.697296245866563,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016218491579462357,
      "loss": 4.6841,
      "mean_token_accuracy": 0.24375319182872773,
      "num_tokens": 149315283.0,
      "step": 86080
    },
    {
      "entropy": 5.368256282806397,
      "epoch": 6.69768527523828,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016216093669771603,
      "loss": 4.5991,
      "mean_token_accuracy": 0.2439291626214981,
      "num_tokens": 149323212.0,
      "step": 86085
    },
    {
      "entropy": 5.421236133575439,
      "epoch": 6.698074304609998,
      "grad_norm": 1.4375,
      "learning_rate": 0.00016213695931296157,
      "loss": 4.7446,
      "mean_token_accuracy": 0.23276620209217072,
      "num_tokens": 149331609.0,
      "step": 86090
    },
    {
      "entropy": 5.409360361099243,
      "epoch": 6.6984633339817155,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00016211298364072417,
      "loss": 4.7125,
      "mean_token_accuracy": 0.2296522468328476,
      "num_tokens": 149340182.0,
      "step": 86095
    },
    {
      "entropy": 5.418511486053466,
      "epoch": 6.698852363353433,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00016208900968136758,
      "loss": 4.7145,
      "mean_token_accuracy": 0.23892219960689545,
      "num_tokens": 149349739.0,
      "step": 86100
    },
    {
      "entropy": 5.4192602157592775,
      "epoch": 6.699241392725151,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00016206503743525553,
      "loss": 4.7012,
      "mean_token_accuracy": 0.23899112790822982,
      "num_tokens": 149358706.0,
      "step": 86105
    },
    {
      "entropy": 5.390712738037109,
      "epoch": 6.6996304220968685,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016204106690275184,
      "loss": 4.6913,
      "mean_token_accuracy": 0.2354682594537735,
      "num_tokens": 149367341.0,
      "step": 86110
    },
    {
      "entropy": 5.29554123878479,
      "epoch": 6.700019451468586,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016201709808422017,
      "loss": 4.52,
      "mean_token_accuracy": 0.2510311409831047,
      "num_tokens": 149375813.0,
      "step": 86115
    },
    {
      "entropy": 5.359553718566895,
      "epoch": 6.700408480840304,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016199313098002411,
      "loss": 4.6276,
      "mean_token_accuracy": 0.2417372062802315,
      "num_tokens": 149384780.0,
      "step": 86120
    },
    {
      "entropy": 5.3716155052185055,
      "epoch": 6.700797510212021,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001619691655905275,
      "loss": 4.5831,
      "mean_token_accuracy": 0.2404239609837532,
      "num_tokens": 149393519.0,
      "step": 86125
    },
    {
      "entropy": 5.410480260848999,
      "epoch": 6.701186539583738,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001619452019160938,
      "loss": 4.6165,
      "mean_token_accuracy": 0.24483379125595092,
      "num_tokens": 149401328.0,
      "step": 86130
    },
    {
      "entropy": 5.436933135986328,
      "epoch": 6.701575568955456,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016192123995708684,
      "loss": 4.6834,
      "mean_token_accuracy": 0.2423110470175743,
      "num_tokens": 149409737.0,
      "step": 86135
    },
    {
      "entropy": 5.429930210113525,
      "epoch": 6.701964598327174,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016189727971386997,
      "loss": 4.722,
      "mean_token_accuracy": 0.2287255823612213,
      "num_tokens": 149418030.0,
      "step": 86140
    },
    {
      "entropy": 5.392004156112671,
      "epoch": 6.7023536276988915,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016187332118680696,
      "loss": 4.6323,
      "mean_token_accuracy": 0.24038649201393128,
      "num_tokens": 149426387.0,
      "step": 86145
    },
    {
      "entropy": 5.309703254699707,
      "epoch": 6.702742657070608,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016184936437626114,
      "loss": 4.5494,
      "mean_token_accuracy": 0.24159204214811325,
      "num_tokens": 149434656.0,
      "step": 86150
    },
    {
      "entropy": 5.306029796600342,
      "epoch": 6.703131686442326,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016182540928259609,
      "loss": 4.5843,
      "mean_token_accuracy": 0.2345937192440033,
      "num_tokens": 149443445.0,
      "step": 86155
    },
    {
      "entropy": 5.351042079925537,
      "epoch": 6.703520715814044,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016180145590617546,
      "loss": 4.5614,
      "mean_token_accuracy": 0.24590293169021607,
      "num_tokens": 149452262.0,
      "step": 86160
    },
    {
      "entropy": 5.368657875061035,
      "epoch": 6.703909745185761,
      "grad_norm": 1.375,
      "learning_rate": 0.00016177750424736242,
      "loss": 4.6734,
      "mean_token_accuracy": 0.23903552293777466,
      "num_tokens": 149460480.0,
      "step": 86165
    },
    {
      "entropy": 5.28786187171936,
      "epoch": 6.704298774557479,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016175355430652065,
      "loss": 4.5946,
      "mean_token_accuracy": 0.25015025585889816,
      "num_tokens": 149469729.0,
      "step": 86170
    },
    {
      "entropy": 5.3523763656616214,
      "epoch": 6.704687803929197,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00016172960608401338,
      "loss": 4.58,
      "mean_token_accuracy": 0.2507786974310875,
      "num_tokens": 149478200.0,
      "step": 86175
    },
    {
      "entropy": 5.389894676208496,
      "epoch": 6.7050768333009145,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016170565958020412,
      "loss": 4.6707,
      "mean_token_accuracy": 0.2401902422308922,
      "num_tokens": 149486746.0,
      "step": 86180
    },
    {
      "entropy": 5.292961311340332,
      "epoch": 6.705465862672632,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016168171479545606,
      "loss": 4.5446,
      "mean_token_accuracy": 0.2491212472319603,
      "num_tokens": 149495550.0,
      "step": 86185
    },
    {
      "entropy": 5.359263372421265,
      "epoch": 6.705854892044349,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016165777173013273,
      "loss": 4.7099,
      "mean_token_accuracy": 0.23626141846179963,
      "num_tokens": 149503950.0,
      "step": 86190
    },
    {
      "entropy": 5.292845630645752,
      "epoch": 6.706243921416067,
      "grad_norm": 1.203125,
      "learning_rate": 0.00016163383038459727,
      "loss": 4.5404,
      "mean_token_accuracy": 0.2490392580628395,
      "num_tokens": 149512197.0,
      "step": 86195
    },
    {
      "entropy": 5.331468534469605,
      "epoch": 6.706632950787784,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016160989075921295,
      "loss": 4.5823,
      "mean_token_accuracy": 0.24503006935119628,
      "num_tokens": 149520663.0,
      "step": 86200
    },
    {
      "entropy": 5.3185656547546385,
      "epoch": 6.707021980159502,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016158595285434307,
      "loss": 4.5664,
      "mean_token_accuracy": 0.24595999121665954,
      "num_tokens": 149529020.0,
      "step": 86205
    },
    {
      "entropy": 5.325,
      "epoch": 6.70741100953122,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016156201667035083,
      "loss": 4.5931,
      "mean_token_accuracy": 0.23842764496803284,
      "num_tokens": 149537945.0,
      "step": 86210
    },
    {
      "entropy": 5.338918876647949,
      "epoch": 6.7078000389029375,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016153808220759948,
      "loss": 4.631,
      "mean_token_accuracy": 0.24312269538640977,
      "num_tokens": 149546395.0,
      "step": 86215
    },
    {
      "entropy": 5.330662250518799,
      "epoch": 6.708189068274654,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016151414946645205,
      "loss": 4.6037,
      "mean_token_accuracy": 0.2547203108668327,
      "num_tokens": 149555131.0,
      "step": 86220
    },
    {
      "entropy": 5.357038497924805,
      "epoch": 6.708578097646372,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001614902184472718,
      "loss": 4.6225,
      "mean_token_accuracy": 0.24005321711301802,
      "num_tokens": 149564735.0,
      "step": 86225
    },
    {
      "entropy": 5.446113872528076,
      "epoch": 6.70896712701809,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001614662891504219,
      "loss": 4.6374,
      "mean_token_accuracy": 0.23702000826597214,
      "num_tokens": 149573210.0,
      "step": 86230
    },
    {
      "entropy": 5.383820390701294,
      "epoch": 6.709356156389807,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016144236157626524,
      "loss": 4.6581,
      "mean_token_accuracy": 0.23509249687194825,
      "num_tokens": 149582385.0,
      "step": 86235
    },
    {
      "entropy": 5.383309412002563,
      "epoch": 6.709745185761525,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016141843572516507,
      "loss": 4.6081,
      "mean_token_accuracy": 0.240153968334198,
      "num_tokens": 149590967.0,
      "step": 86240
    },
    {
      "entropy": 5.321812725067138,
      "epoch": 6.710134215133243,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001613945115974843,
      "loss": 4.5504,
      "mean_token_accuracy": 0.25157203525304794,
      "num_tokens": 149599613.0,
      "step": 86245
    },
    {
      "entropy": 5.47669358253479,
      "epoch": 6.7105232445049605,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016137058919358607,
      "loss": 4.8016,
      "mean_token_accuracy": 0.23862560242414474,
      "num_tokens": 149607900.0,
      "step": 86250
    },
    {
      "entropy": 5.3992087841033936,
      "epoch": 6.710912273876677,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016134666851383322,
      "loss": 4.6278,
      "mean_token_accuracy": 0.2462398052215576,
      "num_tokens": 149616443.0,
      "step": 86255
    },
    {
      "entropy": 5.424568796157837,
      "epoch": 6.711301303248395,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016132274955858882,
      "loss": 4.6863,
      "mean_token_accuracy": 0.23676442950963975,
      "num_tokens": 149624962.0,
      "step": 86260
    },
    {
      "entropy": 5.428032493591308,
      "epoch": 6.711690332620113,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016129883232821575,
      "loss": 4.6669,
      "mean_token_accuracy": 0.23936725854873658,
      "num_tokens": 149634345.0,
      "step": 86265
    },
    {
      "entropy": 5.38296799659729,
      "epoch": 6.71207936199183,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016127491682307694,
      "loss": 4.682,
      "mean_token_accuracy": 0.24103991985321044,
      "num_tokens": 149643514.0,
      "step": 86270
    },
    {
      "entropy": 5.505573654174805,
      "epoch": 6.712468391363548,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016125100304353526,
      "loss": 4.808,
      "mean_token_accuracy": 0.2274618476629257,
      "num_tokens": 149652298.0,
      "step": 86275
    },
    {
      "entropy": 5.346088886260986,
      "epoch": 6.712857420735266,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016122709098995353,
      "loss": 4.515,
      "mean_token_accuracy": 0.25398640632629393,
      "num_tokens": 149660483.0,
      "step": 86280
    },
    {
      "entropy": 5.446690416336059,
      "epoch": 6.713246450106983,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016120318066269464,
      "loss": 4.7529,
      "mean_token_accuracy": 0.2384564146399498,
      "num_tokens": 149669726.0,
      "step": 86285
    },
    {
      "entropy": 5.401085615158081,
      "epoch": 6.7136354794787,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001611792720621212,
      "loss": 4.6738,
      "mean_token_accuracy": 0.2367143377661705,
      "num_tokens": 149678521.0,
      "step": 86290
    },
    {
      "entropy": 5.361565685272216,
      "epoch": 6.714024508850418,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001611553651885963,
      "loss": 4.6054,
      "mean_token_accuracy": 0.24224963188171386,
      "num_tokens": 149687217.0,
      "step": 86295
    },
    {
      "entropy": 5.337229824066162,
      "epoch": 6.714413538222136,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016113146004248247,
      "loss": 4.6251,
      "mean_token_accuracy": 0.23943732529878617,
      "num_tokens": 149696274.0,
      "step": 86300
    },
    {
      "entropy": 5.446482086181641,
      "epoch": 6.714802567593853,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016110755662414257,
      "loss": 4.7229,
      "mean_token_accuracy": 0.2374487563967705,
      "num_tokens": 149704446.0,
      "step": 86305
    },
    {
      "entropy": 5.425869703292847,
      "epoch": 6.715191596965571,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016108365493393924,
      "loss": 4.6713,
      "mean_token_accuracy": 0.24062672406435012,
      "num_tokens": 149712640.0,
      "step": 86310
    },
    {
      "entropy": 5.426177597045898,
      "epoch": 6.715580626337289,
      "grad_norm": 1.28125,
      "learning_rate": 0.000161059754972235,
      "loss": 4.7074,
      "mean_token_accuracy": 0.23139062523841858,
      "num_tokens": 149720937.0,
      "step": 86315
    },
    {
      "entropy": 5.429522752761841,
      "epoch": 6.7159696557090065,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016103585673939275,
      "loss": 4.707,
      "mean_token_accuracy": 0.23763195872306825,
      "num_tokens": 149729337.0,
      "step": 86320
    },
    {
      "entropy": 5.410010576248169,
      "epoch": 6.716358685080723,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016101196023577487,
      "loss": 4.651,
      "mean_token_accuracy": 0.24321259409189225,
      "num_tokens": 149737842.0,
      "step": 86325
    },
    {
      "entropy": 5.297318696975708,
      "epoch": 6.716747714452441,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016098806546174416,
      "loss": 4.549,
      "mean_token_accuracy": 0.2521692648530006,
      "num_tokens": 149747319.0,
      "step": 86330
    },
    {
      "entropy": 5.309335565567016,
      "epoch": 6.717136743824159,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016096417241766298,
      "loss": 4.5255,
      "mean_token_accuracy": 0.2524402871727943,
      "num_tokens": 149755895.0,
      "step": 86335
    },
    {
      "entropy": 5.344257545471192,
      "epoch": 6.717525773195876,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016094028110389408,
      "loss": 4.6593,
      "mean_token_accuracy": 0.2338498279452324,
      "num_tokens": 149764885.0,
      "step": 86340
    },
    {
      "entropy": 5.285638332366943,
      "epoch": 6.717914802567594,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001609163915207998,
      "loss": 4.534,
      "mean_token_accuracy": 0.25285124182701113,
      "num_tokens": 149773877.0,
      "step": 86345
    },
    {
      "entropy": 5.3193916320800785,
      "epoch": 6.718303831939312,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016089250366874274,
      "loss": 4.625,
      "mean_token_accuracy": 0.23865541368722915,
      "num_tokens": 149782761.0,
      "step": 86350
    },
    {
      "entropy": 5.43023910522461,
      "epoch": 6.718692861311029,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00016086861754808525,
      "loss": 4.6994,
      "mean_token_accuracy": 0.23718641996383666,
      "num_tokens": 149791457.0,
      "step": 86355
    },
    {
      "entropy": 5.353921794891358,
      "epoch": 6.719081890682746,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00016084473315918985,
      "loss": 4.6278,
      "mean_token_accuracy": 0.2463632345199585,
      "num_tokens": 149800308.0,
      "step": 86360
    },
    {
      "entropy": 5.514244604110718,
      "epoch": 6.719470920054464,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016082085050241898,
      "loss": 4.8226,
      "mean_token_accuracy": 0.23010505139827728,
      "num_tokens": 149808875.0,
      "step": 86365
    },
    {
      "entropy": 5.385243082046509,
      "epoch": 6.719859949426182,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016079696957813493,
      "loss": 4.5596,
      "mean_token_accuracy": 0.2492030069231987,
      "num_tokens": 149817580.0,
      "step": 86370
    },
    {
      "entropy": 5.382873773574829,
      "epoch": 6.720248978797899,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00016077309038670012,
      "loss": 4.6563,
      "mean_token_accuracy": 0.23919832855463027,
      "num_tokens": 149826188.0,
      "step": 86375
    },
    {
      "entropy": 5.374216318130493,
      "epoch": 6.720638008169617,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016074921292847684,
      "loss": 4.6322,
      "mean_token_accuracy": 0.2425265282392502,
      "num_tokens": 149835306.0,
      "step": 86380
    },
    {
      "entropy": 5.416077947616577,
      "epoch": 6.721027037541335,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016072533720382743,
      "loss": 4.7322,
      "mean_token_accuracy": 0.231866317987442,
      "num_tokens": 149843750.0,
      "step": 86385
    },
    {
      "entropy": 5.344777393341064,
      "epoch": 6.721416066913052,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016070146321311415,
      "loss": 4.6471,
      "mean_token_accuracy": 0.2441350057721138,
      "num_tokens": 149852277.0,
      "step": 86390
    },
    {
      "entropy": 5.382677030563355,
      "epoch": 6.721805096284769,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016067759095669918,
      "loss": 4.6344,
      "mean_token_accuracy": 0.2446226716041565,
      "num_tokens": 149860608.0,
      "step": 86395
    },
    {
      "entropy": 5.4161317348480225,
      "epoch": 6.722194125656487,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00016065372043494485,
      "loss": 4.6617,
      "mean_token_accuracy": 0.24298195540905,
      "num_tokens": 149868943.0,
      "step": 86400
    },
    {
      "entropy": 5.392434310913086,
      "epoch": 6.722583155028205,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016062985164821327,
      "loss": 4.6821,
      "mean_token_accuracy": 0.23374230563640594,
      "num_tokens": 149877907.0,
      "step": 86405
    },
    {
      "entropy": 5.352866983413696,
      "epoch": 6.722972184399922,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001606059845968667,
      "loss": 4.5772,
      "mean_token_accuracy": 0.24947873800992965,
      "num_tokens": 149886360.0,
      "step": 86410
    },
    {
      "entropy": 5.460223627090454,
      "epoch": 6.72336121377164,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001605821192812672,
      "loss": 4.7749,
      "mean_token_accuracy": 0.23298993706703186,
      "num_tokens": 149894934.0,
      "step": 86415
    },
    {
      "entropy": 5.379236173629761,
      "epoch": 6.723750243143357,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016055825570177688,
      "loss": 4.6193,
      "mean_token_accuracy": 0.24513702988624572,
      "num_tokens": 149903274.0,
      "step": 86420
    },
    {
      "entropy": 5.326455354690552,
      "epoch": 6.724139272515075,
      "grad_norm": 1.296875,
      "learning_rate": 0.000160534393858758,
      "loss": 4.6098,
      "mean_token_accuracy": 0.24296549707651138,
      "num_tokens": 149911966.0,
      "step": 86425
    },
    {
      "entropy": 5.420417547225952,
      "epoch": 6.724528301886792,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016051053375257238,
      "loss": 4.6833,
      "mean_token_accuracy": 0.2304786503314972,
      "num_tokens": 149920337.0,
      "step": 86430
    },
    {
      "entropy": 5.329885959625244,
      "epoch": 6.72491733125851,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001604866753835823,
      "loss": 4.6031,
      "mean_token_accuracy": 0.24465828835964204,
      "num_tokens": 149929280.0,
      "step": 86435
    },
    {
      "entropy": 5.306013727188111,
      "epoch": 6.725306360630228,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016046281875214952,
      "loss": 4.5542,
      "mean_token_accuracy": 0.24593015164136886,
      "num_tokens": 149937773.0,
      "step": 86440
    },
    {
      "entropy": 5.447704553604126,
      "epoch": 6.725695390001945,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001604389638586363,
      "loss": 4.7621,
      "mean_token_accuracy": 0.23206159919500352,
      "num_tokens": 149946557.0,
      "step": 86445
    },
    {
      "entropy": 5.372105455398559,
      "epoch": 6.726084419373663,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016041511070340433,
      "loss": 4.5563,
      "mean_token_accuracy": 0.2557765647768974,
      "num_tokens": 149955346.0,
      "step": 86450
    },
    {
      "entropy": 5.410490369796753,
      "epoch": 6.72647344874538,
      "grad_norm": 1.375,
      "learning_rate": 0.0001603912592868158,
      "loss": 4.6701,
      "mean_token_accuracy": 0.24037319123744966,
      "num_tokens": 149964097.0,
      "step": 86455
    },
    {
      "entropy": 5.43944411277771,
      "epoch": 6.726862478117098,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016036740960923237,
      "loss": 4.7067,
      "mean_token_accuracy": 0.23584576398134233,
      "num_tokens": 149973355.0,
      "step": 86460
    },
    {
      "entropy": 5.4766068935394285,
      "epoch": 6.727251507488815,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00016034356167101617,
      "loss": 4.7104,
      "mean_token_accuracy": 0.24522684067487716,
      "num_tokens": 149981280.0,
      "step": 86465
    },
    {
      "entropy": 5.420857572555542,
      "epoch": 6.727640536860533,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016031971547252887,
      "loss": 4.7237,
      "mean_token_accuracy": 0.2279752567410469,
      "num_tokens": 149990261.0,
      "step": 86470
    },
    {
      "entropy": 5.40521092414856,
      "epoch": 6.728029566232251,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016029587101413232,
      "loss": 4.6498,
      "mean_token_accuracy": 0.24155280590057374,
      "num_tokens": 149999636.0,
      "step": 86475
    },
    {
      "entropy": 5.3785608291625975,
      "epoch": 6.728418595603968,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016027202829618846,
      "loss": 4.6335,
      "mean_token_accuracy": 0.24292868375778198,
      "num_tokens": 150008095.0,
      "step": 86480
    },
    {
      "entropy": 5.432597017288208,
      "epoch": 6.728807624975685,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016024818731905882,
      "loss": 4.735,
      "mean_token_accuracy": 0.23055571019649507,
      "num_tokens": 150017294.0,
      "step": 86485
    },
    {
      "entropy": 5.371450281143188,
      "epoch": 6.729196654347403,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016022434808310532,
      "loss": 4.6697,
      "mean_token_accuracy": 0.24857214391231536,
      "num_tokens": 150026686.0,
      "step": 86490
    },
    {
      "entropy": 5.386080074310303,
      "epoch": 6.729585683719121,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016020051058868973,
      "loss": 4.6339,
      "mean_token_accuracy": 0.23878349661827086,
      "num_tokens": 150034691.0,
      "step": 86495
    },
    {
      "entropy": 5.5105753421783445,
      "epoch": 6.729974713090838,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001601766748361736,
      "loss": 4.7967,
      "mean_token_accuracy": 0.22429289817810058,
      "num_tokens": 150043697.0,
      "step": 86500
    },
    {
      "entropy": 5.3892487525939945,
      "epoch": 6.730363742462556,
      "grad_norm": 1.25,
      "learning_rate": 0.00016015284082591873,
      "loss": 4.5944,
      "mean_token_accuracy": 0.2458995684981346,
      "num_tokens": 150052376.0,
      "step": 86505
    },
    {
      "entropy": 5.357676601409912,
      "epoch": 6.730752771834274,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001601290085582866,
      "loss": 4.646,
      "mean_token_accuracy": 0.2475430577993393,
      "num_tokens": 150060649.0,
      "step": 86510
    },
    {
      "entropy": 5.356394529342651,
      "epoch": 6.731141801205991,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016010517803363906,
      "loss": 4.6972,
      "mean_token_accuracy": 0.2411485195159912,
      "num_tokens": 150069628.0,
      "step": 86515
    },
    {
      "entropy": 5.384897661209107,
      "epoch": 6.731530830577709,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016008134925233746,
      "loss": 4.5721,
      "mean_token_accuracy": 0.24705778807401657,
      "num_tokens": 150077755.0,
      "step": 86520
    },
    {
      "entropy": 5.3270580768585205,
      "epoch": 6.731919859949426,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016005752221474358,
      "loss": 4.6633,
      "mean_token_accuracy": 0.24160715639591218,
      "num_tokens": 150086629.0,
      "step": 86525
    },
    {
      "entropy": 5.333341836929321,
      "epoch": 6.732308889321144,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016003369692121875,
      "loss": 4.5703,
      "mean_token_accuracy": 0.24427866339683532,
      "num_tokens": 150095566.0,
      "step": 86530
    },
    {
      "entropy": 5.350177526473999,
      "epoch": 6.732697918692861,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001600098733721247,
      "loss": 4.6146,
      "mean_token_accuracy": 0.24128673374652862,
      "num_tokens": 150103655.0,
      "step": 86535
    },
    {
      "entropy": 5.44404411315918,
      "epoch": 6.733086948064579,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015998605156782264,
      "loss": 4.7254,
      "mean_token_accuracy": 0.23471633046865464,
      "num_tokens": 150112181.0,
      "step": 86540
    },
    {
      "entropy": 5.377424907684326,
      "epoch": 6.733475977436297,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015996223150867428,
      "loss": 4.5735,
      "mean_token_accuracy": 0.2504995182156563,
      "num_tokens": 150120356.0,
      "step": 86545
    },
    {
      "entropy": 5.40149073600769,
      "epoch": 6.733865006808014,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015993841319504102,
      "loss": 4.6564,
      "mean_token_accuracy": 0.2435891568660736,
      "num_tokens": 150129534.0,
      "step": 86550
    },
    {
      "entropy": 5.396664428710937,
      "epoch": 6.734254036179731,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00015991459662728397,
      "loss": 4.6296,
      "mean_token_accuracy": 0.24603921920061111,
      "num_tokens": 150138389.0,
      "step": 86555
    },
    {
      "entropy": 5.34764666557312,
      "epoch": 6.734643065551449,
      "grad_norm": 1.375,
      "learning_rate": 0.00015989078180576495,
      "loss": 4.6085,
      "mean_token_accuracy": 0.2427228569984436,
      "num_tokens": 150147274.0,
      "step": 86560
    },
    {
      "entropy": 5.3338817119598385,
      "epoch": 6.735032094923167,
      "grad_norm": 1.296875,
      "learning_rate": 0.000159866968730845,
      "loss": 4.5508,
      "mean_token_accuracy": 0.2453005999326706,
      "num_tokens": 150156021.0,
      "step": 86565
    },
    {
      "entropy": 5.381415462493896,
      "epoch": 6.735421124294884,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015984315740288563,
      "loss": 4.6109,
      "mean_token_accuracy": 0.23591403067111968,
      "num_tokens": 150164333.0,
      "step": 86570
    },
    {
      "entropy": 5.326209449768067,
      "epoch": 6.735810153666602,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00015981934782224806,
      "loss": 4.6322,
      "mean_token_accuracy": 0.2367313414812088,
      "num_tokens": 150172443.0,
      "step": 86575
    },
    {
      "entropy": 5.379416131973267,
      "epoch": 6.73619918303832,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015979553998929347,
      "loss": 4.695,
      "mean_token_accuracy": 0.2343272313475609,
      "num_tokens": 150181011.0,
      "step": 86580
    },
    {
      "entropy": 5.427183437347412,
      "epoch": 6.736588212410037,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015977173390438327,
      "loss": 4.6929,
      "mean_token_accuracy": 0.2360092356801033,
      "num_tokens": 150189818.0,
      "step": 86585
    },
    {
      "entropy": 5.2997931957244875,
      "epoch": 6.736977241781754,
      "grad_norm": 1.40625,
      "learning_rate": 0.00015974792956787852,
      "loss": 4.5588,
      "mean_token_accuracy": 0.2540013715624809,
      "num_tokens": 150198399.0,
      "step": 86590
    },
    {
      "entropy": 5.3360926628112795,
      "epoch": 6.737366271153472,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001597241269801406,
      "loss": 4.5864,
      "mean_token_accuracy": 0.24660884886980056,
      "num_tokens": 150208095.0,
      "step": 86595
    },
    {
      "entropy": 5.365523481369019,
      "epoch": 6.73775530052519,
      "grad_norm": 1.375,
      "learning_rate": 0.00015970032614153045,
      "loss": 4.6255,
      "mean_token_accuracy": 0.24118990898132325,
      "num_tokens": 150216750.0,
      "step": 86600
    },
    {
      "entropy": 5.33399543762207,
      "epoch": 6.738144329896907,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00015967652705240945,
      "loss": 4.5985,
      "mean_token_accuracy": 0.24316480457782746,
      "num_tokens": 150225446.0,
      "step": 86605
    },
    {
      "entropy": 5.407128238677979,
      "epoch": 6.738533359268625,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001596527297131385,
      "loss": 4.6697,
      "mean_token_accuracy": 0.2347989022731781,
      "num_tokens": 150232586.0,
      "step": 86610
    },
    {
      "entropy": 5.286025428771973,
      "epoch": 6.738922388640343,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00015962893412407885,
      "loss": 4.5945,
      "mean_token_accuracy": 0.23980384618043898,
      "num_tokens": 150240307.0,
      "step": 86615
    },
    {
      "entropy": 5.361106443405151,
      "epoch": 6.7393114180120595,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015960514028559144,
      "loss": 4.6031,
      "mean_token_accuracy": 0.24578217566013336,
      "num_tokens": 150248973.0,
      "step": 86620
    },
    {
      "entropy": 5.499377965927124,
      "epoch": 6.739700447383777,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00015958134819803735,
      "loss": 4.8283,
      "mean_token_accuracy": 0.2253999501466751,
      "num_tokens": 150258299.0,
      "step": 86625
    },
    {
      "entropy": 5.457450008392334,
      "epoch": 6.740089476755495,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00015955755786177766,
      "loss": 4.7441,
      "mean_token_accuracy": 0.23275287002325057,
      "num_tokens": 150266808.0,
      "step": 86630
    },
    {
      "entropy": 5.385130214691162,
      "epoch": 6.7404785061272126,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015953376927717322,
      "loss": 4.6069,
      "mean_token_accuracy": 0.24218972623348237,
      "num_tokens": 150275705.0,
      "step": 86635
    },
    {
      "entropy": 5.424401521682739,
      "epoch": 6.74086753549893,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015950998244458514,
      "loss": 4.6017,
      "mean_token_accuracy": 0.24546981006860732,
      "num_tokens": 150283163.0,
      "step": 86640
    },
    {
      "entropy": 5.463386583328247,
      "epoch": 6.741256564870648,
      "grad_norm": 1.453125,
      "learning_rate": 0.00015948619736437414,
      "loss": 4.6828,
      "mean_token_accuracy": 0.2397756025195122,
      "num_tokens": 150291877.0,
      "step": 86645
    },
    {
      "entropy": 5.347423124313354,
      "epoch": 6.741645594242366,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015946241403690136,
      "loss": 4.5776,
      "mean_token_accuracy": 0.23953987061977386,
      "num_tokens": 150300341.0,
      "step": 86650
    },
    {
      "entropy": 5.342374563217163,
      "epoch": 6.742034623614083,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001594386324625275,
      "loss": 4.6357,
      "mean_token_accuracy": 0.24903225153684616,
      "num_tokens": 150308772.0,
      "step": 86655
    },
    {
      "entropy": 5.4074554443359375,
      "epoch": 6.7424236529858,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015941485264161343,
      "loss": 4.685,
      "mean_token_accuracy": 0.2322802424430847,
      "num_tokens": 150318053.0,
      "step": 86660
    },
    {
      "entropy": 5.3586945056915285,
      "epoch": 6.742812682357518,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015939107457452003,
      "loss": 4.6366,
      "mean_token_accuracy": 0.25096495449543,
      "num_tokens": 150326515.0,
      "step": 86665
    },
    {
      "entropy": 5.396742630004883,
      "epoch": 6.7432017117292355,
      "grad_norm": 1.328125,
      "learning_rate": 0.000159367298261608,
      "loss": 4.6424,
      "mean_token_accuracy": 0.23705615550279618,
      "num_tokens": 150334855.0,
      "step": 86670
    },
    {
      "entropy": 5.462725257873535,
      "epoch": 6.743590741100953,
      "grad_norm": 1.375,
      "learning_rate": 0.00015934352370323827,
      "loss": 4.6783,
      "mean_token_accuracy": 0.24461203068494797,
      "num_tokens": 150343409.0,
      "step": 86675
    },
    {
      "entropy": 5.314132976531982,
      "epoch": 6.743979770472671,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015931975089977134,
      "loss": 4.5765,
      "mean_token_accuracy": 0.2449779525399208,
      "num_tokens": 150352212.0,
      "step": 86680
    },
    {
      "entropy": 5.398190593719482,
      "epoch": 6.744368799844388,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00015929597985156818,
      "loss": 4.6323,
      "mean_token_accuracy": 0.2394014224410057,
      "num_tokens": 150360872.0,
      "step": 86685
    },
    {
      "entropy": 5.365985012054443,
      "epoch": 6.7447578292161054,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015927221055898934,
      "loss": 4.6277,
      "mean_token_accuracy": 0.24483997821807862,
      "num_tokens": 150369052.0,
      "step": 86690
    },
    {
      "entropy": 5.433430337905884,
      "epoch": 6.745146858587823,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00015924844302239545,
      "loss": 4.7372,
      "mean_token_accuracy": 0.23694879710674285,
      "num_tokens": 150377455.0,
      "step": 86695
    },
    {
      "entropy": 5.405974006652832,
      "epoch": 6.745535887959541,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015922467724214728,
      "loss": 4.7331,
      "mean_token_accuracy": 0.23599400818347932,
      "num_tokens": 150385623.0,
      "step": 86700
    },
    {
      "entropy": 5.353680992126465,
      "epoch": 6.7459249173312585,
      "grad_norm": 1.25,
      "learning_rate": 0.0001592009132186053,
      "loss": 4.5929,
      "mean_token_accuracy": 0.24066076129674913,
      "num_tokens": 150394326.0,
      "step": 86705
    },
    {
      "entropy": 5.337457466125488,
      "epoch": 6.746313946702976,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001591771509521302,
      "loss": 4.5303,
      "mean_token_accuracy": 0.24528997093439103,
      "num_tokens": 150403395.0,
      "step": 86710
    },
    {
      "entropy": 5.344397449493409,
      "epoch": 6.746702976074694,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00015915339044308242,
      "loss": 4.5559,
      "mean_token_accuracy": 0.24377324432134628,
      "num_tokens": 150411438.0,
      "step": 86715
    },
    {
      "entropy": 5.327082967758178,
      "epoch": 6.747092005446412,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00015912963169182264,
      "loss": 4.5887,
      "mean_token_accuracy": 0.24230331778526307,
      "num_tokens": 150420952.0,
      "step": 86720
    },
    {
      "entropy": 5.322646522521973,
      "epoch": 6.747481034818128,
      "grad_norm": 1.25,
      "learning_rate": 0.00015910587469871116,
      "loss": 4.4897,
      "mean_token_accuracy": 0.2522676721215248,
      "num_tokens": 150429174.0,
      "step": 86725
    },
    {
      "entropy": 5.366421175003052,
      "epoch": 6.747870064189846,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015908211946410867,
      "loss": 4.5941,
      "mean_token_accuracy": 0.24013547599315643,
      "num_tokens": 150437357.0,
      "step": 86730
    },
    {
      "entropy": 5.437785196304321,
      "epoch": 6.748259093561564,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015905836598837552,
      "loss": 4.7218,
      "mean_token_accuracy": 0.23048072606325148,
      "num_tokens": 150444964.0,
      "step": 86735
    },
    {
      "entropy": 5.379519653320313,
      "epoch": 6.7486481229332815,
      "grad_norm": 1.375,
      "learning_rate": 0.00015903461427187204,
      "loss": 4.7267,
      "mean_token_accuracy": 0.23018602132797242,
      "num_tokens": 150453889.0,
      "step": 86740
    },
    {
      "entropy": 5.383604240417481,
      "epoch": 6.749037152304999,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00015901086431495878,
      "loss": 4.6179,
      "mean_token_accuracy": 0.24353678673505783,
      "num_tokens": 150461748.0,
      "step": 86745
    },
    {
      "entropy": 5.406363201141358,
      "epoch": 6.749426181676717,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000158987116117996,
      "loss": 4.6697,
      "mean_token_accuracy": 0.24028267860412597,
      "num_tokens": 150471021.0,
      "step": 86750
    },
    {
      "entropy": 5.3884134769439695,
      "epoch": 6.749815211048434,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000158963369681344,
      "loss": 4.6404,
      "mean_token_accuracy": 0.2367157071828842,
      "num_tokens": 150480522.0,
      "step": 86755
    },
    {
      "entropy": 5.431360244750977,
      "epoch": 6.750204240420151,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015893962500536325,
      "loss": 4.6701,
      "mean_token_accuracy": 0.24331804513931274,
      "num_tokens": 150490005.0,
      "step": 86760
    },
    {
      "entropy": 5.439209222793579,
      "epoch": 6.750593269791869,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00015891588209041398,
      "loss": 4.6908,
      "mean_token_accuracy": 0.23498098999261857,
      "num_tokens": 150498193.0,
      "step": 86765
    },
    {
      "entropy": 5.425718832015991,
      "epoch": 6.750982299163587,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001588921409368565,
      "loss": 4.6517,
      "mean_token_accuracy": 0.24033116549253464,
      "num_tokens": 150506806.0,
      "step": 86770
    },
    {
      "entropy": 5.308425807952881,
      "epoch": 6.7513713285353045,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00015886840154505084,
      "loss": 4.5348,
      "mean_token_accuracy": 0.2479819029569626,
      "num_tokens": 150515328.0,
      "step": 86775
    },
    {
      "entropy": 5.348219108581543,
      "epoch": 6.751760357907022,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015884466391535745,
      "loss": 4.5752,
      "mean_token_accuracy": 0.2473050534725189,
      "num_tokens": 150523659.0,
      "step": 86780
    },
    {
      "entropy": 5.37948956489563,
      "epoch": 6.75214938727874,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001588209280481363,
      "loss": 4.6868,
      "mean_token_accuracy": 0.23666368573904037,
      "num_tokens": 150532725.0,
      "step": 86785
    },
    {
      "entropy": 5.4109978675842285,
      "epoch": 6.752538416650457,
      "grad_norm": 1.46875,
      "learning_rate": 0.00015879719394374774,
      "loss": 4.6702,
      "mean_token_accuracy": 0.23651717007160186,
      "num_tokens": 150541099.0,
      "step": 86790
    },
    {
      "entropy": 5.36580457687378,
      "epoch": 6.752927446022174,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015877346160255175,
      "loss": 4.6298,
      "mean_token_accuracy": 0.24712107926607133,
      "num_tokens": 150549467.0,
      "step": 86795
    },
    {
      "entropy": 5.3267735004425045,
      "epoch": 6.753316475393892,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015874973102490852,
      "loss": 4.549,
      "mean_token_accuracy": 0.2571324944496155,
      "num_tokens": 150558649.0,
      "step": 86800
    },
    {
      "entropy": 5.36736946105957,
      "epoch": 6.75370550476561,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015872600221117808,
      "loss": 4.5972,
      "mean_token_accuracy": 0.24349643886089326,
      "num_tokens": 150566939.0,
      "step": 86805
    },
    {
      "entropy": 5.36602029800415,
      "epoch": 6.7540945341373275,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015870227516172054,
      "loss": 4.5775,
      "mean_token_accuracy": 0.24661392122507095,
      "num_tokens": 150575642.0,
      "step": 86810
    },
    {
      "entropy": 5.368757152557373,
      "epoch": 6.754483563509045,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001586785498768958,
      "loss": 4.6988,
      "mean_token_accuracy": 0.24269146621227264,
      "num_tokens": 150583726.0,
      "step": 86815
    },
    {
      "entropy": 5.373326778411865,
      "epoch": 6.754872592880762,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015865482635706385,
      "loss": 4.6284,
      "mean_token_accuracy": 0.24545236080884933,
      "num_tokens": 150591976.0,
      "step": 86820
    },
    {
      "entropy": 5.426251602172852,
      "epoch": 6.75526162225248,
      "grad_norm": 1.1875,
      "learning_rate": 0.00015863110460258485,
      "loss": 4.7135,
      "mean_token_accuracy": 0.24481361359357834,
      "num_tokens": 150600835.0,
      "step": 86825
    },
    {
      "entropy": 5.440485048294067,
      "epoch": 6.755650651624197,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015860738461381858,
      "loss": 4.6637,
      "mean_token_accuracy": 0.24940052777528762,
      "num_tokens": 150610386.0,
      "step": 86830
    },
    {
      "entropy": 5.392095232009888,
      "epoch": 6.756039680995915,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00015858366639112502,
      "loss": 4.5824,
      "mean_token_accuracy": 0.23998736590147018,
      "num_tokens": 150618620.0,
      "step": 86835
    },
    {
      "entropy": 5.431262683868408,
      "epoch": 6.756428710367633,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015855994993486394,
      "loss": 4.7206,
      "mean_token_accuracy": 0.23937655985355377,
      "num_tokens": 150627889.0,
      "step": 86840
    },
    {
      "entropy": 5.436983871459961,
      "epoch": 6.7568177397393505,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001585362352453954,
      "loss": 4.7,
      "mean_token_accuracy": 0.24899702221155168,
      "num_tokens": 150636452.0,
      "step": 86845
    },
    {
      "entropy": 5.4389581203460695,
      "epoch": 6.757206769111068,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001585125223230791,
      "loss": 4.6863,
      "mean_token_accuracy": 0.23841559439897536,
      "num_tokens": 150645389.0,
      "step": 86850
    },
    {
      "entropy": 5.432129049301148,
      "epoch": 6.757595798482786,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015848881116827478,
      "loss": 4.6935,
      "mean_token_accuracy": 0.23879595547914506,
      "num_tokens": 150654659.0,
      "step": 86855
    },
    {
      "entropy": 5.390882539749145,
      "epoch": 6.757984827854503,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001584651017813424,
      "loss": 4.6329,
      "mean_token_accuracy": 0.24197384417057038,
      "num_tokens": 150663452.0,
      "step": 86860
    },
    {
      "entropy": 5.528287553787232,
      "epoch": 6.75837385722622,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015844139416264151,
      "loss": 4.7576,
      "mean_token_accuracy": 0.2313082590699196,
      "num_tokens": 150671909.0,
      "step": 86865
    },
    {
      "entropy": 5.3625060558319095,
      "epoch": 6.758762886597938,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015841768831253205,
      "loss": 4.5916,
      "mean_token_accuracy": 0.250088657438755,
      "num_tokens": 150680159.0,
      "step": 86870
    },
    {
      "entropy": 5.378114366531372,
      "epoch": 6.759151915969656,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001583939842313735,
      "loss": 4.5697,
      "mean_token_accuracy": 0.25016418397426604,
      "num_tokens": 150688366.0,
      "step": 86875
    },
    {
      "entropy": 5.36807689666748,
      "epoch": 6.7595409453413735,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001583702819195258,
      "loss": 4.6008,
      "mean_token_accuracy": 0.23929579854011535,
      "num_tokens": 150697284.0,
      "step": 86880
    },
    {
      "entropy": 5.384963512420654,
      "epoch": 6.759929974713091,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015834658137734827,
      "loss": 4.6217,
      "mean_token_accuracy": 0.24373877942562103,
      "num_tokens": 150705719.0,
      "step": 86885
    },
    {
      "entropy": 5.332027769088745,
      "epoch": 6.760319004084808,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015832288260520072,
      "loss": 4.5801,
      "mean_token_accuracy": 0.25135633647441863,
      "num_tokens": 150713855.0,
      "step": 86890
    },
    {
      "entropy": 5.338594102859497,
      "epoch": 6.760708033456526,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001582991856034428,
      "loss": 4.6288,
      "mean_token_accuracy": 0.2425461396574974,
      "num_tokens": 150723033.0,
      "step": 86895
    },
    {
      "entropy": 5.325956964492798,
      "epoch": 6.761097062828243,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015827549037243391,
      "loss": 4.6063,
      "mean_token_accuracy": 0.24316215068101882,
      "num_tokens": 150731193.0,
      "step": 86900
    },
    {
      "entropy": 5.413172388076783,
      "epoch": 6.761486092199961,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00015825179691253378,
      "loss": 4.7017,
      "mean_token_accuracy": 0.23987009823322297,
      "num_tokens": 150740735.0,
      "step": 86905
    },
    {
      "entropy": 5.435426092147827,
      "epoch": 6.761875121571679,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015822810522410163,
      "loss": 4.7474,
      "mean_token_accuracy": 0.22612712234258653,
      "num_tokens": 150749872.0,
      "step": 86910
    },
    {
      "entropy": 5.362718391418457,
      "epoch": 6.7622641509433965,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001582044153074973,
      "loss": 4.6377,
      "mean_token_accuracy": 0.2396436020731926,
      "num_tokens": 150758792.0,
      "step": 86915
    },
    {
      "entropy": 5.386623382568359,
      "epoch": 6.762653180315114,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015818072716307997,
      "loss": 4.6216,
      "mean_token_accuracy": 0.24854274541139604,
      "num_tokens": 150768341.0,
      "step": 86920
    },
    {
      "entropy": 5.405844831466675,
      "epoch": 6.763042209686831,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015815704079120925,
      "loss": 4.6983,
      "mean_token_accuracy": 0.23992435187101363,
      "num_tokens": 150777307.0,
      "step": 86925
    },
    {
      "entropy": 5.446200513839722,
      "epoch": 6.763431239058549,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015813335619224443,
      "loss": 4.6536,
      "mean_token_accuracy": 0.24296968132257463,
      "num_tokens": 150785993.0,
      "step": 86930
    },
    {
      "entropy": 5.409615707397461,
      "epoch": 6.763820268430266,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015810967336654487,
      "loss": 4.6421,
      "mean_token_accuracy": 0.23975160866975784,
      "num_tokens": 150795024.0,
      "step": 86935
    },
    {
      "entropy": 5.368968963623047,
      "epoch": 6.764209297801984,
      "grad_norm": 1.234375,
      "learning_rate": 0.00015808599231447002,
      "loss": 4.6358,
      "mean_token_accuracy": 0.23717820644378662,
      "num_tokens": 150803777.0,
      "step": 86940
    },
    {
      "entropy": 5.4032939910888675,
      "epoch": 6.764598327173702,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015806231303637906,
      "loss": 4.6478,
      "mean_token_accuracy": 0.24559405595064163,
      "num_tokens": 150812230.0,
      "step": 86945
    },
    {
      "entropy": 5.3609741687774655,
      "epoch": 6.7649873565454195,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015803863553263148,
      "loss": 4.6371,
      "mean_token_accuracy": 0.2462459146976471,
      "num_tokens": 150822078.0,
      "step": 86950
    },
    {
      "entropy": 5.464135551452637,
      "epoch": 6.765376385917136,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015801495980358635,
      "loss": 4.6932,
      "mean_token_accuracy": 0.24532705098390578,
      "num_tokens": 150830390.0,
      "step": 86955
    },
    {
      "entropy": 5.353092527389526,
      "epoch": 6.765765415288854,
      "grad_norm": 1.375,
      "learning_rate": 0.000157991285849603,
      "loss": 4.5999,
      "mean_token_accuracy": 0.24576823115348817,
      "num_tokens": 150839094.0,
      "step": 86960
    },
    {
      "entropy": 5.387390661239624,
      "epoch": 6.766154444660572,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015796761367104073,
      "loss": 4.6674,
      "mean_token_accuracy": 0.2510802686214447,
      "num_tokens": 150847440.0,
      "step": 86965
    },
    {
      "entropy": 5.409748268127442,
      "epoch": 6.766543474032289,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015794394326825858,
      "loss": 4.6432,
      "mean_token_accuracy": 0.2342056229710579,
      "num_tokens": 150856079.0,
      "step": 86970
    },
    {
      "entropy": 5.344031715393067,
      "epoch": 6.766932503404007,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015792027464161585,
      "loss": 4.6037,
      "mean_token_accuracy": 0.24318287074565886,
      "num_tokens": 150865016.0,
      "step": 86975
    },
    {
      "entropy": 5.385811471939087,
      "epoch": 6.767321532775725,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015789660779147154,
      "loss": 4.6939,
      "mean_token_accuracy": 0.238667768239975,
      "num_tokens": 150873555.0,
      "step": 86980
    },
    {
      "entropy": 5.441705513000488,
      "epoch": 6.7677105621474425,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015787294271818486,
      "loss": 4.7456,
      "mean_token_accuracy": 0.23239727169275284,
      "num_tokens": 150882348.0,
      "step": 86985
    },
    {
      "entropy": 5.424887561798096,
      "epoch": 6.76809959151916,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001578492794221148,
      "loss": 4.6714,
      "mean_token_accuracy": 0.2386402189731598,
      "num_tokens": 150890396.0,
      "step": 86990
    },
    {
      "entropy": 5.403138160705566,
      "epoch": 6.768488620890877,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001578256179036205,
      "loss": 4.6063,
      "mean_token_accuracy": 0.24374020844697952,
      "num_tokens": 150898691.0,
      "step": 86995
    },
    {
      "entropy": 5.434062147140503,
      "epoch": 6.768877650262595,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015780195816306089,
      "loss": 4.7231,
      "mean_token_accuracy": 0.23543199449777602,
      "num_tokens": 150907071.0,
      "step": 87000
    },
    {
      "epoch": 6.768877650262595,
      "eval_entropy": 5.1254223468328926,
      "eval_loss": 4.857350826263428,
      "eval_mean_token_accuracy": 0.23597580749932337,
      "eval_num_tokens": 150907071.0,
      "eval_runtime": 28.7189,
      "eval_samples_per_second": 1447.059,
      "eval_steps_per_second": 180.891,
      "step": 87000
    },
    {
      "entropy": 5.374092435836792,
      "epoch": 6.769266679634312,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00015777830020079512,
      "loss": 4.6945,
      "mean_token_accuracy": 0.23994003236293793,
      "num_tokens": 150915716.0,
      "step": 87005
    },
    {
      "entropy": 5.332070922851562,
      "epoch": 6.76965570900603,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015775464401718203,
      "loss": 4.5155,
      "mean_token_accuracy": 0.24869565218687056,
      "num_tokens": 150924695.0,
      "step": 87010
    },
    {
      "entropy": 5.321927833557129,
      "epoch": 6.770044738377748,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015773098961258057,
      "loss": 4.5771,
      "mean_token_accuracy": 0.24655727446079254,
      "num_tokens": 150933847.0,
      "step": 87015
    },
    {
      "entropy": 5.393592357635498,
      "epoch": 6.770433767749465,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015770733698734966,
      "loss": 4.6903,
      "mean_token_accuracy": 0.23595406413078307,
      "num_tokens": 150942154.0,
      "step": 87020
    },
    {
      "entropy": 5.3536169052124025,
      "epoch": 6.770822797121182,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015768368614184824,
      "loss": 4.5935,
      "mean_token_accuracy": 0.2424705758690834,
      "num_tokens": 150950903.0,
      "step": 87025
    },
    {
      "entropy": 5.387820243835449,
      "epoch": 6.7712118264929,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015766003707643522,
      "loss": 4.6239,
      "mean_token_accuracy": 0.24669907987117767,
      "num_tokens": 150959569.0,
      "step": 87030
    },
    {
      "entropy": 5.3175304412841795,
      "epoch": 6.771600855864618,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015763638979146937,
      "loss": 4.5938,
      "mean_token_accuracy": 0.24850777685642242,
      "num_tokens": 150967664.0,
      "step": 87035
    },
    {
      "entropy": 5.362477874755859,
      "epoch": 6.771989885236335,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015761274428730942,
      "loss": 4.6579,
      "mean_token_accuracy": 0.2423808306455612,
      "num_tokens": 150976301.0,
      "step": 87040
    },
    {
      "entropy": 5.406316089630127,
      "epoch": 6.772378914608053,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001575891005643143,
      "loss": 4.685,
      "mean_token_accuracy": 0.2395578920841217,
      "num_tokens": 150985706.0,
      "step": 87045
    },
    {
      "entropy": 5.350804185867309,
      "epoch": 6.772767943979771,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015756545862284267,
      "loss": 4.5945,
      "mean_token_accuracy": 0.2478423833847046,
      "num_tokens": 150995126.0,
      "step": 87050
    },
    {
      "entropy": 5.347354364395142,
      "epoch": 6.7731569733514885,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015754181846325334,
      "loss": 4.6177,
      "mean_token_accuracy": 0.24420775026082991,
      "num_tokens": 151003228.0,
      "step": 87055
    },
    {
      "entropy": 5.2783607006072994,
      "epoch": 6.773546002723205,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015751818008590484,
      "loss": 4.4898,
      "mean_token_accuracy": 0.2527284547686577,
      "num_tokens": 151012011.0,
      "step": 87060
    },
    {
      "entropy": 5.278483867645264,
      "epoch": 6.773935032094923,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001574945434911561,
      "loss": 4.5746,
      "mean_token_accuracy": 0.244483382999897,
      "num_tokens": 151021330.0,
      "step": 87065
    },
    {
      "entropy": 5.434381198883057,
      "epoch": 6.774324061466641,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015747090867936555,
      "loss": 4.6858,
      "mean_token_accuracy": 0.24133276790380478,
      "num_tokens": 151030185.0,
      "step": 87070
    },
    {
      "entropy": 5.415054655075073,
      "epoch": 6.774713090838358,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015744727565089196,
      "loss": 4.655,
      "mean_token_accuracy": 0.2371824264526367,
      "num_tokens": 151038487.0,
      "step": 87075
    },
    {
      "entropy": 5.4371226787567135,
      "epoch": 6.775102120210076,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001574236444060938,
      "loss": 4.6727,
      "mean_token_accuracy": 0.24668436944484712,
      "num_tokens": 151046373.0,
      "step": 87080
    },
    {
      "entropy": 5.307097148895264,
      "epoch": 6.775491149581794,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015740001494532962,
      "loss": 4.5295,
      "mean_token_accuracy": 0.2507604345679283,
      "num_tokens": 151055417.0,
      "step": 87085
    },
    {
      "entropy": 5.333750200271607,
      "epoch": 6.775880178953511,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001573763872689582,
      "loss": 4.6721,
      "mean_token_accuracy": 0.23562807440757752,
      "num_tokens": 151064494.0,
      "step": 87090
    },
    {
      "entropy": 5.39428973197937,
      "epoch": 6.776269208325228,
      "grad_norm": 1.40625,
      "learning_rate": 0.00015735276137733776,
      "loss": 4.6406,
      "mean_token_accuracy": 0.2350349709391594,
      "num_tokens": 151073945.0,
      "step": 87095
    },
    {
      "entropy": 5.3464607238769535,
      "epoch": 6.776658237696946,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000157329137270827,
      "loss": 4.5659,
      "mean_token_accuracy": 0.24863184988498688,
      "num_tokens": 151082428.0,
      "step": 87100
    },
    {
      "entropy": 5.385948324203492,
      "epoch": 6.777047267068664,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001573055149497842,
      "loss": 4.6314,
      "mean_token_accuracy": 0.2516795516014099,
      "num_tokens": 151090729.0,
      "step": 87105
    },
    {
      "entropy": 5.379290676116943,
      "epoch": 6.777436296440381,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000157281894414568,
      "loss": 4.6123,
      "mean_token_accuracy": 0.24640195965766906,
      "num_tokens": 151099188.0,
      "step": 87110
    },
    {
      "entropy": 5.3477363109588625,
      "epoch": 6.777825325812099,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00015725827566553659,
      "loss": 4.5961,
      "mean_token_accuracy": 0.24382608979940415,
      "num_tokens": 151107542.0,
      "step": 87115
    },
    {
      "entropy": 5.427003955841064,
      "epoch": 6.778214355183817,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00015723465870304848,
      "loss": 4.7215,
      "mean_token_accuracy": 0.23682317584753038,
      "num_tokens": 151116018.0,
      "step": 87120
    },
    {
      "entropy": 5.353489637374878,
      "epoch": 6.778603384555534,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015721104352746205,
      "loss": 4.621,
      "mean_token_accuracy": 0.24700624495744705,
      "num_tokens": 151124774.0,
      "step": 87125
    },
    {
      "entropy": 5.419941091537476,
      "epoch": 6.778992413927251,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001571874301391354,
      "loss": 4.6738,
      "mean_token_accuracy": 0.24712209552526473,
      "num_tokens": 151133717.0,
      "step": 87130
    },
    {
      "entropy": 5.375555324554443,
      "epoch": 6.779381443298969,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001571638185384271,
      "loss": 4.6351,
      "mean_token_accuracy": 0.23717755526304246,
      "num_tokens": 151141715.0,
      "step": 87135
    },
    {
      "entropy": 5.315714597702026,
      "epoch": 6.779770472670687,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00015714020872569523,
      "loss": 4.577,
      "mean_token_accuracy": 0.24228921681642532,
      "num_tokens": 151150510.0,
      "step": 87140
    },
    {
      "entropy": 5.372259616851807,
      "epoch": 6.780159502042404,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015711660070129817,
      "loss": 4.6182,
      "mean_token_accuracy": 0.25213382244110105,
      "num_tokens": 151159136.0,
      "step": 87145
    },
    {
      "entropy": 5.371782541275024,
      "epoch": 6.780548531414122,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015709299446559395,
      "loss": 4.6316,
      "mean_token_accuracy": 0.2419101968407631,
      "num_tokens": 151167659.0,
      "step": 87150
    },
    {
      "entropy": 5.321930837631226,
      "epoch": 6.780937560785839,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001570693900189409,
      "loss": 4.5753,
      "mean_token_accuracy": 0.24421830624341964,
      "num_tokens": 151177279.0,
      "step": 87155
    },
    {
      "entropy": 5.325595045089722,
      "epoch": 6.781326590157557,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015704578736169721,
      "loss": 4.6039,
      "mean_token_accuracy": 0.2422620713710785,
      "num_tokens": 151185836.0,
      "step": 87160
    },
    {
      "entropy": 5.246536111831665,
      "epoch": 6.781715619529274,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001570221864942209,
      "loss": 4.4996,
      "mean_token_accuracy": 0.2546379044651985,
      "num_tokens": 151195599.0,
      "step": 87165
    },
    {
      "entropy": 5.38866171836853,
      "epoch": 6.782104648900992,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001569985874168702,
      "loss": 4.6585,
      "mean_token_accuracy": 0.23888645470142364,
      "num_tokens": 151203788.0,
      "step": 87170
    },
    {
      "entropy": 5.352604246139526,
      "epoch": 6.78249367827271,
      "grad_norm": 1.375,
      "learning_rate": 0.00015697499013000304,
      "loss": 4.6396,
      "mean_token_accuracy": 0.2408928319811821,
      "num_tokens": 151211660.0,
      "step": 87175
    },
    {
      "entropy": 5.41456708908081,
      "epoch": 6.782882707644427,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015695139463397766,
      "loss": 4.6921,
      "mean_token_accuracy": 0.23083127439022064,
      "num_tokens": 151219893.0,
      "step": 87180
    },
    {
      "entropy": 5.383120393753051,
      "epoch": 6.783271737016145,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015692780092915184,
      "loss": 4.6326,
      "mean_token_accuracy": 0.23402537554502487,
      "num_tokens": 151228865.0,
      "step": 87185
    },
    {
      "entropy": 5.3528313636779785,
      "epoch": 6.783660766387863,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001569042090158838,
      "loss": 4.5913,
      "mean_token_accuracy": 0.24100971221923828,
      "num_tokens": 151236953.0,
      "step": 87190
    },
    {
      "entropy": 5.360649585723877,
      "epoch": 6.78404979575958,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015688061889453136,
      "loss": 4.6219,
      "mean_token_accuracy": 0.24741194397211075,
      "num_tokens": 151245346.0,
      "step": 87195
    },
    {
      "entropy": 5.390010356903076,
      "epoch": 6.784438825131297,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001568570305654526,
      "loss": 4.6761,
      "mean_token_accuracy": 0.24124637097120286,
      "num_tokens": 151254608.0,
      "step": 87200
    },
    {
      "entropy": 5.40191388130188,
      "epoch": 6.784827854503015,
      "grad_norm": 1.375,
      "learning_rate": 0.00015683344402900534,
      "loss": 4.6859,
      "mean_token_accuracy": 0.23909489065408707,
      "num_tokens": 151263687.0,
      "step": 87205
    },
    {
      "entropy": 5.318648481369019,
      "epoch": 6.785216883874733,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001568098592855474,
      "loss": 4.5896,
      "mean_token_accuracy": 0.24328495264053346,
      "num_tokens": 151271856.0,
      "step": 87210
    },
    {
      "entropy": 5.378814220428467,
      "epoch": 6.78560591324645,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015678627633543683,
      "loss": 4.6479,
      "mean_token_accuracy": 0.23840028941631317,
      "num_tokens": 151280010.0,
      "step": 87215
    },
    {
      "entropy": 5.310987281799316,
      "epoch": 6.785994942618168,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001567626951790312,
      "loss": 4.5637,
      "mean_token_accuracy": 0.24741145372390747,
      "num_tokens": 151288086.0,
      "step": 87220
    },
    {
      "entropy": 5.440916967391968,
      "epoch": 6.786383971989885,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001567391158166886,
      "loss": 4.7855,
      "mean_token_accuracy": 0.22621618956327438,
      "num_tokens": 151296866.0,
      "step": 87225
    },
    {
      "entropy": 5.481134080886841,
      "epoch": 6.7867730013616026,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015671553824876672,
      "loss": 4.7345,
      "mean_token_accuracy": 0.23406438231468202,
      "num_tokens": 151305180.0,
      "step": 87230
    },
    {
      "entropy": 5.420098876953125,
      "epoch": 6.78716203073332,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00015669196247562316,
      "loss": 4.678,
      "mean_token_accuracy": 0.23364036083221434,
      "num_tokens": 151315331.0,
      "step": 87235
    },
    {
      "entropy": 5.33068528175354,
      "epoch": 6.787551060105038,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015666838849761588,
      "loss": 4.5864,
      "mean_token_accuracy": 0.24679771661758423,
      "num_tokens": 151324111.0,
      "step": 87240
    },
    {
      "entropy": 5.404269123077393,
      "epoch": 6.787940089476756,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00015664481631510237,
      "loss": 4.6856,
      "mean_token_accuracy": 0.23454857468605042,
      "num_tokens": 151332744.0,
      "step": 87245
    },
    {
      "entropy": 5.373251247406006,
      "epoch": 6.788329118848473,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00015662124592844047,
      "loss": 4.6001,
      "mean_token_accuracy": 0.24750469774007797,
      "num_tokens": 151341302.0,
      "step": 87250
    },
    {
      "entropy": 5.3508179664611815,
      "epoch": 6.788718148220191,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001565976773379877,
      "loss": 4.6341,
      "mean_token_accuracy": 0.23882309794425965,
      "num_tokens": 151349282.0,
      "step": 87255
    },
    {
      "entropy": 5.391197204589844,
      "epoch": 6.789107177591908,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00015657411054410177,
      "loss": 4.7036,
      "mean_token_accuracy": 0.24144572466611863,
      "num_tokens": 151359131.0,
      "step": 87260
    },
    {
      "entropy": 5.368033742904663,
      "epoch": 6.7894962069636255,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001565505455471401,
      "loss": 4.5764,
      "mean_token_accuracy": 0.23894985914230346,
      "num_tokens": 151367724.0,
      "step": 87265
    },
    {
      "entropy": 5.411091566085815,
      "epoch": 6.789885236335343,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001565269823474605,
      "loss": 4.6904,
      "mean_token_accuracy": 0.24330826550722123,
      "num_tokens": 151376446.0,
      "step": 87270
    },
    {
      "entropy": 5.3271385669708256,
      "epoch": 6.790274265707061,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015650342094542031,
      "loss": 4.5684,
      "mean_token_accuracy": 0.24568378180265427,
      "num_tokens": 151385093.0,
      "step": 87275
    },
    {
      "entropy": 5.355845880508423,
      "epoch": 6.790663295078779,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001564798613413771,
      "loss": 4.6819,
      "mean_token_accuracy": 0.23936089426279067,
      "num_tokens": 151393817.0,
      "step": 87280
    },
    {
      "entropy": 5.383494806289673,
      "epoch": 6.791052324450496,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015645630353568836,
      "loss": 4.66,
      "mean_token_accuracy": 0.238224595785141,
      "num_tokens": 151401928.0,
      "step": 87285
    },
    {
      "entropy": 5.3865128517150875,
      "epoch": 6.791441353822213,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00015643274752871145,
      "loss": 4.7049,
      "mean_token_accuracy": 0.2288798362016678,
      "num_tokens": 151410936.0,
      "step": 87290
    },
    {
      "entropy": 5.3475841045379635,
      "epoch": 6.791830383193931,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000156409193320804,
      "loss": 4.6257,
      "mean_token_accuracy": 0.24890607446432114,
      "num_tokens": 151419973.0,
      "step": 87295
    },
    {
      "entropy": 5.449594640731812,
      "epoch": 6.7922194125656485,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001563856409123232,
      "loss": 4.6678,
      "mean_token_accuracy": 0.24042956680059432,
      "num_tokens": 151429082.0,
      "step": 87300
    },
    {
      "entropy": 5.441337728500367,
      "epoch": 6.792608441937366,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001563620903036266,
      "loss": 4.6462,
      "mean_token_accuracy": 0.2414650186896324,
      "num_tokens": 151437249.0,
      "step": 87305
    },
    {
      "entropy": 5.395728969573975,
      "epoch": 6.792997471309084,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00015633854149507142,
      "loss": 4.6965,
      "mean_token_accuracy": 0.2301088333129883,
      "num_tokens": 151445320.0,
      "step": 87310
    },
    {
      "entropy": 5.3012003898620605,
      "epoch": 6.793386500680802,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00015631499448701494,
      "loss": 4.586,
      "mean_token_accuracy": 0.24442399740219117,
      "num_tokens": 151453195.0,
      "step": 87315
    },
    {
      "entropy": 5.415408563613892,
      "epoch": 6.793775530052519,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001562914492798146,
      "loss": 4.7112,
      "mean_token_accuracy": 0.23275848776102065,
      "num_tokens": 151461968.0,
      "step": 87320
    },
    {
      "entropy": 5.410301542282104,
      "epoch": 6.794164559424237,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001562679058738275,
      "loss": 4.7271,
      "mean_token_accuracy": 0.23160205483436586,
      "num_tokens": 151471286.0,
      "step": 87325
    },
    {
      "entropy": 5.427578401565552,
      "epoch": 6.794553588795954,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000156244364269411,
      "loss": 4.6568,
      "mean_token_accuracy": 0.24283526241779327,
      "num_tokens": 151480577.0,
      "step": 87330
    },
    {
      "entropy": 5.4031014919281,
      "epoch": 6.7949426181676715,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015622082446692216,
      "loss": 4.6821,
      "mean_token_accuracy": 0.23542118221521377,
      "num_tokens": 151489691.0,
      "step": 87335
    },
    {
      "entropy": 5.356697845458984,
      "epoch": 6.795331647539389,
      "grad_norm": 1.234375,
      "learning_rate": 0.00015619728646671838,
      "loss": 4.5754,
      "mean_token_accuracy": 0.2472609892487526,
      "num_tokens": 151498004.0,
      "step": 87340
    },
    {
      "entropy": 5.425762796401978,
      "epoch": 6.795720676911107,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001561737502691566,
      "loss": 4.6552,
      "mean_token_accuracy": 0.23818643838167192,
      "num_tokens": 151506661.0,
      "step": 87345
    },
    {
      "entropy": 5.3787792205810545,
      "epoch": 6.796109706282825,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000156150215874594,
      "loss": 4.6085,
      "mean_token_accuracy": 0.24431487470865249,
      "num_tokens": 151515306.0,
      "step": 87350
    },
    {
      "entropy": 5.393277359008789,
      "epoch": 6.796498735654541,
      "grad_norm": 1.25,
      "learning_rate": 0.00015612668328338774,
      "loss": 4.6535,
      "mean_token_accuracy": 0.24524293690919877,
      "num_tokens": 151524319.0,
      "step": 87355
    },
    {
      "entropy": 5.415173196792603,
      "epoch": 6.796887765026259,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015610315249589484,
      "loss": 4.673,
      "mean_token_accuracy": 0.23592498153448105,
      "num_tokens": 151533500.0,
      "step": 87360
    },
    {
      "entropy": 5.360537958145142,
      "epoch": 6.797276794397977,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015607962351247247,
      "loss": 4.5714,
      "mean_token_accuracy": 0.24892601370811462,
      "num_tokens": 151541942.0,
      "step": 87365
    },
    {
      "entropy": 5.418636798858643,
      "epoch": 6.7976658237696945,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001560560963334774,
      "loss": 4.762,
      "mean_token_accuracy": 0.2313599094748497,
      "num_tokens": 151551830.0,
      "step": 87370
    },
    {
      "entropy": 5.310070896148682,
      "epoch": 6.798054853141412,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015603257095926683,
      "loss": 4.5376,
      "mean_token_accuracy": 0.24293670058250427,
      "num_tokens": 151560489.0,
      "step": 87375
    },
    {
      "entropy": 5.378417921066284,
      "epoch": 6.79844388251313,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015600904739019756,
      "loss": 4.7024,
      "mean_token_accuracy": 0.23748379349708557,
      "num_tokens": 151569926.0,
      "step": 87380
    },
    {
      "entropy": 5.376136684417725,
      "epoch": 6.798832911884848,
      "grad_norm": 1.375,
      "learning_rate": 0.0001559855256266267,
      "loss": 4.6371,
      "mean_token_accuracy": 0.23997478932142258,
      "num_tokens": 151578482.0,
      "step": 87385
    },
    {
      "entropy": 5.362591886520386,
      "epoch": 6.799221941256565,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015596200566891103,
      "loss": 4.6352,
      "mean_token_accuracy": 0.24652411490678788,
      "num_tokens": 151587453.0,
      "step": 87390
    },
    {
      "entropy": 5.420590782165528,
      "epoch": 6.799610970628282,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015593848751740738,
      "loss": 4.7098,
      "mean_token_accuracy": 0.23495933562517166,
      "num_tokens": 151596279.0,
      "step": 87395
    },
    {
      "entropy": 5.393957328796387,
      "epoch": 6.8,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015591497117247276,
      "loss": 4.6423,
      "mean_token_accuracy": 0.24249375760555267,
      "num_tokens": 151604871.0,
      "step": 87400
    },
    {
      "entropy": 5.370973300933838,
      "epoch": 6.8003890293717175,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00015589145663446385,
      "loss": 4.6068,
      "mean_token_accuracy": 0.23908052891492843,
      "num_tokens": 151613630.0,
      "step": 87405
    },
    {
      "entropy": 5.438415956497193,
      "epoch": 6.800778058743435,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015586794390373758,
      "loss": 4.7022,
      "mean_token_accuracy": 0.23696373552083969,
      "num_tokens": 151622116.0,
      "step": 87410
    },
    {
      "entropy": 5.373213005065918,
      "epoch": 6.801167088115153,
      "grad_norm": 1.375,
      "learning_rate": 0.00015584443298065055,
      "loss": 4.6669,
      "mean_token_accuracy": 0.23048834055662154,
      "num_tokens": 151630498.0,
      "step": 87415
    },
    {
      "entropy": 5.4005176544189455,
      "epoch": 6.801556117486871,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015582092386555958,
      "loss": 4.6399,
      "mean_token_accuracy": 0.24975327104330064,
      "num_tokens": 151639198.0,
      "step": 87420
    },
    {
      "entropy": 5.4218415260314945,
      "epoch": 6.801945146858587,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001557974165588215,
      "loss": 4.7287,
      "mean_token_accuracy": 0.23959958255290986,
      "num_tokens": 151647568.0,
      "step": 87425
    },
    {
      "entropy": 5.359622812271118,
      "epoch": 6.802334176230305,
      "grad_norm": 1.375,
      "learning_rate": 0.0001557739110607928,
      "loss": 4.6003,
      "mean_token_accuracy": 0.2462053805589676,
      "num_tokens": 151655878.0,
      "step": 87430
    },
    {
      "entropy": 5.4070234298706055,
      "epoch": 6.802723205602023,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015575040737183032,
      "loss": 4.6994,
      "mean_token_accuracy": 0.23679072856903077,
      "num_tokens": 151664573.0,
      "step": 87435
    },
    {
      "entropy": 5.437038326263428,
      "epoch": 6.8031122349737405,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015572690549229054,
      "loss": 4.6915,
      "mean_token_accuracy": 0.230339652299881,
      "num_tokens": 151673533.0,
      "step": 87440
    },
    {
      "entropy": 5.368251800537109,
      "epoch": 6.803501264345458,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001557034054225302,
      "loss": 4.6261,
      "mean_token_accuracy": 0.2411182701587677,
      "num_tokens": 151682241.0,
      "step": 87445
    },
    {
      "entropy": 5.445954847335815,
      "epoch": 6.803890293717176,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001556799071629057,
      "loss": 4.669,
      "mean_token_accuracy": 0.2450113132596016,
      "num_tokens": 151690391.0,
      "step": 87450
    },
    {
      "entropy": 5.36720905303955,
      "epoch": 6.804279323088894,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015565641071377378,
      "loss": 4.5679,
      "mean_token_accuracy": 0.231974233686924,
      "num_tokens": 151698962.0,
      "step": 87455
    },
    {
      "entropy": 5.379471254348755,
      "epoch": 6.80466835246061,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001556329160754908,
      "loss": 4.564,
      "mean_token_accuracy": 0.2477173089981079,
      "num_tokens": 151707491.0,
      "step": 87460
    },
    {
      "entropy": 5.374447965621949,
      "epoch": 6.805057381832328,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00015560942324841342,
      "loss": 4.6683,
      "mean_token_accuracy": 0.24010850340127946,
      "num_tokens": 151716278.0,
      "step": 87465
    },
    {
      "entropy": 5.367658042907715,
      "epoch": 6.805446411204046,
      "grad_norm": 1.46875,
      "learning_rate": 0.000155585932232898,
      "loss": 4.7164,
      "mean_token_accuracy": 0.24166468232870103,
      "num_tokens": 151724221.0,
      "step": 87470
    },
    {
      "entropy": 5.334524917602539,
      "epoch": 6.8058354405757635,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001555624430293009,
      "loss": 4.6031,
      "mean_token_accuracy": 0.23994891792535783,
      "num_tokens": 151732845.0,
      "step": 87475
    },
    {
      "entropy": 5.360730218887329,
      "epoch": 6.806224469947481,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001555389556379787,
      "loss": 4.6887,
      "mean_token_accuracy": 0.2351543426513672,
      "num_tokens": 151741270.0,
      "step": 87480
    },
    {
      "entropy": 5.390143966674804,
      "epoch": 6.806613499319199,
      "grad_norm": 1.234375,
      "learning_rate": 0.00015551547005928756,
      "loss": 4.6083,
      "mean_token_accuracy": 0.2514088585972786,
      "num_tokens": 151749812.0,
      "step": 87485
    },
    {
      "entropy": 5.443428087234497,
      "epoch": 6.807002528690916,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015549198629358414,
      "loss": 4.7227,
      "mean_token_accuracy": 0.2364121377468109,
      "num_tokens": 151757854.0,
      "step": 87490
    },
    {
      "entropy": 5.423127555847168,
      "epoch": 6.807391558062633,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015546850434122451,
      "loss": 4.6589,
      "mean_token_accuracy": 0.24059994965791703,
      "num_tokens": 151766283.0,
      "step": 87495
    },
    {
      "entropy": 5.337667417526245,
      "epoch": 6.807780587434351,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015544502420256518,
      "loss": 4.6129,
      "mean_token_accuracy": 0.24434071630239487,
      "num_tokens": 151774716.0,
      "step": 87500
    },
    {
      "entropy": 5.406131315231323,
      "epoch": 6.808169616806069,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015542154587796231,
      "loss": 4.6317,
      "mean_token_accuracy": 0.23800744712352753,
      "num_tokens": 151783250.0,
      "step": 87505
    },
    {
      "entropy": 5.42042646408081,
      "epoch": 6.8085586461777865,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015539806936777206,
      "loss": 4.7109,
      "mean_token_accuracy": 0.22924352139234544,
      "num_tokens": 151791205.0,
      "step": 87510
    },
    {
      "entropy": 5.435712718963623,
      "epoch": 6.808947675549504,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015537459467235084,
      "loss": 4.6946,
      "mean_token_accuracy": 0.23883578777313233,
      "num_tokens": 151799968.0,
      "step": 87515
    },
    {
      "entropy": 5.394870615005493,
      "epoch": 6.809336704921222,
      "grad_norm": 1.375,
      "learning_rate": 0.00015535112179205466,
      "loss": 4.6234,
      "mean_token_accuracy": 0.24344617426395415,
      "num_tokens": 151808373.0,
      "step": 87520
    },
    {
      "entropy": 5.389349746704101,
      "epoch": 6.80972573429294,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015532765072723986,
      "loss": 4.6567,
      "mean_token_accuracy": 0.24376060515642167,
      "num_tokens": 151817435.0,
      "step": 87525
    },
    {
      "entropy": 5.432136678695679,
      "epoch": 6.810114763664656,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015530418147826237,
      "loss": 4.6768,
      "mean_token_accuracy": 0.2442197397351265,
      "num_tokens": 151825689.0,
      "step": 87530
    },
    {
      "entropy": 5.356233739852906,
      "epoch": 6.810503793036374,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001552807140454785,
      "loss": 4.5859,
      "mean_token_accuracy": 0.24238881170749665,
      "num_tokens": 151834264.0,
      "step": 87535
    },
    {
      "entropy": 5.410935688018799,
      "epoch": 6.810892822408092,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015525724842924417,
      "loss": 4.6499,
      "mean_token_accuracy": 0.24541440308094026,
      "num_tokens": 151842886.0,
      "step": 87540
    },
    {
      "entropy": 5.282352924346924,
      "epoch": 6.8112818517798095,
      "grad_norm": 1.4375,
      "learning_rate": 0.00015523378462991557,
      "loss": 4.578,
      "mean_token_accuracy": 0.24818440526723862,
      "num_tokens": 151851799.0,
      "step": 87545
    },
    {
      "entropy": 5.415251302719116,
      "epoch": 6.811670881151527,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015521032264784857,
      "loss": 4.6832,
      "mean_token_accuracy": 0.23281960636377336,
      "num_tokens": 151860655.0,
      "step": 87550
    },
    {
      "entropy": 5.354683542251587,
      "epoch": 6.812059910523245,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015518686248339932,
      "loss": 4.721,
      "mean_token_accuracy": 0.2395566925406456,
      "num_tokens": 151869660.0,
      "step": 87555
    },
    {
      "entropy": 5.368762063980102,
      "epoch": 6.812448939894962,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015516340413692376,
      "loss": 4.5729,
      "mean_token_accuracy": 0.24459483176469804,
      "num_tokens": 151878521.0,
      "step": 87560
    },
    {
      "entropy": 5.373947238922119,
      "epoch": 6.812837969266679,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001551399476087777,
      "loss": 4.613,
      "mean_token_accuracy": 0.24638134688138963,
      "num_tokens": 151887468.0,
      "step": 87565
    },
    {
      "entropy": 5.384499645233154,
      "epoch": 6.813226998638397,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001551164928993173,
      "loss": 4.5967,
      "mean_token_accuracy": 0.2513629153370857,
      "num_tokens": 151895639.0,
      "step": 87570
    },
    {
      "entropy": 5.403262615203857,
      "epoch": 6.813616028010115,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015509304000889819,
      "loss": 4.6663,
      "mean_token_accuracy": 0.2420704334974289,
      "num_tokens": 151904458.0,
      "step": 87575
    },
    {
      "entropy": 5.359309387207031,
      "epoch": 6.8140050573818325,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015506958893787641,
      "loss": 4.6304,
      "mean_token_accuracy": 0.23891924917697907,
      "num_tokens": 151912256.0,
      "step": 87580
    },
    {
      "entropy": 5.370041704177856,
      "epoch": 6.81439408675355,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015504613968660779,
      "loss": 4.6796,
      "mean_token_accuracy": 0.24173660427331925,
      "num_tokens": 151921043.0,
      "step": 87585
    },
    {
      "entropy": 5.451279735565185,
      "epoch": 6.814783116125268,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015502269225544792,
      "loss": 4.7745,
      "mean_token_accuracy": 0.23179126381874085,
      "num_tokens": 151929334.0,
      "step": 87590
    },
    {
      "entropy": 5.361670684814453,
      "epoch": 6.815172145496985,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015499924664475284,
      "loss": 4.624,
      "mean_token_accuracy": 0.2528132304549217,
      "num_tokens": 151937741.0,
      "step": 87595
    },
    {
      "entropy": 5.449291944503784,
      "epoch": 6.815561174868702,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001549758028548781,
      "loss": 4.6839,
      "mean_token_accuracy": 0.24560977667570114,
      "num_tokens": 151946162.0,
      "step": 87600
    },
    {
      "entropy": 5.424519824981689,
      "epoch": 6.81595020424042,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015495236088617964,
      "loss": 4.638,
      "mean_token_accuracy": 0.2446135014295578,
      "num_tokens": 151954237.0,
      "step": 87605
    },
    {
      "entropy": 5.423345041275025,
      "epoch": 6.816339233612138,
      "grad_norm": 1.453125,
      "learning_rate": 0.00015492892073901292,
      "loss": 4.6461,
      "mean_token_accuracy": 0.23706116527318954,
      "num_tokens": 151962948.0,
      "step": 87610
    },
    {
      "entropy": 5.327799224853516,
      "epoch": 6.8167282629838555,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015490548241373374,
      "loss": 4.5969,
      "mean_token_accuracy": 0.24577128291130065,
      "num_tokens": 151971812.0,
      "step": 87615
    },
    {
      "entropy": 5.392414855957031,
      "epoch": 6.817117292355573,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001548820459106977,
      "loss": 4.6713,
      "mean_token_accuracy": 0.2391129419207573,
      "num_tokens": 151980451.0,
      "step": 87620
    },
    {
      "entropy": 5.446305561065674,
      "epoch": 6.81750632172729,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015485861123026046,
      "loss": 4.6762,
      "mean_token_accuracy": 0.23512422740459443,
      "num_tokens": 151988525.0,
      "step": 87625
    },
    {
      "entropy": 5.435957622528076,
      "epoch": 6.817895351099008,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001548351783727776,
      "loss": 4.6471,
      "mean_token_accuracy": 0.24379929900169373,
      "num_tokens": 151996395.0,
      "step": 87630
    },
    {
      "entropy": 5.43581109046936,
      "epoch": 6.818284380470725,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015481174733860458,
      "loss": 4.7606,
      "mean_token_accuracy": 0.23303811997175217,
      "num_tokens": 152005366.0,
      "step": 87635
    },
    {
      "entropy": 5.380674886703491,
      "epoch": 6.818673409842443,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00015478831812809708,
      "loss": 4.6359,
      "mean_token_accuracy": 0.23668336868286133,
      "num_tokens": 152014676.0,
      "step": 87640
    },
    {
      "entropy": 5.343096542358398,
      "epoch": 6.819062439214161,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015476489074161038,
      "loss": 4.5779,
      "mean_token_accuracy": 0.24872583597898484,
      "num_tokens": 152023599.0,
      "step": 87645
    },
    {
      "entropy": 5.44405517578125,
      "epoch": 6.8194514685858785,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001547414651795002,
      "loss": 4.7203,
      "mean_token_accuracy": 0.23321874737739562,
      "num_tokens": 152031963.0,
      "step": 87650
    },
    {
      "entropy": 5.440528154373169,
      "epoch": 6.819840497957596,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015471804144212182,
      "loss": 4.6806,
      "mean_token_accuracy": 0.2387359395623207,
      "num_tokens": 152040152.0,
      "step": 87655
    },
    {
      "entropy": 5.4771819591522215,
      "epoch": 6.820229527329313,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00015469461952983076,
      "loss": 4.711,
      "mean_token_accuracy": 0.22774491161108018,
      "num_tokens": 152049024.0,
      "step": 87660
    },
    {
      "entropy": 5.388168907165527,
      "epoch": 6.820618556701031,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015467119944298242,
      "loss": 4.629,
      "mean_token_accuracy": 0.2434990733861923,
      "num_tokens": 152057732.0,
      "step": 87665
    },
    {
      "entropy": 5.42332968711853,
      "epoch": 6.821007586072748,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015464778118193197,
      "loss": 4.6726,
      "mean_token_accuracy": 0.24582043439149856,
      "num_tokens": 152065657.0,
      "step": 87670
    },
    {
      "entropy": 5.400262451171875,
      "epoch": 6.821396615444466,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015462436474703495,
      "loss": 4.6709,
      "mean_token_accuracy": 0.2361228033900261,
      "num_tokens": 152074460.0,
      "step": 87675
    },
    {
      "entropy": 5.451297473907471,
      "epoch": 6.821785644816184,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015460095013864657,
      "loss": 4.7723,
      "mean_token_accuracy": 0.24436041116714477,
      "num_tokens": 152082983.0,
      "step": 87680
    },
    {
      "entropy": 5.339689445495606,
      "epoch": 6.8221746741879015,
      "grad_norm": 1.25,
      "learning_rate": 0.0001545775373571221,
      "loss": 4.5643,
      "mean_token_accuracy": 0.24462947100400925,
      "num_tokens": 152091685.0,
      "step": 87685
    },
    {
      "entropy": 5.436685371398926,
      "epoch": 6.822563703559618,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001545541264028169,
      "loss": 4.5894,
      "mean_token_accuracy": 0.24472440928220748,
      "num_tokens": 152100056.0,
      "step": 87690
    },
    {
      "entropy": 5.320098495483398,
      "epoch": 6.822952732931336,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015453071727608607,
      "loss": 4.5917,
      "mean_token_accuracy": 0.24762240648269654,
      "num_tokens": 152108387.0,
      "step": 87695
    },
    {
      "entropy": 5.426633882522583,
      "epoch": 6.823341762303054,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015450730997728496,
      "loss": 4.6638,
      "mean_token_accuracy": 0.24069009125232696,
      "num_tokens": 152116943.0,
      "step": 87700
    },
    {
      "entropy": 5.379858541488647,
      "epoch": 6.823730791674771,
      "grad_norm": 1.40625,
      "learning_rate": 0.00015448390450676852,
      "loss": 4.6125,
      "mean_token_accuracy": 0.25082680881023406,
      "num_tokens": 152125398.0,
      "step": 87705
    },
    {
      "entropy": 5.379056358337403,
      "epoch": 6.824119821046489,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001544605008648921,
      "loss": 4.5698,
      "mean_token_accuracy": 0.2500517204403877,
      "num_tokens": 152133624.0,
      "step": 87710
    },
    {
      "entropy": 5.425550699234009,
      "epoch": 6.824508850418207,
      "grad_norm": 1.375,
      "learning_rate": 0.00015443709905201065,
      "loss": 4.679,
      "mean_token_accuracy": 0.24345652312040328,
      "num_tokens": 152141997.0,
      "step": 87715
    },
    {
      "entropy": 5.380044555664062,
      "epoch": 6.8248978797899245,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015441369906847935,
      "loss": 4.6767,
      "mean_token_accuracy": 0.24345712065696717,
      "num_tokens": 152150152.0,
      "step": 87720
    },
    {
      "entropy": 5.396894645690918,
      "epoch": 6.825286909161642,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015439030091465324,
      "loss": 4.6305,
      "mean_token_accuracy": 0.24957955926656722,
      "num_tokens": 152158766.0,
      "step": 87725
    },
    {
      "entropy": 5.347687911987305,
      "epoch": 6.825675938533359,
      "grad_norm": 1.375,
      "learning_rate": 0.00015436690459088738,
      "loss": 4.6016,
      "mean_token_accuracy": 0.2390442281961441,
      "num_tokens": 152167980.0,
      "step": 87730
    },
    {
      "entropy": 5.372927522659301,
      "epoch": 6.826064967905077,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015434351009753665,
      "loss": 4.665,
      "mean_token_accuracy": 0.2387247934937477,
      "num_tokens": 152176895.0,
      "step": 87735
    },
    {
      "entropy": 5.3585528373718265,
      "epoch": 6.826453997276794,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015432011743495618,
      "loss": 4.5925,
      "mean_token_accuracy": 0.24167822450399398,
      "num_tokens": 152184923.0,
      "step": 87740
    },
    {
      "entropy": 5.361005449295044,
      "epoch": 6.826843026648512,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00015429672660350087,
      "loss": 4.6257,
      "mean_token_accuracy": 0.2453148692846298,
      "num_tokens": 152194261.0,
      "step": 87745
    },
    {
      "entropy": 5.349089813232422,
      "epoch": 6.82723205602023,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00015427333760352542,
      "loss": 4.5846,
      "mean_token_accuracy": 0.24649778455495835,
      "num_tokens": 152203410.0,
      "step": 87750
    },
    {
      "entropy": 5.4112348556518555,
      "epoch": 6.8276210853919475,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001542499504353851,
      "loss": 4.7064,
      "mean_token_accuracy": 0.24222972393035888,
      "num_tokens": 152212099.0,
      "step": 87755
    },
    {
      "entropy": 5.364427661895752,
      "epoch": 6.828010114763664,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015422656509943454,
      "loss": 4.6122,
      "mean_token_accuracy": 0.23617759346961975,
      "num_tokens": 152220433.0,
      "step": 87760
    },
    {
      "entropy": 5.3664626121521,
      "epoch": 6.828399144135382,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015420318159602863,
      "loss": 4.6926,
      "mean_token_accuracy": 0.241220098733902,
      "num_tokens": 152229074.0,
      "step": 87765
    },
    {
      "entropy": 5.404411268234253,
      "epoch": 6.8287881735071,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00015417979992552225,
      "loss": 4.6661,
      "mean_token_accuracy": 0.23936223089694977,
      "num_tokens": 152237862.0,
      "step": 87770
    },
    {
      "entropy": 5.30645260810852,
      "epoch": 6.829177202878817,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015415642008826996,
      "loss": 4.5518,
      "mean_token_accuracy": 0.24576968550682068,
      "num_tokens": 152246308.0,
      "step": 87775
    },
    {
      "entropy": 5.366899299621582,
      "epoch": 6.829566232250535,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015413304208462678,
      "loss": 4.6859,
      "mean_token_accuracy": 0.24269679337739944,
      "num_tokens": 152254333.0,
      "step": 87780
    },
    {
      "entropy": 5.400658178329468,
      "epoch": 6.829955261622253,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001541096659149472,
      "loss": 4.6188,
      "mean_token_accuracy": 0.24316154271364213,
      "num_tokens": 152262993.0,
      "step": 87785
    },
    {
      "entropy": 5.437512016296386,
      "epoch": 6.8303442909939704,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001540862915795861,
      "loss": 4.7198,
      "mean_token_accuracy": 0.23619361370801925,
      "num_tokens": 152271918.0,
      "step": 87790
    },
    {
      "entropy": 5.358557081222534,
      "epoch": 6.830733320365687,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015406291907889803,
      "loss": 4.5836,
      "mean_token_accuracy": 0.25185790807008746,
      "num_tokens": 152280667.0,
      "step": 87795
    },
    {
      "entropy": 5.3118236541748045,
      "epoch": 6.831122349737405,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001540395484132377,
      "loss": 4.5605,
      "mean_token_accuracy": 0.24707349240779877,
      "num_tokens": 152289453.0,
      "step": 87800
    },
    {
      "entropy": 5.3454389572143555,
      "epoch": 6.831511379109123,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015401617958295962,
      "loss": 4.6048,
      "mean_token_accuracy": 0.23756056278944016,
      "num_tokens": 152297844.0,
      "step": 87805
    },
    {
      "entropy": 5.282944393157959,
      "epoch": 6.83190040848084,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015399281258841858,
      "loss": 4.5613,
      "mean_token_accuracy": 0.2535837545990944,
      "num_tokens": 152306614.0,
      "step": 87810
    },
    {
      "entropy": 5.384834480285645,
      "epoch": 6.832289437852558,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001539694474299689,
      "loss": 4.5899,
      "mean_token_accuracy": 0.24623977094888688,
      "num_tokens": 152315772.0,
      "step": 87815
    },
    {
      "entropy": 5.425903606414795,
      "epoch": 6.832678467224276,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001539460841079652,
      "loss": 4.7126,
      "mean_token_accuracy": 0.23233553022146225,
      "num_tokens": 152324219.0,
      "step": 87820
    },
    {
      "entropy": 5.350800275802612,
      "epoch": 6.8330674965959926,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015392272262276208,
      "loss": 4.6711,
      "mean_token_accuracy": 0.24152640998363495,
      "num_tokens": 152333091.0,
      "step": 87825
    },
    {
      "entropy": 5.4330726146698,
      "epoch": 6.83345652596771,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015389936297471383,
      "loss": 4.7115,
      "mean_token_accuracy": 0.23114402443170548,
      "num_tokens": 152341565.0,
      "step": 87830
    },
    {
      "entropy": 5.412523412704468,
      "epoch": 6.833845555339428,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001538760051641751,
      "loss": 4.6239,
      "mean_token_accuracy": 0.24088647216558456,
      "num_tokens": 152350655.0,
      "step": 87835
    },
    {
      "entropy": 5.422820138931274,
      "epoch": 6.834234584711146,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015385264919150015,
      "loss": 4.6317,
      "mean_token_accuracy": 0.2426111027598381,
      "num_tokens": 152359362.0,
      "step": 87840
    },
    {
      "entropy": 5.461972427368164,
      "epoch": 6.834623614082863,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015382929505704346,
      "loss": 4.7718,
      "mean_token_accuracy": 0.22579795271158218,
      "num_tokens": 152367997.0,
      "step": 87845
    },
    {
      "entropy": 5.400615406036377,
      "epoch": 6.835012643454581,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015380594276115936,
      "loss": 4.6272,
      "mean_token_accuracy": 0.24215300679206847,
      "num_tokens": 152375845.0,
      "step": 87850
    },
    {
      "entropy": 5.345032358169556,
      "epoch": 6.835401672826299,
      "grad_norm": 2.09375,
      "learning_rate": 0.00015378259230420212,
      "loss": 4.5816,
      "mean_token_accuracy": 0.24913360625505448,
      "num_tokens": 152385054.0,
      "step": 87855
    },
    {
      "entropy": 5.351049518585205,
      "epoch": 6.835790702198016,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015375924368652617,
      "loss": 4.6294,
      "mean_token_accuracy": 0.2417811319231987,
      "num_tokens": 152393798.0,
      "step": 87860
    },
    {
      "entropy": 5.39056806564331,
      "epoch": 6.836179731569733,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015373589690848566,
      "loss": 4.609,
      "mean_token_accuracy": 0.24121406376361848,
      "num_tokens": 152402193.0,
      "step": 87865
    },
    {
      "entropy": 5.341705799102783,
      "epoch": 6.836568760941451,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015371255197043493,
      "loss": 4.5517,
      "mean_token_accuracy": 0.2551496013998985,
      "num_tokens": 152410752.0,
      "step": 87870
    },
    {
      "entropy": 5.483415222167968,
      "epoch": 6.836957790313169,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00015368920887272813,
      "loss": 4.7671,
      "mean_token_accuracy": 0.22640473991632462,
      "num_tokens": 152419046.0,
      "step": 87875
    },
    {
      "entropy": 5.346845817565918,
      "epoch": 6.837346819684886,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001536658676157196,
      "loss": 4.6064,
      "mean_token_accuracy": 0.23895174860954285,
      "num_tokens": 152428718.0,
      "step": 87880
    },
    {
      "entropy": 5.406897163391113,
      "epoch": 6.837735849056604,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001536425281997633,
      "loss": 4.6854,
      "mean_token_accuracy": 0.24648779332637788,
      "num_tokens": 152437516.0,
      "step": 87885
    },
    {
      "entropy": 5.3680006980896,
      "epoch": 6.838124878428321,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015361919062521344,
      "loss": 4.6093,
      "mean_token_accuracy": 0.24638119488954544,
      "num_tokens": 152445839.0,
      "step": 87890
    },
    {
      "entropy": 5.440985107421875,
      "epoch": 6.8385139078000385,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015359585489242434,
      "loss": 4.7594,
      "mean_token_accuracy": 0.23053200095891951,
      "num_tokens": 152454715.0,
      "step": 87895
    },
    {
      "entropy": 5.482624197006226,
      "epoch": 6.838902937171756,
      "grad_norm": 1.515625,
      "learning_rate": 0.00015357252100174976,
      "loss": 4.7625,
      "mean_token_accuracy": 0.23234462291002272,
      "num_tokens": 152463023.0,
      "step": 87900
    },
    {
      "entropy": 5.466064691543579,
      "epoch": 6.839291966543474,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00015354918895354403,
      "loss": 4.6938,
      "mean_token_accuracy": 0.2424849674105644,
      "num_tokens": 152471296.0,
      "step": 87905
    },
    {
      "entropy": 5.400319576263428,
      "epoch": 6.839680995915192,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015352585874816088,
      "loss": 4.6216,
      "mean_token_accuracy": 0.23863116055727004,
      "num_tokens": 152478997.0,
      "step": 87910
    },
    {
      "entropy": 5.401607465744019,
      "epoch": 6.840070025286909,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015350253038595463,
      "loss": 4.681,
      "mean_token_accuracy": 0.23137616962194443,
      "num_tokens": 152487789.0,
      "step": 87915
    },
    {
      "entropy": 5.442955732345581,
      "epoch": 6.840459054658627,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015347920386727897,
      "loss": 4.7318,
      "mean_token_accuracy": 0.2306816726922989,
      "num_tokens": 152496240.0,
      "step": 87920
    },
    {
      "entropy": 5.349298334121704,
      "epoch": 6.840848084030345,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001534558791924881,
      "loss": 4.5688,
      "mean_token_accuracy": 0.24867893159389495,
      "num_tokens": 152504405.0,
      "step": 87925
    },
    {
      "entropy": 5.378642129898071,
      "epoch": 6.8412371134020615,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00015343255636193583,
      "loss": 4.6453,
      "mean_token_accuracy": 0.23907426446676255,
      "num_tokens": 152513108.0,
      "step": 87930
    },
    {
      "entropy": 5.338487720489502,
      "epoch": 6.841626142773779,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00015340923537597587,
      "loss": 4.6025,
      "mean_token_accuracy": 0.2424149051308632,
      "num_tokens": 152522989.0,
      "step": 87935
    },
    {
      "entropy": 5.385646390914917,
      "epoch": 6.842015172145497,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015338591623496235,
      "loss": 4.7022,
      "mean_token_accuracy": 0.23930055052042007,
      "num_tokens": 152531391.0,
      "step": 87940
    },
    {
      "entropy": 5.3903473854064945,
      "epoch": 6.842404201517215,
      "grad_norm": 1.25,
      "learning_rate": 0.0001533625989392489,
      "loss": 4.6288,
      "mean_token_accuracy": 0.23963913917541504,
      "num_tokens": 152540168.0,
      "step": 87945
    },
    {
      "entropy": 5.364973497390747,
      "epoch": 6.842793230888932,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001533392834891894,
      "loss": 4.6663,
      "mean_token_accuracy": 0.24345590323209762,
      "num_tokens": 152549660.0,
      "step": 87950
    },
    {
      "entropy": 5.3576576709747314,
      "epoch": 6.84318226026065,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015331596988513762,
      "loss": 4.6183,
      "mean_token_accuracy": 0.24452893286943436,
      "num_tokens": 152558471.0,
      "step": 87955
    },
    {
      "entropy": 5.431260395050049,
      "epoch": 6.843571289632367,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015329265812744742,
      "loss": 4.7071,
      "mean_token_accuracy": 0.23741409480571746,
      "num_tokens": 152567687.0,
      "step": 87960
    },
    {
      "entropy": 5.36627163887024,
      "epoch": 6.8439603190040845,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015326934821647238,
      "loss": 4.5771,
      "mean_token_accuracy": 0.23962024301290513,
      "num_tokens": 152576052.0,
      "step": 87965
    },
    {
      "entropy": 5.396063804626465,
      "epoch": 6.844349348375802,
      "grad_norm": 1.25,
      "learning_rate": 0.00015324604015256617,
      "loss": 4.6147,
      "mean_token_accuracy": 0.2478618234395981,
      "num_tokens": 152584697.0,
      "step": 87970
    },
    {
      "entropy": 5.336439895629883,
      "epoch": 6.84473837774752,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001532227339360826,
      "loss": 4.5847,
      "mean_token_accuracy": 0.2527925059199333,
      "num_tokens": 152592982.0,
      "step": 87975
    },
    {
      "entropy": 5.379661941528321,
      "epoch": 6.845127407119238,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001531994295673751,
      "loss": 4.6586,
      "mean_token_accuracy": 0.24514150768518447,
      "num_tokens": 152602108.0,
      "step": 87980
    },
    {
      "entropy": 5.443964576721191,
      "epoch": 6.845516436490955,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015317612704679752,
      "loss": 4.7863,
      "mean_token_accuracy": 0.2256301000714302,
      "num_tokens": 152611578.0,
      "step": 87985
    },
    {
      "entropy": 5.451059865951538,
      "epoch": 6.845905465862673,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001531528263747032,
      "loss": 4.7689,
      "mean_token_accuracy": 0.229328852891922,
      "num_tokens": 152621109.0,
      "step": 87990
    },
    {
      "entropy": 5.441101980209351,
      "epoch": 6.84629449523439,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001531295275514459,
      "loss": 4.6373,
      "mean_token_accuracy": 0.23944544047117233,
      "num_tokens": 152628816.0,
      "step": 87995
    },
    {
      "entropy": 5.3797181129455565,
      "epoch": 6.8466835246061075,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001531062305773789,
      "loss": 4.611,
      "mean_token_accuracy": 0.23547526597976684,
      "num_tokens": 152637393.0,
      "step": 88000
    },
    {
      "entropy": 5.345834445953369,
      "epoch": 6.847072553977825,
      "grad_norm": 1.40625,
      "learning_rate": 0.00015308293545285596,
      "loss": 4.607,
      "mean_token_accuracy": 0.23613484799861909,
      "num_tokens": 152646475.0,
      "step": 88005
    },
    {
      "entropy": 5.478306531906128,
      "epoch": 6.847461583349543,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00015305964217823042,
      "loss": 4.7256,
      "mean_token_accuracy": 0.23159758150577545,
      "num_tokens": 152654453.0,
      "step": 88010
    },
    {
      "entropy": 5.402421140670777,
      "epoch": 6.847850612721261,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015303635075385557,
      "loss": 4.6966,
      "mean_token_accuracy": 0.2322397366166115,
      "num_tokens": 152663696.0,
      "step": 88015
    },
    {
      "entropy": 5.480624723434448,
      "epoch": 6.848239642092978,
      "grad_norm": 1.4375,
      "learning_rate": 0.00015301306118008513,
      "loss": 4.7053,
      "mean_token_accuracy": 0.24171013683080672,
      "num_tokens": 152672991.0,
      "step": 88020
    },
    {
      "entropy": 5.332372379302979,
      "epoch": 6.848628671464695,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015298977345727226,
      "loss": 4.5768,
      "mean_token_accuracy": 0.25284951031208036,
      "num_tokens": 152681582.0,
      "step": 88025
    },
    {
      "entropy": 5.293654584884644,
      "epoch": 6.849017700836413,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001529664875857704,
      "loss": 4.559,
      "mean_token_accuracy": 0.24609773904085158,
      "num_tokens": 152690761.0,
      "step": 88030
    },
    {
      "entropy": 5.401656818389893,
      "epoch": 6.8494067302081305,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015294320356593283,
      "loss": 4.7435,
      "mean_token_accuracy": 0.23611603677272797,
      "num_tokens": 152699509.0,
      "step": 88035
    },
    {
      "entropy": 5.386667680740357,
      "epoch": 6.849795759579848,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00015291992139811295,
      "loss": 4.6464,
      "mean_token_accuracy": 0.24031177312135696,
      "num_tokens": 152707762.0,
      "step": 88040
    },
    {
      "entropy": 5.300319862365723,
      "epoch": 6.850184788951566,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015289664108266395,
      "loss": 4.6041,
      "mean_token_accuracy": 0.24610297083854676,
      "num_tokens": 152715880.0,
      "step": 88045
    },
    {
      "entropy": 5.37862195968628,
      "epoch": 6.850573818323284,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00015287336261993894,
      "loss": 4.6694,
      "mean_token_accuracy": 0.23897553980350494,
      "num_tokens": 152724725.0,
      "step": 88050
    },
    {
      "entropy": 5.385871076583863,
      "epoch": 6.850962847695001,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015285008601029142,
      "loss": 4.6481,
      "mean_token_accuracy": 0.24208159893751144,
      "num_tokens": 152733083.0,
      "step": 88055
    },
    {
      "entropy": 5.453640413284302,
      "epoch": 6.851351877066719,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015282681125407434,
      "loss": 4.6917,
      "mean_token_accuracy": 0.24461700022220612,
      "num_tokens": 152741815.0,
      "step": 88060
    },
    {
      "entropy": 5.330998229980469,
      "epoch": 6.851740906438436,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015280353835164096,
      "loss": 4.5894,
      "mean_token_accuracy": 0.2465410739183426,
      "num_tokens": 152750382.0,
      "step": 88065
    },
    {
      "entropy": 5.384104204177857,
      "epoch": 6.8521299358101535,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015278026730334437,
      "loss": 4.7269,
      "mean_token_accuracy": 0.23490677773952484,
      "num_tokens": 152758882.0,
      "step": 88070
    },
    {
      "entropy": 5.460395956039429,
      "epoch": 6.852518965181871,
      "grad_norm": 1.375,
      "learning_rate": 0.00015275699810953774,
      "loss": 4.6967,
      "mean_token_accuracy": 0.2272198423743248,
      "num_tokens": 152767202.0,
      "step": 88075
    },
    {
      "entropy": 5.404255199432373,
      "epoch": 6.852907994553589,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015273373077057402,
      "loss": 4.6423,
      "mean_token_accuracy": 0.2363704338669777,
      "num_tokens": 152775758.0,
      "step": 88080
    },
    {
      "entropy": 5.426046752929688,
      "epoch": 6.853297023925307,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001527104652868063,
      "loss": 4.6609,
      "mean_token_accuracy": 0.24440321028232576,
      "num_tokens": 152783899.0,
      "step": 88085
    },
    {
      "entropy": 5.395127487182617,
      "epoch": 6.853686053297024,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001526872016585878,
      "loss": 4.6112,
      "mean_token_accuracy": 0.24481153041124343,
      "num_tokens": 152792767.0,
      "step": 88090
    },
    {
      "entropy": 5.4208887100219725,
      "epoch": 6.854075082668741,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015266393988627115,
      "loss": 4.6818,
      "mean_token_accuracy": 0.2405066817998886,
      "num_tokens": 152802235.0,
      "step": 88095
    },
    {
      "entropy": 5.424117136001587,
      "epoch": 6.854464112040459,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015264067997020964,
      "loss": 4.7385,
      "mean_token_accuracy": 0.23192620873451233,
      "num_tokens": 152811297.0,
      "step": 88100
    },
    {
      "entropy": 5.42649507522583,
      "epoch": 6.8548531414121765,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015261742191075594,
      "loss": 4.723,
      "mean_token_accuracy": 0.2274567812681198,
      "num_tokens": 152820816.0,
      "step": 88105
    },
    {
      "entropy": 5.323373556137085,
      "epoch": 6.855242170783894,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00015259416570826315,
      "loss": 4.5634,
      "mean_token_accuracy": 0.252081523835659,
      "num_tokens": 152829336.0,
      "step": 88110
    },
    {
      "entropy": 5.313122940063477,
      "epoch": 6.855631200155612,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000152570911363084,
      "loss": 4.6175,
      "mean_token_accuracy": 0.24802208244800567,
      "num_tokens": 152838776.0,
      "step": 88115
    },
    {
      "entropy": 5.296321249008178,
      "epoch": 6.85602022952733,
      "grad_norm": 1.203125,
      "learning_rate": 0.00015254765887557148,
      "loss": 4.5006,
      "mean_token_accuracy": 0.2550403669476509,
      "num_tokens": 152847863.0,
      "step": 88120
    },
    {
      "entropy": 5.3867145538330075,
      "epoch": 6.856409258899047,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015252440824607832,
      "loss": 4.7165,
      "mean_token_accuracy": 0.23734230697154998,
      "num_tokens": 152855670.0,
      "step": 88125
    },
    {
      "entropy": 5.420558500289917,
      "epoch": 6.856798288270764,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00015250115947495726,
      "loss": 4.695,
      "mean_token_accuracy": 0.23518673777580262,
      "num_tokens": 152864481.0,
      "step": 88130
    },
    {
      "entropy": 5.372547578811646,
      "epoch": 6.857187317642482,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015247791256256123,
      "loss": 4.5955,
      "mean_token_accuracy": 0.23710998445749282,
      "num_tokens": 152873235.0,
      "step": 88135
    },
    {
      "entropy": 5.42131462097168,
      "epoch": 6.8575763470141995,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015245466750924276,
      "loss": 4.6581,
      "mean_token_accuracy": 0.23256264328956605,
      "num_tokens": 152881989.0,
      "step": 88140
    },
    {
      "entropy": 5.340928077697754,
      "epoch": 6.857965376385917,
      "grad_norm": 1.375,
      "learning_rate": 0.00015243142431535473,
      "loss": 4.5256,
      "mean_token_accuracy": 0.2487235888838768,
      "num_tokens": 152890652.0,
      "step": 88145
    },
    {
      "entropy": 5.3610766410827635,
      "epoch": 6.858354405757635,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001524081829812497,
      "loss": 4.5999,
      "mean_token_accuracy": 0.24879216849803926,
      "num_tokens": 152899346.0,
      "step": 88150
    },
    {
      "entropy": 5.43237681388855,
      "epoch": 6.858743435129353,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015238494350728032,
      "loss": 4.7567,
      "mean_token_accuracy": 0.23277739584445953,
      "num_tokens": 152907159.0,
      "step": 88155
    },
    {
      "entropy": 5.407005548477173,
      "epoch": 6.859132464501069,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001523617058937994,
      "loss": 4.6771,
      "mean_token_accuracy": 0.23857893645763398,
      "num_tokens": 152915774.0,
      "step": 88160
    },
    {
      "entropy": 5.405107593536377,
      "epoch": 6.859521493872787,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015233847014115926,
      "loss": 4.6602,
      "mean_token_accuracy": 0.23660048544406892,
      "num_tokens": 152925095.0,
      "step": 88165
    },
    {
      "entropy": 5.351880931854248,
      "epoch": 6.859910523244505,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00015231523624971273,
      "loss": 4.5824,
      "mean_token_accuracy": 0.24751725047826767,
      "num_tokens": 152933835.0,
      "step": 88170
    },
    {
      "entropy": 5.386205720901489,
      "epoch": 6.8602995526162225,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001522920042198121,
      "loss": 4.6983,
      "mean_token_accuracy": 0.24232011437416076,
      "num_tokens": 152941785.0,
      "step": 88175
    },
    {
      "entropy": 5.355608940124512,
      "epoch": 6.86068858198794,
      "grad_norm": 1.25,
      "learning_rate": 0.00015226877405181006,
      "loss": 4.5887,
      "mean_token_accuracy": 0.2437748834490776,
      "num_tokens": 152950721.0,
      "step": 88180
    },
    {
      "entropy": 5.388758611679077,
      "epoch": 6.861077611359658,
      "grad_norm": 1.375,
      "learning_rate": 0.00015224554574605898,
      "loss": 4.753,
      "mean_token_accuracy": 0.23230988532304764,
      "num_tokens": 152959570.0,
      "step": 88185
    },
    {
      "entropy": 5.404854249954224,
      "epoch": 6.861466640731376,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015222231930291143,
      "loss": 4.648,
      "mean_token_accuracy": 0.24176997393369676,
      "num_tokens": 152968187.0,
      "step": 88190
    },
    {
      "entropy": 5.362402105331421,
      "epoch": 6.861855670103093,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015219909472271966,
      "loss": 4.5705,
      "mean_token_accuracy": 0.2397889330983162,
      "num_tokens": 152976993.0,
      "step": 88195
    },
    {
      "entropy": 5.4866022109985355,
      "epoch": 6.86224469947481,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00015217587200583627,
      "loss": 4.7078,
      "mean_token_accuracy": 0.23766548782587052,
      "num_tokens": 152984873.0,
      "step": 88200
    },
    {
      "entropy": 5.411961507797241,
      "epoch": 6.862633728846528,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015215265115261353,
      "loss": 4.6538,
      "mean_token_accuracy": 0.23856997936964036,
      "num_tokens": 152993979.0,
      "step": 88205
    },
    {
      "entropy": 5.373691177368164,
      "epoch": 6.8630227582182455,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00015212943216340373,
      "loss": 4.6535,
      "mean_token_accuracy": 0.23926458060741423,
      "num_tokens": 153002588.0,
      "step": 88210
    },
    {
      "entropy": 5.391956043243408,
      "epoch": 6.863411787589963,
      "grad_norm": 1.453125,
      "learning_rate": 0.00015210621503855917,
      "loss": 4.6178,
      "mean_token_accuracy": 0.2507480472326279,
      "num_tokens": 153010449.0,
      "step": 88215
    },
    {
      "entropy": 5.34791316986084,
      "epoch": 6.863800816961681,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015208299977843221,
      "loss": 4.6365,
      "mean_token_accuracy": 0.23359389305114747,
      "num_tokens": 153019049.0,
      "step": 88220
    },
    {
      "entropy": 5.508857727050781,
      "epoch": 6.864189846333398,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001520597863833752,
      "loss": 4.7719,
      "mean_token_accuracy": 0.23004242777824402,
      "num_tokens": 153027019.0,
      "step": 88225
    },
    {
      "entropy": 5.334723949432373,
      "epoch": 6.864578875705115,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015203657485374022,
      "loss": 4.6154,
      "mean_token_accuracy": 0.2397428721189499,
      "num_tokens": 153036164.0,
      "step": 88230
    },
    {
      "entropy": 5.428294515609741,
      "epoch": 6.864967905076833,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015201336518987942,
      "loss": 4.7008,
      "mean_token_accuracy": 0.23172731101512908,
      "num_tokens": 153044426.0,
      "step": 88235
    },
    {
      "entropy": 5.36576247215271,
      "epoch": 6.865356934448551,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015199015739214515,
      "loss": 4.5719,
      "mean_token_accuracy": 0.24463237226009368,
      "num_tokens": 153053083.0,
      "step": 88240
    },
    {
      "entropy": 5.372410869598388,
      "epoch": 6.8657459638202685,
      "grad_norm": 1.375,
      "learning_rate": 0.00015196695146088934,
      "loss": 4.6483,
      "mean_token_accuracy": 0.24409525990486144,
      "num_tokens": 153061952.0,
      "step": 88245
    },
    {
      "entropy": 5.3827064514160154,
      "epoch": 6.866134993191986,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00015194374739646433,
      "loss": 4.6298,
      "mean_token_accuracy": 0.24847858697175979,
      "num_tokens": 153070668.0,
      "step": 88250
    },
    {
      "entropy": 5.337804317474365,
      "epoch": 6.866524022563704,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000151920545199222,
      "loss": 4.6425,
      "mean_token_accuracy": 0.24685653895139695,
      "num_tokens": 153079231.0,
      "step": 88255
    },
    {
      "entropy": 5.410259056091308,
      "epoch": 6.866913051935422,
      "grad_norm": 1.46875,
      "learning_rate": 0.00015189734486951452,
      "loss": 4.6893,
      "mean_token_accuracy": 0.23687737584114074,
      "num_tokens": 153087401.0,
      "step": 88260
    },
    {
      "entropy": 5.4097654819488525,
      "epoch": 6.867302081307138,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00015187414640769386,
      "loss": 4.6409,
      "mean_token_accuracy": 0.2379336416721344,
      "num_tokens": 153096083.0,
      "step": 88265
    },
    {
      "entropy": 5.356826210021973,
      "epoch": 6.867691110678856,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015185094981411213,
      "loss": 4.6036,
      "mean_token_accuracy": 0.24969720989465713,
      "num_tokens": 153104457.0,
      "step": 88270
    },
    {
      "entropy": 5.406849050521851,
      "epoch": 6.868080140050574,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015182775508912116,
      "loss": 4.6766,
      "mean_token_accuracy": 0.2392251268029213,
      "num_tokens": 153113989.0,
      "step": 88275
    },
    {
      "entropy": 5.35966477394104,
      "epoch": 6.8684691694222915,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015180456223307296,
      "loss": 4.6572,
      "mean_token_accuracy": 0.24257880449295044,
      "num_tokens": 153122486.0,
      "step": 88280
    },
    {
      "entropy": 5.353142881393433,
      "epoch": 6.868858198794009,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001517813712463195,
      "loss": 4.581,
      "mean_token_accuracy": 0.24803954511880874,
      "num_tokens": 153131023.0,
      "step": 88285
    },
    {
      "entropy": 5.3285778045654295,
      "epoch": 6.869247228165727,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00015175818212921256,
      "loss": 4.5875,
      "mean_token_accuracy": 0.25140090435743334,
      "num_tokens": 153138931.0,
      "step": 88290
    },
    {
      "entropy": 5.3311262130737305,
      "epoch": 6.869636257537444,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015173499488210408,
      "loss": 4.5069,
      "mean_token_accuracy": 0.24684303551912307,
      "num_tokens": 153147422.0,
      "step": 88295
    },
    {
      "entropy": 5.392298221588135,
      "epoch": 6.870025286909161,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001517118095053458,
      "loss": 4.6771,
      "mean_token_accuracy": 0.23906631469726564,
      "num_tokens": 153155298.0,
      "step": 88300
    },
    {
      "entropy": 5.474166536331177,
      "epoch": 6.870414316280879,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015168862599928965,
      "loss": 4.6768,
      "mean_token_accuracy": 0.24186852872371672,
      "num_tokens": 153163543.0,
      "step": 88305
    },
    {
      "entropy": 5.399763917922973,
      "epoch": 6.870803345652597,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015166544436428733,
      "loss": 4.6819,
      "mean_token_accuracy": 0.23748457580804824,
      "num_tokens": 153173422.0,
      "step": 88310
    },
    {
      "entropy": 5.45841064453125,
      "epoch": 6.8711923750243145,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015164226460069052,
      "loss": 4.7161,
      "mean_token_accuracy": 0.2328694835305214,
      "num_tokens": 153182226.0,
      "step": 88315
    },
    {
      "entropy": 5.440245294570923,
      "epoch": 6.871581404396032,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015161908670885106,
      "loss": 4.7354,
      "mean_token_accuracy": 0.23037155121564865,
      "num_tokens": 153190992.0,
      "step": 88320
    },
    {
      "entropy": 5.33632173538208,
      "epoch": 6.87197043376775,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015159591068912055,
      "loss": 4.5798,
      "mean_token_accuracy": 0.2459864780306816,
      "num_tokens": 153200359.0,
      "step": 88325
    },
    {
      "entropy": 5.435486602783203,
      "epoch": 6.872359463139467,
      "grad_norm": 1.375,
      "learning_rate": 0.0001515727365418507,
      "loss": 4.64,
      "mean_token_accuracy": 0.24065154790878296,
      "num_tokens": 153208890.0,
      "step": 88330
    },
    {
      "entropy": 5.363480758666992,
      "epoch": 6.872748492511184,
      "grad_norm": 1.4375,
      "learning_rate": 0.00015154956426739302,
      "loss": 4.6872,
      "mean_token_accuracy": 0.24278179705142974,
      "num_tokens": 153217083.0,
      "step": 88335
    },
    {
      "entropy": 5.3804277896881105,
      "epoch": 6.873137521882902,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015152639386609935,
      "loss": 4.6653,
      "mean_token_accuracy": 0.23850482553243638,
      "num_tokens": 153226024.0,
      "step": 88340
    },
    {
      "entropy": 5.38779649734497,
      "epoch": 6.87352655125462,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00015150322533832102,
      "loss": 4.6452,
      "mean_token_accuracy": 0.24300952553749083,
      "num_tokens": 153234364.0,
      "step": 88345
    },
    {
      "entropy": 5.379233932495117,
      "epoch": 6.8739155806263375,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015148005868440966,
      "loss": 4.5847,
      "mean_token_accuracy": 0.2478453665971756,
      "num_tokens": 153243065.0,
      "step": 88350
    },
    {
      "entropy": 5.369056177139282,
      "epoch": 6.874304609998055,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001514568939047169,
      "loss": 4.6209,
      "mean_token_accuracy": 0.23920433968305588,
      "num_tokens": 153251838.0,
      "step": 88355
    },
    {
      "entropy": 5.259841394424439,
      "epoch": 6.874693639369772,
      "grad_norm": 1.375,
      "learning_rate": 0.00015143373099959408,
      "loss": 4.5227,
      "mean_token_accuracy": 0.2548205733299255,
      "num_tokens": 153260144.0,
      "step": 88360
    },
    {
      "entropy": 5.362350940704346,
      "epoch": 6.87508266874149,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001514105699693928,
      "loss": 4.5921,
      "mean_token_accuracy": 0.2408697247505188,
      "num_tokens": 153268403.0,
      "step": 88365
    },
    {
      "entropy": 5.361438941955567,
      "epoch": 6.875471698113207,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00015138741081446427,
      "loss": 4.6148,
      "mean_token_accuracy": 0.2431598886847496,
      "num_tokens": 153277658.0,
      "step": 88370
    },
    {
      "entropy": 5.4265604496002195,
      "epoch": 6.875860727484925,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001513642535351602,
      "loss": 4.6898,
      "mean_token_accuracy": 0.22911168932914733,
      "num_tokens": 153286965.0,
      "step": 88375
    },
    {
      "entropy": 5.373702955245972,
      "epoch": 6.876249756856643,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001513410981318316,
      "loss": 4.6111,
      "mean_token_accuracy": 0.2453094944357872,
      "num_tokens": 153296225.0,
      "step": 88380
    },
    {
      "entropy": 5.433220052719117,
      "epoch": 6.8766387862283604,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001513179446048302,
      "loss": 4.6861,
      "mean_token_accuracy": 0.24074286818504334,
      "num_tokens": 153304558.0,
      "step": 88385
    },
    {
      "entropy": 5.422859764099121,
      "epoch": 6.877027815600078,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001512947929545071,
      "loss": 4.6278,
      "mean_token_accuracy": 0.24786188900470735,
      "num_tokens": 153312853.0,
      "step": 88390
    },
    {
      "entropy": 5.354819917678833,
      "epoch": 6.877416844971796,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015127164318121353,
      "loss": 4.6232,
      "mean_token_accuracy": 0.24665165096521377,
      "num_tokens": 153321351.0,
      "step": 88395
    },
    {
      "entropy": 5.437063264846802,
      "epoch": 6.877805874343513,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001512484952853009,
      "loss": 4.7449,
      "mean_token_accuracy": 0.2301986888051033,
      "num_tokens": 153330543.0,
      "step": 88400
    },
    {
      "entropy": 5.374841022491455,
      "epoch": 6.87819490371523,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015122534926712033,
      "loss": 4.5833,
      "mean_token_accuracy": 0.24829187095165253,
      "num_tokens": 153338280.0,
      "step": 88405
    },
    {
      "entropy": 5.366238832473755,
      "epoch": 6.878583933086948,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00015120220512702316,
      "loss": 4.6543,
      "mean_token_accuracy": 0.23877047896385192,
      "num_tokens": 153347573.0,
      "step": 88410
    },
    {
      "entropy": 5.316925811767578,
      "epoch": 6.878972962458666,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001511790628653603,
      "loss": 4.5607,
      "mean_token_accuracy": 0.24729665368795395,
      "num_tokens": 153355692.0,
      "step": 88415
    },
    {
      "entropy": 5.284665107727051,
      "epoch": 6.879361991830383,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015115592248248326,
      "loss": 4.5906,
      "mean_token_accuracy": 0.23811761289834976,
      "num_tokens": 153364364.0,
      "step": 88420
    },
    {
      "entropy": 5.33720908164978,
      "epoch": 6.879751021202101,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00015113278397874296,
      "loss": 4.6235,
      "mean_token_accuracy": 0.23832522481679916,
      "num_tokens": 153373703.0,
      "step": 88425
    },
    {
      "entropy": 5.423511409759522,
      "epoch": 6.880140050573818,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015110964735449045,
      "loss": 4.6948,
      "mean_token_accuracy": 0.23677553683519365,
      "num_tokens": 153382450.0,
      "step": 88430
    },
    {
      "entropy": 5.422382116317749,
      "epoch": 6.880529079945536,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015108651261007693,
      "loss": 4.6689,
      "mean_token_accuracy": 0.23492585122585297,
      "num_tokens": 153391298.0,
      "step": 88435
    },
    {
      "entropy": 5.438583135604858,
      "epoch": 6.880918109317253,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015106337974585325,
      "loss": 4.7183,
      "mean_token_accuracy": 0.23870889544487,
      "num_tokens": 153399968.0,
      "step": 88440
    },
    {
      "entropy": 5.338947391510009,
      "epoch": 6.881307138688971,
      "grad_norm": 1.375,
      "learning_rate": 0.0001510402487621706,
      "loss": 4.6171,
      "mean_token_accuracy": 0.23897689580917358,
      "num_tokens": 153408509.0,
      "step": 88445
    },
    {
      "entropy": 5.4103001117706295,
      "epoch": 6.881696168060689,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001510171196593798,
      "loss": 4.7112,
      "mean_token_accuracy": 0.2342088595032692,
      "num_tokens": 153416739.0,
      "step": 88450
    },
    {
      "entropy": 5.399285936355591,
      "epoch": 6.882085197432406,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015099399243783193,
      "loss": 4.6047,
      "mean_token_accuracy": 0.24574101716279984,
      "num_tokens": 153424400.0,
      "step": 88455
    },
    {
      "entropy": 5.348751020431519,
      "epoch": 6.882474226804124,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00015097086709787777,
      "loss": 4.6183,
      "mean_token_accuracy": 0.2380060762166977,
      "num_tokens": 153432816.0,
      "step": 88460
    },
    {
      "entropy": 5.315586376190185,
      "epoch": 6.882863256175841,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001509477436398684,
      "loss": 4.5249,
      "mean_token_accuracy": 0.24520407915115355,
      "num_tokens": 153441306.0,
      "step": 88465
    },
    {
      "entropy": 5.324357748031616,
      "epoch": 6.883252285547559,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001509246220641546,
      "loss": 4.5663,
      "mean_token_accuracy": 0.24802208691835403,
      "num_tokens": 153449755.0,
      "step": 88470
    },
    {
      "entropy": 5.436161947250366,
      "epoch": 6.883641314919276,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00015090150237108703,
      "loss": 4.7054,
      "mean_token_accuracy": 0.239231376349926,
      "num_tokens": 153458084.0,
      "step": 88475
    },
    {
      "entropy": 5.2197277545928955,
      "epoch": 6.884030344290994,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015087838456101664,
      "loss": 4.4909,
      "mean_token_accuracy": 0.25081573724746703,
      "num_tokens": 153467389.0,
      "step": 88480
    },
    {
      "entropy": 5.405291795730591,
      "epoch": 6.884419373662712,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00015085526863429426,
      "loss": 4.6619,
      "mean_token_accuracy": 0.2323514476418495,
      "num_tokens": 153476991.0,
      "step": 88485
    },
    {
      "entropy": 5.402490663528442,
      "epoch": 6.884808403034429,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001508321545912706,
      "loss": 4.675,
      "mean_token_accuracy": 0.2405345171689987,
      "num_tokens": 153484907.0,
      "step": 88490
    },
    {
      "entropy": 5.410006809234619,
      "epoch": 6.885197432406146,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015080904243229632,
      "loss": 4.7367,
      "mean_token_accuracy": 0.22872378975152968,
      "num_tokens": 153493758.0,
      "step": 88495
    },
    {
      "entropy": 5.322557020187378,
      "epoch": 6.885586461777864,
      "grad_norm": 2.28125,
      "learning_rate": 0.0001507859321577222,
      "loss": 4.5055,
      "mean_token_accuracy": 0.26338198482990266,
      "num_tokens": 153503236.0,
      "step": 88500
    },
    {
      "entropy": 5.365322875976562,
      "epoch": 6.885975491149582,
      "grad_norm": 1.375,
      "learning_rate": 0.0001507628237678989,
      "loss": 4.6712,
      "mean_token_accuracy": 0.23549790680408478,
      "num_tokens": 153512235.0,
      "step": 88505
    },
    {
      "entropy": 5.358061838150024,
      "epoch": 6.886364520521299,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015073971726317688,
      "loss": 4.673,
      "mean_token_accuracy": 0.23261573761701584,
      "num_tokens": 153520375.0,
      "step": 88510
    },
    {
      "entropy": 5.314078044891358,
      "epoch": 6.886753549893017,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015071661264390697,
      "loss": 4.5405,
      "mean_token_accuracy": 0.24786121547222137,
      "num_tokens": 153528957.0,
      "step": 88515
    },
    {
      "entropy": 5.35184063911438,
      "epoch": 6.887142579264735,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015069350991043956,
      "loss": 4.6625,
      "mean_token_accuracy": 0.24231164455413817,
      "num_tokens": 153538027.0,
      "step": 88520
    },
    {
      "entropy": 5.33264389038086,
      "epoch": 6.887531608636452,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015067040906312538,
      "loss": 4.5537,
      "mean_token_accuracy": 0.2541811317205429,
      "num_tokens": 153546800.0,
      "step": 88525
    },
    {
      "entropy": 5.360409688949585,
      "epoch": 6.88792063800817,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00015064731010231474,
      "loss": 4.6154,
      "mean_token_accuracy": 0.23733845502138137,
      "num_tokens": 153554787.0,
      "step": 88530
    },
    {
      "entropy": 5.410138130187988,
      "epoch": 6.888309667379887,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015062421302835835,
      "loss": 4.6875,
      "mean_token_accuracy": 0.24595928490161895,
      "num_tokens": 153562990.0,
      "step": 88535
    },
    {
      "entropy": 5.366688299179077,
      "epoch": 6.888698696751605,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00015060111784160647,
      "loss": 4.5895,
      "mean_token_accuracy": 0.2460399627685547,
      "num_tokens": 153571480.0,
      "step": 88540
    },
    {
      "entropy": 5.417804002761841,
      "epoch": 6.889087726123322,
      "grad_norm": 1.375,
      "learning_rate": 0.00015057802454240962,
      "loss": 4.6404,
      "mean_token_accuracy": 0.24600243270397187,
      "num_tokens": 153580281.0,
      "step": 88545
    },
    {
      "entropy": 5.3529750347137455,
      "epoch": 6.88947675549504,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015055493313111828,
      "loss": 4.5554,
      "mean_token_accuracy": 0.2372136577963829,
      "num_tokens": 153588535.0,
      "step": 88550
    },
    {
      "entropy": 5.361924076080323,
      "epoch": 6.889865784866758,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001505318436080827,
      "loss": 4.6297,
      "mean_token_accuracy": 0.24531494379043578,
      "num_tokens": 153597183.0,
      "step": 88555
    },
    {
      "entropy": 5.418739175796508,
      "epoch": 6.8902548142384745,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015050875597365338,
      "loss": 4.6873,
      "mean_token_accuracy": 0.23992316275835038,
      "num_tokens": 153605279.0,
      "step": 88560
    },
    {
      "entropy": 5.323651075363159,
      "epoch": 6.890643843610192,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015048567022818042,
      "loss": 4.6088,
      "mean_token_accuracy": 0.24772369265556335,
      "num_tokens": 153613778.0,
      "step": 88565
    },
    {
      "entropy": 5.398797512054443,
      "epoch": 6.89103287298191,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001504625863720143,
      "loss": 4.6958,
      "mean_token_accuracy": 0.23658832162618637,
      "num_tokens": 153623314.0,
      "step": 88570
    },
    {
      "entropy": 5.364537239074707,
      "epoch": 6.891421902353628,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015043950440550514,
      "loss": 4.6143,
      "mean_token_accuracy": 0.23889920562505723,
      "num_tokens": 153631698.0,
      "step": 88575
    },
    {
      "entropy": 5.389769172668457,
      "epoch": 6.891810931725345,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015041642432900336,
      "loss": 4.606,
      "mean_token_accuracy": 0.24734739661216737,
      "num_tokens": 153639976.0,
      "step": 88580
    },
    {
      "entropy": 5.371002149581909,
      "epoch": 6.892199961097063,
      "grad_norm": 1.296875,
      "learning_rate": 0.000150393346142859,
      "loss": 4.5954,
      "mean_token_accuracy": 0.2475140392780304,
      "num_tokens": 153648509.0,
      "step": 88585
    },
    {
      "entropy": 5.300017929077148,
      "epoch": 6.892588990468781,
      "grad_norm": 1.40625,
      "learning_rate": 0.00015037026984742225,
      "loss": 4.5867,
      "mean_token_accuracy": 0.25273074209690094,
      "num_tokens": 153657183.0,
      "step": 88590
    },
    {
      "entropy": 5.32399435043335,
      "epoch": 6.892978019840498,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015034719544304332,
      "loss": 4.5093,
      "mean_token_accuracy": 0.2505064830183983,
      "num_tokens": 153665211.0,
      "step": 88595
    },
    {
      "entropy": 5.3798675537109375,
      "epoch": 6.893367049212215,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015032412293007223,
      "loss": 4.7012,
      "mean_token_accuracy": 0.23629248589277269,
      "num_tokens": 153674145.0,
      "step": 88600
    },
    {
      "entropy": 5.359813642501831,
      "epoch": 6.893756078583933,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001503010523088592,
      "loss": 4.5537,
      "mean_token_accuracy": 0.24048223346471786,
      "num_tokens": 153682335.0,
      "step": 88605
    },
    {
      "entropy": 5.462742567062378,
      "epoch": 6.894145107955651,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015027798357975417,
      "loss": 4.7781,
      "mean_token_accuracy": 0.2250877559185028,
      "num_tokens": 153691118.0,
      "step": 88610
    },
    {
      "entropy": 5.392640924453735,
      "epoch": 6.894534137327368,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001502549167431072,
      "loss": 4.62,
      "mean_token_accuracy": 0.24503267854452132,
      "num_tokens": 153699604.0,
      "step": 88615
    },
    {
      "entropy": 5.379016017913818,
      "epoch": 6.894923166699086,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001502318517992684,
      "loss": 4.6289,
      "mean_token_accuracy": 0.24392159730196,
      "num_tokens": 153708290.0,
      "step": 88620
    },
    {
      "entropy": 5.3818854808807375,
      "epoch": 6.895312196070804,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00015020878874858756,
      "loss": 4.6647,
      "mean_token_accuracy": 0.24295762330293655,
      "num_tokens": 153717815.0,
      "step": 88625
    },
    {
      "entropy": 5.437975072860718,
      "epoch": 6.8957012254425205,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015018572759141485,
      "loss": 4.6778,
      "mean_token_accuracy": 0.24273097217082978,
      "num_tokens": 153726573.0,
      "step": 88630
    },
    {
      "entropy": 5.361535263061524,
      "epoch": 6.896090254814238,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001501626683281,
      "loss": 4.6174,
      "mean_token_accuracy": 0.23608837574720382,
      "num_tokens": 153735205.0,
      "step": 88635
    },
    {
      "entropy": 5.351504135131836,
      "epoch": 6.896479284185956,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015013961095899299,
      "loss": 4.606,
      "mean_token_accuracy": 0.24608503729104997,
      "num_tokens": 153743640.0,
      "step": 88640
    },
    {
      "entropy": 5.453222560882568,
      "epoch": 6.896868313557674,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015011655548444357,
      "loss": 4.6885,
      "mean_token_accuracy": 0.24106051176786422,
      "num_tokens": 153751937.0,
      "step": 88645
    },
    {
      "entropy": 5.416488885879517,
      "epoch": 6.897257342929391,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015009350190480172,
      "loss": 4.6753,
      "mean_token_accuracy": 0.2332911878824234,
      "num_tokens": 153760013.0,
      "step": 88650
    },
    {
      "entropy": 5.406840467453003,
      "epoch": 6.897646372301109,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001500704502204171,
      "loss": 4.6684,
      "mean_token_accuracy": 0.23459284156560897,
      "num_tokens": 153770273.0,
      "step": 88655
    },
    {
      "entropy": 5.388297986984253,
      "epoch": 6.898035401672827,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015004740043163966,
      "loss": 4.5973,
      "mean_token_accuracy": 0.23962300419807434,
      "num_tokens": 153779557.0,
      "step": 88660
    },
    {
      "entropy": 5.403120231628418,
      "epoch": 6.8984244310445435,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015002435253881903,
      "loss": 4.631,
      "mean_token_accuracy": 0.2403094843029976,
      "num_tokens": 153788417.0,
      "step": 88665
    },
    {
      "entropy": 5.378290271759033,
      "epoch": 6.898813460416261,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001500013065423048,
      "loss": 4.6305,
      "mean_token_accuracy": 0.24351128190755844,
      "num_tokens": 153797499.0,
      "step": 88670
    },
    {
      "entropy": 5.314743232727051,
      "epoch": 6.899202489787979,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001499782624424469,
      "loss": 4.5561,
      "mean_token_accuracy": 0.24375635385513306,
      "num_tokens": 153805990.0,
      "step": 88675
    },
    {
      "entropy": 5.455041885375977,
      "epoch": 6.899591519159697,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014995522023959475,
      "loss": 4.7503,
      "mean_token_accuracy": 0.23148346543312073,
      "num_tokens": 153814408.0,
      "step": 88680
    },
    {
      "entropy": 5.41623010635376,
      "epoch": 6.899980548531414,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001499321799340982,
      "loss": 4.698,
      "mean_token_accuracy": 0.23474262058734893,
      "num_tokens": 153823551.0,
      "step": 88685
    },
    {
      "entropy": 5.392134141921997,
      "epoch": 6.900369577903132,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001499091415263068,
      "loss": 4.6944,
      "mean_token_accuracy": 0.23647865056991577,
      "num_tokens": 153832775.0,
      "step": 88690
    },
    {
      "entropy": 5.440068912506104,
      "epoch": 6.900758607274849,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014988610501656997,
      "loss": 4.7012,
      "mean_token_accuracy": 0.23734313547611235,
      "num_tokens": 153841311.0,
      "step": 88695
    },
    {
      "entropy": 5.4537450790405275,
      "epoch": 6.9011476366465665,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014986307040523744,
      "loss": 4.7385,
      "mean_token_accuracy": 0.23397668898105622,
      "num_tokens": 153850542.0,
      "step": 88700
    },
    {
      "entropy": 5.412152576446533,
      "epoch": 6.901536666018284,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014984003769265858,
      "loss": 4.6798,
      "mean_token_accuracy": 0.23768652379512786,
      "num_tokens": 153858840.0,
      "step": 88705
    },
    {
      "entropy": 5.4229193210601805,
      "epoch": 6.901925695390002,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000149817006879183,
      "loss": 4.6352,
      "mean_token_accuracy": 0.240244522690773,
      "num_tokens": 153868019.0,
      "step": 88710
    },
    {
      "entropy": 5.383130836486816,
      "epoch": 6.90231472476172,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014979397796516003,
      "loss": 4.5863,
      "mean_token_accuracy": 0.2476246774196625,
      "num_tokens": 153876637.0,
      "step": 88715
    },
    {
      "entropy": 5.264183044433594,
      "epoch": 6.902703754133437,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014977095095093922,
      "loss": 4.5738,
      "mean_token_accuracy": 0.24169825315475463,
      "num_tokens": 153885272.0,
      "step": 88720
    },
    {
      "entropy": 5.344964647293091,
      "epoch": 6.903092783505155,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001497479258368698,
      "loss": 4.6695,
      "mean_token_accuracy": 0.24257688820362092,
      "num_tokens": 153894645.0,
      "step": 88725
    },
    {
      "entropy": 5.396985769271851,
      "epoch": 6.903481812876873,
      "grad_norm": 1.265625,
      "learning_rate": 0.00014972490262330136,
      "loss": 4.7219,
      "mean_token_accuracy": 0.2377770110964775,
      "num_tokens": 153903319.0,
      "step": 88730
    },
    {
      "entropy": 5.427166509628296,
      "epoch": 6.9038708422485895,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014970188131058305,
      "loss": 4.6862,
      "mean_token_accuracy": 0.2332024395465851,
      "num_tokens": 153912570.0,
      "step": 88735
    },
    {
      "entropy": 5.3763445854187015,
      "epoch": 6.904259871620307,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001496788618990643,
      "loss": 4.6162,
      "mean_token_accuracy": 0.23665273934602737,
      "num_tokens": 153921525.0,
      "step": 88740
    },
    {
      "entropy": 5.463285827636719,
      "epoch": 6.904648900992025,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014965584438909435,
      "loss": 4.7431,
      "mean_token_accuracy": 0.2298821583390236,
      "num_tokens": 153929899.0,
      "step": 88745
    },
    {
      "entropy": 5.459270572662353,
      "epoch": 6.905037930363743,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0001496328287810224,
      "loss": 4.7853,
      "mean_token_accuracy": 0.22772430032491683,
      "num_tokens": 153938147.0,
      "step": 88750
    },
    {
      "entropy": 5.377769041061401,
      "epoch": 6.90542695973546,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001496098150751978,
      "loss": 4.6621,
      "mean_token_accuracy": 0.24288063496351242,
      "num_tokens": 153946540.0,
      "step": 88755
    },
    {
      "entropy": 5.4588418960571286,
      "epoch": 6.905815989107178,
      "grad_norm": 1.234375,
      "learning_rate": 0.00014958680327196964,
      "loss": 4.7277,
      "mean_token_accuracy": 0.23587173372507095,
      "num_tokens": 153955765.0,
      "step": 88760
    },
    {
      "entropy": 5.425323343276977,
      "epoch": 6.906205018478895,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014956379337168716,
      "loss": 4.6473,
      "mean_token_accuracy": 0.24373477846384048,
      "num_tokens": 153964491.0,
      "step": 88765
    },
    {
      "entropy": 5.483047008514404,
      "epoch": 6.9065940478506125,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001495407853746995,
      "loss": 4.7523,
      "mean_token_accuracy": 0.22818804234266282,
      "num_tokens": 153973755.0,
      "step": 88770
    },
    {
      "entropy": 5.374415159225464,
      "epoch": 6.90698307722233,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014951777928135562,
      "loss": 4.6045,
      "mean_token_accuracy": 0.25326897352933886,
      "num_tokens": 153981638.0,
      "step": 88775
    },
    {
      "entropy": 5.362479400634766,
      "epoch": 6.907372106594048,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001494947750920048,
      "loss": 4.6247,
      "mean_token_accuracy": 0.2423771321773529,
      "num_tokens": 153989990.0,
      "step": 88780
    },
    {
      "entropy": 5.415796852111816,
      "epoch": 6.907761135965766,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014947177280699597,
      "loss": 4.6072,
      "mean_token_accuracy": 0.24769791811704636,
      "num_tokens": 153998712.0,
      "step": 88785
    },
    {
      "entropy": 5.392825317382813,
      "epoch": 6.908150165337483,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014944877242667823,
      "loss": 4.6781,
      "mean_token_accuracy": 0.243703892827034,
      "num_tokens": 154007872.0,
      "step": 88790
    },
    {
      "entropy": 5.363379859924317,
      "epoch": 6.908539194709201,
      "grad_norm": 1.265625,
      "learning_rate": 0.00014942577395140046,
      "loss": 4.5588,
      "mean_token_accuracy": 0.24871378242969513,
      "num_tokens": 154017131.0,
      "step": 88795
    },
    {
      "entropy": 5.362335824966431,
      "epoch": 6.908928224080918,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014940277738151176,
      "loss": 4.6069,
      "mean_token_accuracy": 0.2406286746263504,
      "num_tokens": 154025236.0,
      "step": 88800
    },
    {
      "entropy": 5.411101818084717,
      "epoch": 6.9093172534526355,
      "grad_norm": 1.421875,
      "learning_rate": 0.00014937978271736092,
      "loss": 4.7486,
      "mean_token_accuracy": 0.23224181532859803,
      "num_tokens": 154033619.0,
      "step": 88805
    },
    {
      "entropy": 5.4426836490631105,
      "epoch": 6.909706282824353,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014935678995929704,
      "loss": 4.7387,
      "mean_token_accuracy": 0.23622021526098252,
      "num_tokens": 154043214.0,
      "step": 88810
    },
    {
      "entropy": 5.3304368019104,
      "epoch": 6.910095312196071,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00014933379910766882,
      "loss": 4.5603,
      "mean_token_accuracy": 0.2521576344966888,
      "num_tokens": 154051915.0,
      "step": 88815
    },
    {
      "entropy": 5.470569324493408,
      "epoch": 6.910484341567789,
      "grad_norm": 1.546875,
      "learning_rate": 0.00014931081016282516,
      "loss": 4.7359,
      "mean_token_accuracy": 0.23914235532283784,
      "num_tokens": 154060406.0,
      "step": 88820
    },
    {
      "entropy": 5.425373601913452,
      "epoch": 6.910873370939506,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014928782312511498,
      "loss": 4.7062,
      "mean_token_accuracy": 0.23121155947446823,
      "num_tokens": 154068218.0,
      "step": 88825
    },
    {
      "entropy": 5.330532073974609,
      "epoch": 6.911262400311223,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014926483799488694,
      "loss": 4.5581,
      "mean_token_accuracy": 0.24465374648571014,
      "num_tokens": 154076714.0,
      "step": 88830
    },
    {
      "entropy": 5.349448299407959,
      "epoch": 6.911651429682941,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001492418547724899,
      "loss": 4.6414,
      "mean_token_accuracy": 0.237048676609993,
      "num_tokens": 154085457.0,
      "step": 88835
    },
    {
      "entropy": 5.308164501190186,
      "epoch": 6.9120404590546585,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001492188734582725,
      "loss": 4.5091,
      "mean_token_accuracy": 0.24974465370178223,
      "num_tokens": 154094473.0,
      "step": 88840
    },
    {
      "entropy": 5.460560274124146,
      "epoch": 6.912429488426376,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014919589405258354,
      "loss": 4.7241,
      "mean_token_accuracy": 0.24056214988231658,
      "num_tokens": 154104027.0,
      "step": 88845
    },
    {
      "entropy": 5.396836137771606,
      "epoch": 6.912818517798094,
      "grad_norm": 1.421875,
      "learning_rate": 0.00014917291655577164,
      "loss": 4.62,
      "mean_token_accuracy": 0.24340587705373765,
      "num_tokens": 154112264.0,
      "step": 88850
    },
    {
      "entropy": 5.378000068664551,
      "epoch": 6.913207547169812,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001491499409681854,
      "loss": 4.5758,
      "mean_token_accuracy": 0.2447158008813858,
      "num_tokens": 154120628.0,
      "step": 88855
    },
    {
      "entropy": 5.436779499053955,
      "epoch": 6.913596576541529,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001491269672901736,
      "loss": 4.697,
      "mean_token_accuracy": 0.2351957693696022,
      "num_tokens": 154128880.0,
      "step": 88860
    },
    {
      "entropy": 5.406386232376098,
      "epoch": 6.913985605913246,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014910399552208462,
      "loss": 4.6636,
      "mean_token_accuracy": 0.2437896892428398,
      "num_tokens": 154137732.0,
      "step": 88865
    },
    {
      "entropy": 5.392245674133301,
      "epoch": 6.914374635284964,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014908102566426722,
      "loss": 4.732,
      "mean_token_accuracy": 0.23933811038732528,
      "num_tokens": 154146460.0,
      "step": 88870
    },
    {
      "entropy": 5.4168447971344,
      "epoch": 6.9147636646566815,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001490580577170697,
      "loss": 4.6285,
      "mean_token_accuracy": 0.23783452808856964,
      "num_tokens": 154155673.0,
      "step": 88875
    },
    {
      "entropy": 5.365446329116821,
      "epoch": 6.915152694028399,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001490350916808408,
      "loss": 4.569,
      "mean_token_accuracy": 0.2479272663593292,
      "num_tokens": 154164113.0,
      "step": 88880
    },
    {
      "entropy": 5.377678632736206,
      "epoch": 6.915541723400117,
      "grad_norm": 1.5625,
      "learning_rate": 0.00014901212755592886,
      "loss": 4.5499,
      "mean_token_accuracy": 0.24989526271820067,
      "num_tokens": 154172339.0,
      "step": 88885
    },
    {
      "entropy": 5.415231561660766,
      "epoch": 6.9159307527718346,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014898916534268231,
      "loss": 4.7237,
      "mean_token_accuracy": 0.23434912115335466,
      "num_tokens": 154180950.0,
      "step": 88890
    },
    {
      "entropy": 5.403679418563843,
      "epoch": 6.916319782143551,
      "grad_norm": 1.375,
      "learning_rate": 0.0001489662050414497,
      "loss": 4.6958,
      "mean_token_accuracy": 0.2301421731710434,
      "num_tokens": 154189976.0,
      "step": 88895
    },
    {
      "entropy": 5.339669847488404,
      "epoch": 6.916708811515269,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014894324665257923,
      "loss": 4.6941,
      "mean_token_accuracy": 0.23756337612867356,
      "num_tokens": 154199141.0,
      "step": 88900
    },
    {
      "entropy": 5.474231481552124,
      "epoch": 6.917097840886987,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001489202901764195,
      "loss": 4.74,
      "mean_token_accuracy": 0.2249201714992523,
      "num_tokens": 154207490.0,
      "step": 88905
    },
    {
      "entropy": 5.354945850372315,
      "epoch": 6.9174868702587045,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014889733561331854,
      "loss": 4.5651,
      "mean_token_accuracy": 0.24673629105091094,
      "num_tokens": 154215466.0,
      "step": 88910
    },
    {
      "entropy": 5.389419221878052,
      "epoch": 6.917875899630422,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001488743829636249,
      "loss": 4.6201,
      "mean_token_accuracy": 0.2440054953098297,
      "num_tokens": 154223755.0,
      "step": 88915
    },
    {
      "entropy": 5.3603808879852295,
      "epoch": 6.91826492900214,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014885143222768667,
      "loss": 4.633,
      "mean_token_accuracy": 0.23934358954429627,
      "num_tokens": 154232145.0,
      "step": 88920
    },
    {
      "entropy": 5.429491329193115,
      "epoch": 6.9186539583738575,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014882848340585226,
      "loss": 4.6731,
      "mean_token_accuracy": 0.2424626350402832,
      "num_tokens": 154240497.0,
      "step": 88925
    },
    {
      "entropy": 5.463003158569336,
      "epoch": 6.919042987745575,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014880553649846968,
      "loss": 4.6586,
      "mean_token_accuracy": 0.24005881249904631,
      "num_tokens": 154248277.0,
      "step": 88930
    },
    {
      "entropy": 5.438874101638794,
      "epoch": 6.919432017117292,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001487825915058873,
      "loss": 4.6762,
      "mean_token_accuracy": 0.23755115270614624,
      "num_tokens": 154256900.0,
      "step": 88935
    },
    {
      "entropy": 5.45291838645935,
      "epoch": 6.91982104648901,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00014875964842845326,
      "loss": 4.6464,
      "mean_token_accuracy": 0.23750597089529038,
      "num_tokens": 154265696.0,
      "step": 88940
    },
    {
      "entropy": 5.405448961257934,
      "epoch": 6.9202100758607274,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014873670726651538,
      "loss": 4.604,
      "mean_token_accuracy": 0.2416668251156807,
      "num_tokens": 154273642.0,
      "step": 88945
    },
    {
      "entropy": 5.356413412094116,
      "epoch": 6.920599105232445,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001487137680204222,
      "loss": 4.569,
      "mean_token_accuracy": 0.24781796634197234,
      "num_tokens": 154282710.0,
      "step": 88950
    },
    {
      "entropy": 5.3680166721344,
      "epoch": 6.920988134604163,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014869083069052154,
      "loss": 4.6397,
      "mean_token_accuracy": 0.2388213947415352,
      "num_tokens": 154292155.0,
      "step": 88955
    },
    {
      "entropy": 5.375840091705323,
      "epoch": 6.9213771639758805,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014866789527716152,
      "loss": 4.6814,
      "mean_token_accuracy": 0.23245272785425186,
      "num_tokens": 154301384.0,
      "step": 88960
    },
    {
      "entropy": 5.5019118309021,
      "epoch": 6.921766193347597,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00014864496178069008,
      "loss": 4.8468,
      "mean_token_accuracy": 0.22896925657987593,
      "num_tokens": 154310280.0,
      "step": 88965
    },
    {
      "entropy": 5.354854583740234,
      "epoch": 6.922155222719315,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014862203020145522,
      "loss": 4.5109,
      "mean_token_accuracy": 0.2543661996722221,
      "num_tokens": 154318803.0,
      "step": 88970
    },
    {
      "entropy": 5.372418355941773,
      "epoch": 6.922544252091033,
      "grad_norm": 1.375,
      "learning_rate": 0.00014859910053980496,
      "loss": 4.6414,
      "mean_token_accuracy": 0.23671648502349854,
      "num_tokens": 154328008.0,
      "step": 88975
    },
    {
      "entropy": 5.423726034164429,
      "epoch": 6.92293328146275,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014857617279608705,
      "loss": 4.7479,
      "mean_token_accuracy": 0.23334934413433076,
      "num_tokens": 154336521.0,
      "step": 88980
    },
    {
      "entropy": 5.464918184280395,
      "epoch": 6.923322310834468,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001485532469706496,
      "loss": 4.7255,
      "mean_token_accuracy": 0.23467402756214142,
      "num_tokens": 154346488.0,
      "step": 88985
    },
    {
      "entropy": 5.411870670318604,
      "epoch": 6.923711340206186,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001485303230638403,
      "loss": 4.6032,
      "mean_token_accuracy": 0.2423005759716034,
      "num_tokens": 154355114.0,
      "step": 88990
    },
    {
      "entropy": 5.421051740646362,
      "epoch": 6.9241003695779035,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014850740107600708,
      "loss": 4.5996,
      "mean_token_accuracy": 0.2515360161662102,
      "num_tokens": 154364247.0,
      "step": 88995
    },
    {
      "entropy": 5.367387390136718,
      "epoch": 6.92448939894962,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014848448100749766,
      "loss": 4.5719,
      "mean_token_accuracy": 0.24476806521415712,
      "num_tokens": 154373127.0,
      "step": 89000
    },
    {
      "entropy": 5.367836380004883,
      "epoch": 6.924878428321338,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014846156285865998,
      "loss": 4.6545,
      "mean_token_accuracy": 0.23831048756837844,
      "num_tokens": 154381719.0,
      "step": 89005
    },
    {
      "entropy": 5.362370204925537,
      "epoch": 6.925267457693056,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014843864662984158,
      "loss": 4.5698,
      "mean_token_accuracy": 0.248913012444973,
      "num_tokens": 154389292.0,
      "step": 89010
    },
    {
      "entropy": 5.404021453857422,
      "epoch": 6.925656487064773,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014841573232139026,
      "loss": 4.6313,
      "mean_token_accuracy": 0.24815889447927475,
      "num_tokens": 154398109.0,
      "step": 89015
    },
    {
      "entropy": 5.400536823272705,
      "epoch": 6.926045516436491,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001483928199336538,
      "loss": 4.737,
      "mean_token_accuracy": 0.23297124207019806,
      "num_tokens": 154406456.0,
      "step": 89020
    },
    {
      "entropy": 5.34310302734375,
      "epoch": 6.926434545808209,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001483699094669797,
      "loss": 4.66,
      "mean_token_accuracy": 0.23603852093219757,
      "num_tokens": 154415541.0,
      "step": 89025
    },
    {
      "entropy": 5.37505521774292,
      "epoch": 6.926823575179926,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001483470009217158,
      "loss": 4.611,
      "mean_token_accuracy": 0.24516281336545945,
      "num_tokens": 154423914.0,
      "step": 89030
    },
    {
      "entropy": 5.309968662261963,
      "epoch": 6.927212604551643,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014832409429820943,
      "loss": 4.5432,
      "mean_token_accuracy": 0.24795085191726685,
      "num_tokens": 154432422.0,
      "step": 89035
    },
    {
      "entropy": 5.2854986667633055,
      "epoch": 6.927601633923361,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001483011895968084,
      "loss": 4.5448,
      "mean_token_accuracy": 0.24606788754463196,
      "num_tokens": 154441347.0,
      "step": 89040
    },
    {
      "entropy": 5.414838266372681,
      "epoch": 6.927990663295079,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014827828681786019,
      "loss": 4.7312,
      "mean_token_accuracy": 0.23057265281677247,
      "num_tokens": 154449041.0,
      "step": 89045
    },
    {
      "entropy": 5.471208620071411,
      "epoch": 6.928379692666796,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014825538596171218,
      "loss": 4.7599,
      "mean_token_accuracy": 0.24007384330034257,
      "num_tokens": 154457464.0,
      "step": 89050
    },
    {
      "entropy": 5.34762544631958,
      "epoch": 6.928768722038514,
      "grad_norm": 1.421875,
      "learning_rate": 0.000148232487028712,
      "loss": 4.6027,
      "mean_token_accuracy": 0.2399534210562706,
      "num_tokens": 154467043.0,
      "step": 89055
    },
    {
      "entropy": 5.368239974975586,
      "epoch": 6.929157751410232,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014820959001920703,
      "loss": 4.6473,
      "mean_token_accuracy": 0.23976270258426666,
      "num_tokens": 154475673.0,
      "step": 89060
    },
    {
      "entropy": 5.405302906036377,
      "epoch": 6.9295467807819495,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00014818669493354476,
      "loss": 4.7115,
      "mean_token_accuracy": 0.23475291281938554,
      "num_tokens": 154485004.0,
      "step": 89065
    },
    {
      "entropy": 5.426081895828247,
      "epoch": 6.929935810153666,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001481638017720725,
      "loss": 4.6879,
      "mean_token_accuracy": 0.2377195507287979,
      "num_tokens": 154492940.0,
      "step": 89070
    },
    {
      "entropy": 5.415059900283813,
      "epoch": 6.930324839525384,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014814091053513774,
      "loss": 4.7132,
      "mean_token_accuracy": 0.23442801982164382,
      "num_tokens": 154501868.0,
      "step": 89075
    },
    {
      "entropy": 5.449075508117676,
      "epoch": 6.930713868897102,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014811802122308764,
      "loss": 4.6945,
      "mean_token_accuracy": 0.23550050258636473,
      "num_tokens": 154511157.0,
      "step": 89080
    },
    {
      "entropy": 5.425273084640503,
      "epoch": 6.931102898268819,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014809513383626965,
      "loss": 4.6539,
      "mean_token_accuracy": 0.24211314767599107,
      "num_tokens": 154520040.0,
      "step": 89085
    },
    {
      "entropy": 5.3271942138671875,
      "epoch": 6.931491927640537,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001480722483750311,
      "loss": 4.5422,
      "mean_token_accuracy": 0.246901473402977,
      "num_tokens": 154528437.0,
      "step": 89090
    },
    {
      "entropy": 5.3887749195098875,
      "epoch": 6.931880957012254,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014804936483971904,
      "loss": 4.6365,
      "mean_token_accuracy": 0.244598126411438,
      "num_tokens": 154537237.0,
      "step": 89095
    },
    {
      "entropy": 5.414426851272583,
      "epoch": 6.932269986383972,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014802648323068092,
      "loss": 4.7201,
      "mean_token_accuracy": 0.23871537148952485,
      "num_tokens": 154546365.0,
      "step": 89100
    },
    {
      "entropy": 5.468905115127564,
      "epoch": 6.932659015755689,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001480036035482637,
      "loss": 4.7367,
      "mean_token_accuracy": 0.23453299552202225,
      "num_tokens": 154555435.0,
      "step": 89105
    },
    {
      "entropy": 5.376213884353637,
      "epoch": 6.933048045127407,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014798072579281474,
      "loss": 4.5957,
      "mean_token_accuracy": 0.24060264527797698,
      "num_tokens": 154564433.0,
      "step": 89110
    },
    {
      "entropy": 5.389183568954468,
      "epoch": 6.933437074499125,
      "grad_norm": 1.265625,
      "learning_rate": 0.00014795784996468104,
      "loss": 4.6555,
      "mean_token_accuracy": 0.23508816510438918,
      "num_tokens": 154572774.0,
      "step": 89115
    },
    {
      "entropy": 5.3720166206359865,
      "epoch": 6.933826103870842,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014793497606420985,
      "loss": 4.5838,
      "mean_token_accuracy": 0.24755021631717683,
      "num_tokens": 154581697.0,
      "step": 89120
    },
    {
      "entropy": 5.342128944396973,
      "epoch": 6.93421513324256,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014791210409174812,
      "loss": 4.592,
      "mean_token_accuracy": 0.24848069101572037,
      "num_tokens": 154590552.0,
      "step": 89125
    },
    {
      "entropy": 5.415431642532349,
      "epoch": 6.934604162614278,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014788923404764287,
      "loss": 4.6666,
      "mean_token_accuracy": 0.24274978339672088,
      "num_tokens": 154599254.0,
      "step": 89130
    },
    {
      "entropy": 5.390441226959228,
      "epoch": 6.934993191985995,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014786636593224124,
      "loss": 4.6502,
      "mean_token_accuracy": 0.23763030767440796,
      "num_tokens": 154607415.0,
      "step": 89135
    },
    {
      "entropy": 5.307901191711426,
      "epoch": 6.935382221357712,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001478434997458901,
      "loss": 4.5256,
      "mean_token_accuracy": 0.24955178648233414,
      "num_tokens": 154616241.0,
      "step": 89140
    },
    {
      "entropy": 5.407774877548218,
      "epoch": 6.93577125072943,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014782063548893643,
      "loss": 4.714,
      "mean_token_accuracy": 0.24256534576416017,
      "num_tokens": 154625360.0,
      "step": 89145
    },
    {
      "entropy": 5.325503540039063,
      "epoch": 6.936160280101148,
      "grad_norm": 1.25,
      "learning_rate": 0.0001477977731617273,
      "loss": 4.5909,
      "mean_token_accuracy": 0.24459605515003205,
      "num_tokens": 154634113.0,
      "step": 89150
    },
    {
      "entropy": 5.408359622955322,
      "epoch": 6.936549309472865,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001477749127646094,
      "loss": 4.6004,
      "mean_token_accuracy": 0.2532923996448517,
      "num_tokens": 154642465.0,
      "step": 89155
    },
    {
      "entropy": 5.376912546157837,
      "epoch": 6.936938338844583,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00014775205429792978,
      "loss": 4.6709,
      "mean_token_accuracy": 0.24223071634769439,
      "num_tokens": 154652147.0,
      "step": 89160
    },
    {
      "entropy": 5.3498961448669435,
      "epoch": 6.9373273682163,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014772919776203514,
      "loss": 4.6597,
      "mean_token_accuracy": 0.24013046473264693,
      "num_tokens": 154661032.0,
      "step": 89165
    },
    {
      "entropy": 5.330604982376099,
      "epoch": 6.937716397588018,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001477063431572724,
      "loss": 4.5731,
      "mean_token_accuracy": 0.24241770505905152,
      "num_tokens": 154669619.0,
      "step": 89170
    },
    {
      "entropy": 5.3827215194702145,
      "epoch": 6.938105426959735,
      "grad_norm": 1.375,
      "learning_rate": 0.00014768349048398824,
      "loss": 4.6507,
      "mean_token_accuracy": 0.23760260939598082,
      "num_tokens": 154678371.0,
      "step": 89175
    },
    {
      "entropy": 5.397452306747437,
      "epoch": 6.938494456331453,
      "grad_norm": 1.234375,
      "learning_rate": 0.00014766063974252952,
      "loss": 4.6349,
      "mean_token_accuracy": 0.2434176579117775,
      "num_tokens": 154687436.0,
      "step": 89180
    },
    {
      "entropy": 5.413670301437378,
      "epoch": 6.938883485703171,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014763779093324283,
      "loss": 4.7116,
      "mean_token_accuracy": 0.23392346799373626,
      "num_tokens": 154696436.0,
      "step": 89185
    },
    {
      "entropy": 5.301106691360474,
      "epoch": 6.939272515074888,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014761494405647504,
      "loss": 4.5834,
      "mean_token_accuracy": 0.24431550949811937,
      "num_tokens": 154704427.0,
      "step": 89190
    },
    {
      "entropy": 5.436334609985352,
      "epoch": 6.939661544446606,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014759209911257266,
      "loss": 4.6351,
      "mean_token_accuracy": 0.24093306362628936,
      "num_tokens": 154712908.0,
      "step": 89195
    },
    {
      "entropy": 5.389294767379761,
      "epoch": 6.940050573818323,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014756925610188244,
      "loss": 4.5574,
      "mean_token_accuracy": 0.2498522937297821,
      "num_tokens": 154720924.0,
      "step": 89200
    },
    {
      "entropy": 5.375076961517334,
      "epoch": 6.940439603190041,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014754641502475092,
      "loss": 4.6861,
      "mean_token_accuracy": 0.2447671741247177,
      "num_tokens": 154730057.0,
      "step": 89205
    },
    {
      "entropy": 5.354050970077514,
      "epoch": 6.940828632561758,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014752357588152453,
      "loss": 4.6448,
      "mean_token_accuracy": 0.24666367769241332,
      "num_tokens": 154739487.0,
      "step": 89210
    },
    {
      "entropy": 5.401321744918823,
      "epoch": 6.941217661933476,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014750073867255022,
      "loss": 4.6176,
      "mean_token_accuracy": 0.24435776323080063,
      "num_tokens": 154748296.0,
      "step": 89215
    },
    {
      "entropy": 5.315762090682983,
      "epoch": 6.941606691305194,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014747790339817408,
      "loss": 4.5575,
      "mean_token_accuracy": 0.24421963840723038,
      "num_tokens": 154756626.0,
      "step": 89220
    },
    {
      "entropy": 5.4273017883300785,
      "epoch": 6.941995720676911,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001474550700587429,
      "loss": 4.7088,
      "mean_token_accuracy": 0.24053658694028854,
      "num_tokens": 154765050.0,
      "step": 89225
    },
    {
      "entropy": 5.385556077957153,
      "epoch": 6.942384750048628,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014743223865460296,
      "loss": 4.5947,
      "mean_token_accuracy": 0.2428887516260147,
      "num_tokens": 154773096.0,
      "step": 89230
    },
    {
      "entropy": 5.363854932785034,
      "epoch": 6.942773779420346,
      "grad_norm": 1.265625,
      "learning_rate": 0.00014740940918610079,
      "loss": 4.5852,
      "mean_token_accuracy": 0.2441806823015213,
      "num_tokens": 154781484.0,
      "step": 89235
    },
    {
      "entropy": 5.340217876434326,
      "epoch": 6.943162808792064,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001473865816535828,
      "loss": 4.5634,
      "mean_token_accuracy": 0.2563501000404358,
      "num_tokens": 154790605.0,
      "step": 89240
    },
    {
      "entropy": 5.378575277328491,
      "epoch": 6.943551838163781,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001473637560573952,
      "loss": 4.5914,
      "mean_token_accuracy": 0.24379443973302842,
      "num_tokens": 154799472.0,
      "step": 89245
    },
    {
      "entropy": 5.454685974121094,
      "epoch": 6.943940867535499,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014734093239788458,
      "loss": 4.7598,
      "mean_token_accuracy": 0.2374058872461319,
      "num_tokens": 154808220.0,
      "step": 89250
    },
    {
      "entropy": 5.425030899047852,
      "epoch": 6.944329896907217,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014731811067539698,
      "loss": 4.671,
      "mean_token_accuracy": 0.23419081717729567,
      "num_tokens": 154816971.0,
      "step": 89255
    },
    {
      "entropy": 5.335069847106934,
      "epoch": 6.944718926278934,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014729529089027894,
      "loss": 4.631,
      "mean_token_accuracy": 0.24452994465827943,
      "num_tokens": 154826000.0,
      "step": 89260
    },
    {
      "entropy": 5.45575647354126,
      "epoch": 6.945107955650652,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014727247304287649,
      "loss": 4.6808,
      "mean_token_accuracy": 0.2457893654704094,
      "num_tokens": 154834454.0,
      "step": 89265
    },
    {
      "entropy": 5.414825820922852,
      "epoch": 6.945496985022369,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00014724965713353605,
      "loss": 4.6164,
      "mean_token_accuracy": 0.24254976361989974,
      "num_tokens": 154842844.0,
      "step": 89270
    },
    {
      "entropy": 5.400317096710205,
      "epoch": 6.945886014394087,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014722684316260368,
      "loss": 4.6202,
      "mean_token_accuracy": 0.2516010135412216,
      "num_tokens": 154851318.0,
      "step": 89275
    },
    {
      "entropy": 5.3282225131988525,
      "epoch": 6.946275043765804,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014720403113042558,
      "loss": 4.5796,
      "mean_token_accuracy": 0.24145934879779815,
      "num_tokens": 154860537.0,
      "step": 89280
    },
    {
      "entropy": 5.405135440826416,
      "epoch": 6.946664073137522,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014718122103734793,
      "loss": 4.639,
      "mean_token_accuracy": 0.24806546717882155,
      "num_tokens": 154869229.0,
      "step": 89285
    },
    {
      "entropy": 5.419905471801758,
      "epoch": 6.94705310250924,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001471584128837168,
      "loss": 4.6441,
      "mean_token_accuracy": 0.23991436660289764,
      "num_tokens": 154877056.0,
      "step": 89290
    },
    {
      "entropy": 5.354861545562744,
      "epoch": 6.947442131880957,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001471356066698783,
      "loss": 4.5234,
      "mean_token_accuracy": 0.2513883575797081,
      "num_tokens": 154885970.0,
      "step": 89295
    },
    {
      "entropy": 5.307975625991821,
      "epoch": 6.947831161252674,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014711280239617841,
      "loss": 4.562,
      "mean_token_accuracy": 0.24480591714382172,
      "num_tokens": 154895734.0,
      "step": 89300
    },
    {
      "entropy": 5.404718017578125,
      "epoch": 6.948220190624392,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014709000006296326,
      "loss": 4.6196,
      "mean_token_accuracy": 0.24177662134170533,
      "num_tokens": 154904214.0,
      "step": 89305
    },
    {
      "entropy": 5.351935338973999,
      "epoch": 6.94860921999611,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00014706719967057862,
      "loss": 4.6256,
      "mean_token_accuracy": 0.2409597858786583,
      "num_tokens": 154913465.0,
      "step": 89310
    },
    {
      "entropy": 5.352665042877197,
      "epoch": 6.948998249367827,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001470444012193708,
      "loss": 4.6249,
      "mean_token_accuracy": 0.24740462601184846,
      "num_tokens": 154922878.0,
      "step": 89315
    },
    {
      "entropy": 5.310328340530395,
      "epoch": 6.949387278739545,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014702160470968546,
      "loss": 4.5801,
      "mean_token_accuracy": 0.24802251160144806,
      "num_tokens": 154931004.0,
      "step": 89320
    },
    {
      "entropy": 5.372582674026489,
      "epoch": 6.949776308111263,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001469988101418685,
      "loss": 4.5978,
      "mean_token_accuracy": 0.24051104336977006,
      "num_tokens": 154939119.0,
      "step": 89325
    },
    {
      "entropy": 5.43602614402771,
      "epoch": 6.95016533748298,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014697601751626591,
      "loss": 4.6918,
      "mean_token_accuracy": 0.23199718743562697,
      "num_tokens": 154947626.0,
      "step": 89330
    },
    {
      "entropy": 5.408651447296142,
      "epoch": 6.950554366854697,
      "grad_norm": 1.375,
      "learning_rate": 0.00014695322683322344,
      "loss": 4.7332,
      "mean_token_accuracy": 0.233060260117054,
      "num_tokens": 154956337.0,
      "step": 89335
    },
    {
      "entropy": 5.391693592071533,
      "epoch": 6.950943396226415,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000146930438093087,
      "loss": 4.6815,
      "mean_token_accuracy": 0.23635961711406708,
      "num_tokens": 154965096.0,
      "step": 89340
    },
    {
      "entropy": 5.3786499977111815,
      "epoch": 6.951332425598133,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014690765129620227,
      "loss": 4.5756,
      "mean_token_accuracy": 0.24730477035045623,
      "num_tokens": 154973652.0,
      "step": 89345
    },
    {
      "entropy": 5.317218017578125,
      "epoch": 6.95172145496985,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014688486644291504,
      "loss": 4.5086,
      "mean_token_accuracy": 0.24945195019245148,
      "num_tokens": 154981956.0,
      "step": 89350
    },
    {
      "entropy": 5.393576526641846,
      "epoch": 6.952110484341568,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001468620835335711,
      "loss": 4.6486,
      "mean_token_accuracy": 0.23669015616178513,
      "num_tokens": 154990331.0,
      "step": 89355
    },
    {
      "entropy": 5.433332681655884,
      "epoch": 6.952499513713286,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014683930256851604,
      "loss": 4.7003,
      "mean_token_accuracy": 0.23601237535476685,
      "num_tokens": 154998420.0,
      "step": 89360
    },
    {
      "entropy": 5.456935024261474,
      "epoch": 6.9528885430850025,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014681652354809566,
      "loss": 4.7158,
      "mean_token_accuracy": 0.23541837185621262,
      "num_tokens": 155007158.0,
      "step": 89365
    },
    {
      "entropy": 5.381467533111572,
      "epoch": 6.95327757245672,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00014679374647265543,
      "loss": 4.6658,
      "mean_token_accuracy": 0.24312151968479156,
      "num_tokens": 155015920.0,
      "step": 89370
    },
    {
      "entropy": 5.319481229782104,
      "epoch": 6.953666601828438,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00014677097134254107,
      "loss": 4.5621,
      "mean_token_accuracy": 0.24423403590917586,
      "num_tokens": 155024793.0,
      "step": 89375
    },
    {
      "entropy": 5.432153606414795,
      "epoch": 6.954055631200156,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001467481981580981,
      "loss": 4.7542,
      "mean_token_accuracy": 0.2269004240632057,
      "num_tokens": 155034090.0,
      "step": 89380
    },
    {
      "entropy": 5.470992565155029,
      "epoch": 6.954444660571873,
      "grad_norm": 1.875,
      "learning_rate": 0.00014672542691967214,
      "loss": 4.7056,
      "mean_token_accuracy": 0.235000404715538,
      "num_tokens": 155042236.0,
      "step": 89385
    },
    {
      "entropy": 5.447150850296021,
      "epoch": 6.954833689943591,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014670265762760858,
      "loss": 4.6592,
      "mean_token_accuracy": 0.23521073609590532,
      "num_tokens": 155051177.0,
      "step": 89390
    },
    {
      "entropy": 5.404469347000122,
      "epoch": 6.955222719315309,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001466798902822531,
      "loss": 4.6126,
      "mean_token_accuracy": 0.23928658664226532,
      "num_tokens": 155060136.0,
      "step": 89395
    },
    {
      "entropy": 5.353607273101806,
      "epoch": 6.955611748687026,
      "grad_norm": 1.296875,
      "learning_rate": 0.000146657124883951,
      "loss": 4.6726,
      "mean_token_accuracy": 0.23722766786813737,
      "num_tokens": 155068743.0,
      "step": 89400
    },
    {
      "entropy": 5.430614614486695,
      "epoch": 6.956000778058743,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014663436143304764,
      "loss": 4.594,
      "mean_token_accuracy": 0.24776683151721954,
      "num_tokens": 155077034.0,
      "step": 89405
    },
    {
      "entropy": 5.248784971237183,
      "epoch": 6.956389807430461,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001466115999298886,
      "loss": 4.4858,
      "mean_token_accuracy": 0.266329425573349,
      "num_tokens": 155085134.0,
      "step": 89410
    },
    {
      "entropy": 5.427721738815308,
      "epoch": 6.956778836802179,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014658884037481912,
      "loss": 4.6596,
      "mean_token_accuracy": 0.24158472269773484,
      "num_tokens": 155093549.0,
      "step": 89415
    },
    {
      "entropy": 5.381779623031616,
      "epoch": 6.957167866173896,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014656608276818474,
      "loss": 4.6142,
      "mean_token_accuracy": 0.24471794068813324,
      "num_tokens": 155102479.0,
      "step": 89420
    },
    {
      "entropy": 5.262354707717895,
      "epoch": 6.957556895545614,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014654332711033059,
      "loss": 4.5212,
      "mean_token_accuracy": 0.25327264815568923,
      "num_tokens": 155111198.0,
      "step": 89425
    },
    {
      "entropy": 5.40374345779419,
      "epoch": 6.957945924917331,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00014652057340160192,
      "loss": 4.6885,
      "mean_token_accuracy": 0.23478147387504578,
      "num_tokens": 155119545.0,
      "step": 89430
    },
    {
      "entropy": 5.354250049591064,
      "epoch": 6.9583349542890485,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00014649782164234413,
      "loss": 4.58,
      "mean_token_accuracy": 0.24436026811599731,
      "num_tokens": 155127967.0,
      "step": 89435
    },
    {
      "entropy": 5.381389951705932,
      "epoch": 6.958723983660766,
      "grad_norm": 1.203125,
      "learning_rate": 0.00014647507183290228,
      "loss": 4.6607,
      "mean_token_accuracy": 0.2303643062710762,
      "num_tokens": 155136560.0,
      "step": 89440
    },
    {
      "entropy": 5.3719127655029295,
      "epoch": 6.959113013032484,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014645232397362173,
      "loss": 4.6103,
      "mean_token_accuracy": 0.23859770596027374,
      "num_tokens": 155144532.0,
      "step": 89445
    },
    {
      "entropy": 5.399070119857788,
      "epoch": 6.959502042404202,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001464295780648475,
      "loss": 4.7091,
      "mean_token_accuracy": 0.24247010201215743,
      "num_tokens": 155153460.0,
      "step": 89450
    },
    {
      "entropy": 5.373997020721435,
      "epoch": 6.959891071775919,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014640683410692485,
      "loss": 4.6393,
      "mean_token_accuracy": 0.24782025814056396,
      "num_tokens": 155161754.0,
      "step": 89455
    },
    {
      "entropy": 5.351853942871093,
      "epoch": 6.960280101147637,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001463840921001987,
      "loss": 4.5533,
      "mean_token_accuracy": 0.2523351475596428,
      "num_tokens": 155169947.0,
      "step": 89460
    },
    {
      "entropy": 5.274852275848389,
      "epoch": 6.960669130519355,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001463613520450144,
      "loss": 4.5731,
      "mean_token_accuracy": 0.23911781758069992,
      "num_tokens": 155179942.0,
      "step": 89465
    },
    {
      "entropy": 5.483589744567871,
      "epoch": 6.9610581598910715,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014633861394171667,
      "loss": 4.8491,
      "mean_token_accuracy": 0.2204234182834625,
      "num_tokens": 155189468.0,
      "step": 89470
    },
    {
      "entropy": 5.406822776794433,
      "epoch": 6.961447189262789,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014631587779065076,
      "loss": 4.6696,
      "mean_token_accuracy": 0.24305710047483445,
      "num_tokens": 155198617.0,
      "step": 89475
    },
    {
      "entropy": 5.451224088668823,
      "epoch": 6.961836218634507,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014629314359216158,
      "loss": 4.6524,
      "mean_token_accuracy": 0.24407994151115417,
      "num_tokens": 155207022.0,
      "step": 89480
    },
    {
      "entropy": 5.430307149887085,
      "epoch": 6.9622252480062246,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014627041134659407,
      "loss": 4.657,
      "mean_token_accuracy": 0.24445607215166093,
      "num_tokens": 155215749.0,
      "step": 89485
    },
    {
      "entropy": 5.379525756835937,
      "epoch": 6.962614277377942,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014624768105429324,
      "loss": 4.6796,
      "mean_token_accuracy": 0.24149177670478822,
      "num_tokens": 155224335.0,
      "step": 89490
    },
    {
      "entropy": 5.430639171600342,
      "epoch": 6.96300330674966,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001462249527156038,
      "loss": 4.7351,
      "mean_token_accuracy": 0.24728310853242874,
      "num_tokens": 155234211.0,
      "step": 89495
    },
    {
      "entropy": 5.333150482177734,
      "epoch": 6.963392336121377,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014620222633087088,
      "loss": 4.5556,
      "mean_token_accuracy": 0.2527706608176231,
      "num_tokens": 155242640.0,
      "step": 89500
    },
    {
      "entropy": 5.381558656692505,
      "epoch": 6.9637813654930945,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014617950190043912,
      "loss": 4.6332,
      "mean_token_accuracy": 0.23923116326332092,
      "num_tokens": 155251147.0,
      "step": 89505
    },
    {
      "entropy": 5.389881944656372,
      "epoch": 6.964170394864812,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014615677942465333,
      "loss": 4.6036,
      "mean_token_accuracy": 0.2450886145234108,
      "num_tokens": 155259211.0,
      "step": 89510
    },
    {
      "entropy": 5.285770416259766,
      "epoch": 6.96455942423653,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00014613405890385843,
      "loss": 4.5068,
      "mean_token_accuracy": 0.24923093765974044,
      "num_tokens": 155267264.0,
      "step": 89515
    },
    {
      "entropy": 5.3125769138336185,
      "epoch": 6.9649484536082475,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014611134033839897,
      "loss": 4.6061,
      "mean_token_accuracy": 0.24726336896419526,
      "num_tokens": 155276162.0,
      "step": 89520
    },
    {
      "entropy": 5.448733806610107,
      "epoch": 6.965337482979965,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00014608862372861986,
      "loss": 4.7062,
      "mean_token_accuracy": 0.24093364477157592,
      "num_tokens": 155284572.0,
      "step": 89525
    },
    {
      "entropy": 5.413979196548462,
      "epoch": 6.965726512351683,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014606590907486562,
      "loss": 4.6346,
      "mean_token_accuracy": 0.2502425491809845,
      "num_tokens": 155293299.0,
      "step": 89530
    },
    {
      "entropy": 5.3835667133331295,
      "epoch": 6.9661155417234,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00014604319637748106,
      "loss": 4.6428,
      "mean_token_accuracy": 0.24314749240875244,
      "num_tokens": 155301792.0,
      "step": 89535
    },
    {
      "entropy": 5.4710853099823,
      "epoch": 6.9665045710951174,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014602048563681067,
      "loss": 4.7214,
      "mean_token_accuracy": 0.24739219397306442,
      "num_tokens": 155310449.0,
      "step": 89540
    },
    {
      "entropy": 5.399386978149414,
      "epoch": 6.966893600466835,
      "grad_norm": 1.421875,
      "learning_rate": 0.00014599777685319912,
      "loss": 4.6366,
      "mean_token_accuracy": 0.2482352927327156,
      "num_tokens": 155318712.0,
      "step": 89545
    },
    {
      "entropy": 5.385106325149536,
      "epoch": 6.967282629838553,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014597507002699105,
      "loss": 4.6567,
      "mean_token_accuracy": 0.23122390657663344,
      "num_tokens": 155326381.0,
      "step": 89550
    },
    {
      "entropy": 5.295986032485962,
      "epoch": 6.9676716592102705,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014595236515853083,
      "loss": 4.6654,
      "mean_token_accuracy": 0.23752700239419938,
      "num_tokens": 155335669.0,
      "step": 89555
    },
    {
      "entropy": 5.294505882263183,
      "epoch": 6.968060688581988,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014592966224816317,
      "loss": 4.5127,
      "mean_token_accuracy": 0.2599580153822899,
      "num_tokens": 155343796.0,
      "step": 89560
    },
    {
      "entropy": 5.4055983543396,
      "epoch": 6.968449717953705,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001459069612962324,
      "loss": 4.5864,
      "mean_token_accuracy": 0.2497238263487816,
      "num_tokens": 155352857.0,
      "step": 89565
    },
    {
      "entropy": 5.389829158782959,
      "epoch": 6.968838747325423,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014588426230308308,
      "loss": 4.631,
      "mean_token_accuracy": 0.23895051032304765,
      "num_tokens": 155361174.0,
      "step": 89570
    },
    {
      "entropy": 5.3532086372375485,
      "epoch": 6.96922777669714,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014586156526905948,
      "loss": 4.5579,
      "mean_token_accuracy": 0.2474617764353752,
      "num_tokens": 155369441.0,
      "step": 89575
    },
    {
      "entropy": 5.384224987030029,
      "epoch": 6.969616806068858,
      "grad_norm": 1.25,
      "learning_rate": 0.00014583887019450616,
      "loss": 4.7051,
      "mean_token_accuracy": 0.2365070953965187,
      "num_tokens": 155377949.0,
      "step": 89580
    },
    {
      "entropy": 5.301453113555908,
      "epoch": 6.970005835440576,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001458161770797674,
      "loss": 4.4993,
      "mean_token_accuracy": 0.2507901504635811,
      "num_tokens": 155386368.0,
      "step": 89585
    },
    {
      "entropy": 5.340715789794922,
      "epoch": 6.9703948648122935,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014579348592518745,
      "loss": 4.6149,
      "mean_token_accuracy": 0.24846286922693253,
      "num_tokens": 155395662.0,
      "step": 89590
    },
    {
      "entropy": 5.383721923828125,
      "epoch": 6.970783894184011,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001457707967311108,
      "loss": 4.6848,
      "mean_token_accuracy": 0.2395549386739731,
      "num_tokens": 155404805.0,
      "step": 89595
    },
    {
      "entropy": 5.3236593246459964,
      "epoch": 6.971172923555729,
      "grad_norm": 1.375,
      "learning_rate": 0.00014574810949788153,
      "loss": 4.554,
      "mean_token_accuracy": 0.2502008557319641,
      "num_tokens": 155413842.0,
      "step": 89600
    },
    {
      "entropy": 5.478820848464966,
      "epoch": 6.971561952927446,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014572542422584407,
      "loss": 4.6634,
      "mean_token_accuracy": 0.232299542427063,
      "num_tokens": 155422519.0,
      "step": 89605
    },
    {
      "entropy": 5.422950601577758,
      "epoch": 6.971950982299163,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014570274091534234,
      "loss": 4.6641,
      "mean_token_accuracy": 0.2359117940068245,
      "num_tokens": 155430949.0,
      "step": 89610
    },
    {
      "entropy": 5.464028596878052,
      "epoch": 6.972340011670881,
      "grad_norm": 1.484375,
      "learning_rate": 0.00014568005956672086,
      "loss": 4.736,
      "mean_token_accuracy": 0.23836009502410888,
      "num_tokens": 155439681.0,
      "step": 89615
    },
    {
      "entropy": 5.445967531204223,
      "epoch": 6.972729041042599,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001456573801803237,
      "loss": 4.6821,
      "mean_token_accuracy": 0.23620209097862244,
      "num_tokens": 155448224.0,
      "step": 89620
    },
    {
      "entropy": 5.367197179794312,
      "epoch": 6.9731180704143165,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014563470275649482,
      "loss": 4.6381,
      "mean_token_accuracy": 0.2410624235868454,
      "num_tokens": 155457037.0,
      "step": 89625
    },
    {
      "entropy": 5.408782291412353,
      "epoch": 6.973507099786034,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001456120272955785,
      "loss": 4.6872,
      "mean_token_accuracy": 0.23567092567682266,
      "num_tokens": 155466307.0,
      "step": 89630
    },
    {
      "entropy": 5.435531187057495,
      "epoch": 6.973896129157751,
      "grad_norm": 1.5,
      "learning_rate": 0.00014558935379791865,
      "loss": 4.7092,
      "mean_token_accuracy": 0.23930118530988692,
      "num_tokens": 155474455.0,
      "step": 89635
    },
    {
      "entropy": 5.396607112884522,
      "epoch": 6.974285158529469,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014556668226385947,
      "loss": 4.652,
      "mean_token_accuracy": 0.234801022708416,
      "num_tokens": 155483324.0,
      "step": 89640
    },
    {
      "entropy": 5.434888458251953,
      "epoch": 6.974674187901186,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001455440126937448,
      "loss": 4.7074,
      "mean_token_accuracy": 0.24335208535194397,
      "num_tokens": 155492875.0,
      "step": 89645
    },
    {
      "entropy": 5.352641963958741,
      "epoch": 6.975063217272904,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014552134508791874,
      "loss": 4.6194,
      "mean_token_accuracy": 0.24993834495544434,
      "num_tokens": 155501988.0,
      "step": 89650
    },
    {
      "entropy": 5.342614364624024,
      "epoch": 6.975452246644622,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001454986794467251,
      "loss": 4.5765,
      "mean_token_accuracy": 0.24740992933511735,
      "num_tokens": 155509615.0,
      "step": 89655
    },
    {
      "entropy": 5.23600926399231,
      "epoch": 6.9758412760163395,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014547601577050796,
      "loss": 4.5677,
      "mean_token_accuracy": 0.25248008519411086,
      "num_tokens": 155518622.0,
      "step": 89660
    },
    {
      "entropy": 5.384297704696655,
      "epoch": 6.976230305388057,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014545335405961114,
      "loss": 4.6236,
      "mean_token_accuracy": 0.2370568960905075,
      "num_tokens": 155527430.0,
      "step": 89665
    },
    {
      "entropy": 5.363427972793579,
      "epoch": 6.976619334759774,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014543069431437837,
      "loss": 4.5964,
      "mean_token_accuracy": 0.24896981120109557,
      "num_tokens": 155536512.0,
      "step": 89670
    },
    {
      "entropy": 5.418689489364624,
      "epoch": 6.977008364131492,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014540803653515362,
      "loss": 4.6466,
      "mean_token_accuracy": 0.2463422417640686,
      "num_tokens": 155544775.0,
      "step": 89675
    },
    {
      "entropy": 5.4247145652771,
      "epoch": 6.977397393503209,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00014538538072228057,
      "loss": 4.6902,
      "mean_token_accuracy": 0.23839217573404312,
      "num_tokens": 155552782.0,
      "step": 89680
    },
    {
      "entropy": 5.402208757400513,
      "epoch": 6.977786422874927,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014536272687610323,
      "loss": 4.6312,
      "mean_token_accuracy": 0.23542531877756118,
      "num_tokens": 155561448.0,
      "step": 89685
    },
    {
      "entropy": 5.433767318725586,
      "epoch": 6.978175452246645,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00014534007499696506,
      "loss": 4.7184,
      "mean_token_accuracy": 0.23418553918600082,
      "num_tokens": 155571044.0,
      "step": 89690
    },
    {
      "entropy": 5.41443452835083,
      "epoch": 6.9785644816183625,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014531742508520996,
      "loss": 4.7012,
      "mean_token_accuracy": 0.23647561818361282,
      "num_tokens": 155580197.0,
      "step": 89695
    },
    {
      "entropy": 5.429515743255616,
      "epoch": 6.978953510990079,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001452947771411815,
      "loss": 4.7139,
      "mean_token_accuracy": 0.2300940901041031,
      "num_tokens": 155588781.0,
      "step": 89700
    },
    {
      "entropy": 5.477977705001831,
      "epoch": 6.979342540361797,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001452721311652233,
      "loss": 4.7249,
      "mean_token_accuracy": 0.2375420942902565,
      "num_tokens": 155597913.0,
      "step": 89705
    },
    {
      "entropy": 5.418781185150147,
      "epoch": 6.979731569733515,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001452494871576791,
      "loss": 4.6061,
      "mean_token_accuracy": 0.24896115511655809,
      "num_tokens": 155606108.0,
      "step": 89710
    },
    {
      "entropy": 5.467988443374634,
      "epoch": 6.980120599105232,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014522684511889233,
      "loss": 4.7486,
      "mean_token_accuracy": 0.2327537104487419,
      "num_tokens": 155615086.0,
      "step": 89715
    },
    {
      "entropy": 5.4200609683990475,
      "epoch": 6.98050962847695,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001452042050492067,
      "loss": 4.6913,
      "mean_token_accuracy": 0.24197487831115722,
      "num_tokens": 155624006.0,
      "step": 89720
    },
    {
      "entropy": 5.410635995864868,
      "epoch": 6.980898657848668,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014518156694896562,
      "loss": 4.623,
      "mean_token_accuracy": 0.24111818224191667,
      "num_tokens": 155632909.0,
      "step": 89725
    },
    {
      "entropy": 5.440509605407715,
      "epoch": 6.9812876872203855,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014515893081851268,
      "loss": 4.7473,
      "mean_token_accuracy": 0.23411216139793395,
      "num_tokens": 155641803.0,
      "step": 89730
    },
    {
      "entropy": 5.38578314781189,
      "epoch": 6.981676716592102,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014513629665819123,
      "loss": 4.6471,
      "mean_token_accuracy": 0.2460915759205818,
      "num_tokens": 155650101.0,
      "step": 89735
    },
    {
      "entropy": 5.458373546600342,
      "epoch": 6.98206574596382,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00014511366446834484,
      "loss": 4.6558,
      "mean_token_accuracy": 0.2360107734799385,
      "num_tokens": 155659854.0,
      "step": 89740
    },
    {
      "entropy": 5.45317063331604,
      "epoch": 6.982454775335538,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014509103424931684,
      "loss": 4.7231,
      "mean_token_accuracy": 0.24116703122854233,
      "num_tokens": 155668894.0,
      "step": 89745
    },
    {
      "entropy": 5.439077949523925,
      "epoch": 6.982843804707255,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014506840600145055,
      "loss": 4.6678,
      "mean_token_accuracy": 0.23063323944807052,
      "num_tokens": 155677308.0,
      "step": 89750
    },
    {
      "entropy": 5.409154891967773,
      "epoch": 6.983232834078973,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001450457797250895,
      "loss": 4.6853,
      "mean_token_accuracy": 0.23423619121313094,
      "num_tokens": 155685305.0,
      "step": 89755
    },
    {
      "entropy": 5.317077732086181,
      "epoch": 6.983621863450691,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001450231554205768,
      "loss": 4.5573,
      "mean_token_accuracy": 0.24795056134462357,
      "num_tokens": 155694601.0,
      "step": 89760
    },
    {
      "entropy": 5.394887018203735,
      "epoch": 6.984010892822408,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00014500053308825594,
      "loss": 4.6523,
      "mean_token_accuracy": 0.23794064670801163,
      "num_tokens": 155703676.0,
      "step": 89765
    },
    {
      "entropy": 5.390223264694214,
      "epoch": 6.984399922194125,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014497791272847002,
      "loss": 4.6406,
      "mean_token_accuracy": 0.23587863594293595,
      "num_tokens": 155711807.0,
      "step": 89770
    },
    {
      "entropy": 5.318341541290283,
      "epoch": 6.984788951565843,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014495529434156238,
      "loss": 4.5792,
      "mean_token_accuracy": 0.24658452570438386,
      "num_tokens": 155720158.0,
      "step": 89775
    },
    {
      "entropy": 5.402074909210205,
      "epoch": 6.985177980937561,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014493267792787618,
      "loss": 4.6715,
      "mean_token_accuracy": 0.2514621153473854,
      "num_tokens": 155729012.0,
      "step": 89780
    },
    {
      "entropy": 5.372131061553955,
      "epoch": 6.985567010309278,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014491006348775446,
      "loss": 4.6221,
      "mean_token_accuracy": 0.2436370834708214,
      "num_tokens": 155737257.0,
      "step": 89785
    },
    {
      "entropy": 5.40860333442688,
      "epoch": 6.985956039680996,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014488745102154055,
      "loss": 4.6785,
      "mean_token_accuracy": 0.23405275493860245,
      "num_tokens": 155745243.0,
      "step": 89790
    },
    {
      "entropy": 5.341302013397216,
      "epoch": 6.986345069052714,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001448648405295774,
      "loss": 4.6189,
      "mean_token_accuracy": 0.24098528325557708,
      "num_tokens": 155753546.0,
      "step": 89795
    },
    {
      "entropy": 5.367088985443115,
      "epoch": 6.9867340984244315,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00014484223201220823,
      "loss": 4.6904,
      "mean_token_accuracy": 0.23882953226566314,
      "num_tokens": 155762534.0,
      "step": 89800
    },
    {
      "entropy": 5.388047599792481,
      "epoch": 6.987123127796148,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014481962546977596,
      "loss": 4.5921,
      "mean_token_accuracy": 0.24680139273405075,
      "num_tokens": 155770861.0,
      "step": 89805
    },
    {
      "entropy": 5.381996297836304,
      "epoch": 6.987512157167866,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014479702090262368,
      "loss": 4.6661,
      "mean_token_accuracy": 0.24236880987882614,
      "num_tokens": 155779219.0,
      "step": 89810
    },
    {
      "entropy": 5.463217926025391,
      "epoch": 6.987901186539584,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014477441831109444,
      "loss": 4.7777,
      "mean_token_accuracy": 0.23083534091711044,
      "num_tokens": 155787975.0,
      "step": 89815
    },
    {
      "entropy": 5.399364757537842,
      "epoch": 6.988290215911301,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014475181769553103,
      "loss": 4.653,
      "mean_token_accuracy": 0.2414955511689186,
      "num_tokens": 155797580.0,
      "step": 89820
    },
    {
      "entropy": 5.404401254653931,
      "epoch": 6.988679245283019,
      "grad_norm": 1.25,
      "learning_rate": 0.00014472921905627655,
      "loss": 4.6636,
      "mean_token_accuracy": 0.2367745891213417,
      "num_tokens": 155806193.0,
      "step": 89825
    },
    {
      "entropy": 5.400348949432373,
      "epoch": 6.989068274654737,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014470662239367377,
      "loss": 4.682,
      "mean_token_accuracy": 0.23352742642164231,
      "num_tokens": 155815227.0,
      "step": 89830
    },
    {
      "entropy": 5.370688819885254,
      "epoch": 6.989457304026454,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014468402770806565,
      "loss": 4.6212,
      "mean_token_accuracy": 0.23942710757255553,
      "num_tokens": 155823708.0,
      "step": 89835
    },
    {
      "entropy": 5.3683442115783695,
      "epoch": 6.989846333398171,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014466143499979495,
      "loss": 4.6764,
      "mean_token_accuracy": 0.2317187488079071,
      "num_tokens": 155832927.0,
      "step": 89840
    },
    {
      "entropy": 5.314801645278931,
      "epoch": 6.990235362769889,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001446388442692046,
      "loss": 4.6103,
      "mean_token_accuracy": 0.24803401231765748,
      "num_tokens": 155842026.0,
      "step": 89845
    },
    {
      "entropy": 5.387654161453247,
      "epoch": 6.990624392141607,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001446162555166372,
      "loss": 4.6587,
      "mean_token_accuracy": 0.2420601725578308,
      "num_tokens": 155850945.0,
      "step": 89850
    },
    {
      "entropy": 5.4465940475463865,
      "epoch": 6.991013421513324,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00014459366874243568,
      "loss": 4.6667,
      "mean_token_accuracy": 0.23638916313648223,
      "num_tokens": 155858930.0,
      "step": 89855
    },
    {
      "entropy": 5.385081434249878,
      "epoch": 6.991402450885042,
      "grad_norm": 1.21875,
      "learning_rate": 0.00014457108394694268,
      "loss": 4.5969,
      "mean_token_accuracy": 0.24162072241306304,
      "num_tokens": 155868045.0,
      "step": 89860
    },
    {
      "entropy": 5.352211284637451,
      "epoch": 6.99179148025676,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001445485011305008,
      "loss": 4.5844,
      "mean_token_accuracy": 0.24290325939655305,
      "num_tokens": 155877008.0,
      "step": 89865
    },
    {
      "entropy": 5.324769973754883,
      "epoch": 6.992180509628477,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014452592029345286,
      "loss": 4.5192,
      "mean_token_accuracy": 0.2571927890181541,
      "num_tokens": 155885359.0,
      "step": 89870
    },
    {
      "entropy": 5.357456350326538,
      "epoch": 6.992569539000194,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014450334143614125,
      "loss": 4.6962,
      "mean_token_accuracy": 0.23188077807426452,
      "num_tokens": 155893783.0,
      "step": 89875
    },
    {
      "entropy": 5.3238598823547365,
      "epoch": 6.992958568371912,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014448076455890886,
      "loss": 4.5988,
      "mean_token_accuracy": 0.24600047022104263,
      "num_tokens": 155903064.0,
      "step": 89880
    },
    {
      "entropy": 5.440296840667725,
      "epoch": 6.99334759774363,
      "grad_norm": 1.25,
      "learning_rate": 0.00014445818966209818,
      "loss": 4.687,
      "mean_token_accuracy": 0.24332557618618011,
      "num_tokens": 155911783.0,
      "step": 89885
    },
    {
      "entropy": 5.4139715194702145,
      "epoch": 6.993736627115347,
      "grad_norm": 1.453125,
      "learning_rate": 0.00014443561674605158,
      "loss": 4.5698,
      "mean_token_accuracy": 0.24541166722774505,
      "num_tokens": 155919327.0,
      "step": 89890
    },
    {
      "entropy": 5.413807153701782,
      "epoch": 6.994125656487065,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014441304581111174,
      "loss": 4.6813,
      "mean_token_accuracy": 0.24629881232976913,
      "num_tokens": 155927273.0,
      "step": 89895
    },
    {
      "entropy": 5.408115243911743,
      "epoch": 6.994514685858782,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014439047685762104,
      "loss": 4.733,
      "mean_token_accuracy": 0.23326534479856492,
      "num_tokens": 155936211.0,
      "step": 89900
    },
    {
      "entropy": 5.298787784576416,
      "epoch": 6.9949037152305,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014436790988592201,
      "loss": 4.505,
      "mean_token_accuracy": 0.25242430418729783,
      "num_tokens": 155944875.0,
      "step": 89905
    },
    {
      "entropy": 5.321467542648316,
      "epoch": 6.995292744602217,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014434534489635699,
      "loss": 4.5801,
      "mean_token_accuracy": 0.2427438825368881,
      "num_tokens": 155953159.0,
      "step": 89910
    },
    {
      "entropy": 5.459746313095093,
      "epoch": 6.995681773973935,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014432278188926845,
      "loss": 4.7586,
      "mean_token_accuracy": 0.23448698222637177,
      "num_tokens": 155962600.0,
      "step": 89915
    },
    {
      "entropy": 5.275337028503418,
      "epoch": 6.996070803345653,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001443002208649986,
      "loss": 4.5294,
      "mean_token_accuracy": 0.2494208261370659,
      "num_tokens": 155971961.0,
      "step": 89920
    },
    {
      "entropy": 5.4143743991851805,
      "epoch": 6.99645983271737,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014427766182389002,
      "loss": 4.6766,
      "mean_token_accuracy": 0.23622320741415023,
      "num_tokens": 155981096.0,
      "step": 89925
    },
    {
      "entropy": 5.388740587234497,
      "epoch": 6.996848862089088,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014425510476628474,
      "loss": 4.6812,
      "mean_token_accuracy": 0.24523308426141738,
      "num_tokens": 155989355.0,
      "step": 89930
    },
    {
      "entropy": 5.382186555862427,
      "epoch": 6.997237891460806,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001442325496925252,
      "loss": 4.5754,
      "mean_token_accuracy": 0.24558333903551102,
      "num_tokens": 155997823.0,
      "step": 89935
    },
    {
      "entropy": 5.345379447937011,
      "epoch": 6.997626920832523,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001442099966029535,
      "loss": 4.5745,
      "mean_token_accuracy": 0.23580967634916306,
      "num_tokens": 156006105.0,
      "step": 89940
    },
    {
      "entropy": 5.386575412750244,
      "epoch": 6.99801595020424,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014418744549791196,
      "loss": 4.6762,
      "mean_token_accuracy": 0.2360135942697525,
      "num_tokens": 156015759.0,
      "step": 89945
    },
    {
      "entropy": 5.435685920715332,
      "epoch": 6.998404979575958,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014416489637774277,
      "loss": 4.675,
      "mean_token_accuracy": 0.24010681360960007,
      "num_tokens": 156024298.0,
      "step": 89950
    },
    {
      "entropy": 5.368765068054199,
      "epoch": 6.998794008947676,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000144142349242788,
      "loss": 4.6558,
      "mean_token_accuracy": 0.23700524121522903,
      "num_tokens": 156033264.0,
      "step": 89955
    },
    {
      "entropy": 5.425797796249389,
      "epoch": 6.999183038319393,
      "grad_norm": 1.1875,
      "learning_rate": 0.00014411980409338983,
      "loss": 4.7319,
      "mean_token_accuracy": 0.2369768351316452,
      "num_tokens": 156042911.0,
      "step": 89960
    },
    {
      "entropy": 5.350696086883545,
      "epoch": 6.999572067691111,
      "grad_norm": 1.234375,
      "learning_rate": 0.00014409726092989034,
      "loss": 4.6225,
      "mean_token_accuracy": 0.24655075073242189,
      "num_tokens": 156051484.0,
      "step": 89965
    },
    {
      "entropy": 5.4019959449768065,
      "epoch": 6.999961097062828,
      "grad_norm": 1.21875,
      "learning_rate": 0.00014407471975263147,
      "loss": 4.6723,
      "mean_token_accuracy": 0.23713413029909133,
      "num_tokens": 156060196.0,
      "step": 89970
    },
    {
      "entropy": 5.312538411882189,
      "epoch": 7.000311223497374,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014405218056195545,
      "loss": 4.4635,
      "mean_token_accuracy": 0.25627824664115906,
      "num_tokens": 156067765.0,
      "step": 89975
    },
    {
      "entropy": 5.303841829299927,
      "epoch": 7.000700252869092,
      "grad_norm": 1.421875,
      "learning_rate": 0.00014402964335820403,
      "loss": 4.5264,
      "mean_token_accuracy": 0.2511327400803566,
      "num_tokens": 156076160.0,
      "step": 89980
    },
    {
      "entropy": 5.283123445510864,
      "epoch": 7.001089282240809,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00014400710814171943,
      "loss": 4.5048,
      "mean_token_accuracy": 0.25590251982212064,
      "num_tokens": 156084670.0,
      "step": 89985
    },
    {
      "entropy": 5.4067121028900145,
      "epoch": 7.0014783116125265,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001439845749128434,
      "loss": 4.6224,
      "mean_token_accuracy": 0.2503239020705223,
      "num_tokens": 156093914.0,
      "step": 89990
    },
    {
      "entropy": 5.447086811065674,
      "epoch": 7.001867340984244,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014396204367191794,
      "loss": 4.7139,
      "mean_token_accuracy": 0.23153126090765,
      "num_tokens": 156102298.0,
      "step": 89995
    },
    {
      "entropy": 5.40761890411377,
      "epoch": 7.002256370355962,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014393951441928488,
      "loss": 4.5648,
      "mean_token_accuracy": 0.25297785848379134,
      "num_tokens": 156111620.0,
      "step": 90000
    },
    {
      "epoch": 7.002256370355962,
      "eval_entropy": 5.158628855036129,
      "eval_loss": 4.8515472412109375,
      "eval_mean_token_accuracy": 0.23689479727057797,
      "eval_num_tokens": 156111620.0,
      "eval_runtime": 28.7284,
      "eval_samples_per_second": 1446.581,
      "eval_steps_per_second": 180.831,
      "step": 90000
    },
    {
      "entropy": 5.469105863571167,
      "epoch": 7.00264539972768,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014391698715528613,
      "loss": 4.67,
      "mean_token_accuracy": 0.24697525352239608,
      "num_tokens": 156119883.0,
      "step": 90005
    },
    {
      "entropy": 5.399833726882934,
      "epoch": 7.003034429099397,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001438944618802634,
      "loss": 4.6286,
      "mean_token_accuracy": 0.24486246705055237,
      "num_tokens": 156128604.0,
      "step": 90010
    },
    {
      "entropy": 5.39777512550354,
      "epoch": 7.003423458471114,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001438719385945585,
      "loss": 4.639,
      "mean_token_accuracy": 0.2391916587948799,
      "num_tokens": 156136814.0,
      "step": 90015
    },
    {
      "entropy": 5.4153800964355465,
      "epoch": 7.003812487842832,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001438494172985133,
      "loss": 4.6471,
      "mean_token_accuracy": 0.2460146114230156,
      "num_tokens": 156145889.0,
      "step": 90020
    },
    {
      "entropy": 5.363300848007202,
      "epoch": 7.0042015172145495,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001438268979924694,
      "loss": 4.6236,
      "mean_token_accuracy": 0.2503401651978493,
      "num_tokens": 156154627.0,
      "step": 90025
    },
    {
      "entropy": 5.398645210266113,
      "epoch": 7.004590546586267,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001438043806767686,
      "loss": 4.6198,
      "mean_token_accuracy": 0.23911793977022172,
      "num_tokens": 156163508.0,
      "step": 90030
    },
    {
      "entropy": 5.458889532089233,
      "epoch": 7.004979575957985,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014378186535175247,
      "loss": 4.6904,
      "mean_token_accuracy": 0.24258648455142975,
      "num_tokens": 156172819.0,
      "step": 90035
    },
    {
      "entropy": 5.426060152053833,
      "epoch": 7.005368605329703,
      "grad_norm": 1.375,
      "learning_rate": 0.00014375935201776268,
      "loss": 4.6244,
      "mean_token_accuracy": 0.2406049743294716,
      "num_tokens": 156181492.0,
      "step": 90040
    },
    {
      "entropy": 5.3880290508270265,
      "epoch": 7.00575763470142,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014373684067514088,
      "loss": 4.5713,
      "mean_token_accuracy": 0.25495119839906694,
      "num_tokens": 156190642.0,
      "step": 90045
    },
    {
      "entropy": 5.314709281921386,
      "epoch": 7.006146664073137,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001437143313242285,
      "loss": 4.4978,
      "mean_token_accuracy": 0.2605133786797523,
      "num_tokens": 156199751.0,
      "step": 90050
    },
    {
      "entropy": 5.301628255844117,
      "epoch": 7.006535693444855,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014369182396536728,
      "loss": 4.5302,
      "mean_token_accuracy": 0.25309610962867735,
      "num_tokens": 156208332.0,
      "step": 90055
    },
    {
      "entropy": 5.301197719573975,
      "epoch": 7.0069247228165725,
      "grad_norm": 1.375,
      "learning_rate": 0.00014366931859889853,
      "loss": 4.5374,
      "mean_token_accuracy": 0.24901947677135466,
      "num_tokens": 156216977.0,
      "step": 90060
    },
    {
      "entropy": 5.38556694984436,
      "epoch": 7.00731375218829,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014364681522516392,
      "loss": 4.6183,
      "mean_token_accuracy": 0.2463506892323494,
      "num_tokens": 156225844.0,
      "step": 90065
    },
    {
      "entropy": 5.3629655838012695,
      "epoch": 7.007702781560008,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014362431384450474,
      "loss": 4.5333,
      "mean_token_accuracy": 0.2483459457755089,
      "num_tokens": 156233208.0,
      "step": 90070
    },
    {
      "entropy": 5.403060817718506,
      "epoch": 7.008091810931726,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001436018144572625,
      "loss": 4.6052,
      "mean_token_accuracy": 0.24076692759990692,
      "num_tokens": 156241165.0,
      "step": 90075
    },
    {
      "entropy": 5.324351596832275,
      "epoch": 7.008480840303443,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014357931706377868,
      "loss": 4.5797,
      "mean_token_accuracy": 0.24957185983657837,
      "num_tokens": 156250429.0,
      "step": 90080
    },
    {
      "entropy": 5.426231718063354,
      "epoch": 7.00886986967516,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00014355682166439443,
      "loss": 4.7091,
      "mean_token_accuracy": 0.2353959009051323,
      "num_tokens": 156259250.0,
      "step": 90085
    },
    {
      "entropy": 5.409934234619141,
      "epoch": 7.009258899046878,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001435343282594513,
      "loss": 4.6139,
      "mean_token_accuracy": 0.24522257894277572,
      "num_tokens": 156267580.0,
      "step": 90090
    },
    {
      "entropy": 5.387844181060791,
      "epoch": 7.0096479284185955,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001435118368492904,
      "loss": 4.5825,
      "mean_token_accuracy": 0.2457406684756279,
      "num_tokens": 156276434.0,
      "step": 90095
    },
    {
      "entropy": 5.376532745361328,
      "epoch": 7.010036957790313,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014348934743425314,
      "loss": 4.6124,
      "mean_token_accuracy": 0.24795571565628052,
      "num_tokens": 156284984.0,
      "step": 90100
    },
    {
      "entropy": 5.338795328140259,
      "epoch": 7.010425987162031,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014346686001468067,
      "loss": 4.6315,
      "mean_token_accuracy": 0.24564227014780043,
      "num_tokens": 156293728.0,
      "step": 90105
    },
    {
      "entropy": 5.3393354415893555,
      "epoch": 7.010815016533749,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014344437459091434,
      "loss": 4.5518,
      "mean_token_accuracy": 0.25540982484817504,
      "num_tokens": 156303027.0,
      "step": 90110
    },
    {
      "entropy": 5.484247159957886,
      "epoch": 7.011204045905465,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014342189116329512,
      "loss": 4.7259,
      "mean_token_accuracy": 0.23415337055921553,
      "num_tokens": 156311890.0,
      "step": 90115
    },
    {
      "entropy": 5.493964338302613,
      "epoch": 7.011593075277183,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014339940973216438,
      "loss": 4.6875,
      "mean_token_accuracy": 0.24070482105016708,
      "num_tokens": 156320647.0,
      "step": 90120
    },
    {
      "entropy": 5.495654726028443,
      "epoch": 7.011982104648901,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014337693029786312,
      "loss": 4.713,
      "mean_token_accuracy": 0.23850999474525453,
      "num_tokens": 156329909.0,
      "step": 90125
    },
    {
      "entropy": 5.4419536113739015,
      "epoch": 7.0123711340206185,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014335445286073234,
      "loss": 4.6397,
      "mean_token_accuracy": 0.2413053795695305,
      "num_tokens": 156338372.0,
      "step": 90130
    },
    {
      "entropy": 5.332943725585937,
      "epoch": 7.012760163392336,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014333197742111333,
      "loss": 4.4891,
      "mean_token_accuracy": 0.25009157359600065,
      "num_tokens": 156347164.0,
      "step": 90135
    },
    {
      "entropy": 5.385561609268189,
      "epoch": 7.013149192764054,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014330950397934677,
      "loss": 4.659,
      "mean_token_accuracy": 0.24437061697244644,
      "num_tokens": 156356101.0,
      "step": 90140
    },
    {
      "entropy": 5.330642795562744,
      "epoch": 7.013538222135772,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014328703253577407,
      "loss": 4.567,
      "mean_token_accuracy": 0.2484587013721466,
      "num_tokens": 156364275.0,
      "step": 90145
    },
    {
      "entropy": 5.338982057571411,
      "epoch": 7.013927251507488,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014326456309073592,
      "loss": 4.5498,
      "mean_token_accuracy": 0.2531727448105812,
      "num_tokens": 156372995.0,
      "step": 90150
    },
    {
      "entropy": 5.362654018402099,
      "epoch": 7.014316280879206,
      "grad_norm": 1.375,
      "learning_rate": 0.00014324209564457347,
      "loss": 4.5078,
      "mean_token_accuracy": 0.25286509096622467,
      "num_tokens": 156381209.0,
      "step": 90155
    },
    {
      "entropy": 5.2903969287872314,
      "epoch": 7.014705310250924,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014321963019762747,
      "loss": 4.5258,
      "mean_token_accuracy": 0.252904999256134,
      "num_tokens": 156390096.0,
      "step": 90160
    },
    {
      "entropy": 5.370199966430664,
      "epoch": 7.0150943396226415,
      "grad_norm": 1.375,
      "learning_rate": 0.00014319716675023874,
      "loss": 4.6204,
      "mean_token_accuracy": 0.23871349096298217,
      "num_tokens": 156398907.0,
      "step": 90165
    },
    {
      "entropy": 5.360746049880982,
      "epoch": 7.015483368994359,
      "grad_norm": 1.265625,
      "learning_rate": 0.00014317470530274834,
      "loss": 4.638,
      "mean_token_accuracy": 0.2461397171020508,
      "num_tokens": 156407973.0,
      "step": 90170
    },
    {
      "entropy": 5.390788745880127,
      "epoch": 7.015872398366077,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014315224585549686,
      "loss": 4.5708,
      "mean_token_accuracy": 0.252447834610939,
      "num_tokens": 156417096.0,
      "step": 90175
    },
    {
      "entropy": 5.435865831375122,
      "epoch": 7.016261427737795,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014312978840882524,
      "loss": 4.5961,
      "mean_token_accuracy": 0.24610868841409683,
      "num_tokens": 156425339.0,
      "step": 90180
    },
    {
      "entropy": 5.368530654907227,
      "epoch": 7.016650457109511,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014310733296307414,
      "loss": 4.6673,
      "mean_token_accuracy": 0.23838739097118378,
      "num_tokens": 156433782.0,
      "step": 90185
    },
    {
      "entropy": 5.377924203872681,
      "epoch": 7.017039486481229,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014308487951858443,
      "loss": 4.5998,
      "mean_token_accuracy": 0.2441546529531479,
      "num_tokens": 156443034.0,
      "step": 90190
    },
    {
      "entropy": 5.435801458358765,
      "epoch": 7.017428515852947,
      "grad_norm": 1.1875,
      "learning_rate": 0.00014306242807569661,
      "loss": 4.6892,
      "mean_token_accuracy": 0.23872665613889693,
      "num_tokens": 156452628.0,
      "step": 90195
    },
    {
      "entropy": 5.386900472640991,
      "epoch": 7.0178175452246645,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00014303997863475155,
      "loss": 4.5566,
      "mean_token_accuracy": 0.2533245176076889,
      "num_tokens": 156461326.0,
      "step": 90200
    },
    {
      "entropy": 5.406253004074097,
      "epoch": 7.018206574596382,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014301753119608972,
      "loss": 4.6166,
      "mean_token_accuracy": 0.24088977873325348,
      "num_tokens": 156470658.0,
      "step": 90205
    },
    {
      "entropy": 5.39936842918396,
      "epoch": 7.0185956039681,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00014299508576005166,
      "loss": 4.6434,
      "mean_token_accuracy": 0.23767687529325485,
      "num_tokens": 156479885.0,
      "step": 90210
    },
    {
      "entropy": 5.310983943939209,
      "epoch": 7.018984633339818,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014297264232697823,
      "loss": 4.5254,
      "mean_token_accuracy": 0.2549836337566376,
      "num_tokens": 156487667.0,
      "step": 90215
    },
    {
      "entropy": 5.329548072814942,
      "epoch": 7.019373662711534,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014295020089720973,
      "loss": 4.5839,
      "mean_token_accuracy": 0.242526613175869,
      "num_tokens": 156495945.0,
      "step": 90220
    },
    {
      "entropy": 5.382296323776245,
      "epoch": 7.019762692083252,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014292776147108684,
      "loss": 4.6036,
      "mean_token_accuracy": 0.24670106768608094,
      "num_tokens": 156504827.0,
      "step": 90225
    },
    {
      "entropy": 5.38292555809021,
      "epoch": 7.02015172145497,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014290532404894984,
      "loss": 4.6069,
      "mean_token_accuracy": 0.24181021004915237,
      "num_tokens": 156514073.0,
      "step": 90230
    },
    {
      "entropy": 5.407919549942017,
      "epoch": 7.0205407508266875,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014288288863113937,
      "loss": 4.5506,
      "mean_token_accuracy": 0.2500641331076622,
      "num_tokens": 156521691.0,
      "step": 90235
    },
    {
      "entropy": 5.485527276992798,
      "epoch": 7.020929780198405,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001428604552179958,
      "loss": 4.7029,
      "mean_token_accuracy": 0.23322039544582368,
      "num_tokens": 156530272.0,
      "step": 90240
    },
    {
      "entropy": 5.373916292190552,
      "epoch": 7.021318809570123,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014283802380985944,
      "loss": 4.5464,
      "mean_token_accuracy": 0.2565394461154938,
      "num_tokens": 156538564.0,
      "step": 90245
    },
    {
      "entropy": 5.395167398452759,
      "epoch": 7.02170783894184,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014281559440707079,
      "loss": 4.6486,
      "mean_token_accuracy": 0.24414647817611695,
      "num_tokens": 156547715.0,
      "step": 90250
    },
    {
      "entropy": 5.359339094161987,
      "epoch": 7.022096868313557,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014279316700997,
      "loss": 4.5216,
      "mean_token_accuracy": 0.24762504696846008,
      "num_tokens": 156556468.0,
      "step": 90255
    },
    {
      "entropy": 5.399831867218017,
      "epoch": 7.022485897685275,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014277074161889758,
      "loss": 4.5662,
      "mean_token_accuracy": 0.25042266100645066,
      "num_tokens": 156564905.0,
      "step": 90260
    },
    {
      "entropy": 5.378294992446899,
      "epoch": 7.022874927056993,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001427483182341936,
      "loss": 4.6041,
      "mean_token_accuracy": 0.24915739744901658,
      "num_tokens": 156573830.0,
      "step": 90265
    },
    {
      "entropy": 5.352334594726562,
      "epoch": 7.0232639564287105,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014272589685619845,
      "loss": 4.5737,
      "mean_token_accuracy": 0.2502042219042778,
      "num_tokens": 156582451.0,
      "step": 90270
    },
    {
      "entropy": 5.36664228439331,
      "epoch": 7.023652985800428,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001427034774852522,
      "loss": 4.6105,
      "mean_token_accuracy": 0.2399081200361252,
      "num_tokens": 156590793.0,
      "step": 90275
    },
    {
      "entropy": 5.3968676090240475,
      "epoch": 7.024042015172146,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014268106012169514,
      "loss": 4.6064,
      "mean_token_accuracy": 0.23968362659215928,
      "num_tokens": 156599444.0,
      "step": 90280
    },
    {
      "entropy": 5.36963300704956,
      "epoch": 7.024431044543863,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014265864476586742,
      "loss": 4.564,
      "mean_token_accuracy": 0.24803753644227983,
      "num_tokens": 156608338.0,
      "step": 90285
    },
    {
      "entropy": 5.3835653305053714,
      "epoch": 7.02482007391558,
      "grad_norm": 1.421875,
      "learning_rate": 0.00014263623141810906,
      "loss": 4.5618,
      "mean_token_accuracy": 0.25210612267255783,
      "num_tokens": 156616632.0,
      "step": 90290
    },
    {
      "entropy": 5.294729137420655,
      "epoch": 7.025209103287298,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014261382007876028,
      "loss": 4.5303,
      "mean_token_accuracy": 0.2517367318272591,
      "num_tokens": 156624366.0,
      "step": 90295
    },
    {
      "entropy": 5.357906341552734,
      "epoch": 7.025598132659016,
      "grad_norm": 1.328125,
      "learning_rate": 0.000142591410748161,
      "loss": 4.5884,
      "mean_token_accuracy": 0.24427227973937987,
      "num_tokens": 156632536.0,
      "step": 90300
    },
    {
      "entropy": 5.417682552337647,
      "epoch": 7.0259871620307335,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001425690034266514,
      "loss": 4.6521,
      "mean_token_accuracy": 0.24228838831186295,
      "num_tokens": 156641256.0,
      "step": 90305
    },
    {
      "entropy": 5.38389573097229,
      "epoch": 7.026376191402451,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001425465981145713,
      "loss": 4.6108,
      "mean_token_accuracy": 0.24664003551006317,
      "num_tokens": 156650262.0,
      "step": 90310
    },
    {
      "entropy": 5.372488880157471,
      "epoch": 7.026765220774169,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014252419481226085,
      "loss": 4.6357,
      "mean_token_accuracy": 0.23763690888881683,
      "num_tokens": 156658974.0,
      "step": 90315
    },
    {
      "entropy": 5.394075727462768,
      "epoch": 7.027154250145886,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014250179352005987,
      "loss": 4.561,
      "mean_token_accuracy": 0.2505316883325577,
      "num_tokens": 156667754.0,
      "step": 90320
    },
    {
      "entropy": 5.3771830081939695,
      "epoch": 7.027543279517603,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001424793942383082,
      "loss": 4.5348,
      "mean_token_accuracy": 0.23985546082258224,
      "num_tokens": 156675022.0,
      "step": 90325
    },
    {
      "entropy": 5.383935785293579,
      "epoch": 7.027932308889321,
      "grad_norm": 1.375,
      "learning_rate": 0.00014245699696734587,
      "loss": 4.6334,
      "mean_token_accuracy": 0.2465617150068283,
      "num_tokens": 156683232.0,
      "step": 90330
    },
    {
      "entropy": 5.356411933898926,
      "epoch": 7.028321338261039,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001424346017075126,
      "loss": 4.5663,
      "mean_token_accuracy": 0.2501958057284355,
      "num_tokens": 156691844.0,
      "step": 90335
    },
    {
      "entropy": 5.3018553256988525,
      "epoch": 7.0287103676327565,
      "grad_norm": 1.375,
      "learning_rate": 0.0001424122084591482,
      "loss": 4.6023,
      "mean_token_accuracy": 0.2531855657696724,
      "num_tokens": 156700795.0,
      "step": 90340
    },
    {
      "entropy": 5.359295177459717,
      "epoch": 7.029099397004474,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00014238981722259258,
      "loss": 4.5279,
      "mean_token_accuracy": 0.2576060563325882,
      "num_tokens": 156709880.0,
      "step": 90345
    },
    {
      "entropy": 5.408241033554077,
      "epoch": 7.029488426376191,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014236742799818536,
      "loss": 4.5953,
      "mean_token_accuracy": 0.24689457267522813,
      "num_tokens": 156718961.0,
      "step": 90350
    },
    {
      "entropy": 5.48872389793396,
      "epoch": 7.029877455747909,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014234504078626642,
      "loss": 4.7325,
      "mean_token_accuracy": 0.23687449991703033,
      "num_tokens": 156727891.0,
      "step": 90355
    },
    {
      "entropy": 5.391637325286865,
      "epoch": 7.030266485119626,
      "grad_norm": 1.375,
      "learning_rate": 0.00014232265558717521,
      "loss": 4.5918,
      "mean_token_accuracy": 0.24651237428188325,
      "num_tokens": 156736765.0,
      "step": 90360
    },
    {
      "entropy": 5.43891167640686,
      "epoch": 7.030655514491344,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014230027240125166,
      "loss": 4.6652,
      "mean_token_accuracy": 0.23176310062408448,
      "num_tokens": 156745266.0,
      "step": 90365
    },
    {
      "entropy": 5.398260879516601,
      "epoch": 7.031044543863062,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014227789122883512,
      "loss": 4.6028,
      "mean_token_accuracy": 0.2454163283109665,
      "num_tokens": 156754215.0,
      "step": 90370
    },
    {
      "entropy": 5.383979082107544,
      "epoch": 7.0314335732347795,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014225551207026546,
      "loss": 4.6116,
      "mean_token_accuracy": 0.25096548050642015,
      "num_tokens": 156763102.0,
      "step": 90375
    },
    {
      "entropy": 5.333259010314942,
      "epoch": 7.031822602606497,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014223313492588204,
      "loss": 4.5182,
      "mean_token_accuracy": 0.2518729344010353,
      "num_tokens": 156771700.0,
      "step": 90380
    },
    {
      "entropy": 5.384879922866821,
      "epoch": 7.032211631978214,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014221075979602453,
      "loss": 4.5983,
      "mean_token_accuracy": 0.2438817501068115,
      "num_tokens": 156780048.0,
      "step": 90385
    },
    {
      "entropy": 5.352311134338379,
      "epoch": 7.032600661349932,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001421883866810323,
      "loss": 4.6766,
      "mean_token_accuracy": 0.23724692016839982,
      "num_tokens": 156789017.0,
      "step": 90390
    },
    {
      "entropy": 5.437381601333618,
      "epoch": 7.032989690721649,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000142166015581245,
      "loss": 4.6934,
      "mean_token_accuracy": 0.24058008342981338,
      "num_tokens": 156797109.0,
      "step": 90395
    },
    {
      "entropy": 5.411159896850586,
      "epoch": 7.033378720093367,
      "grad_norm": 1.25,
      "learning_rate": 0.00014214364649700202,
      "loss": 4.6891,
      "mean_token_accuracy": 0.24143632054328917,
      "num_tokens": 156805807.0,
      "step": 90400
    },
    {
      "entropy": 5.448505163192749,
      "epoch": 7.033767749465085,
      "grad_norm": 1.421875,
      "learning_rate": 0.00014212127942864251,
      "loss": 4.6169,
      "mean_token_accuracy": 0.23776184022426605,
      "num_tokens": 156814384.0,
      "step": 90405
    },
    {
      "entropy": 5.500348663330078,
      "epoch": 7.0341567788368025,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001420989143765063,
      "loss": 4.7492,
      "mean_token_accuracy": 0.23392697274684907,
      "num_tokens": 156822059.0,
      "step": 90410
    },
    {
      "entropy": 5.309268760681152,
      "epoch": 7.03454580820852,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014207655134093245,
      "loss": 4.5123,
      "mean_token_accuracy": 0.26104619204998014,
      "num_tokens": 156830981.0,
      "step": 90415
    },
    {
      "entropy": 5.3668036460876465,
      "epoch": 7.034934837580237,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001420541903222604,
      "loss": 4.6414,
      "mean_token_accuracy": 0.24169587790966035,
      "num_tokens": 156839396.0,
      "step": 90420
    },
    {
      "entropy": 5.346806430816651,
      "epoch": 7.035323866951955,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014203183132082949,
      "loss": 4.5954,
      "mean_token_accuracy": 0.24926914870738984,
      "num_tokens": 156847426.0,
      "step": 90425
    },
    {
      "entropy": 5.407706165313721,
      "epoch": 7.035712896323672,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014200947433697875,
      "loss": 4.5937,
      "mean_token_accuracy": 0.2482050657272339,
      "num_tokens": 156856502.0,
      "step": 90430
    },
    {
      "entropy": 5.359251403808594,
      "epoch": 7.03610192569539,
      "grad_norm": 1.25,
      "learning_rate": 0.00014198711937104765,
      "loss": 4.5465,
      "mean_token_accuracy": 0.2513579547405243,
      "num_tokens": 156865335.0,
      "step": 90435
    },
    {
      "entropy": 5.326341915130615,
      "epoch": 7.036490955067108,
      "grad_norm": 1.234375,
      "learning_rate": 0.00014196476642337524,
      "loss": 4.5822,
      "mean_token_accuracy": 0.24208869338035582,
      "num_tokens": 156874564.0,
      "step": 90440
    },
    {
      "entropy": 5.37560453414917,
      "epoch": 7.0368799844388255,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00014194241549430085,
      "loss": 4.5999,
      "mean_token_accuracy": 0.24035593271255493,
      "num_tokens": 156882480.0,
      "step": 90445
    },
    {
      "entropy": 5.349340915679932,
      "epoch": 7.037269013810542,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001419200665841634,
      "loss": 4.5332,
      "mean_token_accuracy": 0.25092128068208697,
      "num_tokens": 156891187.0,
      "step": 90450
    },
    {
      "entropy": 5.479745388031006,
      "epoch": 7.03765804318226,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001418977196933022,
      "loss": 4.6647,
      "mean_token_accuracy": 0.24079934656620025,
      "num_tokens": 156899923.0,
      "step": 90455
    },
    {
      "entropy": 5.328784608840943,
      "epoch": 7.038047072553978,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001418753748220562,
      "loss": 4.5281,
      "mean_token_accuracy": 0.24792135059833526,
      "num_tokens": 156908368.0,
      "step": 90460
    },
    {
      "entropy": 5.337542152404785,
      "epoch": 7.038436101925695,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014185303197076455,
      "loss": 4.6179,
      "mean_token_accuracy": 0.2402854934334755,
      "num_tokens": 156916847.0,
      "step": 90465
    },
    {
      "entropy": 5.297685241699218,
      "epoch": 7.038825131297413,
      "grad_norm": 1.375,
      "learning_rate": 0.00014183069113976615,
      "loss": 4.5472,
      "mean_token_accuracy": 0.2490894764661789,
      "num_tokens": 156925304.0,
      "step": 90470
    },
    {
      "entropy": 5.417333745956421,
      "epoch": 7.039214160669131,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014180835232940002,
      "loss": 4.6254,
      "mean_token_accuracy": 0.24444849491119386,
      "num_tokens": 156933687.0,
      "step": 90475
    },
    {
      "entropy": 5.493073749542236,
      "epoch": 7.0396031900408484,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001417860155400053,
      "loss": 4.6792,
      "mean_token_accuracy": 0.2351874515414238,
      "num_tokens": 156941416.0,
      "step": 90480
    },
    {
      "entropy": 5.479555940628051,
      "epoch": 7.039992219412565,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014176368077192065,
      "loss": 4.6449,
      "mean_token_accuracy": 0.23728094398975372,
      "num_tokens": 156950646.0,
      "step": 90485
    },
    {
      "entropy": 5.357027006149292,
      "epoch": 7.040381248784283,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014174134802548514,
      "loss": 4.5832,
      "mean_token_accuracy": 0.24261299520730972,
      "num_tokens": 156959680.0,
      "step": 90490
    },
    {
      "entropy": 5.37088680267334,
      "epoch": 7.040770278156001,
      "grad_norm": 1.375,
      "learning_rate": 0.00014171901730103748,
      "loss": 4.6363,
      "mean_token_accuracy": 0.2495381012558937,
      "num_tokens": 156968200.0,
      "step": 90495
    },
    {
      "entropy": 5.36671724319458,
      "epoch": 7.041159307527718,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001416966885989167,
      "loss": 4.5813,
      "mean_token_accuracy": 0.24611576944589614,
      "num_tokens": 156977257.0,
      "step": 90500
    },
    {
      "entropy": 5.406401109695435,
      "epoch": 7.041548336899436,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014167436191946147,
      "loss": 4.6761,
      "mean_token_accuracy": 0.2424040123820305,
      "num_tokens": 156985280.0,
      "step": 90505
    },
    {
      "entropy": 5.427651309967041,
      "epoch": 7.041937366271154,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014165203726301055,
      "loss": 4.5761,
      "mean_token_accuracy": 0.25598070323467254,
      "num_tokens": 156993866.0,
      "step": 90510
    },
    {
      "entropy": 5.473586702346802,
      "epoch": 7.042326395642871,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014162971462990278,
      "loss": 4.6898,
      "mean_token_accuracy": 0.23484684228897096,
      "num_tokens": 157002441.0,
      "step": 90515
    },
    {
      "entropy": 5.47001838684082,
      "epoch": 7.042715425014588,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014160739402047667,
      "loss": 4.6523,
      "mean_token_accuracy": 0.23967466205358506,
      "num_tokens": 157010548.0,
      "step": 90520
    },
    {
      "entropy": 5.359698390960693,
      "epoch": 7.043104454386306,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001415850754350712,
      "loss": 4.5652,
      "mean_token_accuracy": 0.2526508018374443,
      "num_tokens": 157019878.0,
      "step": 90525
    },
    {
      "entropy": 5.410479307174683,
      "epoch": 7.043493483758024,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014156275887402476,
      "loss": 4.6849,
      "mean_token_accuracy": 0.24247892200946808,
      "num_tokens": 157029042.0,
      "step": 90530
    },
    {
      "entropy": 5.42592420578003,
      "epoch": 7.043882513129741,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014154044433767617,
      "loss": 4.6438,
      "mean_token_accuracy": 0.2448235958814621,
      "num_tokens": 157037837.0,
      "step": 90535
    },
    {
      "entropy": 5.370036172866821,
      "epoch": 7.044271542501459,
      "grad_norm": 1.25,
      "learning_rate": 0.00014151813182636375,
      "loss": 4.5528,
      "mean_token_accuracy": 0.2584438145160675,
      "num_tokens": 157046805.0,
      "step": 90540
    },
    {
      "entropy": 5.377266931533813,
      "epoch": 7.044660571873177,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014149582134042628,
      "loss": 4.6356,
      "mean_token_accuracy": 0.2423542246222496,
      "num_tokens": 157055497.0,
      "step": 90545
    },
    {
      "entropy": 5.396782541275025,
      "epoch": 7.045049601244894,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014147351288020232,
      "loss": 4.6224,
      "mean_token_accuracy": 0.2434399500489235,
      "num_tokens": 157064534.0,
      "step": 90550
    },
    {
      "entropy": 5.303544330596924,
      "epoch": 7.045438630616611,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014145120644603015,
      "loss": 4.5481,
      "mean_token_accuracy": 0.2460574820637703,
      "num_tokens": 157073565.0,
      "step": 90555
    },
    {
      "entropy": 5.385702610015869,
      "epoch": 7.045827659988329,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001414289020382485,
      "loss": 4.5766,
      "mean_token_accuracy": 0.24406556338071822,
      "num_tokens": 157082139.0,
      "step": 90560
    },
    {
      "entropy": 5.45256814956665,
      "epoch": 7.046216689360047,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014140659965719554,
      "loss": 4.7028,
      "mean_token_accuracy": 0.2352342799305916,
      "num_tokens": 157090646.0,
      "step": 90565
    },
    {
      "entropy": 5.369922828674317,
      "epoch": 7.046605718731764,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001413842993032099,
      "loss": 4.5841,
      "mean_token_accuracy": 0.24832893311977386,
      "num_tokens": 157098683.0,
      "step": 90570
    },
    {
      "entropy": 5.411153507232666,
      "epoch": 7.046994748103482,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001413620009766297,
      "loss": 4.6132,
      "mean_token_accuracy": 0.24043766558170318,
      "num_tokens": 157107329.0,
      "step": 90575
    },
    {
      "entropy": 5.307624864578247,
      "epoch": 7.0473837774752,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014133970467779356,
      "loss": 4.5257,
      "mean_token_accuracy": 0.24353550225496293,
      "num_tokens": 157116203.0,
      "step": 90580
    },
    {
      "entropy": 5.327100038528442,
      "epoch": 7.0477728068469165,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014131741040703967,
      "loss": 4.5753,
      "mean_token_accuracy": 0.24991081207990645,
      "num_tokens": 157125797.0,
      "step": 90585
    },
    {
      "entropy": 5.376039743423462,
      "epoch": 7.048161836218634,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001412951181647062,
      "loss": 4.5938,
      "mean_token_accuracy": 0.25237798392772676,
      "num_tokens": 157133718.0,
      "step": 90590
    },
    {
      "entropy": 5.4069324970245365,
      "epoch": 7.048550865590352,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014127282795113158,
      "loss": 4.7458,
      "mean_token_accuracy": 0.24094966053962708,
      "num_tokens": 157142812.0,
      "step": 90595
    },
    {
      "entropy": 5.355803775787353,
      "epoch": 7.04893989496207,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014125053976665386,
      "loss": 4.6003,
      "mean_token_accuracy": 0.24812051653862,
      "num_tokens": 157151024.0,
      "step": 90600
    },
    {
      "entropy": 5.354247903823852,
      "epoch": 7.049328924333787,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014122825361161134,
      "loss": 4.4322,
      "mean_token_accuracy": 0.2580421715974808,
      "num_tokens": 157158653.0,
      "step": 90605
    },
    {
      "entropy": 5.4534299850463865,
      "epoch": 7.049717953705505,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014120596948634213,
      "loss": 4.6445,
      "mean_token_accuracy": 0.25002614706754683,
      "num_tokens": 157167930.0,
      "step": 90610
    },
    {
      "entropy": 5.362213230133056,
      "epoch": 7.050106983077223,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0001411836873911844,
      "loss": 4.5934,
      "mean_token_accuracy": 0.2510505810379982,
      "num_tokens": 157176929.0,
      "step": 90615
    },
    {
      "entropy": 5.3800452709197994,
      "epoch": 7.0504960124489395,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001411614073264763,
      "loss": 4.6221,
      "mean_token_accuracy": 0.2464775711297989,
      "num_tokens": 157185055.0,
      "step": 90620
    },
    {
      "entropy": 5.381664085388183,
      "epoch": 7.050885041820657,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00014113912929255575,
      "loss": 4.6134,
      "mean_token_accuracy": 0.24213008880615233,
      "num_tokens": 157195942.0,
      "step": 90625
    },
    {
      "entropy": 5.38079605102539,
      "epoch": 7.051274071192375,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001411168532897609,
      "loss": 4.609,
      "mean_token_accuracy": 0.24518264085054398,
      "num_tokens": 157204687.0,
      "step": 90630
    },
    {
      "entropy": 5.413254356384277,
      "epoch": 7.051663100564093,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014109457931842965,
      "loss": 4.632,
      "mean_token_accuracy": 0.23882546424865722,
      "num_tokens": 157213340.0,
      "step": 90635
    },
    {
      "entropy": 5.327270126342773,
      "epoch": 7.05205212993581,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001410723073789001,
      "loss": 4.4735,
      "mean_token_accuracy": 0.2556880787014961,
      "num_tokens": 157221418.0,
      "step": 90640
    },
    {
      "entropy": 5.321653413772583,
      "epoch": 7.052441159307528,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00014105003747150998,
      "loss": 4.5578,
      "mean_token_accuracy": 0.25513159185647966,
      "num_tokens": 157229629.0,
      "step": 90645
    },
    {
      "entropy": 5.406086874008179,
      "epoch": 7.052830188679246,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014102776959659747,
      "loss": 4.6348,
      "mean_token_accuracy": 0.24039511233568192,
      "num_tokens": 157238613.0,
      "step": 90650
    },
    {
      "entropy": 5.413584899902344,
      "epoch": 7.0532192180509625,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014100550375450025,
      "loss": 4.6502,
      "mean_token_accuracy": 0.24504314959049225,
      "num_tokens": 157247659.0,
      "step": 90655
    },
    {
      "entropy": 5.403626775741577,
      "epoch": 7.05360824742268,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014098323994555628,
      "loss": 4.6109,
      "mean_token_accuracy": 0.24819845706224442,
      "num_tokens": 157256328.0,
      "step": 90660
    },
    {
      "entropy": 5.456136560440063,
      "epoch": 7.053997276794398,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014096097817010333,
      "loss": 4.6894,
      "mean_token_accuracy": 0.23941798657178878,
      "num_tokens": 157264039.0,
      "step": 90665
    },
    {
      "entropy": 5.384472322463989,
      "epoch": 7.054386306166116,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00014093871842847915,
      "loss": 4.5875,
      "mean_token_accuracy": 0.24930048286914824,
      "num_tokens": 157272633.0,
      "step": 90670
    },
    {
      "entropy": 5.310544776916504,
      "epoch": 7.054775335537833,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014091646072102155,
      "loss": 4.5141,
      "mean_token_accuracy": 0.248792465031147,
      "num_tokens": 157280959.0,
      "step": 90675
    },
    {
      "entropy": 5.437867259979248,
      "epoch": 7.055164364909551,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014089420504806815,
      "loss": 4.6471,
      "mean_token_accuracy": 0.2440081864595413,
      "num_tokens": 157289388.0,
      "step": 90680
    },
    {
      "entropy": 5.290671253204346,
      "epoch": 7.055553394281268,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001408719514099569,
      "loss": 4.4995,
      "mean_token_accuracy": 0.25810355246067046,
      "num_tokens": 157298067.0,
      "step": 90685
    },
    {
      "entropy": 5.361550426483154,
      "epoch": 7.0559424236529855,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014084969980702523,
      "loss": 4.5693,
      "mean_token_accuracy": 0.24985786080360411,
      "num_tokens": 157307238.0,
      "step": 90690
    },
    {
      "entropy": 5.352295684814453,
      "epoch": 7.056331453024703,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001408274502396109,
      "loss": 4.5522,
      "mean_token_accuracy": 0.2535127565264702,
      "num_tokens": 157316321.0,
      "step": 90695
    },
    {
      "entropy": 5.337662982940674,
      "epoch": 7.056720482396421,
      "grad_norm": 1.375,
      "learning_rate": 0.0001408052027080514,
      "loss": 4.6016,
      "mean_token_accuracy": 0.24933214485645294,
      "num_tokens": 157325004.0,
      "step": 90700
    },
    {
      "entropy": 5.354420566558838,
      "epoch": 7.057109511768139,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014078295721268433,
      "loss": 4.5241,
      "mean_token_accuracy": 0.24725250005722046,
      "num_tokens": 157333215.0,
      "step": 90705
    },
    {
      "entropy": 5.41779317855835,
      "epoch": 7.057498541139856,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001407607137538473,
      "loss": 4.6058,
      "mean_token_accuracy": 0.24671480655670167,
      "num_tokens": 157341614.0,
      "step": 90710
    },
    {
      "entropy": 5.341759252548218,
      "epoch": 7.057887570511574,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014073847233187775,
      "loss": 4.5626,
      "mean_token_accuracy": 0.24870250672101973,
      "num_tokens": 157349647.0,
      "step": 90715
    },
    {
      "entropy": 5.347904253005981,
      "epoch": 7.058276599883291,
      "grad_norm": 1.375,
      "learning_rate": 0.00014071623294711322,
      "loss": 4.6088,
      "mean_token_accuracy": 0.24590718746185303,
      "num_tokens": 157358184.0,
      "step": 90720
    },
    {
      "entropy": 5.376938247680664,
      "epoch": 7.0586656292550085,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001406939955998911,
      "loss": 4.6149,
      "mean_token_accuracy": 0.24486315101385117,
      "num_tokens": 157367492.0,
      "step": 90725
    },
    {
      "entropy": 5.401052188873291,
      "epoch": 7.059054658626726,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014067176029054884,
      "loss": 4.5648,
      "mean_token_accuracy": 0.25245814919471743,
      "num_tokens": 157375863.0,
      "step": 90730
    },
    {
      "entropy": 5.412055110931396,
      "epoch": 7.059443687998444,
      "grad_norm": 1.453125,
      "learning_rate": 0.00014064952701942375,
      "loss": 4.6516,
      "mean_token_accuracy": 0.24815927147865297,
      "num_tokens": 157384660.0,
      "step": 90735
    },
    {
      "entropy": 5.35265793800354,
      "epoch": 7.059832717370162,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014062729578685326,
      "loss": 4.6125,
      "mean_token_accuracy": 0.24236167967319489,
      "num_tokens": 157392855.0,
      "step": 90740
    },
    {
      "entropy": 5.343120431900024,
      "epoch": 7.060221746741879,
      "grad_norm": 1.421875,
      "learning_rate": 0.00014060506659317474,
      "loss": 4.5669,
      "mean_token_accuracy": 0.2514739900827408,
      "num_tokens": 157401316.0,
      "step": 90745
    },
    {
      "entropy": 5.334744310379028,
      "epoch": 7.060610776113597,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014058283943872536,
      "loss": 4.6255,
      "mean_token_accuracy": 0.24187921732664108,
      "num_tokens": 157410268.0,
      "step": 90750
    },
    {
      "entropy": 5.34069299697876,
      "epoch": 7.060999805485314,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014056061432384255,
      "loss": 4.5892,
      "mean_token_accuracy": 0.24535911977291108,
      "num_tokens": 157419203.0,
      "step": 90755
    },
    {
      "entropy": 5.448922634124756,
      "epoch": 7.0613888348570315,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001405383912488633,
      "loss": 4.6819,
      "mean_token_accuracy": 0.23608165979385376,
      "num_tokens": 157427740.0,
      "step": 90760
    },
    {
      "entropy": 5.363872814178467,
      "epoch": 7.061777864228749,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014051617021412506,
      "loss": 4.6351,
      "mean_token_accuracy": 0.24915624558925628,
      "num_tokens": 157436202.0,
      "step": 90765
    },
    {
      "entropy": 5.369014835357666,
      "epoch": 7.062166893600467,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014049395121996477,
      "loss": 4.5296,
      "mean_token_accuracy": 0.2525538370013237,
      "num_tokens": 157444549.0,
      "step": 90770
    },
    {
      "entropy": 5.328976058959961,
      "epoch": 7.062555922972185,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001404717342667198,
      "loss": 4.4992,
      "mean_token_accuracy": 0.25200038254261015,
      "num_tokens": 157452713.0,
      "step": 90775
    },
    {
      "entropy": 5.301155281066895,
      "epoch": 7.062944952343902,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014044951935472712,
      "loss": 4.5417,
      "mean_token_accuracy": 0.2477022811770439,
      "num_tokens": 157461463.0,
      "step": 90780
    },
    {
      "entropy": 5.34549036026001,
      "epoch": 7.063333981715619,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014042730648432375,
      "loss": 4.639,
      "mean_token_accuracy": 0.24238620549440384,
      "num_tokens": 157470188.0,
      "step": 90785
    },
    {
      "entropy": 5.4013200283050535,
      "epoch": 7.063723011087337,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001404050956558469,
      "loss": 4.598,
      "mean_token_accuracy": 0.23834707885980605,
      "num_tokens": 157478315.0,
      "step": 90790
    },
    {
      "entropy": 5.31847448348999,
      "epoch": 7.0641120404590545,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001403828868696334,
      "loss": 4.6124,
      "mean_token_accuracy": 0.23967136740684508,
      "num_tokens": 157486937.0,
      "step": 90795
    },
    {
      "entropy": 5.304755020141601,
      "epoch": 7.064501069830772,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001403606801260204,
      "loss": 4.5304,
      "mean_token_accuracy": 0.25286256670951845,
      "num_tokens": 157495676.0,
      "step": 90800
    },
    {
      "entropy": 5.367902612686157,
      "epoch": 7.06489009920249,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014033847542534473,
      "loss": 4.6019,
      "mean_token_accuracy": 0.24580440372228624,
      "num_tokens": 157504265.0,
      "step": 90805
    },
    {
      "entropy": 5.359374332427978,
      "epoch": 7.065279128574208,
      "grad_norm": 1.421875,
      "learning_rate": 0.00014031627276794333,
      "loss": 4.612,
      "mean_token_accuracy": 0.2446424424648285,
      "num_tokens": 157513173.0,
      "step": 90810
    },
    {
      "entropy": 5.372652006149292,
      "epoch": 7.065668157945925,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014029407215415319,
      "loss": 4.5803,
      "mean_token_accuracy": 0.24895976781845092,
      "num_tokens": 157521936.0,
      "step": 90815
    },
    {
      "entropy": 5.314547777175903,
      "epoch": 7.066057187317642,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014027187358431104,
      "loss": 4.5183,
      "mean_token_accuracy": 0.24832783937454223,
      "num_tokens": 157531231.0,
      "step": 90820
    },
    {
      "entropy": 5.411796760559082,
      "epoch": 7.06644621668936,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014024967705875382,
      "loss": 4.6157,
      "mean_token_accuracy": 0.24724265038967133,
      "num_tokens": 157539707.0,
      "step": 90825
    },
    {
      "entropy": 5.472136306762695,
      "epoch": 7.0668352460610775,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00014022748257781824,
      "loss": 4.6869,
      "mean_token_accuracy": 0.23652732819318772,
      "num_tokens": 157547362.0,
      "step": 90830
    },
    {
      "entropy": 5.348248720169067,
      "epoch": 7.067224275432795,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014020529014184115,
      "loss": 4.5868,
      "mean_token_accuracy": 0.24435074180364608,
      "num_tokens": 157556260.0,
      "step": 90835
    },
    {
      "entropy": 5.448028659820556,
      "epoch": 7.067613304804513,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014018309975115918,
      "loss": 4.6682,
      "mean_token_accuracy": 0.23848434835672377,
      "num_tokens": 157564698.0,
      "step": 90840
    },
    {
      "entropy": 5.351601791381836,
      "epoch": 7.068002334176231,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00014016091140610913,
      "loss": 4.5883,
      "mean_token_accuracy": 0.25004030764102936,
      "num_tokens": 157574544.0,
      "step": 90845
    },
    {
      "entropy": 5.304588747024536,
      "epoch": 7.068391363547948,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014013872510702758,
      "loss": 4.4942,
      "mean_token_accuracy": 0.2533778503537178,
      "num_tokens": 157583362.0,
      "step": 90850
    },
    {
      "entropy": 5.299684524536133,
      "epoch": 7.068780392919665,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001401165408542513,
      "loss": 4.4899,
      "mean_token_accuracy": 0.2556603580713272,
      "num_tokens": 157592032.0,
      "step": 90855
    },
    {
      "entropy": 5.339626884460449,
      "epoch": 7.069169422291383,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001400943586481168,
      "loss": 4.5113,
      "mean_token_accuracy": 0.25643721520900725,
      "num_tokens": 157600396.0,
      "step": 90860
    },
    {
      "entropy": 5.399127101898193,
      "epoch": 7.0695584516631005,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001400721784889606,
      "loss": 4.6481,
      "mean_token_accuracy": 0.23925201296806337,
      "num_tokens": 157608919.0,
      "step": 90865
    },
    {
      "entropy": 5.393349361419678,
      "epoch": 7.069947481034818,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014005000037711936,
      "loss": 4.6664,
      "mean_token_accuracy": 0.24181021600961686,
      "num_tokens": 157617702.0,
      "step": 90870
    },
    {
      "entropy": 5.359745597839355,
      "epoch": 7.070336510406536,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014002782431292955,
      "loss": 4.5826,
      "mean_token_accuracy": 0.2516361206769943,
      "num_tokens": 157625999.0,
      "step": 90875
    },
    {
      "entropy": 5.430653762817383,
      "epoch": 7.070725539778254,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014000565029672777,
      "loss": 4.6761,
      "mean_token_accuracy": 0.23965599685907363,
      "num_tokens": 157635231.0,
      "step": 90880
    },
    {
      "entropy": 5.39701018333435,
      "epoch": 7.07111456914997,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001399834783288504,
      "loss": 4.6014,
      "mean_token_accuracy": 0.2444734051823616,
      "num_tokens": 157643763.0,
      "step": 90885
    },
    {
      "entropy": 5.382362794876099,
      "epoch": 7.071503598521688,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001399613084096337,
      "loss": 4.6297,
      "mean_token_accuracy": 0.24666413813829421,
      "num_tokens": 157652464.0,
      "step": 90890
    },
    {
      "entropy": 5.323703956604004,
      "epoch": 7.071892627893406,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013993914053941434,
      "loss": 4.4973,
      "mean_token_accuracy": 0.24931247383356095,
      "num_tokens": 157661399.0,
      "step": 90895
    },
    {
      "entropy": 5.414973878860474,
      "epoch": 7.0722816572651235,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00013991697471852843,
      "loss": 4.6307,
      "mean_token_accuracy": 0.2414214327931404,
      "num_tokens": 157670423.0,
      "step": 90900
    },
    {
      "entropy": 5.420285224914551,
      "epoch": 7.072670686636841,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013989481094731256,
      "loss": 4.6199,
      "mean_token_accuracy": 0.24389638304710387,
      "num_tokens": 157678518.0,
      "step": 90905
    },
    {
      "entropy": 5.299556016921997,
      "epoch": 7.073059716008559,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013987264922610276,
      "loss": 4.5043,
      "mean_token_accuracy": 0.2566379323601723,
      "num_tokens": 157687288.0,
      "step": 90910
    },
    {
      "entropy": 5.3662279605865475,
      "epoch": 7.073448745380277,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013985048955523555,
      "loss": 4.6316,
      "mean_token_accuracy": 0.2530925542116165,
      "num_tokens": 157695772.0,
      "step": 90915
    },
    {
      "entropy": 5.382780313491821,
      "epoch": 7.073837774751993,
      "grad_norm": 1.296875,
      "learning_rate": 0.00013982833193504696,
      "loss": 4.6543,
      "mean_token_accuracy": 0.24570504277944566,
      "num_tokens": 157703973.0,
      "step": 90920
    },
    {
      "entropy": 5.351260232925415,
      "epoch": 7.074226804123711,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013980617636587334,
      "loss": 4.5744,
      "mean_token_accuracy": 0.24799355119466782,
      "num_tokens": 157712846.0,
      "step": 90925
    },
    {
      "entropy": 5.386017370223999,
      "epoch": 7.074615833495429,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013978402284805076,
      "loss": 4.5913,
      "mean_token_accuracy": 0.24599877446889878,
      "num_tokens": 157721675.0,
      "step": 90930
    },
    {
      "entropy": 5.395230770111084,
      "epoch": 7.0750048628671465,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013976187138191548,
      "loss": 4.5443,
      "mean_token_accuracy": 0.25046300888061523,
      "num_tokens": 157729544.0,
      "step": 90935
    },
    {
      "entropy": 5.333476448059082,
      "epoch": 7.075393892238864,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013973972196780346,
      "loss": 4.5381,
      "mean_token_accuracy": 0.2538297578692436,
      "num_tokens": 157738370.0,
      "step": 90940
    },
    {
      "entropy": 5.401645278930664,
      "epoch": 7.075782921610582,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001397175746060509,
      "loss": 4.5598,
      "mean_token_accuracy": 0.25369314551353456,
      "num_tokens": 157746866.0,
      "step": 90945
    },
    {
      "entropy": 5.326844263076782,
      "epoch": 7.0761719509823,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00013969542929699385,
      "loss": 4.5081,
      "mean_token_accuracy": 0.2485993728041649,
      "num_tokens": 157755213.0,
      "step": 90950
    },
    {
      "entropy": 5.293819856643677,
      "epoch": 7.076560980354016,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001396732860409683,
      "loss": 4.5572,
      "mean_token_accuracy": 0.2462384894490242,
      "num_tokens": 157763740.0,
      "step": 90955
    },
    {
      "entropy": 5.33265962600708,
      "epoch": 7.076950009725734,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00013965114483831026,
      "loss": 4.5238,
      "mean_token_accuracy": 0.24613946229219436,
      "num_tokens": 157772412.0,
      "step": 90960
    },
    {
      "entropy": 5.380666971206665,
      "epoch": 7.077339039097452,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013962900568935567,
      "loss": 4.6098,
      "mean_token_accuracy": 0.2461026653647423,
      "num_tokens": 157780839.0,
      "step": 90965
    },
    {
      "entropy": 5.437144374847412,
      "epoch": 7.0777280684691695,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001396068685944404,
      "loss": 4.6763,
      "mean_token_accuracy": 0.23926704823970796,
      "num_tokens": 157789779.0,
      "step": 90970
    },
    {
      "entropy": 5.327627801895142,
      "epoch": 7.078117097840887,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013958473355390045,
      "loss": 4.4659,
      "mean_token_accuracy": 0.2506730929017067,
      "num_tokens": 157797771.0,
      "step": 90975
    },
    {
      "entropy": 5.283209800720215,
      "epoch": 7.078506127212605,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013956260056807157,
      "loss": 4.4529,
      "mean_token_accuracy": 0.2536663100123405,
      "num_tokens": 157806524.0,
      "step": 90980
    },
    {
      "entropy": 5.345575618743896,
      "epoch": 7.078895156584322,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00013954046963728973,
      "loss": 4.5646,
      "mean_token_accuracy": 0.24884819388389587,
      "num_tokens": 157815647.0,
      "step": 90985
    },
    {
      "entropy": 5.376604890823364,
      "epoch": 7.079284185956039,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013951834076189056,
      "loss": 4.5858,
      "mean_token_accuracy": 0.25276903361082076,
      "num_tokens": 157823711.0,
      "step": 90990
    },
    {
      "entropy": 5.448965311050415,
      "epoch": 7.079673215327757,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013949621394221002,
      "loss": 4.7022,
      "mean_token_accuracy": 0.23551566898822784,
      "num_tokens": 157831651.0,
      "step": 90995
    },
    {
      "entropy": 5.275083541870117,
      "epoch": 7.080062244699475,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001394740891785837,
      "loss": 4.4931,
      "mean_token_accuracy": 0.2570167914032936,
      "num_tokens": 157840618.0,
      "step": 91000
    },
    {
      "entropy": 5.373444271087647,
      "epoch": 7.0804512740711925,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013945196647134733,
      "loss": 4.6441,
      "mean_token_accuracy": 0.24536640346050262,
      "num_tokens": 157849964.0,
      "step": 91005
    },
    {
      "entropy": 5.343231534957885,
      "epoch": 7.08084030344291,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00013942984582083673,
      "loss": 4.5104,
      "mean_token_accuracy": 0.24795818328857422,
      "num_tokens": 157858626.0,
      "step": 91010
    },
    {
      "entropy": 5.362875127792359,
      "epoch": 7.081229332814628,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013940772722738736,
      "loss": 4.5692,
      "mean_token_accuracy": 0.24352590441703797,
      "num_tokens": 157866917.0,
      "step": 91015
    },
    {
      "entropy": 5.361588382720948,
      "epoch": 7.081618362186345,
      "grad_norm": 1.296875,
      "learning_rate": 0.00013938561069133497,
      "loss": 4.5879,
      "mean_token_accuracy": 0.2537261828780174,
      "num_tokens": 157875478.0,
      "step": 91020
    },
    {
      "entropy": 5.3179536819458,
      "epoch": 7.082007391558062,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013936349621301503,
      "loss": 4.4982,
      "mean_token_accuracy": 0.25127245783805846,
      "num_tokens": 157883343.0,
      "step": 91025
    },
    {
      "entropy": 5.382967805862426,
      "epoch": 7.08239642092978,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001393413837927632,
      "loss": 4.5887,
      "mean_token_accuracy": 0.24440044015645981,
      "num_tokens": 157891396.0,
      "step": 91030
    },
    {
      "entropy": 5.400006914138794,
      "epoch": 7.082785450301498,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013931927343091487,
      "loss": 4.5948,
      "mean_token_accuracy": 0.24150725305080414,
      "num_tokens": 157899806.0,
      "step": 91035
    },
    {
      "entropy": 5.370210886001587,
      "epoch": 7.0831744796732155,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00013929716512780572,
      "loss": 4.6198,
      "mean_token_accuracy": 0.2406170889735222,
      "num_tokens": 157908272.0,
      "step": 91040
    },
    {
      "entropy": 5.320813512802124,
      "epoch": 7.083563509044933,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013927505888377094,
      "loss": 4.5304,
      "mean_token_accuracy": 0.25785660147666933,
      "num_tokens": 157916875.0,
      "step": 91045
    },
    {
      "entropy": 5.334602308273316,
      "epoch": 7.083952538416651,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001392529546991463,
      "loss": 4.6002,
      "mean_token_accuracy": 0.24974137246608735,
      "num_tokens": 157925544.0,
      "step": 91050
    },
    {
      "entropy": 5.367081546783448,
      "epoch": 7.084341567788368,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013923085257426694,
      "loss": 4.6289,
      "mean_token_accuracy": 0.23865650296211244,
      "num_tokens": 157934321.0,
      "step": 91055
    },
    {
      "entropy": 5.413476276397705,
      "epoch": 7.084730597160085,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013920875250946822,
      "loss": 4.6,
      "mean_token_accuracy": 0.24694847464561462,
      "num_tokens": 157942853.0,
      "step": 91060
    },
    {
      "entropy": 5.372056674957276,
      "epoch": 7.085119626531803,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013918665450508565,
      "loss": 4.5866,
      "mean_token_accuracy": 0.2534540370106697,
      "num_tokens": 157950790.0,
      "step": 91065
    },
    {
      "entropy": 5.429498624801636,
      "epoch": 7.085508655903521,
      "grad_norm": 1.296875,
      "learning_rate": 0.00013916455856145425,
      "loss": 4.6538,
      "mean_token_accuracy": 0.23976635038852692,
      "num_tokens": 157960077.0,
      "step": 91070
    },
    {
      "entropy": 5.4246203899383545,
      "epoch": 7.0858976852752384,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001391424646789096,
      "loss": 4.6497,
      "mean_token_accuracy": 0.2380112275481224,
      "num_tokens": 157968723.0,
      "step": 91075
    },
    {
      "entropy": 5.362253570556641,
      "epoch": 7.086286714646956,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013912037285778686,
      "loss": 4.5361,
      "mean_token_accuracy": 0.25497488528490064,
      "num_tokens": 157977284.0,
      "step": 91080
    },
    {
      "entropy": 5.423880481719971,
      "epoch": 7.086675744018674,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001390982830984211,
      "loss": 4.5923,
      "mean_token_accuracy": 0.24881599843502045,
      "num_tokens": 157985864.0,
      "step": 91085
    },
    {
      "entropy": 5.370779418945313,
      "epoch": 7.087064773390391,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013907619540114762,
      "loss": 4.5236,
      "mean_token_accuracy": 0.24591085612773894,
      "num_tokens": 157994882.0,
      "step": 91090
    },
    {
      "entropy": 5.4062299728393555,
      "epoch": 7.087453802762108,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013905410976630152,
      "loss": 4.6481,
      "mean_token_accuracy": 0.23440274298191072,
      "num_tokens": 158004121.0,
      "step": 91095
    },
    {
      "entropy": 5.329976749420166,
      "epoch": 7.087842832133826,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013903202619421794,
      "loss": 4.5364,
      "mean_token_accuracy": 0.25016292184591293,
      "num_tokens": 158011932.0,
      "step": 91100
    },
    {
      "entropy": 5.311750173568726,
      "epoch": 7.088231861505544,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013900994468523192,
      "loss": 4.5161,
      "mean_token_accuracy": 0.2546752244234085,
      "num_tokens": 158020343.0,
      "step": 91105
    },
    {
      "entropy": 5.373808908462524,
      "epoch": 7.088620890877261,
      "grad_norm": 1.296875,
      "learning_rate": 0.00013898786523967857,
      "loss": 4.5615,
      "mean_token_accuracy": 0.25393116623163225,
      "num_tokens": 158029022.0,
      "step": 91110
    },
    {
      "entropy": 5.436993026733399,
      "epoch": 7.089009920248979,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001389657878578928,
      "loss": 4.6554,
      "mean_token_accuracy": 0.24412131160497666,
      "num_tokens": 158037489.0,
      "step": 91115
    },
    {
      "entropy": 5.363737678527832,
      "epoch": 7.089398949620696,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013894371254020982,
      "loss": 4.5472,
      "mean_token_accuracy": 0.2398570418357849,
      "num_tokens": 158046394.0,
      "step": 91120
    },
    {
      "entropy": 5.353598594665527,
      "epoch": 7.089787978992414,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013892163928696434,
      "loss": 4.5176,
      "mean_token_accuracy": 0.2509632080793381,
      "num_tokens": 158054256.0,
      "step": 91125
    },
    {
      "entropy": 5.365849447250366,
      "epoch": 7.090177008364131,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013889956809849145,
      "loss": 4.5667,
      "mean_token_accuracy": 0.2477084994316101,
      "num_tokens": 158062819.0,
      "step": 91130
    },
    {
      "entropy": 5.402872514724732,
      "epoch": 7.090566037735849,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000138877498975126,
      "loss": 4.6886,
      "mean_token_accuracy": 0.23419993221759797,
      "num_tokens": 158071393.0,
      "step": 91135
    },
    {
      "entropy": 5.4116607189178465,
      "epoch": 7.090955067107567,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013885543191720267,
      "loss": 4.6066,
      "mean_token_accuracy": 0.2406143456697464,
      "num_tokens": 158080194.0,
      "step": 91140
    },
    {
      "entropy": 5.407731533050537,
      "epoch": 7.091344096479284,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013883336692505663,
      "loss": 4.6012,
      "mean_token_accuracy": 0.24365233182907103,
      "num_tokens": 158088461.0,
      "step": 91145
    },
    {
      "entropy": 5.429109668731689,
      "epoch": 7.091733125851002,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013881130399902246,
      "loss": 4.6609,
      "mean_token_accuracy": 0.23886772841215134,
      "num_tokens": 158097077.0,
      "step": 91150
    },
    {
      "entropy": 5.264361381530762,
      "epoch": 7.092122155222719,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00013878924313943502,
      "loss": 4.4757,
      "mean_token_accuracy": 0.25841326266527176,
      "num_tokens": 158105875.0,
      "step": 91155
    },
    {
      "entropy": 5.401818084716797,
      "epoch": 7.092511184594437,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013876718434662906,
      "loss": 4.5882,
      "mean_token_accuracy": 0.25150246322155,
      "num_tokens": 158114187.0,
      "step": 91160
    },
    {
      "entropy": 5.382562255859375,
      "epoch": 7.092900213966154,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013874512762093914,
      "loss": 4.6235,
      "mean_token_accuracy": 0.24551106244325638,
      "num_tokens": 158122827.0,
      "step": 91165
    },
    {
      "entropy": 5.313882112503052,
      "epoch": 7.093289243337872,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013872307296270008,
      "loss": 4.5931,
      "mean_token_accuracy": 0.24442370384931564,
      "num_tokens": 158131210.0,
      "step": 91170
    },
    {
      "entropy": 5.292608165740967,
      "epoch": 7.09367827270959,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013870102037224645,
      "loss": 4.5285,
      "mean_token_accuracy": 0.24674749076366426,
      "num_tokens": 158140190.0,
      "step": 91175
    },
    {
      "entropy": 5.401205348968506,
      "epoch": 7.094067302081307,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013867896984991296,
      "loss": 4.6677,
      "mean_token_accuracy": 0.23697558045387268,
      "num_tokens": 158149409.0,
      "step": 91180
    },
    {
      "entropy": 5.377355241775513,
      "epoch": 7.094456331453025,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013865692139603404,
      "loss": 4.5503,
      "mean_token_accuracy": 0.24438160061836242,
      "num_tokens": 158158546.0,
      "step": 91185
    },
    {
      "entropy": 5.431010484695435,
      "epoch": 7.094845360824742,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001386348750109444,
      "loss": 4.6935,
      "mean_token_accuracy": 0.23980841040611267,
      "num_tokens": 158166872.0,
      "step": 91190
    },
    {
      "entropy": 5.331302070617676,
      "epoch": 7.09523439019646,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013861283069497844,
      "loss": 4.5399,
      "mean_token_accuracy": 0.2549004480242729,
      "num_tokens": 158175723.0,
      "step": 91195
    },
    {
      "entropy": 5.346026134490967,
      "epoch": 7.095623419568177,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013859078844847072,
      "loss": 4.5321,
      "mean_token_accuracy": 0.2528258889913559,
      "num_tokens": 158184378.0,
      "step": 91200
    },
    {
      "entropy": 5.354621124267578,
      "epoch": 7.096012448939895,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013856874827175564,
      "loss": 4.6155,
      "mean_token_accuracy": 0.23974245339632033,
      "num_tokens": 158193217.0,
      "step": 91205
    },
    {
      "entropy": 5.357074022293091,
      "epoch": 7.096401478311613,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001385467101651676,
      "loss": 4.6037,
      "mean_token_accuracy": 0.2478442519903183,
      "num_tokens": 158201761.0,
      "step": 91210
    },
    {
      "entropy": 5.378391170501709,
      "epoch": 7.09679050768333,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001385246741290412,
      "loss": 4.6317,
      "mean_token_accuracy": 0.24663520604372025,
      "num_tokens": 158211560.0,
      "step": 91215
    },
    {
      "entropy": 5.365364265441895,
      "epoch": 7.097179537055047,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001385026401637105,
      "loss": 4.613,
      "mean_token_accuracy": 0.2455908626317978,
      "num_tokens": 158220322.0,
      "step": 91220
    },
    {
      "entropy": 5.432413530349732,
      "epoch": 7.097568566426765,
      "grad_norm": 1.296875,
      "learning_rate": 0.00013848060826951005,
      "loss": 4.7277,
      "mean_token_accuracy": 0.2313808724284172,
      "num_tokens": 158229311.0,
      "step": 91225
    },
    {
      "entropy": 5.356674242019653,
      "epoch": 7.097957595798483,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013845857844677404,
      "loss": 4.489,
      "mean_token_accuracy": 0.25431526005268096,
      "num_tokens": 158238221.0,
      "step": 91230
    },
    {
      "entropy": 5.407193040847778,
      "epoch": 7.0983466251702,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013843655069583683,
      "loss": 4.6397,
      "mean_token_accuracy": 0.24243959188461303,
      "num_tokens": 158246802.0,
      "step": 91235
    },
    {
      "entropy": 5.388497257232666,
      "epoch": 7.098735654541918,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013841452501703262,
      "loss": 4.5886,
      "mean_token_accuracy": 0.24270775616168977,
      "num_tokens": 158255801.0,
      "step": 91240
    },
    {
      "entropy": 5.263103580474853,
      "epoch": 7.099124683913636,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013839250141069547,
      "loss": 4.4746,
      "mean_token_accuracy": 0.25278493762016296,
      "num_tokens": 158263797.0,
      "step": 91245
    },
    {
      "entropy": 5.33375358581543,
      "epoch": 7.099513713285353,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001383704798771598,
      "loss": 4.5627,
      "mean_token_accuracy": 0.24677790105342864,
      "num_tokens": 158271996.0,
      "step": 91250
    },
    {
      "entropy": 5.275630187988281,
      "epoch": 7.09990274265707,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013834846041675957,
      "loss": 4.494,
      "mean_token_accuracy": 0.25560490489006044,
      "num_tokens": 158279985.0,
      "step": 91255
    },
    {
      "entropy": 5.346503877639771,
      "epoch": 7.100291772028788,
      "grad_norm": 1.40625,
      "learning_rate": 0.000138326443029829,
      "loss": 4.549,
      "mean_token_accuracy": 0.24689377397298812,
      "num_tokens": 158287976.0,
      "step": 91260
    },
    {
      "entropy": 5.3862621784210205,
      "epoch": 7.100680801400506,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000138304427716702,
      "loss": 4.6337,
      "mean_token_accuracy": 0.2475188195705414,
      "num_tokens": 158297005.0,
      "step": 91265
    },
    {
      "entropy": 5.316065502166748,
      "epoch": 7.101069830772223,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00013828241447771277,
      "loss": 4.5152,
      "mean_token_accuracy": 0.24952392280101776,
      "num_tokens": 158305633.0,
      "step": 91270
    },
    {
      "entropy": 5.373355054855347,
      "epoch": 7.101458860143941,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001382604033131954,
      "loss": 4.5391,
      "mean_token_accuracy": 0.24722970873117447,
      "num_tokens": 158313387.0,
      "step": 91275
    },
    {
      "entropy": 5.219532251358032,
      "epoch": 7.101847889515659,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013823839422348364,
      "loss": 4.4213,
      "mean_token_accuracy": 0.26534271240234375,
      "num_tokens": 158322074.0,
      "step": 91280
    },
    {
      "entropy": 5.426431512832641,
      "epoch": 7.102236918887376,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013821638720891167,
      "loss": 4.6743,
      "mean_token_accuracy": 0.2369215339422226,
      "num_tokens": 158330878.0,
      "step": 91285
    },
    {
      "entropy": 5.32231125831604,
      "epoch": 7.102625948259093,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00013819438226981323,
      "loss": 4.5401,
      "mean_token_accuracy": 0.24657740890979768,
      "num_tokens": 158339537.0,
      "step": 91290
    },
    {
      "entropy": 5.368477535247803,
      "epoch": 7.103014977630811,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013817237940652239,
      "loss": 4.6219,
      "mean_token_accuracy": 0.24428412169218064,
      "num_tokens": 158348294.0,
      "step": 91295
    },
    {
      "entropy": 5.292247867584228,
      "epoch": 7.103404007002529,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001381503786193728,
      "loss": 4.525,
      "mean_token_accuracy": 0.24932461529970168,
      "num_tokens": 158356760.0,
      "step": 91300
    },
    {
      "entropy": 5.3920176982879635,
      "epoch": 7.103793036374246,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013812837990869847,
      "loss": 4.6449,
      "mean_token_accuracy": 0.24199347645044328,
      "num_tokens": 158365045.0,
      "step": 91305
    },
    {
      "entropy": 5.384446907043457,
      "epoch": 7.104182065745964,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001381063832748331,
      "loss": 4.602,
      "mean_token_accuracy": 0.24329760372638704,
      "num_tokens": 158373489.0,
      "step": 91310
    },
    {
      "entropy": 5.466401481628418,
      "epoch": 7.104571095117682,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013808438871811053,
      "loss": 4.6776,
      "mean_token_accuracy": 0.24135067015886308,
      "num_tokens": 158382614.0,
      "step": 91315
    },
    {
      "entropy": 5.39192943572998,
      "epoch": 7.1049601244893985,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013806239623886439,
      "loss": 4.5621,
      "mean_token_accuracy": 0.24907763600349425,
      "num_tokens": 158390673.0,
      "step": 91320
    },
    {
      "entropy": 5.412025833129883,
      "epoch": 7.105349153861116,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001380404058374284,
      "loss": 4.5675,
      "mean_token_accuracy": 0.2458203613758087,
      "num_tokens": 158399179.0,
      "step": 91325
    },
    {
      "entropy": 5.346821880340576,
      "epoch": 7.105738183232834,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001380184175141363,
      "loss": 4.5364,
      "mean_token_accuracy": 0.25395656526088717,
      "num_tokens": 158407533.0,
      "step": 91330
    },
    {
      "entropy": 5.403315591812134,
      "epoch": 7.106127212604552,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013799643126932154,
      "loss": 4.6183,
      "mean_token_accuracy": 0.24637273848056793,
      "num_tokens": 158416744.0,
      "step": 91335
    },
    {
      "entropy": 5.369313287734985,
      "epoch": 7.106516241976269,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013797444710331804,
      "loss": 4.5387,
      "mean_token_accuracy": 0.25140297710895537,
      "num_tokens": 158425237.0,
      "step": 91340
    },
    {
      "entropy": 5.369378852844238,
      "epoch": 7.106905271347987,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001379524650164591,
      "loss": 4.6334,
      "mean_token_accuracy": 0.23980889916419984,
      "num_tokens": 158434153.0,
      "step": 91345
    },
    {
      "entropy": 5.298630237579346,
      "epoch": 7.107294300719705,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001379304850090785,
      "loss": 4.5227,
      "mean_token_accuracy": 0.24942188560962678,
      "num_tokens": 158442565.0,
      "step": 91350
    },
    {
      "entropy": 5.405721282958984,
      "epoch": 7.1076833300914215,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001379085070815096,
      "loss": 4.6197,
      "mean_token_accuracy": 0.2447480082511902,
      "num_tokens": 158451987.0,
      "step": 91355
    },
    {
      "entropy": 5.3451145648956295,
      "epoch": 7.108072359463139,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001378865312340858,
      "loss": 4.5684,
      "mean_token_accuracy": 0.25439158231019976,
      "num_tokens": 158460339.0,
      "step": 91360
    },
    {
      "entropy": 5.398428153991699,
      "epoch": 7.108461388834857,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013786455746714072,
      "loss": 4.6179,
      "mean_token_accuracy": 0.2410676434636116,
      "num_tokens": 158469276.0,
      "step": 91365
    },
    {
      "entropy": 5.416480302810669,
      "epoch": 7.108850418206575,
      "grad_norm": 1.375,
      "learning_rate": 0.00013784258578100766,
      "loss": 4.5965,
      "mean_token_accuracy": 0.24950332194566727,
      "num_tokens": 158478354.0,
      "step": 91370
    },
    {
      "entropy": 5.413027238845825,
      "epoch": 7.109239447578292,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013782061617602012,
      "loss": 4.6642,
      "mean_token_accuracy": 0.24277035146951675,
      "num_tokens": 158487731.0,
      "step": 91375
    },
    {
      "entropy": 5.391470241546631,
      "epoch": 7.10962847695001,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001377986486525113,
      "loss": 4.5421,
      "mean_token_accuracy": 0.24340494871139526,
      "num_tokens": 158495645.0,
      "step": 91380
    },
    {
      "entropy": 5.33862042427063,
      "epoch": 7.110017506321728,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001377766832108147,
      "loss": 4.5276,
      "mean_token_accuracy": 0.25511841773986815,
      "num_tokens": 158504226.0,
      "step": 91385
    },
    {
      "entropy": 5.465092420578003,
      "epoch": 7.1104065356934445,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00013775471985126342,
      "loss": 4.7213,
      "mean_token_accuracy": 0.24592090994119645,
      "num_tokens": 158513108.0,
      "step": 91390
    },
    {
      "entropy": 5.351700782775879,
      "epoch": 7.110795565065162,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013773275857419093,
      "loss": 4.6539,
      "mean_token_accuracy": 0.23916025161743165,
      "num_tokens": 158522295.0,
      "step": 91395
    },
    {
      "entropy": 5.357116413116455,
      "epoch": 7.11118459443688,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001377107993799303,
      "loss": 4.57,
      "mean_token_accuracy": 0.2410861685872078,
      "num_tokens": 158530657.0,
      "step": 91400
    },
    {
      "entropy": 5.348486137390137,
      "epoch": 7.111573623808598,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013768884226881468,
      "loss": 4.5198,
      "mean_token_accuracy": 0.25296615064144135,
      "num_tokens": 158539775.0,
      "step": 91405
    },
    {
      "entropy": 5.393873453140259,
      "epoch": 7.111962653180315,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013766688724117744,
      "loss": 4.6084,
      "mean_token_accuracy": 0.23653607666492463,
      "num_tokens": 158549151.0,
      "step": 91410
    },
    {
      "entropy": 5.291371583938599,
      "epoch": 7.112351682552033,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00013764493429735153,
      "loss": 4.5447,
      "mean_token_accuracy": 0.24893860071897506,
      "num_tokens": 158558172.0,
      "step": 91415
    },
    {
      "entropy": 5.414524126052856,
      "epoch": 7.112740711923751,
      "grad_norm": 1.5625,
      "learning_rate": 0.00013762298343767023,
      "loss": 4.5976,
      "mean_token_accuracy": 0.2455281913280487,
      "num_tokens": 158566287.0,
      "step": 91420
    },
    {
      "entropy": 5.342790508270264,
      "epoch": 7.1131297412954675,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001376010346624663,
      "loss": 4.5625,
      "mean_token_accuracy": 0.24985070377588273,
      "num_tokens": 158574735.0,
      "step": 91425
    },
    {
      "entropy": 5.438156366348267,
      "epoch": 7.113518770667185,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013757908797207315,
      "loss": 4.6544,
      "mean_token_accuracy": 0.23579265773296357,
      "num_tokens": 158583659.0,
      "step": 91430
    },
    {
      "entropy": 5.377078199386597,
      "epoch": 7.113907800038903,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013755714336682363,
      "loss": 4.6159,
      "mean_token_accuracy": 0.24858437478542328,
      "num_tokens": 158592449.0,
      "step": 91435
    },
    {
      "entropy": 5.319667863845825,
      "epoch": 7.114296829410621,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013753520084705057,
      "loss": 4.5548,
      "mean_token_accuracy": 0.25604947209358214,
      "num_tokens": 158601153.0,
      "step": 91440
    },
    {
      "entropy": 5.443948078155517,
      "epoch": 7.114685858782338,
      "grad_norm": 1.484375,
      "learning_rate": 0.00013751326041308714,
      "loss": 4.6312,
      "mean_token_accuracy": 0.24449303150177001,
      "num_tokens": 158609137.0,
      "step": 91445
    },
    {
      "entropy": 5.423870086669922,
      "epoch": 7.115074888154056,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013749132206526606,
      "loss": 4.6571,
      "mean_token_accuracy": 0.24130951017141342,
      "num_tokens": 158618392.0,
      "step": 91450
    },
    {
      "entropy": 5.3648756980896,
      "epoch": 7.115463917525773,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013746938580392037,
      "loss": 4.646,
      "mean_token_accuracy": 0.24565006792545319,
      "num_tokens": 158626991.0,
      "step": 91455
    },
    {
      "entropy": 5.351207828521728,
      "epoch": 7.1158529468974905,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001374474516293828,
      "loss": 4.5491,
      "mean_token_accuracy": 0.25301655381917953,
      "num_tokens": 158635557.0,
      "step": 91460
    },
    {
      "entropy": 5.352166175842285,
      "epoch": 7.116241976269208,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00013742551954198628,
      "loss": 4.5383,
      "mean_token_accuracy": 0.25258086919784545,
      "num_tokens": 158643460.0,
      "step": 91465
    },
    {
      "entropy": 5.3396888256073,
      "epoch": 7.116631005640926,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013740358954206344,
      "loss": 4.6013,
      "mean_token_accuracy": 0.24118566513061523,
      "num_tokens": 158652349.0,
      "step": 91470
    },
    {
      "entropy": 5.38053412437439,
      "epoch": 7.117020035012644,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013738166162994715,
      "loss": 4.5955,
      "mean_token_accuracy": 0.25439736992120743,
      "num_tokens": 158661153.0,
      "step": 91475
    },
    {
      "entropy": 5.4333079814910885,
      "epoch": 7.117409064384361,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013735973580597017,
      "loss": 4.6088,
      "mean_token_accuracy": 0.24635385423898698,
      "num_tokens": 158669310.0,
      "step": 91480
    },
    {
      "entropy": 5.365189456939698,
      "epoch": 7.117798093756079,
      "grad_norm": 1.40625,
      "learning_rate": 0.000137337812070465,
      "loss": 4.6233,
      "mean_token_accuracy": 0.24315168112516403,
      "num_tokens": 158677541.0,
      "step": 91485
    },
    {
      "entropy": 5.419115257263184,
      "epoch": 7.118187123127796,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013731589042376453,
      "loss": 4.6476,
      "mean_token_accuracy": 0.24928930699825286,
      "num_tokens": 158686505.0,
      "step": 91490
    },
    {
      "entropy": 5.44438943862915,
      "epoch": 7.1185761524995135,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001372939708662012,
      "loss": 4.6878,
      "mean_token_accuracy": 0.23970138728618623,
      "num_tokens": 158694681.0,
      "step": 91495
    },
    {
      "entropy": 5.279157209396362,
      "epoch": 7.118965181871231,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013727205339810776,
      "loss": 4.5316,
      "mean_token_accuracy": 0.24674390405416488,
      "num_tokens": 158703112.0,
      "step": 91500
    },
    {
      "entropy": 5.387183809280396,
      "epoch": 7.119354211242949,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001372501380198166,
      "loss": 4.6036,
      "mean_token_accuracy": 0.24845709055662155,
      "num_tokens": 158712015.0,
      "step": 91505
    },
    {
      "entropy": 5.4620733737945555,
      "epoch": 7.119743240614667,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013722822473166043,
      "loss": 4.6754,
      "mean_token_accuracy": 0.23750420957803725,
      "num_tokens": 158720211.0,
      "step": 91510
    },
    {
      "entropy": 5.40203628540039,
      "epoch": 7.120132269986384,
      "grad_norm": 1.46875,
      "learning_rate": 0.00013720631353397167,
      "loss": 4.611,
      "mean_token_accuracy": 0.2411225214600563,
      "num_tokens": 158728123.0,
      "step": 91515
    },
    {
      "entropy": 5.353665494918824,
      "epoch": 7.120521299358102,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013718440442708274,
      "loss": 4.5523,
      "mean_token_accuracy": 0.25296953320503235,
      "num_tokens": 158736046.0,
      "step": 91520
    },
    {
      "entropy": 5.392257833480835,
      "epoch": 7.120910328729819,
      "grad_norm": 1.265625,
      "learning_rate": 0.00013716249741132613,
      "loss": 4.6288,
      "mean_token_accuracy": 0.24305288642644882,
      "num_tokens": 158744341.0,
      "step": 91525
    },
    {
      "entropy": 5.368483257293701,
      "epoch": 7.1212993581015365,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013714059248703418,
      "loss": 4.5889,
      "mean_token_accuracy": 0.25273632258176804,
      "num_tokens": 158751930.0,
      "step": 91530
    },
    {
      "entropy": 5.388767099380493,
      "epoch": 7.121688387473254,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013711868965453933,
      "loss": 4.6264,
      "mean_token_accuracy": 0.2471355140209198,
      "num_tokens": 158761718.0,
      "step": 91535
    },
    {
      "entropy": 5.367141008377075,
      "epoch": 7.122077416844972,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013709678891417394,
      "loss": 4.5957,
      "mean_token_accuracy": 0.24287642538547516,
      "num_tokens": 158770595.0,
      "step": 91540
    },
    {
      "entropy": 5.358305263519287,
      "epoch": 7.12246644621669,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00013707489026627023,
      "loss": 4.5563,
      "mean_token_accuracy": 0.24979768544435502,
      "num_tokens": 158779489.0,
      "step": 91545
    },
    {
      "entropy": 5.372180604934693,
      "epoch": 7.122855475588407,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013705299371116061,
      "loss": 4.6096,
      "mean_token_accuracy": 0.24220267087221145,
      "num_tokens": 158788610.0,
      "step": 91550
    },
    {
      "entropy": 5.468094205856323,
      "epoch": 7.123244504960124,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013703109924917718,
      "loss": 4.6774,
      "mean_token_accuracy": 0.24878009557723998,
      "num_tokens": 158796798.0,
      "step": 91555
    },
    {
      "entropy": 5.333374643325806,
      "epoch": 7.123633534331842,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013700920688065226,
      "loss": 4.6124,
      "mean_token_accuracy": 0.2475343719124794,
      "num_tokens": 158805210.0,
      "step": 91560
    },
    {
      "entropy": 5.316061210632324,
      "epoch": 7.1240225637035595,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013698731660591795,
      "loss": 4.5105,
      "mean_token_accuracy": 0.2544756501913071,
      "num_tokens": 158813506.0,
      "step": 91565
    },
    {
      "entropy": 5.341976356506348,
      "epoch": 7.124411593075277,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001369654284253065,
      "loss": 4.5515,
      "mean_token_accuracy": 0.24659589678049088,
      "num_tokens": 158822004.0,
      "step": 91570
    },
    {
      "entropy": 5.296399259567261,
      "epoch": 7.124800622446995,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013694354233914987,
      "loss": 4.5459,
      "mean_token_accuracy": 0.25388205200433733,
      "num_tokens": 158831020.0,
      "step": 91575
    },
    {
      "entropy": 5.361878204345703,
      "epoch": 7.1251896518187126,
      "grad_norm": 1.484375,
      "learning_rate": 0.00013692165834778034,
      "loss": 4.5698,
      "mean_token_accuracy": 0.24970492124557495,
      "num_tokens": 158840246.0,
      "step": 91580
    },
    {
      "entropy": 5.3834075927734375,
      "epoch": 7.12557868119043,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013689977645152979,
      "loss": 4.6409,
      "mean_token_accuracy": 0.2410392165184021,
      "num_tokens": 158848789.0,
      "step": 91585
    },
    {
      "entropy": 5.385695362091065,
      "epoch": 7.125967710562147,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013687789665073043,
      "loss": 4.697,
      "mean_token_accuracy": 0.23960563093423842,
      "num_tokens": 158857444.0,
      "step": 91590
    },
    {
      "entropy": 5.395308208465576,
      "epoch": 7.126356739933865,
      "grad_norm": 1.578125,
      "learning_rate": 0.00013685601894571412,
      "loss": 4.6156,
      "mean_token_accuracy": 0.2439314290881157,
      "num_tokens": 158866738.0,
      "step": 91595
    },
    {
      "entropy": 5.352168464660645,
      "epoch": 7.1267457693055825,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001368341433368127,
      "loss": 4.4701,
      "mean_token_accuracy": 0.25717429220676424,
      "num_tokens": 158875004.0,
      "step": 91600
    },
    {
      "entropy": 5.424878168106079,
      "epoch": 7.1271347986773,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001368122698243584,
      "loss": 4.6956,
      "mean_token_accuracy": 0.24251196533441544,
      "num_tokens": 158884487.0,
      "step": 91605
    },
    {
      "entropy": 5.405588674545288,
      "epoch": 7.127523828049018,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013679039840868286,
      "loss": 4.6227,
      "mean_token_accuracy": 0.24260909855365753,
      "num_tokens": 158892945.0,
      "step": 91610
    },
    {
      "entropy": 5.442149448394775,
      "epoch": 7.1279128574207355,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001367685290901181,
      "loss": 4.653,
      "mean_token_accuracy": 0.2500215947628021,
      "num_tokens": 158901512.0,
      "step": 91615
    },
    {
      "entropy": 5.3508460998535154,
      "epoch": 7.128301886792453,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013674666186899593,
      "loss": 4.5627,
      "mean_token_accuracy": 0.25110318660736086,
      "num_tokens": 158910612.0,
      "step": 91620
    },
    {
      "entropy": 5.436287212371826,
      "epoch": 7.12869091616417,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013672479674564801,
      "loss": 4.6955,
      "mean_token_accuracy": 0.22931321412324907,
      "num_tokens": 158919736.0,
      "step": 91625
    },
    {
      "entropy": 5.433459234237671,
      "epoch": 7.129079945535888,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013670293372040632,
      "loss": 4.7014,
      "mean_token_accuracy": 0.24178088456392288,
      "num_tokens": 158929689.0,
      "step": 91630
    },
    {
      "entropy": 5.3421934127807615,
      "epoch": 7.1294689749076054,
      "grad_norm": 1.375,
      "learning_rate": 0.0001366810727936024,
      "loss": 4.5319,
      "mean_token_accuracy": 0.25077295303344727,
      "num_tokens": 158938506.0,
      "step": 91635
    },
    {
      "entropy": 5.3192039966583256,
      "epoch": 7.129858004279323,
      "grad_norm": 1.296875,
      "learning_rate": 0.00013665921396556813,
      "loss": 4.5465,
      "mean_token_accuracy": 0.2537343993782997,
      "num_tokens": 158947659.0,
      "step": 91640
    },
    {
      "entropy": 5.419354867935181,
      "epoch": 7.130247033651041,
      "grad_norm": 1.390625,
      "learning_rate": 0.000136637357236635,
      "loss": 4.6419,
      "mean_token_accuracy": 0.24435766637325287,
      "num_tokens": 158956253.0,
      "step": 91645
    },
    {
      "entropy": 5.4129722118377686,
      "epoch": 7.1306360630227585,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001366155026071348,
      "loss": 4.6429,
      "mean_token_accuracy": 0.24408211410045624,
      "num_tokens": 158965481.0,
      "step": 91650
    },
    {
      "entropy": 5.324102163314819,
      "epoch": 7.131025092394475,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013659365007739904,
      "loss": 4.523,
      "mean_token_accuracy": 0.2508287787437439,
      "num_tokens": 158973867.0,
      "step": 91655
    },
    {
      "entropy": 5.359076833724975,
      "epoch": 7.131414121766193,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00013657179964775946,
      "loss": 4.5593,
      "mean_token_accuracy": 0.25037179589271547,
      "num_tokens": 158982034.0,
      "step": 91660
    },
    {
      "entropy": 5.343366098403931,
      "epoch": 7.131803151137911,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00013654995131854737,
      "loss": 4.5665,
      "mean_token_accuracy": 0.25332073122262955,
      "num_tokens": 158991382.0,
      "step": 91665
    },
    {
      "entropy": 5.36948790550232,
      "epoch": 7.132192180509628,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001365281050900944,
      "loss": 4.5874,
      "mean_token_accuracy": 0.24257041662931442,
      "num_tokens": 158999492.0,
      "step": 91670
    },
    {
      "entropy": 5.413803243637085,
      "epoch": 7.132581209881346,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013650626096273208,
      "loss": 4.6626,
      "mean_token_accuracy": 0.2473751425743103,
      "num_tokens": 159009117.0,
      "step": 91675
    },
    {
      "entropy": 5.287208509445191,
      "epoch": 7.132970239253064,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013648441893679175,
      "loss": 4.4786,
      "mean_token_accuracy": 0.2520866826176643,
      "num_tokens": 159017560.0,
      "step": 91680
    },
    {
      "entropy": 5.411993408203125,
      "epoch": 7.1333592686247815,
      "grad_norm": 1.515625,
      "learning_rate": 0.00013646257901260504,
      "loss": 4.5794,
      "mean_token_accuracy": 0.24998858124017714,
      "num_tokens": 159025917.0,
      "step": 91685
    },
    {
      "entropy": 5.348544025421143,
      "epoch": 7.133748297996498,
      "grad_norm": 1.28125,
      "learning_rate": 0.00013644074119050303,
      "loss": 4.6068,
      "mean_token_accuracy": 0.2583964139223099,
      "num_tokens": 159035540.0,
      "step": 91690
    },
    {
      "entropy": 5.441460609436035,
      "epoch": 7.134137327368216,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013641890547081738,
      "loss": 4.6697,
      "mean_token_accuracy": 0.23759443014860154,
      "num_tokens": 159043976.0,
      "step": 91695
    },
    {
      "entropy": 5.393807363510132,
      "epoch": 7.134526356739934,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013639707185387923,
      "loss": 4.6118,
      "mean_token_accuracy": 0.24634259790182114,
      "num_tokens": 159052369.0,
      "step": 91700
    },
    {
      "entropy": 5.347597980499268,
      "epoch": 7.134915386111651,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001363752403400198,
      "loss": 4.523,
      "mean_token_accuracy": 0.24707057774066926,
      "num_tokens": 159061962.0,
      "step": 91705
    },
    {
      "entropy": 5.354585647583008,
      "epoch": 7.135304415483369,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013635341092957054,
      "loss": 4.565,
      "mean_token_accuracy": 0.24991392344236374,
      "num_tokens": 159070922.0,
      "step": 91710
    },
    {
      "entropy": 5.445838260650635,
      "epoch": 7.135693444855087,
      "grad_norm": 1.375,
      "learning_rate": 0.00013633158362286256,
      "loss": 4.7184,
      "mean_token_accuracy": 0.23609377294778824,
      "num_tokens": 159080004.0,
      "step": 91715
    },
    {
      "entropy": 5.3380670070648195,
      "epoch": 7.1360824742268045,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001363097584202271,
      "loss": 4.5566,
      "mean_token_accuracy": 0.2535250782966614,
      "num_tokens": 159089417.0,
      "step": 91720
    },
    {
      "entropy": 5.403206014633179,
      "epoch": 7.136471503598521,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001362879353219953,
      "loss": 4.6679,
      "mean_token_accuracy": 0.23650895357131957,
      "num_tokens": 159098463.0,
      "step": 91725
    },
    {
      "entropy": 5.2806315422058105,
      "epoch": 7.136860532970239,
      "grad_norm": 1.375,
      "learning_rate": 0.00013626611432849834,
      "loss": 4.5207,
      "mean_token_accuracy": 0.253005550801754,
      "num_tokens": 159107453.0,
      "step": 91730
    },
    {
      "entropy": 5.387668561935425,
      "epoch": 7.137249562341957,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013624429544006718,
      "loss": 4.6066,
      "mean_token_accuracy": 0.23736293464899064,
      "num_tokens": 159116101.0,
      "step": 91735
    },
    {
      "entropy": 5.278861713409424,
      "epoch": 7.137638591713674,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013622247865703303,
      "loss": 4.5272,
      "mean_token_accuracy": 0.25451577752828597,
      "num_tokens": 159125422.0,
      "step": 91740
    },
    {
      "entropy": 5.395887804031372,
      "epoch": 7.138027621085392,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013620066397972692,
      "loss": 4.5914,
      "mean_token_accuracy": 0.24109362661838532,
      "num_tokens": 159133428.0,
      "step": 91745
    },
    {
      "entropy": 5.338425350189209,
      "epoch": 7.13841665045711,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013617885140847975,
      "loss": 4.5358,
      "mean_token_accuracy": 0.2553575739264488,
      "num_tokens": 159142300.0,
      "step": 91750
    },
    {
      "entropy": 5.386847734451294,
      "epoch": 7.1388056798288275,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001361570409436226,
      "loss": 4.5519,
      "mean_token_accuracy": 0.24778425544500352,
      "num_tokens": 159150732.0,
      "step": 91755
    },
    {
      "entropy": 5.302852010726928,
      "epoch": 7.139194709200544,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013613523258548634,
      "loss": 4.487,
      "mean_token_accuracy": 0.26098258048295975,
      "num_tokens": 159159291.0,
      "step": 91760
    },
    {
      "entropy": 5.332905244827271,
      "epoch": 7.139583738572262,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001361134263344019,
      "loss": 4.6084,
      "mean_token_accuracy": 0.2485702157020569,
      "num_tokens": 159167858.0,
      "step": 91765
    },
    {
      "entropy": 5.3920135498046875,
      "epoch": 7.13997276794398,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013609162219070016,
      "loss": 4.6733,
      "mean_token_accuracy": 0.24138628989458083,
      "num_tokens": 159175683.0,
      "step": 91770
    },
    {
      "entropy": 5.372470855712891,
      "epoch": 7.140361797315697,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013606982015471197,
      "loss": 4.5851,
      "mean_token_accuracy": 0.24423436671495438,
      "num_tokens": 159184809.0,
      "step": 91775
    },
    {
      "entropy": 5.407551574707031,
      "epoch": 7.140750826687415,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00013604802022676815,
      "loss": 4.6284,
      "mean_token_accuracy": 0.24580636918544768,
      "num_tokens": 159193019.0,
      "step": 91780
    },
    {
      "entropy": 5.4136851787567135,
      "epoch": 7.141139856059133,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013602622240719943,
      "loss": 4.6645,
      "mean_token_accuracy": 0.24887157380580902,
      "num_tokens": 159201601.0,
      "step": 91785
    },
    {
      "entropy": 5.349868440628052,
      "epoch": 7.14152888543085,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001360044266963366,
      "loss": 4.534,
      "mean_token_accuracy": 0.2553478732705116,
      "num_tokens": 159209311.0,
      "step": 91790
    },
    {
      "entropy": 5.379832935333252,
      "epoch": 7.141917914802567,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013598263309451032,
      "loss": 4.5889,
      "mean_token_accuracy": 0.24965229332447053,
      "num_tokens": 159218009.0,
      "step": 91795
    },
    {
      "entropy": 5.29295802116394,
      "epoch": 7.142306944174285,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013596084160205128,
      "loss": 4.4619,
      "mean_token_accuracy": 0.25085591822862624,
      "num_tokens": 159226196.0,
      "step": 91800
    },
    {
      "entropy": 5.360033321380615,
      "epoch": 7.142695973546003,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013593905221929019,
      "loss": 4.6153,
      "mean_token_accuracy": 0.24143189936876297,
      "num_tokens": 159235433.0,
      "step": 91805
    },
    {
      "entropy": 5.418444299697876,
      "epoch": 7.14308500291772,
      "grad_norm": 1.515625,
      "learning_rate": 0.00013591726494655771,
      "loss": 4.6031,
      "mean_token_accuracy": 0.2509622544050217,
      "num_tokens": 159242977.0,
      "step": 91810
    },
    {
      "entropy": 5.374507856369019,
      "epoch": 7.143474032289438,
      "grad_norm": 1.375,
      "learning_rate": 0.00013589547978418439,
      "loss": 4.6451,
      "mean_token_accuracy": 0.24340640753507614,
      "num_tokens": 159251514.0,
      "step": 91815
    },
    {
      "entropy": 5.347555923461914,
      "epoch": 7.143863061661156,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001358736967325006,
      "loss": 4.5553,
      "mean_token_accuracy": 0.2536451920866966,
      "num_tokens": 159261151.0,
      "step": 91820
    },
    {
      "entropy": 5.284817886352539,
      "epoch": 7.144252091032873,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001358519157918372,
      "loss": 4.5278,
      "mean_token_accuracy": 0.2518927201628685,
      "num_tokens": 159269930.0,
      "step": 91825
    },
    {
      "entropy": 5.3484782695770265,
      "epoch": 7.14464112040459,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001358301369625244,
      "loss": 4.6188,
      "mean_token_accuracy": 0.24383719265460968,
      "num_tokens": 159278801.0,
      "step": 91830
    },
    {
      "entropy": 5.312946224212647,
      "epoch": 7.145030149776308,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013580836024489277,
      "loss": 4.579,
      "mean_token_accuracy": 0.24584972113370895,
      "num_tokens": 159288201.0,
      "step": 91835
    },
    {
      "entropy": 5.33912410736084,
      "epoch": 7.145419179148026,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013578658563927272,
      "loss": 4.5933,
      "mean_token_accuracy": 0.24312245845794678,
      "num_tokens": 159297415.0,
      "step": 91840
    },
    {
      "entropy": 5.328149366378784,
      "epoch": 7.145808208519743,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013576481314599468,
      "loss": 4.5151,
      "mean_token_accuracy": 0.2482807904481888,
      "num_tokens": 159306286.0,
      "step": 91845
    },
    {
      "entropy": 5.379630899429321,
      "epoch": 7.146197237891461,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013574304276538893,
      "loss": 4.5873,
      "mean_token_accuracy": 0.24731138348579407,
      "num_tokens": 159314869.0,
      "step": 91850
    },
    {
      "entropy": 5.345496797561646,
      "epoch": 7.146586267263178,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013572127449778594,
      "loss": 4.5351,
      "mean_token_accuracy": 0.25402589440345763,
      "num_tokens": 159323322.0,
      "step": 91855
    },
    {
      "entropy": 5.375813007354736,
      "epoch": 7.146975296634896,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001356995083435159,
      "loss": 4.6858,
      "mean_token_accuracy": 0.23946799635887145,
      "num_tokens": 159331477.0,
      "step": 91860
    },
    {
      "entropy": 5.370027112960815,
      "epoch": 7.147364326006613,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013567774430290906,
      "loss": 4.6716,
      "mean_token_accuracy": 0.23948404043912888,
      "num_tokens": 159341107.0,
      "step": 91865
    },
    {
      "entropy": 5.3276674270629885,
      "epoch": 7.147753355378331,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013565598237629567,
      "loss": 4.5711,
      "mean_token_accuracy": 0.2388285994529724,
      "num_tokens": 159350308.0,
      "step": 91870
    },
    {
      "entropy": 5.2721117496490475,
      "epoch": 7.148142384750049,
      "grad_norm": 1.375,
      "learning_rate": 0.00013563422256400597,
      "loss": 4.4401,
      "mean_token_accuracy": 0.2631159543991089,
      "num_tokens": 159358890.0,
      "step": 91875
    },
    {
      "entropy": 5.35380072593689,
      "epoch": 7.148531414121766,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013561246486637019,
      "loss": 4.5375,
      "mean_token_accuracy": 0.25105722397565844,
      "num_tokens": 159367990.0,
      "step": 91880
    },
    {
      "entropy": 5.3977538585662845,
      "epoch": 7.148920443493484,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013559070928371834,
      "loss": 4.6567,
      "mean_token_accuracy": 0.24644645005464555,
      "num_tokens": 159376989.0,
      "step": 91885
    },
    {
      "entropy": 5.432134008407592,
      "epoch": 7.149309472865201,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013556895581638066,
      "loss": 4.6811,
      "mean_token_accuracy": 0.2360985904932022,
      "num_tokens": 159385942.0,
      "step": 91890
    },
    {
      "entropy": 5.419182634353637,
      "epoch": 7.149698502236919,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001355472044646871,
      "loss": 4.6954,
      "mean_token_accuracy": 0.2404509037733078,
      "num_tokens": 159394889.0,
      "step": 91895
    },
    {
      "entropy": 5.323710489273071,
      "epoch": 7.150087531608636,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013552545522896772,
      "loss": 4.6236,
      "mean_token_accuracy": 0.24577766507864,
      "num_tokens": 159404090.0,
      "step": 91900
    },
    {
      "entropy": 5.419246625900269,
      "epoch": 7.150476560980354,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001355037081095526,
      "loss": 4.6927,
      "mean_token_accuracy": 0.2423545390367508,
      "num_tokens": 159413326.0,
      "step": 91905
    },
    {
      "entropy": 5.266530466079712,
      "epoch": 7.150865590352072,
      "grad_norm": 1.28125,
      "learning_rate": 0.00013548196310677162,
      "loss": 4.4994,
      "mean_token_accuracy": 0.25059216022491454,
      "num_tokens": 159422576.0,
      "step": 91910
    },
    {
      "entropy": 5.456769800186157,
      "epoch": 7.151254619723789,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001354602202209549,
      "loss": 4.659,
      "mean_token_accuracy": 0.2418781891465187,
      "num_tokens": 159430914.0,
      "step": 91915
    },
    {
      "entropy": 5.326454496383667,
      "epoch": 7.151643649095507,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001354384794524321,
      "loss": 4.4827,
      "mean_token_accuracy": 0.25647252202034,
      "num_tokens": 159439280.0,
      "step": 91920
    },
    {
      "entropy": 5.397288799285889,
      "epoch": 7.152032678467224,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013541674080153336,
      "loss": 4.6033,
      "mean_token_accuracy": 0.23796941936016083,
      "num_tokens": 159448101.0,
      "step": 91925
    },
    {
      "entropy": 5.345769643783569,
      "epoch": 7.152421707838942,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00013539500426858832,
      "loss": 4.5851,
      "mean_token_accuracy": 0.2511184364557266,
      "num_tokens": 159456995.0,
      "step": 91930
    },
    {
      "entropy": 5.446032333374023,
      "epoch": 7.152810737210659,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001353732698539269,
      "loss": 4.6556,
      "mean_token_accuracy": 0.23929524272680283,
      "num_tokens": 159465778.0,
      "step": 91935
    },
    {
      "entropy": 5.339037179946899,
      "epoch": 7.153199766582377,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013535153755787888,
      "loss": 4.5352,
      "mean_token_accuracy": 0.25619996190071104,
      "num_tokens": 159474277.0,
      "step": 91940
    },
    {
      "entropy": 5.401755237579346,
      "epoch": 7.153588795954095,
      "grad_norm": 1.4375,
      "learning_rate": 0.000135329807380774,
      "loss": 4.6812,
      "mean_token_accuracy": 0.24394504278898238,
      "num_tokens": 159482190.0,
      "step": 91945
    },
    {
      "entropy": 5.3622260093688965,
      "epoch": 7.153977825325812,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013530807932294204,
      "loss": 4.5521,
      "mean_token_accuracy": 0.24665105044841767,
      "num_tokens": 159490547.0,
      "step": 91950
    },
    {
      "entropy": 5.422585201263428,
      "epoch": 7.15436685469753,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013528635338471256,
      "loss": 4.6784,
      "mean_token_accuracy": 0.2374989092350006,
      "num_tokens": 159499342.0,
      "step": 91955
    },
    {
      "entropy": 5.330280685424805,
      "epoch": 7.154755884069247,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00013526462956641534,
      "loss": 4.5251,
      "mean_token_accuracy": 0.24916384369134903,
      "num_tokens": 159507984.0,
      "step": 91960
    },
    {
      "entropy": 5.448762655258179,
      "epoch": 7.155144913440965,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001352429078683799,
      "loss": 4.6903,
      "mean_token_accuracy": 0.2348872035741806,
      "num_tokens": 159516923.0,
      "step": 91965
    },
    {
      "entropy": 5.388234615325928,
      "epoch": 7.155533942812682,
      "grad_norm": 1.46875,
      "learning_rate": 0.000135221188290936,
      "loss": 4.6052,
      "mean_token_accuracy": 0.24421861171722412,
      "num_tokens": 159525383.0,
      "step": 91970
    },
    {
      "entropy": 5.352590847015381,
      "epoch": 7.1559229721844,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013519947083441303,
      "loss": 4.5447,
      "mean_token_accuracy": 0.24830097407102586,
      "num_tokens": 159533718.0,
      "step": 91975
    },
    {
      "entropy": 5.429497957229614,
      "epoch": 7.156312001556118,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001351777554991405,
      "loss": 4.7015,
      "mean_token_accuracy": 0.23887073397636413,
      "num_tokens": 159541781.0,
      "step": 91980
    },
    {
      "entropy": 5.3368628978729244,
      "epoch": 7.156701030927835,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00013515604228544808,
      "loss": 4.6258,
      "mean_token_accuracy": 0.2414705440402031,
      "num_tokens": 159550784.0,
      "step": 91985
    },
    {
      "entropy": 5.322275686264038,
      "epoch": 7.157090060299552,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013513433119366505,
      "loss": 4.4944,
      "mean_token_accuracy": 0.25704392343759536,
      "num_tokens": 159559445.0,
      "step": 91990
    },
    {
      "entropy": 5.346365022659302,
      "epoch": 7.15747908967127,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013511262222412103,
      "loss": 4.5362,
      "mean_token_accuracy": 0.2544390827417374,
      "num_tokens": 159568056.0,
      "step": 91995
    },
    {
      "entropy": 5.3697751522064205,
      "epoch": 7.157868119042988,
      "grad_norm": 1.375,
      "learning_rate": 0.00013509091537714519,
      "loss": 4.6145,
      "mean_token_accuracy": 0.24309487789869308,
      "num_tokens": 159576102.0,
      "step": 92000
    },
    {
      "entropy": 5.370607185363769,
      "epoch": 7.158257148414705,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00013506921065306704,
      "loss": 4.5203,
      "mean_token_accuracy": 0.2478438451886177,
      "num_tokens": 159583876.0,
      "step": 92005
    },
    {
      "entropy": 5.367839050292969,
      "epoch": 7.158646177786423,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013504750805221598,
      "loss": 4.5082,
      "mean_token_accuracy": 0.25539582073688505,
      "num_tokens": 159592111.0,
      "step": 92010
    },
    {
      "entropy": 5.290379905700684,
      "epoch": 7.159035207158141,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013502580757492112,
      "loss": 4.5605,
      "mean_token_accuracy": 0.2559082478284836,
      "num_tokens": 159600639.0,
      "step": 92015
    },
    {
      "entropy": 5.267931938171387,
      "epoch": 7.159424236529858,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013500410922151197,
      "loss": 4.5649,
      "mean_token_accuracy": 0.2536333501338959,
      "num_tokens": 159608327.0,
      "step": 92020
    },
    {
      "entropy": 5.2552543640136715,
      "epoch": 7.159813265901575,
      "grad_norm": 1.375,
      "learning_rate": 0.0001349824129923175,
      "loss": 4.4561,
      "mean_token_accuracy": 0.26401792615652087,
      "num_tokens": 159616918.0,
      "step": 92025
    },
    {
      "entropy": 5.339044904708862,
      "epoch": 7.160202295273293,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013496071888766714,
      "loss": 4.5958,
      "mean_token_accuracy": 0.24941682815551758,
      "num_tokens": 159625815.0,
      "step": 92030
    },
    {
      "entropy": 5.354481410980225,
      "epoch": 7.160591324645011,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013493902690788994,
      "loss": 4.5802,
      "mean_token_accuracy": 0.25219210237264633,
      "num_tokens": 159634404.0,
      "step": 92035
    },
    {
      "entropy": 5.375811815261841,
      "epoch": 7.160980354016728,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013491733705331516,
      "loss": 4.5756,
      "mean_token_accuracy": 0.24587667137384414,
      "num_tokens": 159643014.0,
      "step": 92040
    },
    {
      "entropy": 5.388228178024292,
      "epoch": 7.161369383388446,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00013489564932427168,
      "loss": 4.6414,
      "mean_token_accuracy": 0.24523551166057586,
      "num_tokens": 159650891.0,
      "step": 92045
    },
    {
      "entropy": 5.318236207962036,
      "epoch": 7.161758412760164,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00013487396372108883,
      "loss": 4.5837,
      "mean_token_accuracy": 0.24198245108127595,
      "num_tokens": 159660116.0,
      "step": 92050
    },
    {
      "entropy": 5.407441425323486,
      "epoch": 7.162147442131881,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00013485228024409556,
      "loss": 4.5513,
      "mean_token_accuracy": 0.2468734234571457,
      "num_tokens": 159668686.0,
      "step": 92055
    },
    {
      "entropy": 5.297761583328247,
      "epoch": 7.162536471503598,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013483059889362082,
      "loss": 4.5002,
      "mean_token_accuracy": 0.26113088726997374,
      "num_tokens": 159677742.0,
      "step": 92060
    },
    {
      "entropy": 5.394374179840088,
      "epoch": 7.162925500875316,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001348089196699936,
      "loss": 4.6825,
      "mean_token_accuracy": 0.2414155900478363,
      "num_tokens": 159686211.0,
      "step": 92065
    },
    {
      "entropy": 5.4024735450744625,
      "epoch": 7.163314530247034,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013478724257354289,
      "loss": 4.5943,
      "mean_token_accuracy": 0.24288256764411925,
      "num_tokens": 159694687.0,
      "step": 92070
    },
    {
      "entropy": 5.321637105941773,
      "epoch": 7.163703559618751,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013476556760459767,
      "loss": 4.5442,
      "mean_token_accuracy": 0.257716666162014,
      "num_tokens": 159702460.0,
      "step": 92075
    },
    {
      "entropy": 5.321578693389893,
      "epoch": 7.164092588990469,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013474389476348677,
      "loss": 4.5668,
      "mean_token_accuracy": 0.25082063525915144,
      "num_tokens": 159710291.0,
      "step": 92080
    },
    {
      "entropy": 5.45791335105896,
      "epoch": 7.164481618362187,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013472222405053896,
      "loss": 4.673,
      "mean_token_accuracy": 0.2369900405406952,
      "num_tokens": 159718682.0,
      "step": 92085
    },
    {
      "entropy": 5.328112840652466,
      "epoch": 7.164870647733904,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013470055546608317,
      "loss": 4.5262,
      "mean_token_accuracy": 0.2469850480556488,
      "num_tokens": 159726928.0,
      "step": 92090
    },
    {
      "entropy": 5.300587606430054,
      "epoch": 7.165259677105621,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001346788890104481,
      "loss": 4.5209,
      "mean_token_accuracy": 0.25638154000043867,
      "num_tokens": 159735369.0,
      "step": 92095
    },
    {
      "entropy": 5.433917999267578,
      "epoch": 7.165648706477339,
      "grad_norm": 1.46875,
      "learning_rate": 0.00013465722468396262,
      "loss": 4.6284,
      "mean_token_accuracy": 0.24544442147016526,
      "num_tokens": 159744263.0,
      "step": 92100
    },
    {
      "entropy": 5.375982332229614,
      "epoch": 7.166037735849057,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013463556248695526,
      "loss": 4.6053,
      "mean_token_accuracy": 0.2510353371500969,
      "num_tokens": 159753481.0,
      "step": 92105
    },
    {
      "entropy": 5.3766186237335205,
      "epoch": 7.166426765220774,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013461390241975492,
      "loss": 4.5884,
      "mean_token_accuracy": 0.24433238804340363,
      "num_tokens": 159762758.0,
      "step": 92110
    },
    {
      "entropy": 5.341560983657837,
      "epoch": 7.166815794592492,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013459224448269014,
      "loss": 4.5946,
      "mean_token_accuracy": 0.24225471615791322,
      "num_tokens": 159772381.0,
      "step": 92115
    },
    {
      "entropy": 5.411349153518676,
      "epoch": 7.16720482396421,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00013457058867608957,
      "loss": 4.6758,
      "mean_token_accuracy": 0.24554115831851958,
      "num_tokens": 159781901.0,
      "step": 92120
    },
    {
      "entropy": 5.389523983001709,
      "epoch": 7.1675938533359265,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013454893500028175,
      "loss": 4.5918,
      "mean_token_accuracy": 0.24954482764005662,
      "num_tokens": 159789741.0,
      "step": 92125
    },
    {
      "entropy": 5.4893652439117435,
      "epoch": 7.167982882707644,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001345272834555954,
      "loss": 4.7645,
      "mean_token_accuracy": 0.23119542002677917,
      "num_tokens": 159798360.0,
      "step": 92130
    },
    {
      "entropy": 5.425929117202759,
      "epoch": 7.168371912079362,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013450563404235883,
      "loss": 4.5696,
      "mean_token_accuracy": 0.24779837280511857,
      "num_tokens": 159806474.0,
      "step": 92135
    },
    {
      "entropy": 5.354390287399292,
      "epoch": 7.16876094145108,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013448398676090063,
      "loss": 4.5613,
      "mean_token_accuracy": 0.25062927752733233,
      "num_tokens": 159815332.0,
      "step": 92140
    },
    {
      "entropy": 5.342707109451294,
      "epoch": 7.169149970822797,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001344623416115494,
      "loss": 4.6024,
      "mean_token_accuracy": 0.24676621854305267,
      "num_tokens": 159824452.0,
      "step": 92145
    },
    {
      "entropy": 5.28343243598938,
      "epoch": 7.169539000194515,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001344406985946333,
      "loss": 4.4655,
      "mean_token_accuracy": 0.25851436257362365,
      "num_tokens": 159834107.0,
      "step": 92150
    },
    {
      "entropy": 5.3631617546081545,
      "epoch": 7.169928029566233,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013441905771048102,
      "loss": 4.6042,
      "mean_token_accuracy": 0.24699197113513946,
      "num_tokens": 159843199.0,
      "step": 92155
    },
    {
      "entropy": 5.348460912704468,
      "epoch": 7.1703170589379495,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013439741895942073,
      "loss": 4.5534,
      "mean_token_accuracy": 0.2526869237422943,
      "num_tokens": 159851440.0,
      "step": 92160
    },
    {
      "entropy": 5.366093111038208,
      "epoch": 7.170706088309667,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00013437578234178077,
      "loss": 4.6101,
      "mean_token_accuracy": 0.2501327842473984,
      "num_tokens": 159860308.0,
      "step": 92165
    },
    {
      "entropy": 5.405993270874023,
      "epoch": 7.171095117681385,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013435414785788956,
      "loss": 4.6275,
      "mean_token_accuracy": 0.247701296210289,
      "num_tokens": 159869281.0,
      "step": 92170
    },
    {
      "entropy": 5.349754571914673,
      "epoch": 7.1714841470531026,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013433251550807522,
      "loss": 4.5448,
      "mean_token_accuracy": 0.24572580307722092,
      "num_tokens": 159878522.0,
      "step": 92175
    },
    {
      "entropy": 5.34875521659851,
      "epoch": 7.17187317642482,
      "grad_norm": 1.375,
      "learning_rate": 0.00013431088529266614,
      "loss": 4.53,
      "mean_token_accuracy": 0.2569475769996643,
      "num_tokens": 159886717.0,
      "step": 92180
    },
    {
      "entropy": 5.375426149368286,
      "epoch": 7.172262205796538,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013428925721199037,
      "loss": 4.5822,
      "mean_token_accuracy": 0.25048134475946426,
      "num_tokens": 159895244.0,
      "step": 92185
    },
    {
      "entropy": 5.3331298828125,
      "epoch": 7.172651235168255,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001342676312663762,
      "loss": 4.5599,
      "mean_token_accuracy": 0.24866369664669036,
      "num_tokens": 159903650.0,
      "step": 92190
    },
    {
      "entropy": 5.370479726791382,
      "epoch": 7.1730402645399725,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00013424600745615166,
      "loss": 4.5385,
      "mean_token_accuracy": 0.2464562475681305,
      "num_tokens": 159912210.0,
      "step": 92195
    },
    {
      "entropy": 5.371892976760864,
      "epoch": 7.17342929391169,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013422438578164497,
      "loss": 4.6599,
      "mean_token_accuracy": 0.24194591492414474,
      "num_tokens": 159920731.0,
      "step": 92200
    },
    {
      "entropy": 5.322563123703003,
      "epoch": 7.173818323283408,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001342027662431842,
      "loss": 4.5449,
      "mean_token_accuracy": 0.2501654803752899,
      "num_tokens": 159929320.0,
      "step": 92205
    },
    {
      "entropy": 5.301455068588257,
      "epoch": 7.1742073526551255,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013418114884109728,
      "loss": 4.5737,
      "mean_token_accuracy": 0.24366295486688613,
      "num_tokens": 159937939.0,
      "step": 92210
    },
    {
      "entropy": 5.3282218933105465,
      "epoch": 7.174596382026843,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013415953357571236,
      "loss": 4.5467,
      "mean_token_accuracy": 0.2466237261891365,
      "num_tokens": 159946304.0,
      "step": 92215
    },
    {
      "entropy": 5.296998834609985,
      "epoch": 7.174985411398561,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001341379204473573,
      "loss": 4.5716,
      "mean_token_accuracy": 0.2444770559668541,
      "num_tokens": 159955289.0,
      "step": 92220
    },
    {
      "entropy": 5.320829629898071,
      "epoch": 7.175374440770278,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001341163094563601,
      "loss": 4.5579,
      "mean_token_accuracy": 0.24642224311828614,
      "num_tokens": 159964131.0,
      "step": 92225
    },
    {
      "entropy": 5.352261686325074,
      "epoch": 7.1757634701419954,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001340947006030487,
      "loss": 4.5922,
      "mean_token_accuracy": 0.2512372300028801,
      "num_tokens": 159973048.0,
      "step": 92230
    },
    {
      "entropy": 5.317491817474365,
      "epoch": 7.176152499513713,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013407309388775092,
      "loss": 4.5111,
      "mean_token_accuracy": 0.25626313388347627,
      "num_tokens": 159981529.0,
      "step": 92235
    },
    {
      "entropy": 5.3932582378387455,
      "epoch": 7.176541528885431,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013405148931079468,
      "loss": 4.6515,
      "mean_token_accuracy": 0.23996201306581497,
      "num_tokens": 159989764.0,
      "step": 92240
    },
    {
      "entropy": 5.3304518699646,
      "epoch": 7.1769305582571485,
      "grad_norm": 1.296875,
      "learning_rate": 0.00013402988687250767,
      "loss": 4.557,
      "mean_token_accuracy": 0.25237902998924255,
      "num_tokens": 159998322.0,
      "step": 92245
    },
    {
      "entropy": 5.375513219833374,
      "epoch": 7.177319587628866,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001340082865732178,
      "loss": 4.5863,
      "mean_token_accuracy": 0.25139415413141253,
      "num_tokens": 160006750.0,
      "step": 92250
    },
    {
      "entropy": 5.379061126708985,
      "epoch": 7.177708617000584,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013398668841325267,
      "loss": 4.5793,
      "mean_token_accuracy": 0.24378471225500106,
      "num_tokens": 160015673.0,
      "step": 92255
    },
    {
      "entropy": 5.3945722579956055,
      "epoch": 7.178097646372301,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013396509239294022,
      "loss": 4.5989,
      "mean_token_accuracy": 0.24884821474552155,
      "num_tokens": 160024695.0,
      "step": 92260
    },
    {
      "entropy": 5.419507312774658,
      "epoch": 7.178486675744018,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013394349851260785,
      "loss": 4.691,
      "mean_token_accuracy": 0.24319444000720977,
      "num_tokens": 160033822.0,
      "step": 92265
    },
    {
      "entropy": 5.302937412261963,
      "epoch": 7.178875705115736,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013392190677258354,
      "loss": 4.5067,
      "mean_token_accuracy": 0.25397036373615267,
      "num_tokens": 160042687.0,
      "step": 92270
    },
    {
      "entropy": 5.312562704086304,
      "epoch": 7.179264734487454,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013390031717319472,
      "loss": 4.5039,
      "mean_token_accuracy": 0.2555666819214821,
      "num_tokens": 160050487.0,
      "step": 92275
    },
    {
      "entropy": 5.4166792869567875,
      "epoch": 7.1796537638591715,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013387872971476895,
      "loss": 4.6447,
      "mean_token_accuracy": 0.25058740973472593,
      "num_tokens": 160058446.0,
      "step": 92280
    },
    {
      "entropy": 5.38418288230896,
      "epoch": 7.180042793230889,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001338571443976339,
      "loss": 4.6493,
      "mean_token_accuracy": 0.24443479031324386,
      "num_tokens": 160067062.0,
      "step": 92285
    },
    {
      "entropy": 5.309890651702881,
      "epoch": 7.180431822602607,
      "grad_norm": 1.3125,
      "learning_rate": 0.000133835561222117,
      "loss": 4.5688,
      "mean_token_accuracy": 0.24733113050460814,
      "num_tokens": 160075644.0,
      "step": 92290
    },
    {
      "entropy": 5.401846075057984,
      "epoch": 7.180820851974324,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001338139801885458,
      "loss": 4.6492,
      "mean_token_accuracy": 0.24133193194866182,
      "num_tokens": 160084050.0,
      "step": 92295
    },
    {
      "entropy": 5.317452812194825,
      "epoch": 7.181209881346041,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001337924012972476,
      "loss": 4.5575,
      "mean_token_accuracy": 0.24576072096824647,
      "num_tokens": 160093325.0,
      "step": 92300
    },
    {
      "entropy": 5.417838382720947,
      "epoch": 7.181598910717759,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013377082454855017,
      "loss": 4.6511,
      "mean_token_accuracy": 0.24910195618867875,
      "num_tokens": 160101909.0,
      "step": 92305
    },
    {
      "entropy": 5.401565265655518,
      "epoch": 7.181987940089477,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00013374924994278049,
      "loss": 4.6463,
      "mean_token_accuracy": 0.23961753398180008,
      "num_tokens": 160110188.0,
      "step": 92310
    },
    {
      "entropy": 5.318426179885864,
      "epoch": 7.1823769694611945,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00013372767748026627,
      "loss": 4.5538,
      "mean_token_accuracy": 0.24642678052186967,
      "num_tokens": 160119274.0,
      "step": 92315
    },
    {
      "entropy": 5.277544355392456,
      "epoch": 7.182765998832912,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013370610716133464,
      "loss": 4.4784,
      "mean_token_accuracy": 0.25864783823490145,
      "num_tokens": 160127539.0,
      "step": 92320
    },
    {
      "entropy": 5.3370684623718265,
      "epoch": 7.183155028204629,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001336845389863129,
      "loss": 4.5288,
      "mean_token_accuracy": 0.2553452059626579,
      "num_tokens": 160136614.0,
      "step": 92325
    },
    {
      "entropy": 5.305613470077515,
      "epoch": 7.183544057576347,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001336629729555284,
      "loss": 4.5238,
      "mean_token_accuracy": 0.2538721993565559,
      "num_tokens": 160145359.0,
      "step": 92330
    },
    {
      "entropy": 5.3404582977294925,
      "epoch": 7.183933086948064,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00013364140906930823,
      "loss": 4.6267,
      "mean_token_accuracy": 0.2443299651145935,
      "num_tokens": 160153157.0,
      "step": 92335
    },
    {
      "entropy": 5.3828661918640135,
      "epoch": 7.184322116319782,
      "grad_norm": 1.53125,
      "learning_rate": 0.00013361984732797978,
      "loss": 4.6179,
      "mean_token_accuracy": 0.24346195310354232,
      "num_tokens": 160161469.0,
      "step": 92340
    },
    {
      "entropy": 5.362735939025879,
      "epoch": 7.1847111456915,
      "grad_norm": 1.5,
      "learning_rate": 0.00013359828773187003,
      "loss": 4.5985,
      "mean_token_accuracy": 0.23637529760599135,
      "num_tokens": 160169377.0,
      "step": 92345
    },
    {
      "entropy": 5.3138617992401125,
      "epoch": 7.1851001750632175,
      "grad_norm": 1.375,
      "learning_rate": 0.0001335767302813063,
      "loss": 4.5898,
      "mean_token_accuracy": 0.24501776397228242,
      "num_tokens": 160178003.0,
      "step": 92350
    },
    {
      "entropy": 5.289225053787232,
      "epoch": 7.185489204434935,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013355517497661557,
      "loss": 4.523,
      "mean_token_accuracy": 0.2454012230038643,
      "num_tokens": 160186141.0,
      "step": 92355
    },
    {
      "entropy": 5.349445819854736,
      "epoch": 7.185878233806652,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001335336218181249,
      "loss": 4.5291,
      "mean_token_accuracy": 0.25032408386468885,
      "num_tokens": 160194183.0,
      "step": 92360
    },
    {
      "entropy": 5.3792760372161865,
      "epoch": 7.18626726317837,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001335120708061614,
      "loss": 4.6612,
      "mean_token_accuracy": 0.24206367880105972,
      "num_tokens": 160202581.0,
      "step": 92365
    },
    {
      "entropy": 5.434961175918579,
      "epoch": 7.186656292550087,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001334905219410519,
      "loss": 4.6254,
      "mean_token_accuracy": 0.23878999650478364,
      "num_tokens": 160210797.0,
      "step": 92370
    },
    {
      "entropy": 5.306635332107544,
      "epoch": 7.187045321921805,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013346897522312361,
      "loss": 4.479,
      "mean_token_accuracy": 0.25838774889707566,
      "num_tokens": 160220058.0,
      "step": 92375
    },
    {
      "entropy": 5.354110956192017,
      "epoch": 7.187434351293523,
      "grad_norm": 1.375,
      "learning_rate": 0.00013344743065270333,
      "loss": 4.6277,
      "mean_token_accuracy": 0.24415798336267472,
      "num_tokens": 160228688.0,
      "step": 92380
    },
    {
      "entropy": 5.367720222473144,
      "epoch": 7.1878233806652405,
      "grad_norm": 1.53125,
      "learning_rate": 0.000133425888230118,
      "loss": 4.5161,
      "mean_token_accuracy": 0.25048488974571226,
      "num_tokens": 160236917.0,
      "step": 92385
    },
    {
      "entropy": 5.412647819519043,
      "epoch": 7.188212410036958,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001334043479556944,
      "loss": 4.6298,
      "mean_token_accuracy": 0.24219121783971786,
      "num_tokens": 160245012.0,
      "step": 92390
    },
    {
      "entropy": 5.321695947647095,
      "epoch": 7.188601439408675,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013338280982975954,
      "loss": 4.4519,
      "mean_token_accuracy": 0.25519401133060454,
      "num_tokens": 160252911.0,
      "step": 92395
    },
    {
      "entropy": 5.386152458190918,
      "epoch": 7.188990468780393,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013336127385264005,
      "loss": 4.6124,
      "mean_token_accuracy": 0.23880428075790405,
      "num_tokens": 160261073.0,
      "step": 92400
    },
    {
      "entropy": 5.358757734298706,
      "epoch": 7.18937949815211,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00013333974002466278,
      "loss": 4.5912,
      "mean_token_accuracy": 0.24331521540880202,
      "num_tokens": 160269533.0,
      "step": 92405
    },
    {
      "entropy": 5.360108470916748,
      "epoch": 7.189768527523828,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001333182083461546,
      "loss": 4.6146,
      "mean_token_accuracy": 0.24993547946214675,
      "num_tokens": 160278219.0,
      "step": 92410
    },
    {
      "entropy": 5.344538450241089,
      "epoch": 7.190157556895546,
      "grad_norm": 1.28125,
      "learning_rate": 0.00013329667881744196,
      "loss": 4.633,
      "mean_token_accuracy": 0.24352982342243196,
      "num_tokens": 160286798.0,
      "step": 92415
    },
    {
      "entropy": 5.371330642700196,
      "epoch": 7.1905465862672635,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013327515143885177,
      "loss": 4.6301,
      "mean_token_accuracy": 0.23921996504068374,
      "num_tokens": 160295422.0,
      "step": 92420
    },
    {
      "entropy": 5.298847818374634,
      "epoch": 7.19093561563898,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013325362621071052,
      "loss": 4.5581,
      "mean_token_accuracy": 0.25083294212818147,
      "num_tokens": 160304255.0,
      "step": 92425
    },
    {
      "entropy": 5.404986906051636,
      "epoch": 7.191324645010698,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00013323210313334494,
      "loss": 4.6462,
      "mean_token_accuracy": 0.24684948772192,
      "num_tokens": 160313012.0,
      "step": 92430
    },
    {
      "entropy": 5.428472995758057,
      "epoch": 7.191713674382416,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001332105822070816,
      "loss": 4.6941,
      "mean_token_accuracy": 0.2322460249066353,
      "num_tokens": 160322576.0,
      "step": 92435
    },
    {
      "entropy": 5.393605422973633,
      "epoch": 7.192102703754133,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013318906343224684,
      "loss": 4.6179,
      "mean_token_accuracy": 0.24242098182439803,
      "num_tokens": 160331612.0,
      "step": 92440
    },
    {
      "entropy": 5.376457405090332,
      "epoch": 7.192491733125851,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00013316754680916745,
      "loss": 4.6537,
      "mean_token_accuracy": 0.24048831313848495,
      "num_tokens": 160340110.0,
      "step": 92445
    },
    {
      "entropy": 5.388484001159668,
      "epoch": 7.192880762497569,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001331460323381697,
      "loss": 4.5989,
      "mean_token_accuracy": 0.24702430963516236,
      "num_tokens": 160348036.0,
      "step": 92450
    },
    {
      "entropy": 5.363025856018067,
      "epoch": 7.1932697918692865,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001331245200195802,
      "loss": 4.5248,
      "mean_token_accuracy": 0.2451356902718544,
      "num_tokens": 160356456.0,
      "step": 92455
    },
    {
      "entropy": 5.374103116989136,
      "epoch": 7.193658821241003,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013310300985372523,
      "loss": 4.5508,
      "mean_token_accuracy": 0.247349314391613,
      "num_tokens": 160365108.0,
      "step": 92460
    },
    {
      "entropy": 5.392385005950928,
      "epoch": 7.194047850612721,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001330815018409312,
      "loss": 4.7014,
      "mean_token_accuracy": 0.23145233541727067,
      "num_tokens": 160374439.0,
      "step": 92465
    },
    {
      "entropy": 5.359987354278564,
      "epoch": 7.194436879984439,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013305999598152462,
      "loss": 4.6076,
      "mean_token_accuracy": 0.2418766126036644,
      "num_tokens": 160383097.0,
      "step": 92470
    },
    {
      "entropy": 5.405289173126221,
      "epoch": 7.194825909356156,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013303849227583154,
      "loss": 4.6421,
      "mean_token_accuracy": 0.23963939845561982,
      "num_tokens": 160391819.0,
      "step": 92475
    },
    {
      "entropy": 5.400410556793213,
      "epoch": 7.195214938727874,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001330169907241785,
      "loss": 4.5768,
      "mean_token_accuracy": 0.24823762029409407,
      "num_tokens": 160400248.0,
      "step": 92480
    },
    {
      "entropy": 5.3757466793060305,
      "epoch": 7.195603968099592,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013299549132689154,
      "loss": 4.5744,
      "mean_token_accuracy": 0.25009443461894987,
      "num_tokens": 160408975.0,
      "step": 92485
    },
    {
      "entropy": 5.446235609054566,
      "epoch": 7.1959929974713095,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001329739940842971,
      "loss": 4.7113,
      "mean_token_accuracy": 0.23513687700033187,
      "num_tokens": 160417011.0,
      "step": 92490
    },
    {
      "entropy": 5.379382038116455,
      "epoch": 7.196382026843026,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00013295249899672112,
      "loss": 4.6215,
      "mean_token_accuracy": 0.23636218458414077,
      "num_tokens": 160426365.0,
      "step": 92495
    },
    {
      "entropy": 5.44202446937561,
      "epoch": 7.196771056214744,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013293100606448993,
      "loss": 4.7075,
      "mean_token_accuracy": 0.2350318178534508,
      "num_tokens": 160434963.0,
      "step": 92500
    },
    {
      "entropy": 5.388013172149658,
      "epoch": 7.197160085586462,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00013290951528792958,
      "loss": 4.6138,
      "mean_token_accuracy": 0.24473245888948442,
      "num_tokens": 160443609.0,
      "step": 92505
    },
    {
      "entropy": 5.42171688079834,
      "epoch": 7.197549114958179,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001328880266673662,
      "loss": 4.6717,
      "mean_token_accuracy": 0.23804708421230317,
      "num_tokens": 160451939.0,
      "step": 92510
    },
    {
      "entropy": 5.425670337677002,
      "epoch": 7.197938144329897,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001328665402031259,
      "loss": 4.682,
      "mean_token_accuracy": 0.23748090565204621,
      "num_tokens": 160461124.0,
      "step": 92515
    },
    {
      "entropy": 5.344083499908447,
      "epoch": 7.198327173701615,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013284505589553446,
      "loss": 4.6,
      "mean_token_accuracy": 0.2454697757959366,
      "num_tokens": 160470152.0,
      "step": 92520
    },
    {
      "entropy": 5.336736392974854,
      "epoch": 7.198716203073332,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013282357374491815,
      "loss": 4.5952,
      "mean_token_accuracy": 0.24086117297410964,
      "num_tokens": 160477930.0,
      "step": 92525
    },
    {
      "entropy": 5.37874813079834,
      "epoch": 7.199105232445049,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013280209375160265,
      "loss": 4.6339,
      "mean_token_accuracy": 0.25125999450683595,
      "num_tokens": 160486945.0,
      "step": 92530
    },
    {
      "entropy": 5.3494977951049805,
      "epoch": 7.199494261816767,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001327806159159142,
      "loss": 4.6009,
      "mean_token_accuracy": 0.24929711371660232,
      "num_tokens": 160496063.0,
      "step": 92535
    },
    {
      "entropy": 5.449994087219238,
      "epoch": 7.199883291188485,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013275914023817853,
      "loss": 4.7441,
      "mean_token_accuracy": 0.24051253050565718,
      "num_tokens": 160505148.0,
      "step": 92540
    },
    {
      "entropy": 5.4145416736602785,
      "epoch": 7.200272320560202,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013273766671872145,
      "loss": 4.6459,
      "mean_token_accuracy": 0.2421468198299408,
      "num_tokens": 160513575.0,
      "step": 92545
    },
    {
      "entropy": 5.410236644744873,
      "epoch": 7.20066134993192,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013271619535786891,
      "loss": 4.606,
      "mean_token_accuracy": 0.24200905859470367,
      "num_tokens": 160521228.0,
      "step": 92550
    },
    {
      "entropy": 5.360778045654297,
      "epoch": 7.201050379303638,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00013269472615594659,
      "loss": 4.5547,
      "mean_token_accuracy": 0.2508039534091949,
      "num_tokens": 160529868.0,
      "step": 92555
    },
    {
      "entropy": 5.4860191822052,
      "epoch": 7.201439408675355,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013267325911328032,
      "loss": 4.7198,
      "mean_token_accuracy": 0.23879399299621581,
      "num_tokens": 160538777.0,
      "step": 92560
    },
    {
      "entropy": 5.399104166030884,
      "epoch": 7.201828438047072,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013265179423019576,
      "loss": 4.602,
      "mean_token_accuracy": 0.24598790109157562,
      "num_tokens": 160547905.0,
      "step": 92565
    },
    {
      "entropy": 5.376680946350097,
      "epoch": 7.20221746741879,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013263033150701873,
      "loss": 4.5256,
      "mean_token_accuracy": 0.25060620009899137,
      "num_tokens": 160556483.0,
      "step": 92570
    },
    {
      "entropy": 5.4358710765838625,
      "epoch": 7.202606496790508,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013260887094407471,
      "loss": 4.7057,
      "mean_token_accuracy": 0.22855093330144882,
      "num_tokens": 160564740.0,
      "step": 92575
    },
    {
      "entropy": 5.376035118103028,
      "epoch": 7.202995526162225,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013258741254168954,
      "loss": 4.6548,
      "mean_token_accuracy": 0.23600477278232573,
      "num_tokens": 160573670.0,
      "step": 92580
    },
    {
      "entropy": 5.403229665756226,
      "epoch": 7.203384555533943,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00013256595630018863,
      "loss": 4.6776,
      "mean_token_accuracy": 0.23997832387685775,
      "num_tokens": 160582622.0,
      "step": 92585
    },
    {
      "entropy": 5.330080270767212,
      "epoch": 7.203773584905661,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013254450221989773,
      "loss": 4.5162,
      "mean_token_accuracy": 0.250907164812088,
      "num_tokens": 160591274.0,
      "step": 92590
    },
    {
      "entropy": 5.316724538803101,
      "epoch": 7.204162614277378,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001325230503011422,
      "loss": 4.5828,
      "mean_token_accuracy": 0.2536097079515457,
      "num_tokens": 160599648.0,
      "step": 92595
    },
    {
      "entropy": 5.32495231628418,
      "epoch": 7.204551643649095,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001325016005442475,
      "loss": 4.539,
      "mean_token_accuracy": 0.2522721663117409,
      "num_tokens": 160607490.0,
      "step": 92600
    },
    {
      "entropy": 5.39938907623291,
      "epoch": 7.204940673020813,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013248015294953935,
      "loss": 4.6294,
      "mean_token_accuracy": 0.24212455451488496,
      "num_tokens": 160615870.0,
      "step": 92605
    },
    {
      "entropy": 5.36916561126709,
      "epoch": 7.205329702392531,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013245870751734296,
      "loss": 4.5409,
      "mean_token_accuracy": 0.24709579944610596,
      "num_tokens": 160624123.0,
      "step": 92610
    },
    {
      "entropy": 5.340246152877808,
      "epoch": 7.205718731764248,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001324372642479839,
      "loss": 4.5857,
      "mean_token_accuracy": 0.2515995889902115,
      "num_tokens": 160633322.0,
      "step": 92615
    },
    {
      "entropy": 5.459691572189331,
      "epoch": 7.206107761135966,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001324158231417874,
      "loss": 4.7535,
      "mean_token_accuracy": 0.2346562400460243,
      "num_tokens": 160641865.0,
      "step": 92620
    },
    {
      "entropy": 5.3782069206237795,
      "epoch": 7.206496790507684,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00013239438419907878,
      "loss": 4.6097,
      "mean_token_accuracy": 0.25325213372707367,
      "num_tokens": 160651156.0,
      "step": 92625
    },
    {
      "entropy": 5.326267576217651,
      "epoch": 7.206885819879401,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013237294742018347,
      "loss": 4.5231,
      "mean_token_accuracy": 0.2503870651125908,
      "num_tokens": 160659741.0,
      "step": 92630
    },
    {
      "entropy": 5.43643012046814,
      "epoch": 7.207274849251118,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00013235151280542663,
      "loss": 4.6451,
      "mean_token_accuracy": 0.24275488406419754,
      "num_tokens": 160668974.0,
      "step": 92635
    },
    {
      "entropy": 5.376981258392334,
      "epoch": 7.207663878622836,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00013233008035513355,
      "loss": 4.6052,
      "mean_token_accuracy": 0.24843924343585969,
      "num_tokens": 160677739.0,
      "step": 92640
    },
    {
      "entropy": 5.372924852371216,
      "epoch": 7.208052907994554,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001323086500696294,
      "loss": 4.6404,
      "mean_token_accuracy": 0.24831223338842393,
      "num_tokens": 160686121.0,
      "step": 92645
    },
    {
      "entropy": 5.463928270339966,
      "epoch": 7.208441937366271,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013228722194923943,
      "loss": 4.7402,
      "mean_token_accuracy": 0.2355212613940239,
      "num_tokens": 160695868.0,
      "step": 92650
    },
    {
      "entropy": 5.423352909088135,
      "epoch": 7.208830966737989,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001322657959942886,
      "loss": 4.6383,
      "mean_token_accuracy": 0.24114654958248138,
      "num_tokens": 160704865.0,
      "step": 92655
    },
    {
      "entropy": 5.384922456741333,
      "epoch": 7.209219996109706,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013224437220510227,
      "loss": 4.5759,
      "mean_token_accuracy": 0.25256330966949464,
      "num_tokens": 160713560.0,
      "step": 92660
    },
    {
      "entropy": 5.303297328948974,
      "epoch": 7.209609025481424,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013222295058200526,
      "loss": 4.5238,
      "mean_token_accuracy": 0.26012155413627625,
      "num_tokens": 160723653.0,
      "step": 92665
    },
    {
      "entropy": 5.3452128887176515,
      "epoch": 7.209998054853141,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013220153112532268,
      "loss": 4.5135,
      "mean_token_accuracy": 0.24808261543512344,
      "num_tokens": 160731830.0,
      "step": 92670
    },
    {
      "entropy": 5.464673662185669,
      "epoch": 7.210387084224859,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001321801138353797,
      "loss": 4.6914,
      "mean_token_accuracy": 0.24500210136175155,
      "num_tokens": 160740281.0,
      "step": 92675
    },
    {
      "entropy": 5.325867223739624,
      "epoch": 7.210776113596577,
      "grad_norm": 1.296875,
      "learning_rate": 0.00013215869871250114,
      "loss": 4.4868,
      "mean_token_accuracy": 0.25682717710733416,
      "num_tokens": 160748895.0,
      "step": 92680
    },
    {
      "entropy": 5.223721885681153,
      "epoch": 7.211165142968294,
      "grad_norm": 1.421875,
      "learning_rate": 0.000132137285757012,
      "loss": 4.3677,
      "mean_token_accuracy": 0.2657358780503273,
      "num_tokens": 160757989.0,
      "step": 92685
    },
    {
      "entropy": 5.2762144088745115,
      "epoch": 7.211554172340012,
      "grad_norm": 1.21875,
      "learning_rate": 0.00013211587496923704,
      "loss": 4.4553,
      "mean_token_accuracy": 0.2630413696169853,
      "num_tokens": 160767585.0,
      "step": 92690
    },
    {
      "entropy": 5.34943470954895,
      "epoch": 7.211943201711729,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001320944663495014,
      "loss": 4.5735,
      "mean_token_accuracy": 0.2493397504091263,
      "num_tokens": 160776007.0,
      "step": 92695
    },
    {
      "entropy": 5.3272223472595215,
      "epoch": 7.212332231083447,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00013207305989812974,
      "loss": 4.5622,
      "mean_token_accuracy": 0.25202645659446715,
      "num_tokens": 160784325.0,
      "step": 92700
    },
    {
      "entropy": 5.410389614105225,
      "epoch": 7.212721260455164,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00013205165561544686,
      "loss": 4.6621,
      "mean_token_accuracy": 0.24264848977327347,
      "num_tokens": 160792804.0,
      "step": 92705
    },
    {
      "entropy": 5.315902376174927,
      "epoch": 7.213110289826882,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013203025350177762,
      "loss": 4.5078,
      "mean_token_accuracy": 0.24905136972665787,
      "num_tokens": 160801287.0,
      "step": 92710
    },
    {
      "entropy": 5.40170693397522,
      "epoch": 7.2134993191986,
      "grad_norm": 1.46875,
      "learning_rate": 0.00013200885355744665,
      "loss": 4.645,
      "mean_token_accuracy": 0.23643695414066315,
      "num_tokens": 160809564.0,
      "step": 92715
    },
    {
      "entropy": 5.2941618919372555,
      "epoch": 7.213888348570317,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013198745578277884,
      "loss": 4.5287,
      "mean_token_accuracy": 0.25108041167259215,
      "num_tokens": 160817779.0,
      "step": 92720
    },
    {
      "entropy": 5.420750904083252,
      "epoch": 7.214277377942035,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0001319660601780987,
      "loss": 4.6994,
      "mean_token_accuracy": 0.24000463634729385,
      "num_tokens": 160825733.0,
      "step": 92725
    },
    {
      "entropy": 5.272272205352783,
      "epoch": 7.214666407313752,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001319446667437309,
      "loss": 4.4731,
      "mean_token_accuracy": 0.25085352808237077,
      "num_tokens": 160834787.0,
      "step": 92730
    },
    {
      "entropy": 5.284072303771973,
      "epoch": 7.21505543668547,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001319232754800001,
      "loss": 4.5315,
      "mean_token_accuracy": 0.2474823623895645,
      "num_tokens": 160843499.0,
      "step": 92735
    },
    {
      "entropy": 5.428152656555175,
      "epoch": 7.215444466057187,
      "grad_norm": 1.5,
      "learning_rate": 0.0001319018863872309,
      "loss": 4.6825,
      "mean_token_accuracy": 0.24184058904647826,
      "num_tokens": 160852074.0,
      "step": 92740
    },
    {
      "entropy": 5.383679437637329,
      "epoch": 7.215833495428905,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001318804994657478,
      "loss": 4.6061,
      "mean_token_accuracy": 0.24806584268808365,
      "num_tokens": 160860608.0,
      "step": 92745
    },
    {
      "entropy": 5.413587379455566,
      "epoch": 7.216222524800623,
      "grad_norm": 1.375,
      "learning_rate": 0.0001318591147158753,
      "loss": 4.6118,
      "mean_token_accuracy": 0.24971940964460373,
      "num_tokens": 160869217.0,
      "step": 92750
    },
    {
      "entropy": 5.4731425762176515,
      "epoch": 7.21661155417234,
      "grad_norm": 1.546875,
      "learning_rate": 0.000131837732137938,
      "loss": 4.6795,
      "mean_token_accuracy": 0.244609996676445,
      "num_tokens": 160877222.0,
      "step": 92755
    },
    {
      "entropy": 5.4110054016113285,
      "epoch": 7.217000583544057,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013181635173226008,
      "loss": 4.6053,
      "mean_token_accuracy": 0.23623024076223373,
      "num_tokens": 160885876.0,
      "step": 92760
    },
    {
      "entropy": 5.395467805862427,
      "epoch": 7.217389612915775,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013179497349916625,
      "loss": 4.655,
      "mean_token_accuracy": 0.23777615576982497,
      "num_tokens": 160894295.0,
      "step": 92765
    },
    {
      "entropy": 5.3823799133300785,
      "epoch": 7.2177786422874926,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001317735974389807,
      "loss": 4.6428,
      "mean_token_accuracy": 0.23799294233322144,
      "num_tokens": 160903503.0,
      "step": 92770
    },
    {
      "entropy": 5.372606039047241,
      "epoch": 7.21816767165921,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013175222355202792,
      "loss": 4.5702,
      "mean_token_accuracy": 0.24836839288473128,
      "num_tokens": 160912060.0,
      "step": 92775
    },
    {
      "entropy": 5.3608283519744875,
      "epoch": 7.218556701030928,
      "grad_norm": 1.25,
      "learning_rate": 0.00013173085183863212,
      "loss": 4.5851,
      "mean_token_accuracy": 0.24413824677467347,
      "num_tokens": 160920777.0,
      "step": 92780
    },
    {
      "entropy": 5.4142923831939695,
      "epoch": 7.218945730402646,
      "grad_norm": 1.375,
      "learning_rate": 0.00013170948229911756,
      "loss": 4.6413,
      "mean_token_accuracy": 0.2441364511847496,
      "num_tokens": 160929377.0,
      "step": 92785
    },
    {
      "entropy": 5.333751392364502,
      "epoch": 7.219334759774363,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001316881149338086,
      "loss": 4.475,
      "mean_token_accuracy": 0.2568366974592209,
      "num_tokens": 160937127.0,
      "step": 92790
    },
    {
      "entropy": 5.359666061401367,
      "epoch": 7.21972378914608,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00013166674974302928,
      "loss": 4.6167,
      "mean_token_accuracy": 0.248251248896122,
      "num_tokens": 160946239.0,
      "step": 92795
    },
    {
      "entropy": 5.2943785190582275,
      "epoch": 7.220112818517798,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001316453867271039,
      "loss": 4.5511,
      "mean_token_accuracy": 0.2480513021349907,
      "num_tokens": 160954518.0,
      "step": 92800
    },
    {
      "entropy": 5.4204699993133545,
      "epoch": 7.2205018478895155,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013162402588635664,
      "loss": 4.6742,
      "mean_token_accuracy": 0.2394874021410942,
      "num_tokens": 160962653.0,
      "step": 92805
    },
    {
      "entropy": 5.392215585708618,
      "epoch": 7.220890877261233,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013160266722111167,
      "loss": 4.6087,
      "mean_token_accuracy": 0.2468222752213478,
      "num_tokens": 160971130.0,
      "step": 92810
    },
    {
      "entropy": 5.335342741012573,
      "epoch": 7.221279906632951,
      "grad_norm": 1.375,
      "learning_rate": 0.00013158131073169295,
      "loss": 4.5735,
      "mean_token_accuracy": 0.24484187960624695,
      "num_tokens": 160980174.0,
      "step": 92815
    },
    {
      "entropy": 5.368626070022583,
      "epoch": 7.221668936004669,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001315599564184245,
      "loss": 4.5578,
      "mean_token_accuracy": 0.24539077430963516,
      "num_tokens": 160988937.0,
      "step": 92820
    },
    {
      "entropy": 5.4161317348480225,
      "epoch": 7.222057965376386,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001315386042816305,
      "loss": 4.6108,
      "mean_token_accuracy": 0.24908680319786072,
      "num_tokens": 160997962.0,
      "step": 92825
    },
    {
      "entropy": 5.455827379226685,
      "epoch": 7.222446994748103,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001315172543216347,
      "loss": 4.6649,
      "mean_token_accuracy": 0.2327004998922348,
      "num_tokens": 161006876.0,
      "step": 92830
    },
    {
      "entropy": 5.306347751617432,
      "epoch": 7.222836024119821,
      "grad_norm": 1.5,
      "learning_rate": 0.00013149590653876135,
      "loss": 4.544,
      "mean_token_accuracy": 0.25749264657497406,
      "num_tokens": 161014801.0,
      "step": 92835
    },
    {
      "entropy": 5.282178068161011,
      "epoch": 7.2232250534915385,
      "grad_norm": 1.265625,
      "learning_rate": 0.00013147456093333407,
      "loss": 4.5286,
      "mean_token_accuracy": 0.24960041642189026,
      "num_tokens": 161023740.0,
      "step": 92840
    },
    {
      "entropy": 5.344103050231934,
      "epoch": 7.223614082863256,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013145321750567695,
      "loss": 4.5602,
      "mean_token_accuracy": 0.2493737295269966,
      "num_tokens": 161032457.0,
      "step": 92845
    },
    {
      "entropy": 5.35132737159729,
      "epoch": 7.224003112234974,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00013143187625611375,
      "loss": 4.6286,
      "mean_token_accuracy": 0.24160081148147583,
      "num_tokens": 161041131.0,
      "step": 92850
    },
    {
      "entropy": 5.392951345443725,
      "epoch": 7.224392141606692,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00013141053718496837,
      "loss": 4.6978,
      "mean_token_accuracy": 0.23786876499652862,
      "num_tokens": 161049711.0,
      "step": 92855
    },
    {
      "entropy": 5.431059455871582,
      "epoch": 7.224781170978408,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013138920029256445,
      "loss": 4.6754,
      "mean_token_accuracy": 0.2304826036095619,
      "num_tokens": 161058128.0,
      "step": 92860
    },
    {
      "entropy": 5.298126649856568,
      "epoch": 7.225170200350126,
      "grad_norm": 1.265625,
      "learning_rate": 0.00013136786557922585,
      "loss": 4.4735,
      "mean_token_accuracy": 0.2624309331178665,
      "num_tokens": 161066700.0,
      "step": 92865
    },
    {
      "entropy": 5.280375528335571,
      "epoch": 7.225559229721844,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001313465330452764,
      "loss": 4.4998,
      "mean_token_accuracy": 0.2503114178776741,
      "num_tokens": 161075020.0,
      "step": 92870
    },
    {
      "entropy": 5.369185066223144,
      "epoch": 7.2259482590935615,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013132520269103947,
      "loss": 4.557,
      "mean_token_accuracy": 0.25066802501678465,
      "num_tokens": 161084471.0,
      "step": 92875
    },
    {
      "entropy": 5.400961637496948,
      "epoch": 7.226337288465279,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013130387451683908,
      "loss": 4.6336,
      "mean_token_accuracy": 0.24023017436265945,
      "num_tokens": 161094123.0,
      "step": 92880
    },
    {
      "entropy": 5.4293739795684814,
      "epoch": 7.226726317836997,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001312825485229986,
      "loss": 4.6018,
      "mean_token_accuracy": 0.24661445021629333,
      "num_tokens": 161102501.0,
      "step": 92885
    },
    {
      "entropy": 5.336338233947754,
      "epoch": 7.227115347208715,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013126122470984172,
      "loss": 4.6071,
      "mean_token_accuracy": 0.24096372872591018,
      "num_tokens": 161110982.0,
      "step": 92890
    },
    {
      "entropy": 5.384690856933593,
      "epoch": 7.227504376580431,
      "grad_norm": 1.34375,
      "learning_rate": 0.000131239903077692,
      "loss": 4.6165,
      "mean_token_accuracy": 0.2364717811346054,
      "num_tokens": 161119333.0,
      "step": 92895
    },
    {
      "entropy": 5.305839872360229,
      "epoch": 7.227893405952149,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013121858362687285,
      "loss": 4.5347,
      "mean_token_accuracy": 0.247261443734169,
      "num_tokens": 161127580.0,
      "step": 92900
    },
    {
      "entropy": 5.2867790222167965,
      "epoch": 7.228282435323867,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013119726635770797,
      "loss": 4.5011,
      "mean_token_accuracy": 0.2608225464820862,
      "num_tokens": 161136114.0,
      "step": 92905
    },
    {
      "entropy": 5.32884955406189,
      "epoch": 7.2286714646955845,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001311759512705205,
      "loss": 4.4929,
      "mean_token_accuracy": 0.2572451338171959,
      "num_tokens": 161144012.0,
      "step": 92910
    },
    {
      "entropy": 5.396964073181152,
      "epoch": 7.229060494067302,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013115463836563425,
      "loss": 4.6859,
      "mean_token_accuracy": 0.2353859469294548,
      "num_tokens": 161152656.0,
      "step": 92915
    },
    {
      "entropy": 5.363293218612671,
      "epoch": 7.22944952343902,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00013113332764337225,
      "loss": 4.6301,
      "mean_token_accuracy": 0.2407084360718727,
      "num_tokens": 161161818.0,
      "step": 92920
    },
    {
      "entropy": 5.415321350097656,
      "epoch": 7.229838552810738,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013111201910405809,
      "loss": 4.6021,
      "mean_token_accuracy": 0.24947014600038528,
      "num_tokens": 161170179.0,
      "step": 92925
    },
    {
      "entropy": 5.400695371627807,
      "epoch": 7.230227582182454,
      "grad_norm": 1.296875,
      "learning_rate": 0.00013109071274801498,
      "loss": 4.6366,
      "mean_token_accuracy": 0.24518899172544478,
      "num_tokens": 161179262.0,
      "step": 92930
    },
    {
      "entropy": 5.384346342086792,
      "epoch": 7.230616611554172,
      "grad_norm": 1.234375,
      "learning_rate": 0.00013106940857556624,
      "loss": 4.6223,
      "mean_token_accuracy": 0.24664367884397506,
      "num_tokens": 161188413.0,
      "step": 92935
    },
    {
      "entropy": 5.397604608535767,
      "epoch": 7.23100564092589,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001310481065870352,
      "loss": 4.5576,
      "mean_token_accuracy": 0.25470231771469115,
      "num_tokens": 161195703.0,
      "step": 92940
    },
    {
      "entropy": 5.431394433975219,
      "epoch": 7.2313946702976075,
      "grad_norm": 1.484375,
      "learning_rate": 0.00013102680678274493,
      "loss": 4.6618,
      "mean_token_accuracy": 0.2478412300348282,
      "num_tokens": 161204055.0,
      "step": 92945
    },
    {
      "entropy": 5.355821800231934,
      "epoch": 7.231783699669325,
      "grad_norm": 1.28125,
      "learning_rate": 0.00013100550916301883,
      "loss": 4.5704,
      "mean_token_accuracy": 0.24886807799339294,
      "num_tokens": 161212081.0,
      "step": 92950
    },
    {
      "entropy": 5.381619882583618,
      "epoch": 7.232172729041043,
      "grad_norm": 1.296875,
      "learning_rate": 0.00013098421372817983,
      "loss": 4.6271,
      "mean_token_accuracy": 0.2370375081896782,
      "num_tokens": 161221282.0,
      "step": 92955
    },
    {
      "entropy": 5.400146245956421,
      "epoch": 7.232561758412761,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013096292047855127,
      "loss": 4.6812,
      "mean_token_accuracy": 0.23739182353019714,
      "num_tokens": 161230979.0,
      "step": 92960
    },
    {
      "entropy": 5.329092073440552,
      "epoch": 7.232950787784477,
      "grad_norm": 1.484375,
      "learning_rate": 0.00013094162941445603,
      "loss": 4.53,
      "mean_token_accuracy": 0.2553288385272026,
      "num_tokens": 161239148.0,
      "step": 92965
    },
    {
      "entropy": 5.329554891586303,
      "epoch": 7.233339817156195,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001309203405362173,
      "loss": 4.5338,
      "mean_token_accuracy": 0.2553807407617569,
      "num_tokens": 161247808.0,
      "step": 92970
    },
    {
      "entropy": 5.3601250648498535,
      "epoch": 7.233728846527913,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013089905384415818,
      "loss": 4.5478,
      "mean_token_accuracy": 0.2537130728363991,
      "num_tokens": 161256784.0,
      "step": 92975
    },
    {
      "entropy": 5.359031295776367,
      "epoch": 7.2341178758996305,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001308777693386014,
      "loss": 4.6172,
      "mean_token_accuracy": 0.24877254217863082,
      "num_tokens": 161265773.0,
      "step": 92980
    },
    {
      "entropy": 5.402850914001465,
      "epoch": 7.234506905271348,
      "grad_norm": 1.296875,
      "learning_rate": 0.00013085648701987017,
      "loss": 4.6568,
      "mean_token_accuracy": 0.2348551020026207,
      "num_tokens": 161274676.0,
      "step": 92985
    },
    {
      "entropy": 5.424603700637817,
      "epoch": 7.234895934643066,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013083520688828725,
      "loss": 4.7122,
      "mean_token_accuracy": 0.23791230618953704,
      "num_tokens": 161283256.0,
      "step": 92990
    },
    {
      "entropy": 5.427352094650269,
      "epoch": 7.235284964014783,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00013081392894417565,
      "loss": 4.6658,
      "mean_token_accuracy": 0.24176755994558335,
      "num_tokens": 161292240.0,
      "step": 92995
    },
    {
      "entropy": 5.42385048866272,
      "epoch": 7.2356739933865,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001307926531878582,
      "loss": 4.5891,
      "mean_token_accuracy": 0.2505391076207161,
      "num_tokens": 161300817.0,
      "step": 93000
    },
    {
      "epoch": 7.2356739933865,
      "eval_entropy": 5.117161943455862,
      "eval_loss": 4.848870277404785,
      "eval_mean_token_accuracy": 0.23732912991773175,
      "eval_num_tokens": 161300817.0,
      "eval_runtime": 28.6143,
      "eval_samples_per_second": 1452.352,
      "eval_steps_per_second": 181.553,
      "step": 93000
    },
    {
      "entropy": 5.3514879703521725,
      "epoch": 7.236063022758218,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001307713796196577,
      "loss": 4.6209,
      "mean_token_accuracy": 0.24437925964593887,
      "num_tokens": 161309032.0,
      "step": 93005
    },
    {
      "entropy": 5.355354833602905,
      "epoch": 7.236452052129936,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013075010823989695,
      "loss": 4.6686,
      "mean_token_accuracy": 0.2417302832007408,
      "num_tokens": 161317913.0,
      "step": 93010
    },
    {
      "entropy": 5.318651580810547,
      "epoch": 7.2368410815016535,
      "grad_norm": 1.25,
      "learning_rate": 0.00013072883904889865,
      "loss": 4.5286,
      "mean_token_accuracy": 0.25664413422346116,
      "num_tokens": 161326533.0,
      "step": 93015
    },
    {
      "entropy": 5.372893524169922,
      "epoch": 7.237230110873371,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013070757204698566,
      "loss": 4.5536,
      "mean_token_accuracy": 0.2523710384964943,
      "num_tokens": 161335817.0,
      "step": 93020
    },
    {
      "entropy": 5.325049495697021,
      "epoch": 7.237619140245089,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013068630723448056,
      "loss": 4.5762,
      "mean_token_accuracy": 0.24989894926548004,
      "num_tokens": 161344169.0,
      "step": 93025
    },
    {
      "entropy": 5.361840486526489,
      "epoch": 7.238008169616806,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001306650446117061,
      "loss": 4.5403,
      "mean_token_accuracy": 0.25378065556287766,
      "num_tokens": 161354791.0,
      "step": 93030
    },
    {
      "entropy": 5.311911153793335,
      "epoch": 7.238397198988523,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013064378417898477,
      "loss": 4.5343,
      "mean_token_accuracy": 0.2469213515520096,
      "num_tokens": 161363175.0,
      "step": 93035
    },
    {
      "entropy": 5.371565961837769,
      "epoch": 7.238786228360241,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013062252593663934,
      "loss": 4.573,
      "mean_token_accuracy": 0.24588743597269058,
      "num_tokens": 161371543.0,
      "step": 93040
    },
    {
      "entropy": 5.461717939376831,
      "epoch": 7.239175257731959,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013060126988499222,
      "loss": 4.7475,
      "mean_token_accuracy": 0.23391314297914506,
      "num_tokens": 161380048.0,
      "step": 93045
    },
    {
      "entropy": 5.383434247970581,
      "epoch": 7.2395642871036765,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000130580016024366,
      "loss": 4.6698,
      "mean_token_accuracy": 0.243982295691967,
      "num_tokens": 161389382.0,
      "step": 93050
    },
    {
      "entropy": 5.3832862854003904,
      "epoch": 7.239953316475394,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013055876435508323,
      "loss": 4.6168,
      "mean_token_accuracy": 0.2482757732272148,
      "num_tokens": 161398294.0,
      "step": 93055
    },
    {
      "entropy": 5.324546813964844,
      "epoch": 7.240342345847111,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013053751487746622,
      "loss": 4.528,
      "mean_token_accuracy": 0.24760883599519729,
      "num_tokens": 161406433.0,
      "step": 93060
    },
    {
      "entropy": 5.252767419815063,
      "epoch": 7.240731375218829,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013051626759183748,
      "loss": 4.4182,
      "mean_token_accuracy": 0.26350547969341276,
      "num_tokens": 161414780.0,
      "step": 93065
    },
    {
      "entropy": 5.325701808929443,
      "epoch": 7.241120404590546,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013049502249851944,
      "loss": 4.5325,
      "mean_token_accuracy": 0.2576692759990692,
      "num_tokens": 161423666.0,
      "step": 93070
    },
    {
      "entropy": 5.395560884475708,
      "epoch": 7.241509433962264,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00013047377959783449,
      "loss": 4.6398,
      "mean_token_accuracy": 0.24152043759822844,
      "num_tokens": 161432062.0,
      "step": 93075
    },
    {
      "entropy": 5.318017816543579,
      "epoch": 7.241898463333982,
      "grad_norm": 1.375,
      "learning_rate": 0.00013045253889010488,
      "loss": 4.5613,
      "mean_token_accuracy": 0.24792979210615157,
      "num_tokens": 161440206.0,
      "step": 93080
    },
    {
      "entropy": 5.324124336242676,
      "epoch": 7.2422874927056995,
      "grad_norm": 1.28125,
      "learning_rate": 0.00013043130037565284,
      "loss": 4.6001,
      "mean_token_accuracy": 0.24896049052476882,
      "num_tokens": 161448993.0,
      "step": 93085
    },
    {
      "entropy": 5.261946725845337,
      "epoch": 7.242676522077417,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00013041006405480082,
      "loss": 4.4521,
      "mean_token_accuracy": 0.2586507469415665,
      "num_tokens": 161457782.0,
      "step": 93090
    },
    {
      "entropy": 5.46934666633606,
      "epoch": 7.243065551449134,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00013038882992787084,
      "loss": 4.7216,
      "mean_token_accuracy": 0.23379883468151091,
      "num_tokens": 161465864.0,
      "step": 93095
    },
    {
      "entropy": 5.385258579254151,
      "epoch": 7.243454580820852,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013036759799518526,
      "loss": 4.6274,
      "mean_token_accuracy": 0.23793036937713624,
      "num_tokens": 161474023.0,
      "step": 93100
    },
    {
      "entropy": 5.3992713451385494,
      "epoch": 7.243843610192569,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013034636825706613,
      "loss": 4.7121,
      "mean_token_accuracy": 0.23404035717248917,
      "num_tokens": 161482914.0,
      "step": 93105
    },
    {
      "entropy": 5.391446876525879,
      "epoch": 7.244232639564287,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001303251407138357,
      "loss": 4.6088,
      "mean_token_accuracy": 0.24092815220355987,
      "num_tokens": 161491662.0,
      "step": 93110
    },
    {
      "entropy": 5.441685485839844,
      "epoch": 7.244621668936005,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013030391536581588,
      "loss": 4.6644,
      "mean_token_accuracy": 0.23856064975261687,
      "num_tokens": 161500218.0,
      "step": 93115
    },
    {
      "entropy": 5.336873388290405,
      "epoch": 7.2450106983077225,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013028269221332896,
      "loss": 4.5461,
      "mean_token_accuracy": 0.25544821172952653,
      "num_tokens": 161508847.0,
      "step": 93120
    },
    {
      "entropy": 5.223408842086792,
      "epoch": 7.24539972767944,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00013026147125669668,
      "loss": 4.4332,
      "mean_token_accuracy": 0.25919142067432405,
      "num_tokens": 161516454.0,
      "step": 93125
    },
    {
      "entropy": 5.307506895065307,
      "epoch": 7.245788757051157,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013024025249624128,
      "loss": 4.5523,
      "mean_token_accuracy": 0.245629745721817,
      "num_tokens": 161524174.0,
      "step": 93130
    },
    {
      "entropy": 5.445923328399658,
      "epoch": 7.246177786422875,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013021903593228468,
      "loss": 4.7423,
      "mean_token_accuracy": 0.23531498908996581,
      "num_tokens": 161533340.0,
      "step": 93135
    },
    {
      "entropy": 5.439331436157227,
      "epoch": 7.246566815794592,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001301978215651487,
      "loss": 4.7689,
      "mean_token_accuracy": 0.23754979819059371,
      "num_tokens": 161542703.0,
      "step": 93140
    },
    {
      "entropy": 5.393563222885132,
      "epoch": 7.24695584516631,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00013017660939515535,
      "loss": 4.6218,
      "mean_token_accuracy": 0.24551842510700225,
      "num_tokens": 161550818.0,
      "step": 93145
    },
    {
      "entropy": 5.367584133148194,
      "epoch": 7.247344874538028,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013015539942262642,
      "loss": 4.631,
      "mean_token_accuracy": 0.2403274118900299,
      "num_tokens": 161559897.0,
      "step": 93150
    },
    {
      "entropy": 5.349287366867065,
      "epoch": 7.2477339039097455,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013013419164788383,
      "loss": 4.628,
      "mean_token_accuracy": 0.23320333361625672,
      "num_tokens": 161568711.0,
      "step": 93155
    },
    {
      "entropy": 5.352972841262817,
      "epoch": 7.248122933281463,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001301129860712492,
      "loss": 4.5707,
      "mean_token_accuracy": 0.24405422508716584,
      "num_tokens": 161577485.0,
      "step": 93160
    },
    {
      "entropy": 5.368986797332764,
      "epoch": 7.24851196265318,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013009178269304448,
      "loss": 4.5951,
      "mean_token_accuracy": 0.24606360495090485,
      "num_tokens": 161586349.0,
      "step": 93165
    },
    {
      "entropy": 5.369545936584473,
      "epoch": 7.248900992024898,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001300705815135913,
      "loss": 4.5946,
      "mean_token_accuracy": 0.25031156092882156,
      "num_tokens": 161595382.0,
      "step": 93170
    },
    {
      "entropy": 5.31597638130188,
      "epoch": 7.249290021396615,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001300493825332113,
      "loss": 4.6062,
      "mean_token_accuracy": 0.24337042719125748,
      "num_tokens": 161604570.0,
      "step": 93175
    },
    {
      "entropy": 5.378845834732056,
      "epoch": 7.249679050768333,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001300281857522263,
      "loss": 4.6176,
      "mean_token_accuracy": 0.24753223806619645,
      "num_tokens": 161613301.0,
      "step": 93180
    },
    {
      "entropy": 5.42596664428711,
      "epoch": 7.250068080140051,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001300069911709577,
      "loss": 4.6668,
      "mean_token_accuracy": 0.24765763133764268,
      "num_tokens": 161621612.0,
      "step": 93185
    },
    {
      "entropy": 5.2650309085845945,
      "epoch": 7.2504571095117685,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001299857987897274,
      "loss": 4.5693,
      "mean_token_accuracy": 0.24019169360399245,
      "num_tokens": 161630092.0,
      "step": 93190
    },
    {
      "entropy": 5.325260639190674,
      "epoch": 7.250846138883485,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001299646086088566,
      "loss": 4.5347,
      "mean_token_accuracy": 0.2549667716026306,
      "num_tokens": 161638348.0,
      "step": 93195
    },
    {
      "entropy": 5.4125528812408445,
      "epoch": 7.251235168255203,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001299434206286671,
      "loss": 4.6874,
      "mean_token_accuracy": 0.23951479494571687,
      "num_tokens": 161647375.0,
      "step": 93200
    },
    {
      "entropy": 5.377710342407227,
      "epoch": 7.251624197626921,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012992223484948036,
      "loss": 4.5985,
      "mean_token_accuracy": 0.24648675620555877,
      "num_tokens": 161656320.0,
      "step": 93205
    },
    {
      "entropy": 5.307938766479492,
      "epoch": 7.252013226998638,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012990105127161773,
      "loss": 4.5499,
      "mean_token_accuracy": 0.24703573137521745,
      "num_tokens": 161664310.0,
      "step": 93210
    },
    {
      "entropy": 5.4520317077636715,
      "epoch": 7.252402256370356,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012987986989540076,
      "loss": 4.6943,
      "mean_token_accuracy": 0.23632523119449617,
      "num_tokens": 161672297.0,
      "step": 93215
    },
    {
      "entropy": 5.26498646736145,
      "epoch": 7.252791285742074,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012985869072115072,
      "loss": 4.4512,
      "mean_token_accuracy": 0.2665297329425812,
      "num_tokens": 161681088.0,
      "step": 93220
    },
    {
      "entropy": 5.268145418167114,
      "epoch": 7.2531803151137915,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001298375137491891,
      "loss": 4.5084,
      "mean_token_accuracy": 0.25905380100011827,
      "num_tokens": 161689927.0,
      "step": 93225
    },
    {
      "entropy": 5.2751048564910885,
      "epoch": 7.253569344485508,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012981633897983708,
      "loss": 4.501,
      "mean_token_accuracy": 0.25445732325315473,
      "num_tokens": 161698665.0,
      "step": 93230
    },
    {
      "entropy": 5.360054016113281,
      "epoch": 7.253958373857226,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001297951664134161,
      "loss": 4.5967,
      "mean_token_accuracy": 0.242214173078537,
      "num_tokens": 161707668.0,
      "step": 93235
    },
    {
      "entropy": 5.408990812301636,
      "epoch": 7.254347403228944,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012977399605024732,
      "loss": 4.6586,
      "mean_token_accuracy": 0.2456701695919037,
      "num_tokens": 161717079.0,
      "step": 93240
    },
    {
      "entropy": 5.374428558349609,
      "epoch": 7.254736432600661,
      "grad_norm": 1.59375,
      "learning_rate": 0.00012975282789065202,
      "loss": 4.5713,
      "mean_token_accuracy": 0.25754389762878416,
      "num_tokens": 161724910.0,
      "step": 93245
    },
    {
      "entropy": 5.368446016311646,
      "epoch": 7.255125461972379,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012973166193495139,
      "loss": 4.6176,
      "mean_token_accuracy": 0.23908609449863433,
      "num_tokens": 161733044.0,
      "step": 93250
    },
    {
      "entropy": 5.30822377204895,
      "epoch": 7.255514491344097,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012971049818346646,
      "loss": 4.6154,
      "mean_token_accuracy": 0.24119185358285905,
      "num_tokens": 161741793.0,
      "step": 93255
    },
    {
      "entropy": 5.346643543243408,
      "epoch": 7.2559035207158145,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001296893366365186,
      "loss": 4.5515,
      "mean_token_accuracy": 0.25351144969463346,
      "num_tokens": 161750659.0,
      "step": 93260
    },
    {
      "entropy": 5.305310010910034,
      "epoch": 7.256292550087531,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001296681772944286,
      "loss": 4.5477,
      "mean_token_accuracy": 0.24555520415306092,
      "num_tokens": 161760217.0,
      "step": 93265
    },
    {
      "entropy": 5.269532346725464,
      "epoch": 7.256681579459249,
      "grad_norm": 1.28125,
      "learning_rate": 0.00012964702015751784,
      "loss": 4.5702,
      "mean_token_accuracy": 0.25058519691228864,
      "num_tokens": 161770484.0,
      "step": 93270
    },
    {
      "entropy": 5.401564502716065,
      "epoch": 7.257070608830967,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001296258652261072,
      "loss": 4.6364,
      "mean_token_accuracy": 0.24727412015199662,
      "num_tokens": 161778634.0,
      "step": 93275
    },
    {
      "entropy": 5.444714164733886,
      "epoch": 7.257459638202684,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012960471250051758,
      "loss": 4.6645,
      "mean_token_accuracy": 0.24420372396707535,
      "num_tokens": 161787233.0,
      "step": 93280
    },
    {
      "entropy": 5.430482959747314,
      "epoch": 7.257848667574402,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012958356198107014,
      "loss": 4.6042,
      "mean_token_accuracy": 0.24489883035421373,
      "num_tokens": 161795839.0,
      "step": 93285
    },
    {
      "entropy": 5.316358613967895,
      "epoch": 7.25823769694612,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012956241366808558,
      "loss": 4.525,
      "mean_token_accuracy": 0.25635805875062945,
      "num_tokens": 161805439.0,
      "step": 93290
    },
    {
      "entropy": 5.3432715892791744,
      "epoch": 7.2586267263178375,
      "grad_norm": 1.5,
      "learning_rate": 0.000129541267561885,
      "loss": 4.5847,
      "mean_token_accuracy": 0.2515170007944107,
      "num_tokens": 161813988.0,
      "step": 93295
    },
    {
      "entropy": 5.3332335472106935,
      "epoch": 7.259015755689554,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001295201236627891,
      "loss": 4.5933,
      "mean_token_accuracy": 0.24958566278219224,
      "num_tokens": 161822614.0,
      "step": 93300
    },
    {
      "entropy": 5.404656600952149,
      "epoch": 7.259404785061272,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001294989819711188,
      "loss": 4.6196,
      "mean_token_accuracy": 0.25517726242542266,
      "num_tokens": 161830993.0,
      "step": 93305
    },
    {
      "entropy": 5.351178503036499,
      "epoch": 7.25979381443299,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001294778424871948,
      "loss": 4.5996,
      "mean_token_accuracy": 0.24369374215602874,
      "num_tokens": 161839096.0,
      "step": 93310
    },
    {
      "entropy": 5.277793884277344,
      "epoch": 7.260182843804707,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012945670521133802,
      "loss": 4.4574,
      "mean_token_accuracy": 0.2554700434207916,
      "num_tokens": 161847951.0,
      "step": 93315
    },
    {
      "entropy": 5.367852687835693,
      "epoch": 7.260571873176425,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012943557014386898,
      "loss": 4.6289,
      "mean_token_accuracy": 0.2517382323741913,
      "num_tokens": 161856228.0,
      "step": 93320
    },
    {
      "entropy": 5.414979028701782,
      "epoch": 7.260960902548143,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012941443728510855,
      "loss": 4.682,
      "mean_token_accuracy": 0.24199674874544144,
      "num_tokens": 161864890.0,
      "step": 93325
    },
    {
      "entropy": 5.409875535964966,
      "epoch": 7.26134993191986,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012939330663537725,
      "loss": 4.6231,
      "mean_token_accuracy": 0.24039355516433716,
      "num_tokens": 161872864.0,
      "step": 93330
    },
    {
      "entropy": 5.369995546340943,
      "epoch": 7.261738961291577,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012937217819499575,
      "loss": 4.6239,
      "mean_token_accuracy": 0.24475659281015397,
      "num_tokens": 161881142.0,
      "step": 93335
    },
    {
      "entropy": 5.394140148162842,
      "epoch": 7.262127990663295,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012935105196428475,
      "loss": 4.679,
      "mean_token_accuracy": 0.24490994364023208,
      "num_tokens": 161890103.0,
      "step": 93340
    },
    {
      "entropy": 5.344478559494019,
      "epoch": 7.262517020035013,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012932992794356462,
      "loss": 4.5089,
      "mean_token_accuracy": 0.2574861317873001,
      "num_tokens": 161898495.0,
      "step": 93345
    },
    {
      "entropy": 5.40095362663269,
      "epoch": 7.26290604940673,
      "grad_norm": 1.28125,
      "learning_rate": 0.00012930880613315607,
      "loss": 4.618,
      "mean_token_accuracy": 0.24594858288764954,
      "num_tokens": 161906600.0,
      "step": 93350
    },
    {
      "entropy": 5.400445365905762,
      "epoch": 7.263295078778448,
      "grad_norm": 1.328125,
      "learning_rate": 0.00012928768653337947,
      "loss": 4.6579,
      "mean_token_accuracy": 0.2433096319437027,
      "num_tokens": 161914967.0,
      "step": 93355
    },
    {
      "entropy": 5.447196054458618,
      "epoch": 7.263684108150166,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012926656914455525,
      "loss": 4.7206,
      "mean_token_accuracy": 0.2378399446606636,
      "num_tokens": 161923699.0,
      "step": 93360
    },
    {
      "entropy": 5.4417140007019045,
      "epoch": 7.2640731375218826,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012924545396700394,
      "loss": 4.6627,
      "mean_token_accuracy": 0.2346501886844635,
      "num_tokens": 161931912.0,
      "step": 93365
    },
    {
      "entropy": 5.378168106079102,
      "epoch": 7.2644621668936,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001292243410010458,
      "loss": 4.6069,
      "mean_token_accuracy": 0.24344323128461837,
      "num_tokens": 161940405.0,
      "step": 93370
    },
    {
      "entropy": 5.3612902641296385,
      "epoch": 7.264851196265318,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012920323024700133,
      "loss": 4.5939,
      "mean_token_accuracy": 0.25001506209373475,
      "num_tokens": 161948943.0,
      "step": 93375
    },
    {
      "entropy": 5.330392694473266,
      "epoch": 7.265240225637036,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012918212170519072,
      "loss": 4.5455,
      "mean_token_accuracy": 0.25500842183828354,
      "num_tokens": 161956951.0,
      "step": 93380
    },
    {
      "entropy": 5.349444723129272,
      "epoch": 7.265629255008753,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001291610153759344,
      "loss": 4.5728,
      "mean_token_accuracy": 0.24413973689079285,
      "num_tokens": 161965267.0,
      "step": 93385
    },
    {
      "entropy": 5.342901563644409,
      "epoch": 7.266018284380471,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001291399112595525,
      "loss": 4.5404,
      "mean_token_accuracy": 0.2545383632183075,
      "num_tokens": 161972788.0,
      "step": 93390
    },
    {
      "entropy": 5.377701616287231,
      "epoch": 7.266407313752188,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00012911880935636522,
      "loss": 4.6302,
      "mean_token_accuracy": 0.2427387297153473,
      "num_tokens": 161982273.0,
      "step": 93395
    },
    {
      "entropy": 5.354363441467285,
      "epoch": 7.2667963431239055,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012909770966669292,
      "loss": 4.5795,
      "mean_token_accuracy": 0.24795849174261092,
      "num_tokens": 161991222.0,
      "step": 93400
    },
    {
      "entropy": 5.397664976119995,
      "epoch": 7.267185372495623,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001290766121908556,
      "loss": 4.6531,
      "mean_token_accuracy": 0.24305486530065537,
      "num_tokens": 162000329.0,
      "step": 93405
    },
    {
      "entropy": 5.274490547180176,
      "epoch": 7.267574401867341,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012905551692917345,
      "loss": 4.4913,
      "mean_token_accuracy": 0.26493613570928576,
      "num_tokens": 162009119.0,
      "step": 93410
    },
    {
      "entropy": 5.487717485427856,
      "epoch": 7.267963431239059,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001290344238819665,
      "loss": 4.7127,
      "mean_token_accuracy": 0.24073898494243623,
      "num_tokens": 162017743.0,
      "step": 93415
    },
    {
      "entropy": 5.394275426864624,
      "epoch": 7.268352460610776,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012901333304955487,
      "loss": 4.629,
      "mean_token_accuracy": 0.2452261745929718,
      "num_tokens": 162026091.0,
      "step": 93420
    },
    {
      "entropy": 5.308052253723145,
      "epoch": 7.268741489982494,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012899224443225853,
      "loss": 4.6075,
      "mean_token_accuracy": 0.24434154331684113,
      "num_tokens": 162035366.0,
      "step": 93425
    },
    {
      "entropy": 5.379477071762085,
      "epoch": 7.269130519354211,
      "grad_norm": 1.265625,
      "learning_rate": 0.00012897115803039752,
      "loss": 4.625,
      "mean_token_accuracy": 0.2426157847046852,
      "num_tokens": 162044229.0,
      "step": 93430
    },
    {
      "entropy": 5.313031435012817,
      "epoch": 7.2695195487259285,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00012895007384429178,
      "loss": 4.5771,
      "mean_token_accuracy": 0.251587949693203,
      "num_tokens": 162053650.0,
      "step": 93435
    },
    {
      "entropy": 5.3820047855377195,
      "epoch": 7.269908578097646,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0001289289918742611,
      "loss": 4.5603,
      "mean_token_accuracy": 0.2584817498922348,
      "num_tokens": 162063033.0,
      "step": 93440
    },
    {
      "entropy": 5.300290203094482,
      "epoch": 7.270297607469364,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001289079121206256,
      "loss": 4.6089,
      "mean_token_accuracy": 0.25136625170707705,
      "num_tokens": 162072650.0,
      "step": 93445
    },
    {
      "entropy": 5.480862855911255,
      "epoch": 7.270686636841082,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012888683458370492,
      "loss": 4.7159,
      "mean_token_accuracy": 0.2363581359386444,
      "num_tokens": 162081106.0,
      "step": 93450
    },
    {
      "entropy": 5.341135501861572,
      "epoch": 7.271075666212799,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000128865759263819,
      "loss": 4.5825,
      "mean_token_accuracy": 0.2511118933558464,
      "num_tokens": 162089652.0,
      "step": 93455
    },
    {
      "entropy": 5.300408220291137,
      "epoch": 7.271464695584517,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012884468616128749,
      "loss": 4.4903,
      "mean_token_accuracy": 0.25939127802848816,
      "num_tokens": 162097791.0,
      "step": 93460
    },
    {
      "entropy": 5.4042867660522464,
      "epoch": 7.271853724956234,
      "grad_norm": 1.328125,
      "learning_rate": 0.00012882361527643033,
      "loss": 4.6572,
      "mean_token_accuracy": 0.24083279818296432,
      "num_tokens": 162105997.0,
      "step": 93465
    },
    {
      "entropy": 5.2898821353912355,
      "epoch": 7.2722427543279515,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012880254660956714,
      "loss": 4.5549,
      "mean_token_accuracy": 0.25347730964422227,
      "num_tokens": 162114403.0,
      "step": 93470
    },
    {
      "entropy": 5.361013793945313,
      "epoch": 7.272631783699669,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001287814801610176,
      "loss": 4.6227,
      "mean_token_accuracy": 0.24590259194374084,
      "num_tokens": 162123511.0,
      "step": 93475
    },
    {
      "entropy": 5.313603353500366,
      "epoch": 7.273020813071387,
      "grad_norm": 1.5,
      "learning_rate": 0.0001287604159311014,
      "loss": 4.5608,
      "mean_token_accuracy": 0.2522977113723755,
      "num_tokens": 162132050.0,
      "step": 93480
    },
    {
      "entropy": 5.377662801742554,
      "epoch": 7.273409842443105,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012873935392013804,
      "loss": 4.5725,
      "mean_token_accuracy": 0.2519827738404274,
      "num_tokens": 162140110.0,
      "step": 93485
    },
    {
      "entropy": 5.3529825687408445,
      "epoch": 7.273798871814822,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012871829412844726,
      "loss": 4.6137,
      "mean_token_accuracy": 0.24381026178598403,
      "num_tokens": 162148759.0,
      "step": 93490
    },
    {
      "entropy": 5.314110565185547,
      "epoch": 7.27418790118654,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001286972365563485,
      "loss": 4.5418,
      "mean_token_accuracy": 0.25211970657110216,
      "num_tokens": 162156913.0,
      "step": 93495
    },
    {
      "entropy": 5.402555131912232,
      "epoch": 7.274576930558257,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012867618120416136,
      "loss": 4.6426,
      "mean_token_accuracy": 0.24447511583566667,
      "num_tokens": 162166514.0,
      "step": 93500
    },
    {
      "entropy": 5.361934995651245,
      "epoch": 7.2749659599299745,
      "grad_norm": 1.25,
      "learning_rate": 0.0001286551280722052,
      "loss": 4.5827,
      "mean_token_accuracy": 0.25179579854011536,
      "num_tokens": 162175474.0,
      "step": 93505
    },
    {
      "entropy": 5.385845804214478,
      "epoch": 7.275354989301692,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012863407716079962,
      "loss": 4.6607,
      "mean_token_accuracy": 0.24399127960205078,
      "num_tokens": 162184485.0,
      "step": 93510
    },
    {
      "entropy": 5.426646566390991,
      "epoch": 7.27574401867341,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00012861302847026393,
      "loss": 4.7119,
      "mean_token_accuracy": 0.23900941908359527,
      "num_tokens": 162193251.0,
      "step": 93515
    },
    {
      "entropy": 5.445105791091919,
      "epoch": 7.276133048045128,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001285919820009175,
      "loss": 4.6795,
      "mean_token_accuracy": 0.24055661261081696,
      "num_tokens": 162201676.0,
      "step": 93520
    },
    {
      "entropy": 5.372499465942383,
      "epoch": 7.276522077416845,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012857093775307975,
      "loss": 4.6222,
      "mean_token_accuracy": 0.24343554079532623,
      "num_tokens": 162210531.0,
      "step": 93525
    },
    {
      "entropy": 5.3678178787231445,
      "epoch": 7.276911106788562,
      "grad_norm": 1.375,
      "learning_rate": 0.00012854989572706982,
      "loss": 4.5958,
      "mean_token_accuracy": 0.2453724279999733,
      "num_tokens": 162219733.0,
      "step": 93530
    },
    {
      "entropy": 5.416921949386596,
      "epoch": 7.27730013616028,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012852885592320724,
      "loss": 4.6539,
      "mean_token_accuracy": 0.247358737885952,
      "num_tokens": 162228793.0,
      "step": 93535
    },
    {
      "entropy": 5.408203601837158,
      "epoch": 7.2776891655319975,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001285078183418111,
      "loss": 4.6426,
      "mean_token_accuracy": 0.24388593435287476,
      "num_tokens": 162237927.0,
      "step": 93540
    },
    {
      "entropy": 5.370429944992066,
      "epoch": 7.278078194903715,
      "grad_norm": 1.375,
      "learning_rate": 0.0001284867829832007,
      "loss": 4.5506,
      "mean_token_accuracy": 0.2460001990199089,
      "num_tokens": 162246581.0,
      "step": 93545
    },
    {
      "entropy": 5.329010486602783,
      "epoch": 7.278467224275433,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012846574984769522,
      "loss": 4.6015,
      "mean_token_accuracy": 0.2515176981687546,
      "num_tokens": 162255193.0,
      "step": 93550
    },
    {
      "entropy": 5.365826749801636,
      "epoch": 7.278856253647151,
      "grad_norm": 1.375,
      "learning_rate": 0.0001284447189356136,
      "loss": 4.5395,
      "mean_token_accuracy": 0.24620168656110764,
      "num_tokens": 162263684.0,
      "step": 93555
    },
    {
      "entropy": 5.435653924942017,
      "epoch": 7.279245283018868,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012842369024727523,
      "loss": 4.7185,
      "mean_token_accuracy": 0.2361210137605667,
      "num_tokens": 162272137.0,
      "step": 93560
    },
    {
      "entropy": 5.3710143089294435,
      "epoch": 7.279634312390585,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012840266378299898,
      "loss": 4.6362,
      "mean_token_accuracy": 0.24721072614192963,
      "num_tokens": 162280802.0,
      "step": 93565
    },
    {
      "entropy": 5.383084154129028,
      "epoch": 7.280023341762303,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012838163954310405,
      "loss": 4.5932,
      "mean_token_accuracy": 0.2464805468916893,
      "num_tokens": 162288758.0,
      "step": 93570
    },
    {
      "entropy": 5.454525327682495,
      "epoch": 7.2804123711340205,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012836061752790929,
      "loss": 4.6637,
      "mean_token_accuracy": 0.23232460021972656,
      "num_tokens": 162297324.0,
      "step": 93575
    },
    {
      "entropy": 5.333440685272217,
      "epoch": 7.280801400505738,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012833959773773384,
      "loss": 4.5429,
      "mean_token_accuracy": 0.2443806752562523,
      "num_tokens": 162305309.0,
      "step": 93580
    },
    {
      "entropy": 5.34431414604187,
      "epoch": 7.281190429877456,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00012831858017289647,
      "loss": 4.5616,
      "mean_token_accuracy": 0.2477089822292328,
      "num_tokens": 162314364.0,
      "step": 93585
    },
    {
      "entropy": 5.323424434661865,
      "epoch": 7.281579459249174,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001282975648337163,
      "loss": 4.5361,
      "mean_token_accuracy": 0.24524455964565278,
      "num_tokens": 162322821.0,
      "step": 93590
    },
    {
      "entropy": 5.3863636493682865,
      "epoch": 7.28196848862089,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012827655172051194,
      "loss": 4.6566,
      "mean_token_accuracy": 0.23839305192232133,
      "num_tokens": 162331443.0,
      "step": 93595
    },
    {
      "entropy": 5.3014092445373535,
      "epoch": 7.282357517992608,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001282555408336024,
      "loss": 4.4503,
      "mean_token_accuracy": 0.25749406069517133,
      "num_tokens": 162339560.0,
      "step": 93600
    },
    {
      "entropy": 5.3786157131195065,
      "epoch": 7.282746547364326,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012823453217330655,
      "loss": 4.6588,
      "mean_token_accuracy": 0.24155599474906922,
      "num_tokens": 162348207.0,
      "step": 93605
    },
    {
      "entropy": 5.362637281417847,
      "epoch": 7.2831355767360435,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000128213525739943,
      "loss": 4.533,
      "mean_token_accuracy": 0.25910462290048597,
      "num_tokens": 162356336.0,
      "step": 93610
    },
    {
      "entropy": 5.384792995452881,
      "epoch": 7.283524606107761,
      "grad_norm": 1.375,
      "learning_rate": 0.0001281925215338306,
      "loss": 4.5894,
      "mean_token_accuracy": 0.24695801138877868,
      "num_tokens": 162364619.0,
      "step": 93615
    },
    {
      "entropy": 5.3533693790435795,
      "epoch": 7.283913635479479,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012817151955528796,
      "loss": 4.611,
      "mean_token_accuracy": 0.24998821914196015,
      "num_tokens": 162373526.0,
      "step": 93620
    },
    {
      "entropy": 5.266763210296631,
      "epoch": 7.284302664851197,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001281505198046339,
      "loss": 4.4793,
      "mean_token_accuracy": 0.25037487149238585,
      "num_tokens": 162381981.0,
      "step": 93625
    },
    {
      "entropy": 5.378292608261108,
      "epoch": 7.284691694222914,
      "grad_norm": 1.40625,
      "learning_rate": 0.000128129522282187,
      "loss": 4.5752,
      "mean_token_accuracy": 0.24290703386068344,
      "num_tokens": 162390003.0,
      "step": 93630
    },
    {
      "entropy": 5.343885803222657,
      "epoch": 7.285080723594631,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001281085269882657,
      "loss": 4.5152,
      "mean_token_accuracy": 0.2545665234327316,
      "num_tokens": 162398699.0,
      "step": 93635
    },
    {
      "entropy": 5.254929733276367,
      "epoch": 7.285469752966349,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001280875339231888,
      "loss": 4.5085,
      "mean_token_accuracy": 0.25459733307361604,
      "num_tokens": 162407268.0,
      "step": 93640
    },
    {
      "entropy": 5.33442907333374,
      "epoch": 7.2858587823380665,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00012806654308727462,
      "loss": 4.5619,
      "mean_token_accuracy": 0.25379718840122223,
      "num_tokens": 162416439.0,
      "step": 93645
    },
    {
      "entropy": 5.334119462966919,
      "epoch": 7.286247811709784,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001280455544808419,
      "loss": 4.6599,
      "mean_token_accuracy": 0.23769280463457107,
      "num_tokens": 162425348.0,
      "step": 93650
    },
    {
      "entropy": 5.231319999694824,
      "epoch": 7.286636841081502,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001280245681042089,
      "loss": 4.4369,
      "mean_token_accuracy": 0.264738492667675,
      "num_tokens": 162433790.0,
      "step": 93655
    },
    {
      "entropy": 5.3592901706695555,
      "epoch": 7.28702587045322,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012800358395769412,
      "loss": 4.6262,
      "mean_token_accuracy": 0.23720459043979644,
      "num_tokens": 162442843.0,
      "step": 93660
    },
    {
      "entropy": 5.296915817260742,
      "epoch": 7.287414899824936,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001279826020416161,
      "loss": 4.5451,
      "mean_token_accuracy": 0.25168350636959075,
      "num_tokens": 162451888.0,
      "step": 93665
    },
    {
      "entropy": 5.427657747268677,
      "epoch": 7.287803929196654,
      "grad_norm": 1.453125,
      "learning_rate": 0.000127961622356293,
      "loss": 4.645,
      "mean_token_accuracy": 0.24339783042669297,
      "num_tokens": 162460744.0,
      "step": 93670
    },
    {
      "entropy": 5.3516443252563475,
      "epoch": 7.288192958568372,
      "grad_norm": 1.375,
      "learning_rate": 0.00012794064490204333,
      "loss": 4.5796,
      "mean_token_accuracy": 0.24696160554885865,
      "num_tokens": 162469115.0,
      "step": 93675
    },
    {
      "entropy": 5.3333296298980715,
      "epoch": 7.2885819879400895,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012791966967918523,
      "loss": 4.5525,
      "mean_token_accuracy": 0.24592935293912888,
      "num_tokens": 162477598.0,
      "step": 93680
    },
    {
      "entropy": 5.39199538230896,
      "epoch": 7.288971017311807,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001278986966880371,
      "loss": 4.603,
      "mean_token_accuracy": 0.24908401370048522,
      "num_tokens": 162485880.0,
      "step": 93685
    },
    {
      "entropy": 5.343863248825073,
      "epoch": 7.289360046683525,
      "grad_norm": 1.375,
      "learning_rate": 0.00012787772592891703,
      "loss": 4.5583,
      "mean_token_accuracy": 0.24881335496902465,
      "num_tokens": 162493784.0,
      "step": 93690
    },
    {
      "entropy": 5.274034309387207,
      "epoch": 7.289749076055243,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001278567574021434,
      "loss": 4.492,
      "mean_token_accuracy": 0.26425474137067795,
      "num_tokens": 162502630.0,
      "step": 93695
    },
    {
      "entropy": 5.392499828338623,
      "epoch": 7.290138105426959,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012783579110803417,
      "loss": 4.6321,
      "mean_token_accuracy": 0.24215712100267411,
      "num_tokens": 162511250.0,
      "step": 93700
    },
    {
      "entropy": 5.3459662914276125,
      "epoch": 7.290527134798677,
      "grad_norm": 1.3125,
      "learning_rate": 0.00012781482704690768,
      "loss": 4.5469,
      "mean_token_accuracy": 0.24679122865200043,
      "num_tokens": 162519979.0,
      "step": 93705
    },
    {
      "entropy": 5.380033254623413,
      "epoch": 7.290916164170395,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012779386521908194,
      "loss": 4.5684,
      "mean_token_accuracy": 0.25042191594839097,
      "num_tokens": 162529128.0,
      "step": 93710
    },
    {
      "entropy": 5.376624202728271,
      "epoch": 7.2913051935421125,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012777290562487488,
      "loss": 4.632,
      "mean_token_accuracy": 0.24071722477674484,
      "num_tokens": 162537933.0,
      "step": 93715
    },
    {
      "entropy": 5.38666262626648,
      "epoch": 7.29169422291383,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012775194826460467,
      "loss": 4.6551,
      "mean_token_accuracy": 0.2382311850786209,
      "num_tokens": 162546971.0,
      "step": 93720
    },
    {
      "entropy": 5.3483030796051025,
      "epoch": 7.292083252285548,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012773099313858927,
      "loss": 4.6207,
      "mean_token_accuracy": 0.23936288356781005,
      "num_tokens": 162555811.0,
      "step": 93725
    },
    {
      "entropy": 5.413087606430054,
      "epoch": 7.292472281657265,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001277100402471466,
      "loss": 4.6195,
      "mean_token_accuracy": 0.24625207036733626,
      "num_tokens": 162564824.0,
      "step": 93730
    },
    {
      "entropy": 5.352571821212768,
      "epoch": 7.292861311028982,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012768908959059477,
      "loss": 4.4897,
      "mean_token_accuracy": 0.2502407118678093,
      "num_tokens": 162572317.0,
      "step": 93735
    },
    {
      "entropy": 5.366762447357178,
      "epoch": 7.2932503404007,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012766814116925142,
      "loss": 4.5876,
      "mean_token_accuracy": 0.24714734852313996,
      "num_tokens": 162580606.0,
      "step": 93740
    },
    {
      "entropy": 5.291939640045166,
      "epoch": 7.293639369772418,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012764719498343464,
      "loss": 4.5375,
      "mean_token_accuracy": 0.25012839436531065,
      "num_tokens": 162589694.0,
      "step": 93745
    },
    {
      "entropy": 5.277828884124756,
      "epoch": 7.2940283991441355,
      "grad_norm": 1.46875,
      "learning_rate": 0.000127626251033462,
      "loss": 4.5048,
      "mean_token_accuracy": 0.25645933002233506,
      "num_tokens": 162597867.0,
      "step": 93750
    },
    {
      "entropy": 5.355414819717407,
      "epoch": 7.294417428515853,
      "grad_norm": 1.3125,
      "learning_rate": 0.00012760530931965153,
      "loss": 4.6515,
      "mean_token_accuracy": 0.2421593874692917,
      "num_tokens": 162606685.0,
      "step": 93755
    },
    {
      "entropy": 5.443055438995361,
      "epoch": 7.294806457887571,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012758436984232083,
      "loss": 4.7573,
      "mean_token_accuracy": 0.2316468194127083,
      "num_tokens": 162615565.0,
      "step": 93760
    },
    {
      "entropy": 5.336056756973266,
      "epoch": 7.295195487259288,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012756343260178775,
      "loss": 4.514,
      "mean_token_accuracy": 0.2550398349761963,
      "num_tokens": 162623595.0,
      "step": 93765
    },
    {
      "entropy": 5.394806623458862,
      "epoch": 7.295584516631005,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012754249759836984,
      "loss": 4.5785,
      "mean_token_accuracy": 0.24844130575656892,
      "num_tokens": 162631797.0,
      "step": 93770
    },
    {
      "entropy": 5.414483547210693,
      "epoch": 7.295973546002723,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001275215648323849,
      "loss": 4.5759,
      "mean_token_accuracy": 0.24719517379999162,
      "num_tokens": 162640333.0,
      "step": 93775
    },
    {
      "entropy": 5.459576940536499,
      "epoch": 7.296362575374441,
      "grad_norm": 1.375,
      "learning_rate": 0.00012750063430415036,
      "loss": 4.7288,
      "mean_token_accuracy": 0.2347236081957817,
      "num_tokens": 162649162.0,
      "step": 93780
    },
    {
      "entropy": 5.338545227050782,
      "epoch": 7.2967516047461585,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012747970601398407,
      "loss": 4.5808,
      "mean_token_accuracy": 0.2431936651468277,
      "num_tokens": 162657782.0,
      "step": 93785
    },
    {
      "entropy": 5.383843183517456,
      "epoch": 7.297140634117876,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001274587799622034,
      "loss": 4.5841,
      "mean_token_accuracy": 0.2429990604519844,
      "num_tokens": 162666252.0,
      "step": 93790
    },
    {
      "entropy": 5.334454774856567,
      "epoch": 7.297529663489594,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012743785614912576,
      "loss": 4.5983,
      "mean_token_accuracy": 0.2513753712177277,
      "num_tokens": 162675328.0,
      "step": 93795
    },
    {
      "entropy": 5.343678569793701,
      "epoch": 7.297918692861311,
      "grad_norm": 1.296875,
      "learning_rate": 0.00012741693457506893,
      "loss": 4.5701,
      "mean_token_accuracy": 0.24483009576797485,
      "num_tokens": 162683933.0,
      "step": 93800
    },
    {
      "entropy": 5.399790668487549,
      "epoch": 7.298307722233028,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012739601524035013,
      "loss": 4.661,
      "mean_token_accuracy": 0.24368840903043748,
      "num_tokens": 162692937.0,
      "step": 93805
    },
    {
      "entropy": 5.268685674667358,
      "epoch": 7.298696751604746,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012737509814528697,
      "loss": 4.4896,
      "mean_token_accuracy": 0.25580201148986814,
      "num_tokens": 162702568.0,
      "step": 93810
    },
    {
      "entropy": 5.427707147598267,
      "epoch": 7.299085780976464,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001273541832901967,
      "loss": 4.6869,
      "mean_token_accuracy": 0.232516285777092,
      "num_tokens": 162710826.0,
      "step": 93815
    },
    {
      "entropy": 5.417893266677856,
      "epoch": 7.2994748103481815,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012733327067539663,
      "loss": 4.6342,
      "mean_token_accuracy": 0.24155422151088715,
      "num_tokens": 162720066.0,
      "step": 93820
    },
    {
      "entropy": 5.341436147689819,
      "epoch": 7.299863839719899,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001273123603012042,
      "loss": 4.6146,
      "mean_token_accuracy": 0.24349660128355027,
      "num_tokens": 162729164.0,
      "step": 93825
    },
    {
      "entropy": 5.359644794464112,
      "epoch": 7.300252869091617,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001272914521679366,
      "loss": 4.5884,
      "mean_token_accuracy": 0.23897091895341874,
      "num_tokens": 162737897.0,
      "step": 93830
    },
    {
      "entropy": 5.350044679641724,
      "epoch": 7.300641898463334,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012727054627591116,
      "loss": 4.6103,
      "mean_token_accuracy": 0.2448935851454735,
      "num_tokens": 162747153.0,
      "step": 93835
    },
    {
      "entropy": 5.355051422119141,
      "epoch": 7.301030927835051,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000127249642625445,
      "loss": 4.5954,
      "mean_token_accuracy": 0.2443198710680008,
      "num_tokens": 162755658.0,
      "step": 93840
    },
    {
      "entropy": 5.401859712600708,
      "epoch": 7.301419957206769,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012722874121685534,
      "loss": 4.5909,
      "mean_token_accuracy": 0.248111492395401,
      "num_tokens": 162763254.0,
      "step": 93845
    },
    {
      "entropy": 5.388643264770508,
      "epoch": 7.301808986578487,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012720784205045934,
      "loss": 4.5735,
      "mean_token_accuracy": 0.24981037974357606,
      "num_tokens": 162771609.0,
      "step": 93850
    },
    {
      "entropy": 5.313520288467407,
      "epoch": 7.3021980159502045,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012718694512657414,
      "loss": 4.5516,
      "mean_token_accuracy": 0.2485866814851761,
      "num_tokens": 162780077.0,
      "step": 93855
    },
    {
      "entropy": 5.348296213150024,
      "epoch": 7.302587045321922,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012716605044551672,
      "loss": 4.6002,
      "mean_token_accuracy": 0.2487013131380081,
      "num_tokens": 162788393.0,
      "step": 93860
    },
    {
      "entropy": 5.47361650466919,
      "epoch": 7.302976074693639,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012714515800760416,
      "loss": 4.6503,
      "mean_token_accuracy": 0.2346815899014473,
      "num_tokens": 162796549.0,
      "step": 93865
    },
    {
      "entropy": 5.404931592941284,
      "epoch": 7.303365104065357,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012712426781315357,
      "loss": 4.6618,
      "mean_token_accuracy": 0.2411891594529152,
      "num_tokens": 162805401.0,
      "step": 93870
    },
    {
      "entropy": 5.405546951293945,
      "epoch": 7.303754133437074,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001271033798624818,
      "loss": 4.663,
      "mean_token_accuracy": 0.2372954547405243,
      "num_tokens": 162814108.0,
      "step": 93875
    },
    {
      "entropy": 5.3753767013549805,
      "epoch": 7.304143162808792,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012708249415590588,
      "loss": 4.5822,
      "mean_token_accuracy": 0.24128783941268922,
      "num_tokens": 162822373.0,
      "step": 93880
    },
    {
      "entropy": 5.428216600418091,
      "epoch": 7.30453219218051,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001270616106937426,
      "loss": 4.6502,
      "mean_token_accuracy": 0.2408711791038513,
      "num_tokens": 162830772.0,
      "step": 93885
    },
    {
      "entropy": 5.318695974349976,
      "epoch": 7.3049212215522275,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000127040729476309,
      "loss": 4.5697,
      "mean_token_accuracy": 0.24698079377412796,
      "num_tokens": 162839992.0,
      "step": 93890
    },
    {
      "entropy": 5.443019151687622,
      "epoch": 7.305310250923945,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012701985050392185,
      "loss": 4.6612,
      "mean_token_accuracy": 0.24559315145015717,
      "num_tokens": 162848949.0,
      "step": 93895
    },
    {
      "entropy": 5.4101423740386965,
      "epoch": 7.305699280295662,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012699897377689784,
      "loss": 4.5763,
      "mean_token_accuracy": 0.2527389720082283,
      "num_tokens": 162857084.0,
      "step": 93900
    },
    {
      "entropy": 5.419390392303467,
      "epoch": 7.30608830966738,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012697809929555388,
      "loss": 4.643,
      "mean_token_accuracy": 0.2433452323079109,
      "num_tokens": 162866207.0,
      "step": 93905
    },
    {
      "entropy": 5.350918388366699,
      "epoch": 7.306477339039097,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012695722706020663,
      "loss": 4.5796,
      "mean_token_accuracy": 0.2502825438976288,
      "num_tokens": 162874016.0,
      "step": 93910
    },
    {
      "entropy": 5.303159570693969,
      "epoch": 7.306866368410815,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012693635707117291,
      "loss": 4.5474,
      "mean_token_accuracy": 0.24673781394958497,
      "num_tokens": 162883143.0,
      "step": 93915
    },
    {
      "entropy": 5.227814435958862,
      "epoch": 7.307255397782533,
      "grad_norm": 1.296875,
      "learning_rate": 0.00012691548932876917,
      "loss": 4.4446,
      "mean_token_accuracy": 0.2518265262246132,
      "num_tokens": 162891548.0,
      "step": 93920
    },
    {
      "entropy": 5.356292247772217,
      "epoch": 7.30764442715425,
      "grad_norm": 1.375,
      "learning_rate": 0.00012689462383331222,
      "loss": 4.6097,
      "mean_token_accuracy": 0.24886824786663056,
      "num_tokens": 162899866.0,
      "step": 93925
    },
    {
      "entropy": 5.289357233047485,
      "epoch": 7.308033456525967,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00012687376058511868,
      "loss": 4.5236,
      "mean_token_accuracy": 0.25425864160060885,
      "num_tokens": 162908887.0,
      "step": 93930
    },
    {
      "entropy": 5.338607740402222,
      "epoch": 7.308422485897685,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000126852899584505,
      "loss": 4.5665,
      "mean_token_accuracy": 0.24844253063201904,
      "num_tokens": 162916806.0,
      "step": 93935
    },
    {
      "entropy": 5.387065696716308,
      "epoch": 7.308811515269403,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012683204083178786,
      "loss": 4.6334,
      "mean_token_accuracy": 0.2447605848312378,
      "num_tokens": 162925767.0,
      "step": 93940
    },
    {
      "entropy": 5.346244239807129,
      "epoch": 7.30920054464112,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012681118432728353,
      "loss": 4.5741,
      "mean_token_accuracy": 0.24965574443340302,
      "num_tokens": 162934240.0,
      "step": 93945
    },
    {
      "entropy": 5.400351238250733,
      "epoch": 7.309589574012838,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012679033007130872,
      "loss": 4.623,
      "mean_token_accuracy": 0.24179142266511916,
      "num_tokens": 162943838.0,
      "step": 93950
    },
    {
      "entropy": 5.389816474914551,
      "epoch": 7.309978603384556,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012676947806417965,
      "loss": 4.5988,
      "mean_token_accuracy": 0.247365865111351,
      "num_tokens": 162952097.0,
      "step": 93955
    },
    {
      "entropy": 5.352571105957031,
      "epoch": 7.310367632756273,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012674862830621293,
      "loss": 4.6207,
      "mean_token_accuracy": 0.23930107802152634,
      "num_tokens": 162960648.0,
      "step": 93960
    },
    {
      "entropy": 5.415088701248169,
      "epoch": 7.310756662127991,
      "grad_norm": 1.3125,
      "learning_rate": 0.00012672778079772474,
      "loss": 4.6664,
      "mean_token_accuracy": 0.23732315599918366,
      "num_tokens": 162969119.0,
      "step": 93965
    },
    {
      "entropy": 5.400457429885864,
      "epoch": 7.311145691499708,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001267069355390315,
      "loss": 4.6006,
      "mean_token_accuracy": 0.24561842978000642,
      "num_tokens": 162978287.0,
      "step": 93970
    },
    {
      "entropy": 5.31756739616394,
      "epoch": 7.311534720871426,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012668609253044952,
      "loss": 4.5793,
      "mean_token_accuracy": 0.244005386531353,
      "num_tokens": 162986720.0,
      "step": 93975
    },
    {
      "entropy": 5.383456897735596,
      "epoch": 7.311923750243143,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012666525177229494,
      "loss": 4.5717,
      "mean_token_accuracy": 0.24326060861349105,
      "num_tokens": 162994774.0,
      "step": 93980
    },
    {
      "entropy": 5.374219179153442,
      "epoch": 7.312312779614861,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012664441326488412,
      "loss": 4.6565,
      "mean_token_accuracy": 0.24258837550878526,
      "num_tokens": 163003274.0,
      "step": 93985
    },
    {
      "entropy": 5.329337549209595,
      "epoch": 7.312701808986579,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00012662357700853315,
      "loss": 4.5939,
      "mean_token_accuracy": 0.24615825563669205,
      "num_tokens": 163012431.0,
      "step": 93990
    },
    {
      "entropy": 5.364320373535156,
      "epoch": 7.313090838358296,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012660274300355822,
      "loss": 4.6734,
      "mean_token_accuracy": 0.23605892360210418,
      "num_tokens": 163021394.0,
      "step": 93995
    },
    {
      "entropy": 5.436716938018799,
      "epoch": 7.313479867730013,
      "grad_norm": 1.21875,
      "learning_rate": 0.00012658191125027547,
      "loss": 4.6766,
      "mean_token_accuracy": 0.23759645968675613,
      "num_tokens": 163030512.0,
      "step": 94000
    },
    {
      "entropy": 5.4383526802062985,
      "epoch": 7.313868897101731,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012656108174900104,
      "loss": 4.6133,
      "mean_token_accuracy": 0.23983788043260573,
      "num_tokens": 163038427.0,
      "step": 94005
    },
    {
      "entropy": 5.267654800415039,
      "epoch": 7.314257926473449,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012654025450005096,
      "loss": 4.4597,
      "mean_token_accuracy": 0.2626446411013603,
      "num_tokens": 163046923.0,
      "step": 94010
    },
    {
      "entropy": 5.379391145706177,
      "epoch": 7.314646955845166,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012651942950374115,
      "loss": 4.6518,
      "mean_token_accuracy": 0.2375599578022957,
      "num_tokens": 163055720.0,
      "step": 94015
    },
    {
      "entropy": 5.347490978240967,
      "epoch": 7.315035985216884,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001264986067603877,
      "loss": 4.5721,
      "mean_token_accuracy": 0.24959192276000977,
      "num_tokens": 163065132.0,
      "step": 94020
    },
    {
      "entropy": 5.328269052505493,
      "epoch": 7.315425014588602,
      "grad_norm": 1.265625,
      "learning_rate": 0.00012647778627030652,
      "loss": 4.5607,
      "mean_token_accuracy": 0.24698563814163207,
      "num_tokens": 163074426.0,
      "step": 94025
    },
    {
      "entropy": 5.330621957778931,
      "epoch": 7.315814043960319,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012645696803381353,
      "loss": 4.5675,
      "mean_token_accuracy": 0.24733350723981856,
      "num_tokens": 163082609.0,
      "step": 94030
    },
    {
      "entropy": 5.339424562454224,
      "epoch": 7.316203073332036,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012643615205122462,
      "loss": 4.504,
      "mean_token_accuracy": 0.24648047983646393,
      "num_tokens": 163090494.0,
      "step": 94035
    },
    {
      "entropy": 5.366439914703369,
      "epoch": 7.316592102703754,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012641533832285568,
      "loss": 4.6407,
      "mean_token_accuracy": 0.24788683503866196,
      "num_tokens": 163098332.0,
      "step": 94040
    },
    {
      "entropy": 5.321274757385254,
      "epoch": 7.316981132075472,
      "grad_norm": 1.28125,
      "learning_rate": 0.00012639452684902239,
      "loss": 4.549,
      "mean_token_accuracy": 0.2535044848918915,
      "num_tokens": 163107479.0,
      "step": 94045
    },
    {
      "entropy": 5.33533992767334,
      "epoch": 7.317370161447189,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012637371763004072,
      "loss": 4.5586,
      "mean_token_accuracy": 0.24317685663700103,
      "num_tokens": 163115903.0,
      "step": 94050
    },
    {
      "entropy": 5.361750078201294,
      "epoch": 7.317759190818907,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001263529106662263,
      "loss": 4.638,
      "mean_token_accuracy": 0.24287838339805604,
      "num_tokens": 163124601.0,
      "step": 94055
    },
    {
      "entropy": 5.410828161239624,
      "epoch": 7.318148220190625,
      "grad_norm": 1.28125,
      "learning_rate": 0.00012633210595789474,
      "loss": 4.6524,
      "mean_token_accuracy": 0.23831938505172728,
      "num_tokens": 163133985.0,
      "step": 94060
    },
    {
      "entropy": 5.398840379714966,
      "epoch": 7.3185372495623415,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012631130350536203,
      "loss": 4.6584,
      "mean_token_accuracy": 0.24547156542539597,
      "num_tokens": 163142452.0,
      "step": 94065
    },
    {
      "entropy": 5.444306564331055,
      "epoch": 7.318926278934059,
      "grad_norm": 1.375,
      "learning_rate": 0.00012629050330894347,
      "loss": 4.6758,
      "mean_token_accuracy": 0.24658536463975905,
      "num_tokens": 163151033.0,
      "step": 94070
    },
    {
      "entropy": 5.360301923751831,
      "epoch": 7.319315308305777,
      "grad_norm": 1.328125,
      "learning_rate": 0.00012626970536895493,
      "loss": 4.5736,
      "mean_token_accuracy": 0.2563576400279999,
      "num_tokens": 163160256.0,
      "step": 94075
    },
    {
      "entropy": 5.463718223571777,
      "epoch": 7.319704337677495,
      "grad_norm": 1.375,
      "learning_rate": 0.00012624890968571185,
      "loss": 4.7094,
      "mean_token_accuracy": 0.24417247027158737,
      "num_tokens": 163168417.0,
      "step": 94080
    },
    {
      "entropy": 5.3682310581207275,
      "epoch": 7.320093367049212,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012622811625952982,
      "loss": 4.5311,
      "mean_token_accuracy": 0.24549825638532638,
      "num_tokens": 163177160.0,
      "step": 94085
    },
    {
      "entropy": 5.4310572147369385,
      "epoch": 7.32048239642093,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00012620732509072437,
      "loss": 4.663,
      "mean_token_accuracy": 0.23133173435926438,
      "num_tokens": 163186205.0,
      "step": 94090
    },
    {
      "entropy": 5.369527530670166,
      "epoch": 7.320871425792648,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012618653617961086,
      "loss": 4.618,
      "mean_token_accuracy": 0.24943159967660905,
      "num_tokens": 163194726.0,
      "step": 94095
    },
    {
      "entropy": 5.3082798480987545,
      "epoch": 7.3212604551643645,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012616574952650485,
      "loss": 4.5021,
      "mean_token_accuracy": 0.2549077033996582,
      "num_tokens": 163202694.0,
      "step": 94100
    },
    {
      "entropy": 5.318939065933227,
      "epoch": 7.321649484536082,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001261449651317216,
      "loss": 4.5992,
      "mean_token_accuracy": 0.251708360016346,
      "num_tokens": 163211574.0,
      "step": 94105
    },
    {
      "entropy": 5.341650390625,
      "epoch": 7.3220385139078,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001261241829955767,
      "loss": 4.5663,
      "mean_token_accuracy": 0.24781087934970855,
      "num_tokens": 163220963.0,
      "step": 94110
    },
    {
      "entropy": 5.419349241256714,
      "epoch": 7.322427543279518,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012610340311838524,
      "loss": 4.6437,
      "mean_token_accuracy": 0.24247910231351852,
      "num_tokens": 163229876.0,
      "step": 94115
    },
    {
      "entropy": 5.412204599380493,
      "epoch": 7.322816572651235,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012608262550046278,
      "loss": 4.6784,
      "mean_token_accuracy": 0.23569331020116807,
      "num_tokens": 163238613.0,
      "step": 94120
    },
    {
      "entropy": 5.426954746246338,
      "epoch": 7.323205602022953,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012606185014212438,
      "loss": 4.67,
      "mean_token_accuracy": 0.23521170020103455,
      "num_tokens": 163247192.0,
      "step": 94125
    },
    {
      "entropy": 5.361254453659058,
      "epoch": 7.323594631394671,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012604107704368528,
      "loss": 4.5071,
      "mean_token_accuracy": 0.2524102583527565,
      "num_tokens": 163255363.0,
      "step": 94130
    },
    {
      "entropy": 5.388315629959107,
      "epoch": 7.3239836607663875,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00012602030620546088,
      "loss": 4.5705,
      "mean_token_accuracy": 0.25677483081817626,
      "num_tokens": 163264090.0,
      "step": 94135
    },
    {
      "entropy": 5.39150595664978,
      "epoch": 7.324372690138105,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001259995376277661,
      "loss": 4.6589,
      "mean_token_accuracy": 0.242257459461689,
      "num_tokens": 163273009.0,
      "step": 94140
    },
    {
      "entropy": 5.325036287307739,
      "epoch": 7.324761719509823,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001259787713109163,
      "loss": 4.5296,
      "mean_token_accuracy": 0.2529591202735901,
      "num_tokens": 163280876.0,
      "step": 94145
    },
    {
      "entropy": 5.312546157836914,
      "epoch": 7.325150748881541,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001259580072552263,
      "loss": 4.5661,
      "mean_token_accuracy": 0.24881526827812195,
      "num_tokens": 163290107.0,
      "step": 94150
    },
    {
      "entropy": 5.377563953399658,
      "epoch": 7.325539778253258,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012593724546101142,
      "loss": 4.5669,
      "mean_token_accuracy": 0.2510275781154633,
      "num_tokens": 163298591.0,
      "step": 94155
    },
    {
      "entropy": 5.317975091934204,
      "epoch": 7.325928807624976,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012591648592858652,
      "loss": 4.4879,
      "mean_token_accuracy": 0.25029309988021853,
      "num_tokens": 163308299.0,
      "step": 94160
    },
    {
      "entropy": 5.385346698760986,
      "epoch": 7.326317836996694,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001258957286582667,
      "loss": 4.6843,
      "mean_token_accuracy": 0.23791890144348143,
      "num_tokens": 163317475.0,
      "step": 94165
    },
    {
      "entropy": 5.416301488876343,
      "epoch": 7.3267068663684105,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012587497365036692,
      "loss": 4.6091,
      "mean_token_accuracy": 0.24556662142276764,
      "num_tokens": 163326089.0,
      "step": 94170
    },
    {
      "entropy": 5.297440528869629,
      "epoch": 7.327095895740128,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012585422090520193,
      "loss": 4.5224,
      "mean_token_accuracy": 0.248697167634964,
      "num_tokens": 163334153.0,
      "step": 94175
    },
    {
      "entropy": 5.348762893676758,
      "epoch": 7.327484925111846,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001258334704230869,
      "loss": 4.6036,
      "mean_token_accuracy": 0.24213108867406846,
      "num_tokens": 163343815.0,
      "step": 94180
    },
    {
      "entropy": 5.400892686843872,
      "epoch": 7.327873954483564,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001258127222043364,
      "loss": 4.6109,
      "mean_token_accuracy": 0.2370713546872139,
      "num_tokens": 163351510.0,
      "step": 94185
    },
    {
      "entropy": 5.372803831100464,
      "epoch": 7.328262983855281,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012579197624926547,
      "loss": 4.5917,
      "mean_token_accuracy": 0.24456847608089446,
      "num_tokens": 163360503.0,
      "step": 94190
    },
    {
      "entropy": 5.3610621929168705,
      "epoch": 7.328652013226999,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012577123255818874,
      "loss": 4.6401,
      "mean_token_accuracy": 0.2394181415438652,
      "num_tokens": 163369705.0,
      "step": 94195
    },
    {
      "entropy": 5.362016201019287,
      "epoch": 7.329041042598716,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000125750491131421,
      "loss": 4.6383,
      "mean_token_accuracy": 0.2416483938694,
      "num_tokens": 163379111.0,
      "step": 94200
    },
    {
      "entropy": 5.401790761947632,
      "epoch": 7.3294300719704335,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00012572975196927717,
      "loss": 4.6487,
      "mean_token_accuracy": 0.24734949916601182,
      "num_tokens": 163387901.0,
      "step": 94205
    },
    {
      "entropy": 5.395131587982178,
      "epoch": 7.329819101342151,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001257090150720716,
      "loss": 4.5457,
      "mean_token_accuracy": 0.25504214465618136,
      "num_tokens": 163397195.0,
      "step": 94210
    },
    {
      "entropy": 5.331527519226074,
      "epoch": 7.330208130713869,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001256882804401192,
      "loss": 4.5101,
      "mean_token_accuracy": 0.25379130244255066,
      "num_tokens": 163405939.0,
      "step": 94215
    },
    {
      "entropy": 5.282521677017212,
      "epoch": 7.330597160085587,
      "grad_norm": 1.25,
      "learning_rate": 0.00012566754807373442,
      "loss": 4.5519,
      "mean_token_accuracy": 0.25476140081882476,
      "num_tokens": 163415703.0,
      "step": 94220
    },
    {
      "entropy": 5.331199693679809,
      "epoch": 7.330986189457304,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012564681797323202,
      "loss": 4.5469,
      "mean_token_accuracy": 0.24518670439720153,
      "num_tokens": 163423971.0,
      "step": 94225
    },
    {
      "entropy": 5.408124399185181,
      "epoch": 7.331375218829022,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012562609013892632,
      "loss": 4.642,
      "mean_token_accuracy": 0.23427651971578597,
      "num_tokens": 163432551.0,
      "step": 94230
    },
    {
      "entropy": 5.351016092300415,
      "epoch": 7.331764248200739,
      "grad_norm": 1.203125,
      "learning_rate": 0.00012560536457113203,
      "loss": 4.5898,
      "mean_token_accuracy": 0.24789413064718246,
      "num_tokens": 163441657.0,
      "step": 94235
    },
    {
      "entropy": 5.358830451965332,
      "epoch": 7.3321532775724565,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001255846412701635,
      "loss": 4.6501,
      "mean_token_accuracy": 0.24453209936618805,
      "num_tokens": 163450431.0,
      "step": 94240
    },
    {
      "entropy": 5.362634801864624,
      "epoch": 7.332542306944174,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012556392023633527,
      "loss": 4.5956,
      "mean_token_accuracy": 0.24651786237955092,
      "num_tokens": 163459025.0,
      "step": 94245
    },
    {
      "entropy": 5.350490665435791,
      "epoch": 7.332931336315892,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012554320146996169,
      "loss": 4.5985,
      "mean_token_accuracy": 0.2480500340461731,
      "num_tokens": 163467267.0,
      "step": 94250
    },
    {
      "entropy": 5.410300445556641,
      "epoch": 7.33332036568761,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001255224849713571,
      "loss": 4.6405,
      "mean_token_accuracy": 0.2414240226149559,
      "num_tokens": 163475939.0,
      "step": 94255
    },
    {
      "entropy": 5.3582274436950685,
      "epoch": 7.333709395059327,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012550177074083585,
      "loss": 4.5409,
      "mean_token_accuracy": 0.2526076093316078,
      "num_tokens": 163483664.0,
      "step": 94260
    },
    {
      "entropy": 5.418527507781983,
      "epoch": 7.334098424431044,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001254810587787123,
      "loss": 4.615,
      "mean_token_accuracy": 0.24115387052297593,
      "num_tokens": 163492601.0,
      "step": 94265
    },
    {
      "entropy": 5.370981931686401,
      "epoch": 7.334487453802762,
      "grad_norm": 1.5625,
      "learning_rate": 0.00012546034908530074,
      "loss": 4.6348,
      "mean_token_accuracy": 0.23939983248710633,
      "num_tokens": 163500262.0,
      "step": 94270
    },
    {
      "entropy": 5.352333927154541,
      "epoch": 7.3348764831744795,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001254396416609154,
      "loss": 4.5374,
      "mean_token_accuracy": 0.25349895656108856,
      "num_tokens": 163508836.0,
      "step": 94275
    },
    {
      "entropy": 5.37919545173645,
      "epoch": 7.335265512546197,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012541893650587035,
      "loss": 4.573,
      "mean_token_accuracy": 0.23996481895446778,
      "num_tokens": 163516800.0,
      "step": 94280
    },
    {
      "entropy": 5.376312017440796,
      "epoch": 7.335654541917915,
      "grad_norm": 1.328125,
      "learning_rate": 0.00012539823362047998,
      "loss": 4.6171,
      "mean_token_accuracy": 0.2384917214512825,
      "num_tokens": 163525217.0,
      "step": 94285
    },
    {
      "entropy": 5.315344142913818,
      "epoch": 7.336043571289633,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012537753300505817,
      "loss": 4.5251,
      "mean_token_accuracy": 0.24736091792583464,
      "num_tokens": 163534847.0,
      "step": 94290
    },
    {
      "entropy": 5.353728485107422,
      "epoch": 7.33643260066135,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012535683465991922,
      "loss": 4.6002,
      "mean_token_accuracy": 0.24443187564611435,
      "num_tokens": 163543391.0,
      "step": 94295
    },
    {
      "entropy": 5.323368263244629,
      "epoch": 7.336821630033067,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001253361385853771,
      "loss": 4.5712,
      "mean_token_accuracy": 0.2500728666782379,
      "num_tokens": 163551601.0,
      "step": 94300
    },
    {
      "entropy": 5.377783679962159,
      "epoch": 7.337210659404785,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001253154447817459,
      "loss": 4.6348,
      "mean_token_accuracy": 0.24851650446653367,
      "num_tokens": 163560618.0,
      "step": 94305
    },
    {
      "entropy": 5.363304519653321,
      "epoch": 7.3375996887765025,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012529475324933947,
      "loss": 4.6216,
      "mean_token_accuracy": 0.24460252076387407,
      "num_tokens": 163569729.0,
      "step": 94310
    },
    {
      "entropy": 5.381510162353516,
      "epoch": 7.33798871814822,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012527406398847203,
      "loss": 4.615,
      "mean_token_accuracy": 0.24908929765224458,
      "num_tokens": 163578834.0,
      "step": 94315
    },
    {
      "entropy": 5.314643526077271,
      "epoch": 7.338377747519938,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012525337699945726,
      "loss": 4.5469,
      "mean_token_accuracy": 0.25252521187067034,
      "num_tokens": 163587068.0,
      "step": 94320
    },
    {
      "entropy": 5.380915594100952,
      "epoch": 7.338766776891656,
      "grad_norm": 1.375,
      "learning_rate": 0.00012523269228260908,
      "loss": 4.528,
      "mean_token_accuracy": 0.2542674109339714,
      "num_tokens": 163595453.0,
      "step": 94325
    },
    {
      "entropy": 5.329588508605957,
      "epoch": 7.339155806263373,
      "grad_norm": 1.3125,
      "learning_rate": 0.00012521200983824158,
      "loss": 4.5293,
      "mean_token_accuracy": 0.2535392001271248,
      "num_tokens": 163604124.0,
      "step": 94330
    },
    {
      "entropy": 5.335412073135376,
      "epoch": 7.33954483563509,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001251913296666683,
      "loss": 4.5561,
      "mean_token_accuracy": 0.24718367159366608,
      "num_tokens": 163612722.0,
      "step": 94335
    },
    {
      "entropy": 5.279577827453613,
      "epoch": 7.339933865006808,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012517065176820326,
      "loss": 4.5523,
      "mean_token_accuracy": 0.25031495094299316,
      "num_tokens": 163621580.0,
      "step": 94340
    },
    {
      "entropy": 5.440175294876099,
      "epoch": 7.3403228943785255,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012514997614315997,
      "loss": 4.6736,
      "mean_token_accuracy": 0.23754564076662063,
      "num_tokens": 163631118.0,
      "step": 94345
    },
    {
      "entropy": 5.378798961639404,
      "epoch": 7.340711923750243,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012512930279185236,
      "loss": 4.6019,
      "mean_token_accuracy": 0.24993743598461152,
      "num_tokens": 163639783.0,
      "step": 94350
    },
    {
      "entropy": 5.360303544998169,
      "epoch": 7.341100953121961,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012510863171459407,
      "loss": 4.6152,
      "mean_token_accuracy": 0.24297598898410797,
      "num_tokens": 163648335.0,
      "step": 94355
    },
    {
      "entropy": 5.408425569534302,
      "epoch": 7.341489982493679,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012508796291169864,
      "loss": 4.621,
      "mean_token_accuracy": 0.24012731462717057,
      "num_tokens": 163656750.0,
      "step": 94360
    },
    {
      "entropy": 5.44068284034729,
      "epoch": 7.341879011865396,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001250672963834798,
      "loss": 4.6615,
      "mean_token_accuracy": 0.24300095736980437,
      "num_tokens": 163664652.0,
      "step": 94365
    },
    {
      "entropy": 5.381797409057617,
      "epoch": 7.342268041237113,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012504663213025108,
      "loss": 4.6276,
      "mean_token_accuracy": 0.24616905450820922,
      "num_tokens": 163673677.0,
      "step": 94370
    },
    {
      "entropy": 5.447887659072876,
      "epoch": 7.342657070608831,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012502597015232606,
      "loss": 4.6823,
      "mean_token_accuracy": 0.2420792907476425,
      "num_tokens": 163682803.0,
      "step": 94375
    },
    {
      "entropy": 5.361238670349121,
      "epoch": 7.3430460999805485,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012500531045001818,
      "loss": 4.6304,
      "mean_token_accuracy": 0.23542912304401398,
      "num_tokens": 163691144.0,
      "step": 94380
    },
    {
      "entropy": 5.389438772201538,
      "epoch": 7.343435129352266,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012498465302364104,
      "loss": 4.615,
      "mean_token_accuracy": 0.2446685627102852,
      "num_tokens": 163699634.0,
      "step": 94385
    },
    {
      "entropy": 5.395974445343017,
      "epoch": 7.343824158723984,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012496399787350796,
      "loss": 4.5663,
      "mean_token_accuracy": 0.24455775320529938,
      "num_tokens": 163707906.0,
      "step": 94390
    },
    {
      "entropy": 5.38205099105835,
      "epoch": 7.344213188095702,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001249433449999324,
      "loss": 4.6061,
      "mean_token_accuracy": 0.2479534462094307,
      "num_tokens": 163717164.0,
      "step": 94395
    },
    {
      "entropy": 5.381329202651978,
      "epoch": 7.344602217467418,
      "grad_norm": 1.375,
      "learning_rate": 0.00012492269440322784,
      "loss": 4.6582,
      "mean_token_accuracy": 0.2376382291316986,
      "num_tokens": 163726874.0,
      "step": 94400
    },
    {
      "entropy": 5.312429809570313,
      "epoch": 7.344991246839136,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012490204608370738,
      "loss": 4.5521,
      "mean_token_accuracy": 0.24370041042566298,
      "num_tokens": 163735689.0,
      "step": 94405
    },
    {
      "entropy": 5.372054624557495,
      "epoch": 7.345380276210854,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001248814000416846,
      "loss": 4.598,
      "mean_token_accuracy": 0.24875753968954087,
      "num_tokens": 163744350.0,
      "step": 94410
    },
    {
      "entropy": 5.342721319198608,
      "epoch": 7.3457693055825715,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012486075627747255,
      "loss": 4.5349,
      "mean_token_accuracy": 0.25628087371587754,
      "num_tokens": 163752296.0,
      "step": 94415
    },
    {
      "entropy": 5.349948596954346,
      "epoch": 7.346158334954289,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012484011479138463,
      "loss": 4.6009,
      "mean_token_accuracy": 0.24388409405946732,
      "num_tokens": 163761143.0,
      "step": 94420
    },
    {
      "entropy": 5.31762580871582,
      "epoch": 7.346547364326007,
      "grad_norm": 1.296875,
      "learning_rate": 0.00012481947558373385,
      "loss": 4.5885,
      "mean_token_accuracy": 0.24912268966436385,
      "num_tokens": 163769965.0,
      "step": 94425
    },
    {
      "entropy": 5.326842546463013,
      "epoch": 7.3469363936977246,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012479883865483362,
      "loss": 4.5249,
      "mean_token_accuracy": 0.2545724675059319,
      "num_tokens": 163777822.0,
      "step": 94430
    },
    {
      "entropy": 5.334474658966064,
      "epoch": 7.347325423069441,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012477820400499687,
      "loss": 4.6357,
      "mean_token_accuracy": 0.23473903387784958,
      "num_tokens": 163786795.0,
      "step": 94435
    },
    {
      "entropy": 5.316759061813355,
      "epoch": 7.347714452441159,
      "grad_norm": 1.3125,
      "learning_rate": 0.00012475757163453677,
      "loss": 4.5321,
      "mean_token_accuracy": 0.246415813267231,
      "num_tokens": 163795980.0,
      "step": 94440
    },
    {
      "entropy": 5.4038928031921385,
      "epoch": 7.348103481812877,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012473694154376642,
      "loss": 4.6589,
      "mean_token_accuracy": 0.24382228553295135,
      "num_tokens": 163805160.0,
      "step": 94445
    },
    {
      "entropy": 5.384531831741333,
      "epoch": 7.3484925111845945,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012471631373299876,
      "loss": 4.6097,
      "mean_token_accuracy": 0.2441108986735344,
      "num_tokens": 163814429.0,
      "step": 94450
    },
    {
      "entropy": 5.387227773666382,
      "epoch": 7.348881540556312,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0001246956882025469,
      "loss": 4.6286,
      "mean_token_accuracy": 0.24336243271827698,
      "num_tokens": 163822868.0,
      "step": 94455
    },
    {
      "entropy": 5.4389585018157955,
      "epoch": 7.34927056992803,
      "grad_norm": 1.28125,
      "learning_rate": 0.00012467506495272357,
      "loss": 4.6981,
      "mean_token_accuracy": 0.24395828396081926,
      "num_tokens": 163832244.0,
      "step": 94460
    },
    {
      "entropy": 5.3698138236999515,
      "epoch": 7.349659599299747,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012465444398384204,
      "loss": 4.6461,
      "mean_token_accuracy": 0.2423446848988533,
      "num_tokens": 163841232.0,
      "step": 94465
    },
    {
      "entropy": 5.408413600921631,
      "epoch": 7.350048628671464,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012463382529621502,
      "loss": 4.5994,
      "mean_token_accuracy": 0.2496738150715828,
      "num_tokens": 163849860.0,
      "step": 94470
    },
    {
      "entropy": 5.363593244552613,
      "epoch": 7.350437658043182,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001246132088901553,
      "loss": 4.531,
      "mean_token_accuracy": 0.25939404517412185,
      "num_tokens": 163858278.0,
      "step": 94475
    },
    {
      "entropy": 5.389538097381592,
      "epoch": 7.3508266874149,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012459259476597584,
      "loss": 4.6414,
      "mean_token_accuracy": 0.24487038999795913,
      "num_tokens": 163866383.0,
      "step": 94480
    },
    {
      "entropy": 5.3618577003479,
      "epoch": 7.3512157167866174,
      "grad_norm": 1.375,
      "learning_rate": 0.00012457198292398927,
      "loss": 4.5584,
      "mean_token_accuracy": 0.2512852609157562,
      "num_tokens": 163875169.0,
      "step": 94485
    },
    {
      "entropy": 5.4225725650787355,
      "epoch": 7.351604746158335,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001245513733645085,
      "loss": 4.6833,
      "mean_token_accuracy": 0.24572090357542037,
      "num_tokens": 163884427.0,
      "step": 94490
    },
    {
      "entropy": 5.334325742721558,
      "epoch": 7.351993775530053,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012453076608784613,
      "loss": 4.5177,
      "mean_token_accuracy": 0.2518465220928192,
      "num_tokens": 163892692.0,
      "step": 94495
    },
    {
      "entropy": 5.320864915847778,
      "epoch": 7.3523828049017705,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012451016109431493,
      "loss": 4.5717,
      "mean_token_accuracy": 0.2518074333667755,
      "num_tokens": 163902057.0,
      "step": 94500
    },
    {
      "entropy": 5.34097695350647,
      "epoch": 7.352771834273487,
      "grad_norm": 1.28125,
      "learning_rate": 0.00012448955838422745,
      "loss": 4.5687,
      "mean_token_accuracy": 0.24783486723899842,
      "num_tokens": 163910649.0,
      "step": 94505
    },
    {
      "entropy": 5.488706588745117,
      "epoch": 7.353160863645205,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001244689579578964,
      "loss": 4.7753,
      "mean_token_accuracy": 0.23000237345695496,
      "num_tokens": 163919969.0,
      "step": 94510
    },
    {
      "entropy": 5.3812949657440186,
      "epoch": 7.353549893016923,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012444835981563436,
      "loss": 4.647,
      "mean_token_accuracy": 0.24544771760702133,
      "num_tokens": 163929561.0,
      "step": 94515
    },
    {
      "entropy": 5.411035490036011,
      "epoch": 7.35393892238864,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012442776395775375,
      "loss": 4.6238,
      "mean_token_accuracy": 0.24927684515714646,
      "num_tokens": 163938595.0,
      "step": 94520
    },
    {
      "entropy": 5.382682132720947,
      "epoch": 7.354327951760358,
      "grad_norm": 1.375,
      "learning_rate": 0.00012440717038456715,
      "loss": 4.5634,
      "mean_token_accuracy": 0.25016694217920304,
      "num_tokens": 163947466.0,
      "step": 94525
    },
    {
      "entropy": 5.367576026916504,
      "epoch": 7.354716981132076,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012438657909638707,
      "loss": 4.5599,
      "mean_token_accuracy": 0.24404421597719192,
      "num_tokens": 163955632.0,
      "step": 94530
    },
    {
      "entropy": 5.381631326675415,
      "epoch": 7.355106010503793,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012436599009352597,
      "loss": 4.6116,
      "mean_token_accuracy": 0.24691084176301956,
      "num_tokens": 163964566.0,
      "step": 94535
    },
    {
      "entropy": 5.271064329147339,
      "epoch": 7.35549503987551,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012434540337629617,
      "loss": 4.5131,
      "mean_token_accuracy": 0.25496512204408645,
      "num_tokens": 163972665.0,
      "step": 94540
    },
    {
      "entropy": 5.358954191207886,
      "epoch": 7.355884069247228,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012432481894501012,
      "loss": 4.5897,
      "mean_token_accuracy": 0.24608606994152069,
      "num_tokens": 163980594.0,
      "step": 94545
    },
    {
      "entropy": 5.38698410987854,
      "epoch": 7.356273098618946,
      "grad_norm": 1.5,
      "learning_rate": 0.00012430423679998012,
      "loss": 4.6691,
      "mean_token_accuracy": 0.23326386064291,
      "num_tokens": 163988882.0,
      "step": 94550
    },
    {
      "entropy": 5.338452911376953,
      "epoch": 7.356662127990663,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012428365694151843,
      "loss": 4.5067,
      "mean_token_accuracy": 0.2581985771656036,
      "num_tokens": 163997204.0,
      "step": 94555
    },
    {
      "entropy": 5.394546461105347,
      "epoch": 7.357051157362381,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0001242630793699374,
      "loss": 4.6317,
      "mean_token_accuracy": 0.24523754715919494,
      "num_tokens": 164005414.0,
      "step": 94560
    },
    {
      "entropy": 5.389801931381226,
      "epoch": 7.357440186734099,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012424250408554918,
      "loss": 4.5986,
      "mean_token_accuracy": 0.2513165161013603,
      "num_tokens": 164013911.0,
      "step": 94565
    },
    {
      "entropy": 5.28177227973938,
      "epoch": 7.357829216105816,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012422193108866606,
      "loss": 4.5307,
      "mean_token_accuracy": 0.25732963234186174,
      "num_tokens": 164023072.0,
      "step": 94570
    },
    {
      "entropy": 5.368022394180298,
      "epoch": 7.358218245477533,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012420136037960005,
      "loss": 4.619,
      "mean_token_accuracy": 0.24548924565315247,
      "num_tokens": 164031205.0,
      "step": 94575
    },
    {
      "entropy": 5.335402536392212,
      "epoch": 7.358607274849251,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012418079195866346,
      "loss": 4.5286,
      "mean_token_accuracy": 0.2510799154639244,
      "num_tokens": 164040068.0,
      "step": 94580
    },
    {
      "entropy": 5.328437423706054,
      "epoch": 7.358996304220969,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001241602258261682,
      "loss": 4.6126,
      "mean_token_accuracy": 0.2355450227856636,
      "num_tokens": 164049027.0,
      "step": 94585
    },
    {
      "entropy": 5.439224195480347,
      "epoch": 7.359385333592686,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001241396619824265,
      "loss": 4.7106,
      "mean_token_accuracy": 0.2392003446817398,
      "num_tokens": 164058733.0,
      "step": 94590
    },
    {
      "entropy": 5.30728850364685,
      "epoch": 7.359774362964404,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012411910042775033,
      "loss": 4.5122,
      "mean_token_accuracy": 0.25841052681207655,
      "num_tokens": 164066950.0,
      "step": 94595
    },
    {
      "entropy": 5.443985891342163,
      "epoch": 7.360163392336121,
      "grad_norm": 1.375,
      "learning_rate": 0.0001240985411624516,
      "loss": 4.7146,
      "mean_token_accuracy": 0.23773619383573533,
      "num_tokens": 164074983.0,
      "step": 94600
    },
    {
      "entropy": 5.335656929016113,
      "epoch": 7.360552421707839,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001240779841868424,
      "loss": 4.5575,
      "mean_token_accuracy": 0.2531688541173935,
      "num_tokens": 164084156.0,
      "step": 94605
    },
    {
      "entropy": 5.387413263320923,
      "epoch": 7.360941451079556,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012405742950123447,
      "loss": 4.6379,
      "mean_token_accuracy": 0.2388383373618126,
      "num_tokens": 164092467.0,
      "step": 94610
    },
    {
      "entropy": 5.272861099243164,
      "epoch": 7.361330480451274,
      "grad_norm": 1.3125,
      "learning_rate": 0.00012403687710593994,
      "loss": 4.4593,
      "mean_token_accuracy": 0.264107283949852,
      "num_tokens": 164100956.0,
      "step": 94615
    },
    {
      "entropy": 5.319189643859863,
      "epoch": 7.361719509822992,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012401632700127037,
      "loss": 4.5363,
      "mean_token_accuracy": 0.25721351355314254,
      "num_tokens": 164108908.0,
      "step": 94620
    },
    {
      "entropy": 5.408845138549805,
      "epoch": 7.362108539194709,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012399577918753785,
      "loss": 4.567,
      "mean_token_accuracy": 0.24320504516363145,
      "num_tokens": 164117013.0,
      "step": 94625
    },
    {
      "entropy": 5.343125343322754,
      "epoch": 7.362497568566427,
      "grad_norm": 1.484375,
      "learning_rate": 0.000123975233665054,
      "loss": 4.5386,
      "mean_token_accuracy": 0.24477118402719497,
      "num_tokens": 164124833.0,
      "step": 94630
    },
    {
      "entropy": 5.3228271484375,
      "epoch": 7.362886597938144,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012395469043413054,
      "loss": 4.6495,
      "mean_token_accuracy": 0.24897656142711638,
      "num_tokens": 164133890.0,
      "step": 94635
    },
    {
      "entropy": 5.2942994117736815,
      "epoch": 7.363275627309862,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001239341494950793,
      "loss": 4.614,
      "mean_token_accuracy": 0.2438293516635895,
      "num_tokens": 164142800.0,
      "step": 94640
    },
    {
      "entropy": 5.316054153442383,
      "epoch": 7.363664656681579,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012391361084821182,
      "loss": 4.5418,
      "mean_token_accuracy": 0.2431260421872139,
      "num_tokens": 164151325.0,
      "step": 94645
    },
    {
      "entropy": 5.309487342834473,
      "epoch": 7.364053686053297,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012389307449383992,
      "loss": 4.5341,
      "mean_token_accuracy": 0.2551542088389397,
      "num_tokens": 164159812.0,
      "step": 94650
    },
    {
      "entropy": 5.390709066390992,
      "epoch": 7.364442715425015,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000123872540432275,
      "loss": 4.6264,
      "mean_token_accuracy": 0.2445986896753311,
      "num_tokens": 164168628.0,
      "step": 94655
    },
    {
      "entropy": 5.322995042800903,
      "epoch": 7.364831744796732,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012385200866382875,
      "loss": 4.4827,
      "mean_token_accuracy": 0.25955532789230346,
      "num_tokens": 164177297.0,
      "step": 94660
    },
    {
      "entropy": 5.361328887939453,
      "epoch": 7.36522077416845,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012383147918881274,
      "loss": 4.556,
      "mean_token_accuracy": 0.24996531158685684,
      "num_tokens": 164185561.0,
      "step": 94665
    },
    {
      "entropy": 5.343436002731323,
      "epoch": 7.365609803540167,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012381095200753836,
      "loss": 4.5647,
      "mean_token_accuracy": 0.24244602769613266,
      "num_tokens": 164194028.0,
      "step": 94670
    },
    {
      "entropy": 5.404474210739136,
      "epoch": 7.365998832911885,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00012379042712031723,
      "loss": 4.7273,
      "mean_token_accuracy": 0.23863156735897065,
      "num_tokens": 164202475.0,
      "step": 94675
    },
    {
      "entropy": 5.29662446975708,
      "epoch": 7.366387862283602,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00012376990452746064,
      "loss": 4.5558,
      "mean_token_accuracy": 0.24755702614784242,
      "num_tokens": 164210954.0,
      "step": 94680
    },
    {
      "entropy": 5.382462835311889,
      "epoch": 7.36677689165532,
      "grad_norm": 1.3125,
      "learning_rate": 0.00012374938422928005,
      "loss": 4.6273,
      "mean_token_accuracy": 0.23223475366830826,
      "num_tokens": 164220393.0,
      "step": 94685
    },
    {
      "entropy": 5.288799476623535,
      "epoch": 7.367165921027038,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001237288662260868,
      "loss": 4.5402,
      "mean_token_accuracy": 0.2502361938357353,
      "num_tokens": 164229958.0,
      "step": 94690
    },
    {
      "entropy": 5.311399459838867,
      "epoch": 7.367554950398755,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00012370835051819226,
      "loss": 4.5208,
      "mean_token_accuracy": 0.25485492050647734,
      "num_tokens": 164238267.0,
      "step": 94695
    },
    {
      "entropy": 5.364652490615844,
      "epoch": 7.367943979770473,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012368783710590766,
      "loss": 4.6192,
      "mean_token_accuracy": 0.250770303606987,
      "num_tokens": 164246901.0,
      "step": 94700
    },
    {
      "entropy": 5.2456567764282225,
      "epoch": 7.36833300914219,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012366732598954435,
      "loss": 4.4882,
      "mean_token_accuracy": 0.247441166639328,
      "num_tokens": 164255351.0,
      "step": 94705
    },
    {
      "entropy": 5.40628490447998,
      "epoch": 7.368722038513908,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001236468171694135,
      "loss": 4.6534,
      "mean_token_accuracy": 0.2465667247772217,
      "num_tokens": 164263665.0,
      "step": 94710
    },
    {
      "entropy": 5.338277053833008,
      "epoch": 7.369111067885625,
      "grad_norm": 1.28125,
      "learning_rate": 0.00012362631064582623,
      "loss": 4.6254,
      "mean_token_accuracy": 0.2427482023835182,
      "num_tokens": 164273568.0,
      "step": 94715
    },
    {
      "entropy": 5.400489616394043,
      "epoch": 7.369500097257343,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012360580641909382,
      "loss": 4.5584,
      "mean_token_accuracy": 0.24988534599542617,
      "num_tokens": 164281538.0,
      "step": 94720
    },
    {
      "entropy": 5.398919534683228,
      "epoch": 7.369889126629061,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001235853044895272,
      "loss": 4.6415,
      "mean_token_accuracy": 0.2396869882941246,
      "num_tokens": 164290419.0,
      "step": 94725
    },
    {
      "entropy": 5.393545007705688,
      "epoch": 7.370278156000778,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012356480485743772,
      "loss": 4.6315,
      "mean_token_accuracy": 0.24184854328632355,
      "num_tokens": 164299023.0,
      "step": 94730
    },
    {
      "entropy": 5.344344902038574,
      "epoch": 7.370667185372495,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00012354430752313628,
      "loss": 4.5887,
      "mean_token_accuracy": 0.24575280994176865,
      "num_tokens": 164307003.0,
      "step": 94735
    },
    {
      "entropy": 5.348322582244873,
      "epoch": 7.371056214744213,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012352381248693383,
      "loss": 4.6784,
      "mean_token_accuracy": 0.2435499832034111,
      "num_tokens": 164316502.0,
      "step": 94740
    },
    {
      "entropy": 5.32273383140564,
      "epoch": 7.371445244115931,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012350331974914154,
      "loss": 4.5398,
      "mean_token_accuracy": 0.24413011372089385,
      "num_tokens": 164324437.0,
      "step": 94745
    },
    {
      "entropy": 5.347464609146118,
      "epoch": 7.371834273487648,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012348282931007011,
      "loss": 4.5217,
      "mean_token_accuracy": 0.2476254478096962,
      "num_tokens": 164333012.0,
      "step": 94750
    },
    {
      "entropy": 5.298477745056152,
      "epoch": 7.372223302859366,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001234623411700307,
      "loss": 4.5663,
      "mean_token_accuracy": 0.24725541323423386,
      "num_tokens": 164341491.0,
      "step": 94755
    },
    {
      "entropy": 5.302954769134521,
      "epoch": 7.372612332231084,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000123441855329334,
      "loss": 4.5484,
      "mean_token_accuracy": 0.24250570982694625,
      "num_tokens": 164349813.0,
      "step": 94760
    },
    {
      "entropy": 5.316489028930664,
      "epoch": 7.373001361602801,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012342137178829096,
      "loss": 4.5569,
      "mean_token_accuracy": 0.24775611013174056,
      "num_tokens": 164358364.0,
      "step": 94765
    },
    {
      "entropy": 5.343932628631592,
      "epoch": 7.373390390974518,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012340089054721227,
      "loss": 4.6078,
      "mean_token_accuracy": 0.2511534675955772,
      "num_tokens": 164366174.0,
      "step": 94770
    },
    {
      "entropy": 5.211862945556641,
      "epoch": 7.373779420346236,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012338041160640884,
      "loss": 4.3841,
      "mean_token_accuracy": 0.2656616732478142,
      "num_tokens": 164375693.0,
      "step": 94775
    },
    {
      "entropy": 5.40116057395935,
      "epoch": 7.374168449717954,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012335993496619127,
      "loss": 4.5928,
      "mean_token_accuracy": 0.24313049614429474,
      "num_tokens": 164383786.0,
      "step": 94780
    },
    {
      "entropy": 5.223314237594605,
      "epoch": 7.374557479089671,
      "grad_norm": 1.375,
      "learning_rate": 0.00012333946062687042,
      "loss": 4.4429,
      "mean_token_accuracy": 0.26262580305337907,
      "num_tokens": 164392404.0,
      "step": 94785
    },
    {
      "entropy": 5.390042781829834,
      "epoch": 7.374946508461389,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012331898858875678,
      "loss": 4.6593,
      "mean_token_accuracy": 0.24416809380054474,
      "num_tokens": 164401224.0,
      "step": 94790
    },
    {
      "entropy": 5.41112060546875,
      "epoch": 7.375335537833107,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012329851885216104,
      "loss": 4.6278,
      "mean_token_accuracy": 0.24610287845134735,
      "num_tokens": 164409969.0,
      "step": 94795
    },
    {
      "entropy": 5.342955541610718,
      "epoch": 7.3757245672048235,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001232780514173939,
      "loss": 4.5713,
      "mean_token_accuracy": 0.25124277025461195,
      "num_tokens": 164418777.0,
      "step": 94800
    },
    {
      "entropy": 5.343933820724487,
      "epoch": 7.376113596576541,
      "grad_norm": 1.328125,
      "learning_rate": 0.00012325758628476576,
      "loss": 4.5342,
      "mean_token_accuracy": 0.2541049376130104,
      "num_tokens": 164427225.0,
      "step": 94805
    },
    {
      "entropy": 5.358779430389404,
      "epoch": 7.376502625948259,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012323712345458735,
      "loss": 4.6286,
      "mean_token_accuracy": 0.24401804953813552,
      "num_tokens": 164435918.0,
      "step": 94810
    },
    {
      "entropy": 5.331714248657226,
      "epoch": 7.376891655319977,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012321666292716896,
      "loss": 4.6523,
      "mean_token_accuracy": 0.2398310974240303,
      "num_tokens": 164444812.0,
      "step": 94815
    },
    {
      "entropy": 5.412523412704468,
      "epoch": 7.377280684691694,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001231962047028211,
      "loss": 4.613,
      "mean_token_accuracy": 0.24533528089523315,
      "num_tokens": 164453733.0,
      "step": 94820
    },
    {
      "entropy": 5.3308412551879885,
      "epoch": 7.377669714063412,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012317574878185426,
      "loss": 4.5312,
      "mean_token_accuracy": 0.2561042934656143,
      "num_tokens": 164462588.0,
      "step": 94825
    },
    {
      "entropy": 5.345142936706543,
      "epoch": 7.37805874343513,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012315529516457874,
      "loss": 4.5715,
      "mean_token_accuracy": 0.2516396030783653,
      "num_tokens": 164470838.0,
      "step": 94830
    },
    {
      "entropy": 5.311398363113403,
      "epoch": 7.378447772806847,
      "grad_norm": 1.453125,
      "learning_rate": 0.000123134843851305,
      "loss": 4.5616,
      "mean_token_accuracy": 0.24898409247398376,
      "num_tokens": 164479797.0,
      "step": 94835
    },
    {
      "entropy": 5.341183614730835,
      "epoch": 7.378836802178564,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012311439484234316,
      "loss": 4.6093,
      "mean_token_accuracy": 0.2485154688358307,
      "num_tokens": 164488471.0,
      "step": 94840
    },
    {
      "entropy": 5.269777870178222,
      "epoch": 7.379225831550282,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012309394813800375,
      "loss": 4.4695,
      "mean_token_accuracy": 0.26335751414299013,
      "num_tokens": 164497804.0,
      "step": 94845
    },
    {
      "entropy": 5.384412050247192,
      "epoch": 7.379614860922,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001230735037385968,
      "loss": 4.6608,
      "mean_token_accuracy": 0.2360770508646965,
      "num_tokens": 164506053.0,
      "step": 94850
    },
    {
      "entropy": 5.386424160003662,
      "epoch": 7.380003890293717,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012305306164443264,
      "loss": 4.6285,
      "mean_token_accuracy": 0.2494660422205925,
      "num_tokens": 164514210.0,
      "step": 94855
    },
    {
      "entropy": 5.407201766967773,
      "epoch": 7.380392919665435,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012303262185582147,
      "loss": 4.6585,
      "mean_token_accuracy": 0.2391506552696228,
      "num_tokens": 164523331.0,
      "step": 94860
    },
    {
      "entropy": 5.317540788650513,
      "epoch": 7.380781949037153,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012301218437307332,
      "loss": 4.5875,
      "mean_token_accuracy": 0.2460081860423088,
      "num_tokens": 164531414.0,
      "step": 94865
    },
    {
      "entropy": 5.39361572265625,
      "epoch": 7.3811709784088695,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001229917491964984,
      "loss": 4.6709,
      "mean_token_accuracy": 0.23448848277330397,
      "num_tokens": 164540788.0,
      "step": 94870
    },
    {
      "entropy": 5.332365226745606,
      "epoch": 7.381560007780587,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001229713163264067,
      "loss": 4.617,
      "mean_token_accuracy": 0.2457575887441635,
      "num_tokens": 164550432.0,
      "step": 94875
    },
    {
      "entropy": 5.466070652008057,
      "epoch": 7.381949037152305,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012295088576310838,
      "loss": 4.7199,
      "mean_token_accuracy": 0.2345777913928032,
      "num_tokens": 164559009.0,
      "step": 94880
    },
    {
      "entropy": 5.345713138580322,
      "epoch": 7.382338066524023,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012293045750691323,
      "loss": 4.5812,
      "mean_token_accuracy": 0.24821756333112716,
      "num_tokens": 164567811.0,
      "step": 94885
    },
    {
      "entropy": 5.401643991470337,
      "epoch": 7.38272709589574,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001229100315581315,
      "loss": 4.6503,
      "mean_token_accuracy": 0.23876250982284547,
      "num_tokens": 164576114.0,
      "step": 94890
    },
    {
      "entropy": 5.42862377166748,
      "epoch": 7.383116125267458,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001228896079170729,
      "loss": 4.7854,
      "mean_token_accuracy": 0.23500944077968597,
      "num_tokens": 164584994.0,
      "step": 94895
    },
    {
      "entropy": 5.4134509563446045,
      "epoch": 7.383505154639176,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012286918658404734,
      "loss": 4.6169,
      "mean_token_accuracy": 0.2459103912115097,
      "num_tokens": 164594064.0,
      "step": 94900
    },
    {
      "entropy": 5.412442016601562,
      "epoch": 7.3838941840108925,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001228487675593648,
      "loss": 4.6104,
      "mean_token_accuracy": 0.24115063846111298,
      "num_tokens": 164602786.0,
      "step": 94905
    },
    {
      "entropy": 5.341301965713501,
      "epoch": 7.38428321338261,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00012282835084333497,
      "loss": 4.5499,
      "mean_token_accuracy": 0.25320925265550615,
      "num_tokens": 164610748.0,
      "step": 94910
    },
    {
      "entropy": 5.310116720199585,
      "epoch": 7.384672242754328,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012280793643626775,
      "loss": 4.595,
      "mean_token_accuracy": 0.24526146948337554,
      "num_tokens": 164619657.0,
      "step": 94915
    },
    {
      "entropy": 5.453195238113404,
      "epoch": 7.385061272126046,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012278752433847283,
      "loss": 4.7097,
      "mean_token_accuracy": 0.24205150157213212,
      "num_tokens": 164627787.0,
      "step": 94920
    },
    {
      "entropy": 5.284538745880127,
      "epoch": 7.385450301497763,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012276711455025996,
      "loss": 4.5206,
      "mean_token_accuracy": 0.2544983208179474,
      "num_tokens": 164636164.0,
      "step": 94925
    },
    {
      "entropy": 5.395350551605224,
      "epoch": 7.385839330869481,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012274670707193886,
      "loss": 4.6282,
      "mean_token_accuracy": 0.24236219972372056,
      "num_tokens": 164645181.0,
      "step": 94930
    },
    {
      "entropy": 5.31056227684021,
      "epoch": 7.386228360241198,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001227263019038191,
      "loss": 4.5499,
      "mean_token_accuracy": 0.2581384971737862,
      "num_tokens": 164654666.0,
      "step": 94935
    },
    {
      "entropy": 5.422539949417114,
      "epoch": 7.3866173896129155,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001227058990462104,
      "loss": 4.7644,
      "mean_token_accuracy": 0.23850566446781157,
      "num_tokens": 164663096.0,
      "step": 94940
    },
    {
      "entropy": 5.2748127460479735,
      "epoch": 7.387006418984633,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001226854984994222,
      "loss": 4.5244,
      "mean_token_accuracy": 0.2507762089371681,
      "num_tokens": 164671980.0,
      "step": 94945
    },
    {
      "entropy": 5.270105457305908,
      "epoch": 7.387395448356351,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001226651002637642,
      "loss": 4.4041,
      "mean_token_accuracy": 0.267032690346241,
      "num_tokens": 164680175.0,
      "step": 94950
    },
    {
      "entropy": 5.367155647277832,
      "epoch": 7.387784477728069,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012264470433954577,
      "loss": 4.5825,
      "mean_token_accuracy": 0.2519281834363937,
      "num_tokens": 164688565.0,
      "step": 94955
    },
    {
      "entropy": 5.319106340408325,
      "epoch": 7.388173507099786,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012262431072707653,
      "loss": 4.5838,
      "mean_token_accuracy": 0.24923646003007888,
      "num_tokens": 164696770.0,
      "step": 94960
    },
    {
      "entropy": 5.365648317337036,
      "epoch": 7.388562536471504,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001226039194266658,
      "loss": 4.5557,
      "mean_token_accuracy": 0.2549635708332062,
      "num_tokens": 164705605.0,
      "step": 94965
    },
    {
      "entropy": 5.424297285079956,
      "epoch": 7.388951565843221,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012258353043862304,
      "loss": 4.6785,
      "mean_token_accuracy": 0.242553947865963,
      "num_tokens": 164714055.0,
      "step": 94970
    },
    {
      "entropy": 5.332719373703003,
      "epoch": 7.3893405952149385,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012256314376325758,
      "loss": 4.5453,
      "mean_token_accuracy": 0.2546596437692642,
      "num_tokens": 164723567.0,
      "step": 94975
    },
    {
      "entropy": 5.4244149684906,
      "epoch": 7.389729624586656,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012254275940087886,
      "loss": 4.6452,
      "mean_token_accuracy": 0.24257283806800842,
      "num_tokens": 164732720.0,
      "step": 94980
    },
    {
      "entropy": 5.397849035263062,
      "epoch": 7.390118653958374,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001225223773517961,
      "loss": 4.6205,
      "mean_token_accuracy": 0.23756182193756104,
      "num_tokens": 164741161.0,
      "step": 94985
    },
    {
      "entropy": 5.423972511291504,
      "epoch": 7.390507683330092,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001225019976163184,
      "loss": 4.6596,
      "mean_token_accuracy": 0.23251235187053682,
      "num_tokens": 164749384.0,
      "step": 94990
    },
    {
      "entropy": 5.350456094741821,
      "epoch": 7.390896712701809,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012248162019475533,
      "loss": 4.5295,
      "mean_token_accuracy": 0.2533285692334175,
      "num_tokens": 164757641.0,
      "step": 94995
    },
    {
      "entropy": 5.283652210235596,
      "epoch": 7.391285742073527,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012246124508741584,
      "loss": 4.495,
      "mean_token_accuracy": 0.25785348415374754,
      "num_tokens": 164766106.0,
      "step": 95000
    },
    {
      "entropy": 5.343594884872436,
      "epoch": 7.391674771445244,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001224408722946092,
      "loss": 4.6195,
      "mean_token_accuracy": 0.24606130570173262,
      "num_tokens": 164775172.0,
      "step": 95005
    },
    {
      "entropy": 5.36291880607605,
      "epoch": 7.3920638008169615,
      "grad_norm": 1.3125,
      "learning_rate": 0.00012242050181664456,
      "loss": 4.5982,
      "mean_token_accuracy": 0.24788551032543182,
      "num_tokens": 164784165.0,
      "step": 95010
    },
    {
      "entropy": 5.427055788040161,
      "epoch": 7.392452830188679,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012240013365383083,
      "loss": 4.6907,
      "mean_token_accuracy": 0.24425375759601592,
      "num_tokens": 164793246.0,
      "step": 95015
    },
    {
      "entropy": 5.302855539321899,
      "epoch": 7.392841859560397,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012237976780647727,
      "loss": 4.557,
      "mean_token_accuracy": 0.24770202934741975,
      "num_tokens": 164802259.0,
      "step": 95020
    },
    {
      "entropy": 5.373119115829468,
      "epoch": 7.3932308889321146,
      "grad_norm": 1.375,
      "learning_rate": 0.00012235940427489278,
      "loss": 4.5939,
      "mean_token_accuracy": 0.24522648602724076,
      "num_tokens": 164810755.0,
      "step": 95025
    },
    {
      "entropy": 5.3369721412658695,
      "epoch": 7.393619918303832,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001223390430593864,
      "loss": 4.5961,
      "mean_token_accuracy": 0.24127318114042282,
      "num_tokens": 164819231.0,
      "step": 95030
    },
    {
      "entropy": 5.352329397201538,
      "epoch": 7.39400894767555,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012231868416026702,
      "loss": 4.5977,
      "mean_token_accuracy": 0.24860863089561464,
      "num_tokens": 164828006.0,
      "step": 95035
    },
    {
      "entropy": 5.382353353500366,
      "epoch": 7.394397977047267,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012229832757784366,
      "loss": 4.5938,
      "mean_token_accuracy": 0.25271220207214357,
      "num_tokens": 164836806.0,
      "step": 95040
    },
    {
      "entropy": 5.285083627700805,
      "epoch": 7.3947870064189845,
      "grad_norm": 1.328125,
      "learning_rate": 0.00012227797331242502,
      "loss": 4.5277,
      "mean_token_accuracy": 0.24739310443401336,
      "num_tokens": 164845258.0,
      "step": 95045
    },
    {
      "entropy": 5.2635438442230225,
      "epoch": 7.395176035790702,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012225762136432015,
      "loss": 4.5664,
      "mean_token_accuracy": 0.24888639748096467,
      "num_tokens": 164854242.0,
      "step": 95050
    },
    {
      "entropy": 5.338901138305664,
      "epoch": 7.39556506516242,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001222372717338377,
      "loss": 4.6262,
      "mean_token_accuracy": 0.243410062789917,
      "num_tokens": 164863799.0,
      "step": 95055
    },
    {
      "entropy": 5.398573732376098,
      "epoch": 7.3959540945341375,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012221692442128648,
      "loss": 4.614,
      "mean_token_accuracy": 0.24854376465082167,
      "num_tokens": 164872371.0,
      "step": 95060
    },
    {
      "entropy": 5.3273396492004395,
      "epoch": 7.396343123905855,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012219657942697532,
      "loss": 4.5711,
      "mean_token_accuracy": 0.24667229652404785,
      "num_tokens": 164881118.0,
      "step": 95065
    },
    {
      "entropy": 5.30540714263916,
      "epoch": 7.396732153277572,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001221762367512128,
      "loss": 4.5499,
      "mean_token_accuracy": 0.25120449215173724,
      "num_tokens": 164889896.0,
      "step": 95070
    },
    {
      "entropy": 5.387515592575073,
      "epoch": 7.39712118264929,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012215589639430772,
      "loss": 4.694,
      "mean_token_accuracy": 0.2404867574572563,
      "num_tokens": 164898455.0,
      "step": 95075
    },
    {
      "entropy": 5.336265945434571,
      "epoch": 7.3975102120210074,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012213555835656855,
      "loss": 4.5923,
      "mean_token_accuracy": 0.24309456944465638,
      "num_tokens": 164907556.0,
      "step": 95080
    },
    {
      "entropy": 5.227671432495117,
      "epoch": 7.397899241392725,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012211522263830407,
      "loss": 4.4235,
      "mean_token_accuracy": 0.26744659096002577,
      "num_tokens": 164916312.0,
      "step": 95085
    },
    {
      "entropy": 5.372040033340454,
      "epoch": 7.398288270764443,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012209488923982275,
      "loss": 4.5553,
      "mean_token_accuracy": 0.24639788419008254,
      "num_tokens": 164924761.0,
      "step": 95090
    },
    {
      "entropy": 5.394526147842408,
      "epoch": 7.3986773001361605,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012207455816143302,
      "loss": 4.673,
      "mean_token_accuracy": 0.24732169955968858,
      "num_tokens": 164933781.0,
      "step": 95095
    },
    {
      "entropy": 5.378658819198608,
      "epoch": 7.399066329507878,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012205422940344351,
      "loss": 4.6141,
      "mean_token_accuracy": 0.24725998491048812,
      "num_tokens": 164942133.0,
      "step": 95100
    },
    {
      "entropy": 5.392763662338257,
      "epoch": 7.399455358879595,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001220339029661626,
      "loss": 4.6581,
      "mean_token_accuracy": 0.24658043533563614,
      "num_tokens": 164950489.0,
      "step": 95105
    },
    {
      "entropy": 5.380920553207398,
      "epoch": 7.399844388251313,
      "grad_norm": 1.328125,
      "learning_rate": 0.00012201357884989878,
      "loss": 4.5999,
      "mean_token_accuracy": 0.25295215249061587,
      "num_tokens": 164958892.0,
      "step": 95110
    },
    {
      "entropy": 5.381988906860352,
      "epoch": 7.40023341762303,
      "grad_norm": 1.328125,
      "learning_rate": 0.00012199325705496033,
      "loss": 4.5035,
      "mean_token_accuracy": 0.2544448062777519,
      "num_tokens": 164966771.0,
      "step": 95115
    },
    {
      "entropy": 5.397501516342163,
      "epoch": 7.400622446994748,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012197293758165578,
      "loss": 4.6395,
      "mean_token_accuracy": 0.2481552019715309,
      "num_tokens": 164975315.0,
      "step": 95120
    },
    {
      "entropy": 5.34716911315918,
      "epoch": 7.401011476366466,
      "grad_norm": 1.3125,
      "learning_rate": 0.00012195262043029324,
      "loss": 4.5855,
      "mean_token_accuracy": 0.24119466096162795,
      "num_tokens": 164984519.0,
      "step": 95125
    },
    {
      "entropy": 5.311775207519531,
      "epoch": 7.4014005057381835,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012193230560118109,
      "loss": 4.6291,
      "mean_token_accuracy": 0.23924950361251832,
      "num_tokens": 164993749.0,
      "step": 95130
    },
    {
      "entropy": 5.3775371551513675,
      "epoch": 7.4017895351099,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00012191199309462763,
      "loss": 4.6452,
      "mean_token_accuracy": 0.24290068149566652,
      "num_tokens": 165002380.0,
      "step": 95135
    },
    {
      "entropy": 5.29567666053772,
      "epoch": 7.402178564481618,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012189168291094092,
      "loss": 4.4718,
      "mean_token_accuracy": 0.253648068010807,
      "num_tokens": 165010613.0,
      "step": 95140
    },
    {
      "entropy": 5.448227977752685,
      "epoch": 7.402567593853336,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012187137505042934,
      "loss": 4.6962,
      "mean_token_accuracy": 0.2373233899474144,
      "num_tokens": 165018863.0,
      "step": 95145
    },
    {
      "entropy": 5.420655012130737,
      "epoch": 7.402956623225053,
      "grad_norm": 1.296875,
      "learning_rate": 0.00012185106951340081,
      "loss": 4.6346,
      "mean_token_accuracy": 0.2388949513435364,
      "num_tokens": 165027331.0,
      "step": 95150
    },
    {
      "entropy": 5.350565004348755,
      "epoch": 7.403345652596771,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012183076630016363,
      "loss": 4.5062,
      "mean_token_accuracy": 0.2611319452524185,
      "num_tokens": 165036690.0,
      "step": 95155
    },
    {
      "entropy": 5.334790802001953,
      "epoch": 7.403734681968489,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012181046541102569,
      "loss": 4.5606,
      "mean_token_accuracy": 0.24774931222200394,
      "num_tokens": 165044909.0,
      "step": 95160
    },
    {
      "entropy": 5.355012989044189,
      "epoch": 7.4041237113402065,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001217901668462952,
      "loss": 4.5877,
      "mean_token_accuracy": 0.24723540097475052,
      "num_tokens": 165053965.0,
      "step": 95165
    },
    {
      "entropy": 5.349900388717652,
      "epoch": 7.404512740711923,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012176987060628004,
      "loss": 4.6182,
      "mean_token_accuracy": 0.24845463931560516,
      "num_tokens": 165062782.0,
      "step": 95170
    },
    {
      "entropy": 5.434891414642334,
      "epoch": 7.404901770083641,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012174957669128818,
      "loss": 4.6576,
      "mean_token_accuracy": 0.23709941357374192,
      "num_tokens": 165071697.0,
      "step": 95175
    },
    {
      "entropy": 5.4279248237609865,
      "epoch": 7.405290799455359,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012172928510162761,
      "loss": 4.7387,
      "mean_token_accuracy": 0.2408421203494072,
      "num_tokens": 165081200.0,
      "step": 95180
    },
    {
      "entropy": 5.360916662216186,
      "epoch": 7.405679828827076,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012170899583760611,
      "loss": 4.5959,
      "mean_token_accuracy": 0.24289380013942719,
      "num_tokens": 165090070.0,
      "step": 95185
    },
    {
      "entropy": 5.384060001373291,
      "epoch": 7.406068858198794,
      "grad_norm": 1.5,
      "learning_rate": 0.0001216887088995316,
      "loss": 4.6002,
      "mean_token_accuracy": 0.24444034844636917,
      "num_tokens": 165098429.0,
      "step": 95190
    },
    {
      "entropy": 5.367804431915284,
      "epoch": 7.406457887570512,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00012166842428771201,
      "loss": 4.625,
      "mean_token_accuracy": 0.24509040862321854,
      "num_tokens": 165108645.0,
      "step": 95195
    },
    {
      "entropy": 5.399759817123413,
      "epoch": 7.4068469169422295,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012164814200245496,
      "loss": 4.5879,
      "mean_token_accuracy": 0.24854462444782258,
      "num_tokens": 165116807.0,
      "step": 95200
    },
    {
      "entropy": 5.372501039505005,
      "epoch": 7.407235946313946,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012162786204406832,
      "loss": 4.6393,
      "mean_token_accuracy": 0.24301631152629852,
      "num_tokens": 165126080.0,
      "step": 95205
    },
    {
      "entropy": 5.339708852767944,
      "epoch": 7.407624975685664,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001216075844128597,
      "loss": 4.5758,
      "mean_token_accuracy": 0.2411342442035675,
      "num_tokens": 165135177.0,
      "step": 95210
    },
    {
      "entropy": 5.395641803741455,
      "epoch": 7.408014005057382,
      "grad_norm": 1.375,
      "learning_rate": 0.0001215873091091369,
      "loss": 4.6786,
      "mean_token_accuracy": 0.23944492638111115,
      "num_tokens": 165144055.0,
      "step": 95215
    },
    {
      "entropy": 5.304111051559448,
      "epoch": 7.408403034429099,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012156703613320744,
      "loss": 4.5157,
      "mean_token_accuracy": 0.2535321518778801,
      "num_tokens": 165152604.0,
      "step": 95220
    },
    {
      "entropy": 5.366166877746582,
      "epoch": 7.408792063800817,
      "grad_norm": 1.5,
      "learning_rate": 0.00012154676548537905,
      "loss": 4.5824,
      "mean_token_accuracy": 0.24678106158971785,
      "num_tokens": 165160909.0,
      "step": 95225
    },
    {
      "entropy": 5.306424951553344,
      "epoch": 7.409181093172535,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00012152649716595919,
      "loss": 4.4929,
      "mean_token_accuracy": 0.2608799129724503,
      "num_tokens": 165170243.0,
      "step": 95230
    },
    {
      "entropy": 5.366546106338501,
      "epoch": 7.4095701225442525,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012150623117525553,
      "loss": 4.5955,
      "mean_token_accuracy": 0.24251407384872437,
      "num_tokens": 165178654.0,
      "step": 95235
    },
    {
      "entropy": 5.310669660568237,
      "epoch": 7.409959151915969,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012148596751357541,
      "loss": 4.5532,
      "mean_token_accuracy": 0.2533928230404854,
      "num_tokens": 165187971.0,
      "step": 95240
    },
    {
      "entropy": 5.3183252811431885,
      "epoch": 7.410348181287687,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012146570618122648,
      "loss": 4.5254,
      "mean_token_accuracy": 0.24964617043733597,
      "num_tokens": 165196321.0,
      "step": 95245
    },
    {
      "entropy": 5.409750938415527,
      "epoch": 7.410737210659405,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012144544717851607,
      "loss": 4.6164,
      "mean_token_accuracy": 0.24363637417554856,
      "num_tokens": 165205290.0,
      "step": 95250
    },
    {
      "entropy": 5.365794467926025,
      "epoch": 7.411126240031122,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012142519050575147,
      "loss": 4.6245,
      "mean_token_accuracy": 0.24321460127830505,
      "num_tokens": 165214869.0,
      "step": 95255
    },
    {
      "entropy": 5.3992938041687015,
      "epoch": 7.41151526940284,
      "grad_norm": 1.5,
      "learning_rate": 0.00012140493616324029,
      "loss": 4.6107,
      "mean_token_accuracy": 0.24865069091320038,
      "num_tokens": 165222911.0,
      "step": 95260
    },
    {
      "entropy": 5.350206899642944,
      "epoch": 7.411904298774558,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012138468415128966,
      "loss": 4.5736,
      "mean_token_accuracy": 0.24446709752082824,
      "num_tokens": 165231784.0,
      "step": 95265
    },
    {
      "entropy": 5.337639188766479,
      "epoch": 7.412293328146275,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012136443447020703,
      "loss": 4.5812,
      "mean_token_accuracy": 0.2519434094429016,
      "num_tokens": 165241193.0,
      "step": 95270
    },
    {
      "entropy": 5.335296726226806,
      "epoch": 7.412682357517992,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012134418712029948,
      "loss": 4.5649,
      "mean_token_accuracy": 0.2500278651714325,
      "num_tokens": 165249737.0,
      "step": 95275
    },
    {
      "entropy": 5.305316305160522,
      "epoch": 7.41307138688971,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012132394210187442,
      "loss": 4.4905,
      "mean_token_accuracy": 0.25234214663505555,
      "num_tokens": 165258935.0,
      "step": 95280
    },
    {
      "entropy": 5.368996429443359,
      "epoch": 7.413460416261428,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00012130369941523893,
      "loss": 4.6624,
      "mean_token_accuracy": 0.2401558369398117,
      "num_tokens": 165268236.0,
      "step": 95285
    },
    {
      "entropy": 5.30174446105957,
      "epoch": 7.413849445633145,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012128345906070007,
      "loss": 4.587,
      "mean_token_accuracy": 0.2445603147149086,
      "num_tokens": 165278014.0,
      "step": 95290
    },
    {
      "entropy": 5.321513032913208,
      "epoch": 7.414238475004863,
      "grad_norm": 1.265625,
      "learning_rate": 0.00012126322103856514,
      "loss": 4.6269,
      "mean_token_accuracy": 0.24154505580663682,
      "num_tokens": 165287210.0,
      "step": 95295
    },
    {
      "entropy": 5.369547414779663,
      "epoch": 7.414627504376581,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012124298534914108,
      "loss": 4.6022,
      "mean_token_accuracy": 0.24651080667972564,
      "num_tokens": 165295811.0,
      "step": 95300
    },
    {
      "entropy": 5.4268190383911135,
      "epoch": 7.415016533748298,
      "grad_norm": 1.265625,
      "learning_rate": 0.00012122275199273505,
      "loss": 4.6373,
      "mean_token_accuracy": 0.2454831197857857,
      "num_tokens": 165305586.0,
      "step": 95305
    },
    {
      "entropy": 5.34252028465271,
      "epoch": 7.415405563120015,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012120252096965391,
      "loss": 4.5759,
      "mean_token_accuracy": 0.24288100600242615,
      "num_tokens": 165314538.0,
      "step": 95310
    },
    {
      "entropy": 5.449515247344971,
      "epoch": 7.415794592491733,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001211822922802048,
      "loss": 4.6833,
      "mean_token_accuracy": 0.23488292694091797,
      "num_tokens": 165322783.0,
      "step": 95315
    },
    {
      "entropy": 5.4323282718658445,
      "epoch": 7.416183621863451,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001211620659246945,
      "loss": 4.7112,
      "mean_token_accuracy": 0.23923946171998978,
      "num_tokens": 165331640.0,
      "step": 95320
    },
    {
      "entropy": 5.408011960983276,
      "epoch": 7.416572651235168,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012114184190343001,
      "loss": 4.6592,
      "mean_token_accuracy": 0.2387697383761406,
      "num_tokens": 165340301.0,
      "step": 95325
    },
    {
      "entropy": 5.347905063629151,
      "epoch": 7.416961680606886,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012112162021671824,
      "loss": 4.5926,
      "mean_token_accuracy": 0.24666311144828795,
      "num_tokens": 165348336.0,
      "step": 95330
    },
    {
      "entropy": 5.360327100753784,
      "epoch": 7.417350709978604,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001211014008648659,
      "loss": 4.6322,
      "mean_token_accuracy": 0.23925192803144454,
      "num_tokens": 165357821.0,
      "step": 95335
    },
    {
      "entropy": 5.323946762084961,
      "epoch": 7.417739739350321,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001210811838481799,
      "loss": 4.5524,
      "mean_token_accuracy": 0.24809849709272386,
      "num_tokens": 165365850.0,
      "step": 95340
    },
    {
      "entropy": 5.321326112747192,
      "epoch": 7.418128768722038,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012106096916696689,
      "loss": 4.5835,
      "mean_token_accuracy": 0.24456214904785156,
      "num_tokens": 165374734.0,
      "step": 95345
    },
    {
      "entropy": 5.381647253036499,
      "epoch": 7.418517798093756,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012104075682153371,
      "loss": 4.6586,
      "mean_token_accuracy": 0.23327119052410125,
      "num_tokens": 165383444.0,
      "step": 95350
    },
    {
      "entropy": 5.3334815979003904,
      "epoch": 7.418906827465474,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012102054681218695,
      "loss": 4.5656,
      "mean_token_accuracy": 0.25162993669509887,
      "num_tokens": 165392856.0,
      "step": 95355
    },
    {
      "entropy": 5.349500322341919,
      "epoch": 7.419295856837191,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012100033913923337,
      "loss": 4.571,
      "mean_token_accuracy": 0.2486968457698822,
      "num_tokens": 165401340.0,
      "step": 95360
    },
    {
      "entropy": 5.340807390213013,
      "epoch": 7.419684886208909,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001209801338029795,
      "loss": 4.5077,
      "mean_token_accuracy": 0.2533141955733299,
      "num_tokens": 165409553.0,
      "step": 95365
    },
    {
      "entropy": 5.3534300327301025,
      "epoch": 7.420073915580627,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012095993080373191,
      "loss": 4.6287,
      "mean_token_accuracy": 0.2430339425802231,
      "num_tokens": 165418383.0,
      "step": 95370
    },
    {
      "entropy": 5.300458526611328,
      "epoch": 7.420462944952344,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012093973014179727,
      "loss": 4.5262,
      "mean_token_accuracy": 0.2545942232012749,
      "num_tokens": 165427118.0,
      "step": 95375
    },
    {
      "entropy": 5.401031017303467,
      "epoch": 7.420851974324061,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012091953181748191,
      "loss": 4.6781,
      "mean_token_accuracy": 0.23979877978563308,
      "num_tokens": 165435447.0,
      "step": 95380
    },
    {
      "entropy": 5.311152601242066,
      "epoch": 7.421241003695779,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012089933583109249,
      "loss": 4.5366,
      "mean_token_accuracy": 0.2533645197749138,
      "num_tokens": 165443744.0,
      "step": 95385
    },
    {
      "entropy": 5.37006311416626,
      "epoch": 7.421630033067497,
      "grad_norm": 1.21875,
      "learning_rate": 0.00012087914218293531,
      "loss": 4.6071,
      "mean_token_accuracy": 0.25137185752391816,
      "num_tokens": 165452855.0,
      "step": 95390
    },
    {
      "entropy": 5.405819129943848,
      "epoch": 7.422019062439214,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012085895087331684,
      "loss": 4.5676,
      "mean_token_accuracy": 0.24899381995201111,
      "num_tokens": 165461027.0,
      "step": 95395
    },
    {
      "entropy": 5.367409420013428,
      "epoch": 7.422408091810932,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012083876190254351,
      "loss": 4.6618,
      "mean_token_accuracy": 0.24195487350225447,
      "num_tokens": 165469879.0,
      "step": 95400
    },
    {
      "entropy": 5.349075508117676,
      "epoch": 7.422797121182649,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012081857527092155,
      "loss": 4.6179,
      "mean_token_accuracy": 0.24573010802268982,
      "num_tokens": 165478793.0,
      "step": 95405
    },
    {
      "entropy": 5.34460220336914,
      "epoch": 7.423186150554367,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012079839097875734,
      "loss": 4.5964,
      "mean_token_accuracy": 0.2502253815531731,
      "num_tokens": 165486578.0,
      "step": 95410
    },
    {
      "entropy": 5.269533491134643,
      "epoch": 7.423575179926084,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00012077820902635706,
      "loss": 4.4815,
      "mean_token_accuracy": 0.25794602930545807,
      "num_tokens": 165495795.0,
      "step": 95415
    },
    {
      "entropy": 5.285582733154297,
      "epoch": 7.423964209297802,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012075802941402707,
      "loss": 4.4881,
      "mean_token_accuracy": 0.2512006416916847,
      "num_tokens": 165504340.0,
      "step": 95420
    },
    {
      "entropy": 5.373102712631225,
      "epoch": 7.42435323866952,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012073785214207338,
      "loss": 4.5765,
      "mean_token_accuracy": 0.24481986761093139,
      "num_tokens": 165512923.0,
      "step": 95425
    },
    {
      "entropy": 5.347365856170654,
      "epoch": 7.424742268041237,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012071767721080234,
      "loss": 4.5741,
      "mean_token_accuracy": 0.23958317935466766,
      "num_tokens": 165521305.0,
      "step": 95430
    },
    {
      "entropy": 5.4263581275939945,
      "epoch": 7.425131297412955,
      "grad_norm": 1.375,
      "learning_rate": 0.0001206975046205199,
      "loss": 4.6527,
      "mean_token_accuracy": 0.24770910888910294,
      "num_tokens": 165529349.0,
      "step": 95435
    },
    {
      "entropy": 5.403101587295533,
      "epoch": 7.425520326784672,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001206773343715323,
      "loss": 4.6595,
      "mean_token_accuracy": 0.23513127267360687,
      "num_tokens": 165539191.0,
      "step": 95440
    },
    {
      "entropy": 5.386759662628174,
      "epoch": 7.42590935615639,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012065716646414555,
      "loss": 4.6018,
      "mean_token_accuracy": 0.2391947865486145,
      "num_tokens": 165547801.0,
      "step": 95445
    },
    {
      "entropy": 5.344751071929932,
      "epoch": 7.426298385528107,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012063700089866556,
      "loss": 4.6005,
      "mean_token_accuracy": 0.24736715108156204,
      "num_tokens": 165555889.0,
      "step": 95450
    },
    {
      "entropy": 5.365631437301635,
      "epoch": 7.426687414899825,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012061683767539836,
      "loss": 4.5367,
      "mean_token_accuracy": 0.2534312218427658,
      "num_tokens": 165563709.0,
      "step": 95455
    },
    {
      "entropy": 5.315147447586059,
      "epoch": 7.427076444271543,
      "grad_norm": 1.25,
      "learning_rate": 0.00012059667679464994,
      "loss": 4.5598,
      "mean_token_accuracy": 0.24474878758192062,
      "num_tokens": 165572843.0,
      "step": 95460
    },
    {
      "entropy": 5.477127027511597,
      "epoch": 7.42746547364326,
      "grad_norm": 1.328125,
      "learning_rate": 0.00012057651825672628,
      "loss": 4.7301,
      "mean_token_accuracy": 0.2277245193719864,
      "num_tokens": 165581993.0,
      "step": 95465
    },
    {
      "entropy": 5.394501829147339,
      "epoch": 7.427854503014977,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012055636206193314,
      "loss": 4.5787,
      "mean_token_accuracy": 0.24280174523591996,
      "num_tokens": 165590874.0,
      "step": 95470
    },
    {
      "entropy": 5.418191385269165,
      "epoch": 7.428243532386695,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001205362082105763,
      "loss": 4.6373,
      "mean_token_accuracy": 0.24172065258026124,
      "num_tokens": 165599305.0,
      "step": 95475
    },
    {
      "entropy": 5.3658219337463375,
      "epoch": 7.428632561758413,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012051605670296174,
      "loss": 4.6121,
      "mean_token_accuracy": 0.25252673029899597,
      "num_tokens": 165608101.0,
      "step": 95480
    },
    {
      "entropy": 5.299737596511841,
      "epoch": 7.42902159113013,
      "grad_norm": 1.296875,
      "learning_rate": 0.00012049590753939505,
      "loss": 4.5499,
      "mean_token_accuracy": 0.2471553310751915,
      "num_tokens": 165617306.0,
      "step": 95485
    },
    {
      "entropy": 5.466805267333984,
      "epoch": 7.429410620501848,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012047576072018212,
      "loss": 4.7095,
      "mean_token_accuracy": 0.2512857973575592,
      "num_tokens": 165625343.0,
      "step": 95490
    },
    {
      "entropy": 5.430821895599365,
      "epoch": 7.429799649873566,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012045561624562853,
      "loss": 4.6443,
      "mean_token_accuracy": 0.2439176321029663,
      "num_tokens": 165633597.0,
      "step": 95495
    },
    {
      "entropy": 5.42846999168396,
      "epoch": 7.430188679245283,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012043547411603999,
      "loss": 4.65,
      "mean_token_accuracy": 0.24697925746440888,
      "num_tokens": 165642373.0,
      "step": 95500
    },
    {
      "entropy": 5.281901693344116,
      "epoch": 7.430577708617,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012041533433172208,
      "loss": 4.4814,
      "mean_token_accuracy": 0.2630492985248566,
      "num_tokens": 165650619.0,
      "step": 95505
    },
    {
      "entropy": 5.334194898605347,
      "epoch": 7.430966737988718,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012039519689298049,
      "loss": 4.5959,
      "mean_token_accuracy": 0.2436278060078621,
      "num_tokens": 165658982.0,
      "step": 95510
    },
    {
      "entropy": 5.343293142318726,
      "epoch": 7.431355767360436,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00012037506180012064,
      "loss": 4.5994,
      "mean_token_accuracy": 0.253036929666996,
      "num_tokens": 165668551.0,
      "step": 95515
    },
    {
      "entropy": 5.383273315429688,
      "epoch": 7.431744796732153,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012035492905344813,
      "loss": 4.6171,
      "mean_token_accuracy": 0.24843266904354094,
      "num_tokens": 165677926.0,
      "step": 95520
    },
    {
      "entropy": 5.38317985534668,
      "epoch": 7.432133826103871,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001203347986532685,
      "loss": 4.6536,
      "mean_token_accuracy": 0.24124307930469513,
      "num_tokens": 165687090.0,
      "step": 95525
    },
    {
      "entropy": 5.370198059082031,
      "epoch": 7.432522855475589,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012031467059988702,
      "loss": 4.547,
      "mean_token_accuracy": 0.25004076212644577,
      "num_tokens": 165695057.0,
      "step": 95530
    },
    {
      "entropy": 5.353267860412598,
      "epoch": 7.432911884847306,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012029454489360929,
      "loss": 4.5897,
      "mean_token_accuracy": 0.24847921282052993,
      "num_tokens": 165703664.0,
      "step": 95535
    },
    {
      "entropy": 5.397789716720581,
      "epoch": 7.433300914219023,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012027442153474054,
      "loss": 4.6919,
      "mean_token_accuracy": 0.23618103563785553,
      "num_tokens": 165712221.0,
      "step": 95540
    },
    {
      "entropy": 5.3736083030700685,
      "epoch": 7.433689943590741,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012025430052358625,
      "loss": 4.6105,
      "mean_token_accuracy": 0.2380044773221016,
      "num_tokens": 165720728.0,
      "step": 95545
    },
    {
      "entropy": 5.30751051902771,
      "epoch": 7.434078972962459,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012023418186045164,
      "loss": 4.4851,
      "mean_token_accuracy": 0.25109519213438036,
      "num_tokens": 165729361.0,
      "step": 95550
    },
    {
      "entropy": 5.364740705490112,
      "epoch": 7.434468002334176,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012021406554564194,
      "loss": 4.6251,
      "mean_token_accuracy": 0.2447954922914505,
      "num_tokens": 165737890.0,
      "step": 95555
    },
    {
      "entropy": 5.427466964721679,
      "epoch": 7.434857031705894,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012019395157946245,
      "loss": 4.6899,
      "mean_token_accuracy": 0.23436451703310013,
      "num_tokens": 165746357.0,
      "step": 95560
    },
    {
      "entropy": 5.416901731491089,
      "epoch": 7.435246061077612,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001201738399622183,
      "loss": 4.5995,
      "mean_token_accuracy": 0.24728408008813857,
      "num_tokens": 165754484.0,
      "step": 95565
    },
    {
      "entropy": 5.36837739944458,
      "epoch": 7.435635090449329,
      "grad_norm": 1.3125,
      "learning_rate": 0.00012015373069421474,
      "loss": 4.6161,
      "mean_token_accuracy": 0.25039729923009874,
      "num_tokens": 165763937.0,
      "step": 95570
    },
    {
      "entropy": 5.385457181930542,
      "epoch": 7.436024119821046,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012013362377575683,
      "loss": 4.6171,
      "mean_token_accuracy": 0.24384610801935197,
      "num_tokens": 165772907.0,
      "step": 95575
    },
    {
      "entropy": 5.3442497730255125,
      "epoch": 7.436413149192764,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012011351920714972,
      "loss": 4.5537,
      "mean_token_accuracy": 0.253333705663681,
      "num_tokens": 165781675.0,
      "step": 95580
    },
    {
      "entropy": 5.308239364624024,
      "epoch": 7.436802178564482,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012009341698869839,
      "loss": 4.4638,
      "mean_token_accuracy": 0.2530152678489685,
      "num_tokens": 165790245.0,
      "step": 95585
    },
    {
      "entropy": 5.334100818634033,
      "epoch": 7.437191207936199,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012007331712070787,
      "loss": 4.5183,
      "mean_token_accuracy": 0.2582956373691559,
      "num_tokens": 165798049.0,
      "step": 95590
    },
    {
      "entropy": 5.303201007843017,
      "epoch": 7.437580237307917,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012005321960348325,
      "loss": 4.5947,
      "mean_token_accuracy": 0.24198537617921828,
      "num_tokens": 165806550.0,
      "step": 95595
    },
    {
      "entropy": 5.298541927337647,
      "epoch": 7.437969266679635,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00012003312443732932,
      "loss": 4.5607,
      "mean_token_accuracy": 0.24498239606618882,
      "num_tokens": 165815008.0,
      "step": 95600
    },
    {
      "entropy": 5.375693416595459,
      "epoch": 7.4383582960513515,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012001303162255114,
      "loss": 4.6453,
      "mean_token_accuracy": 0.24309327751398085,
      "num_tokens": 165823980.0,
      "step": 95605
    },
    {
      "entropy": 5.524424028396607,
      "epoch": 7.438747325423069,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011999294115945347,
      "loss": 4.7343,
      "mean_token_accuracy": 0.24227204769849778,
      "num_tokens": 165831895.0,
      "step": 95610
    },
    {
      "entropy": 5.458871507644654,
      "epoch": 7.439136354794787,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011997285304834127,
      "loss": 4.6755,
      "mean_token_accuracy": 0.23641749918460847,
      "num_tokens": 165840810.0,
      "step": 95615
    },
    {
      "entropy": 5.3983148574829105,
      "epoch": 7.4395253841665046,
      "grad_norm": 1.328125,
      "learning_rate": 0.00011995276728951919,
      "loss": 4.589,
      "mean_token_accuracy": 0.24555610716342927,
      "num_tokens": 165849149.0,
      "step": 95620
    },
    {
      "entropy": 5.431836652755737,
      "epoch": 7.439914413538222,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011993268388329216,
      "loss": 4.6328,
      "mean_token_accuracy": 0.2510825663805008,
      "num_tokens": 165858122.0,
      "step": 95625
    },
    {
      "entropy": 5.404377794265747,
      "epoch": 7.44030344290994,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011991260282996483,
      "loss": 4.6624,
      "mean_token_accuracy": 0.24317525029182435,
      "num_tokens": 165867296.0,
      "step": 95630
    },
    {
      "entropy": 5.367813396453857,
      "epoch": 7.440692472281658,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011989252412984183,
      "loss": 4.5842,
      "mean_token_accuracy": 0.2521876871585846,
      "num_tokens": 165875660.0,
      "step": 95635
    },
    {
      "entropy": 5.416261148452759,
      "epoch": 7.4410815016533745,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011987244778322797,
      "loss": 4.6175,
      "mean_token_accuracy": 0.23864393532276154,
      "num_tokens": 165883529.0,
      "step": 95640
    },
    {
      "entropy": 5.396754550933838,
      "epoch": 7.441470531025092,
      "grad_norm": 1.375,
      "learning_rate": 0.00011985237379042775,
      "loss": 4.7279,
      "mean_token_accuracy": 0.23898280113935472,
      "num_tokens": 165892612.0,
      "step": 95645
    },
    {
      "entropy": 5.317754554748535,
      "epoch": 7.44185956039681,
      "grad_norm": 1.328125,
      "learning_rate": 0.00011983230215174586,
      "loss": 4.5539,
      "mean_token_accuracy": 0.25641758292913436,
      "num_tokens": 165901247.0,
      "step": 95650
    },
    {
      "entropy": 5.444772672653198,
      "epoch": 7.4422485897685275,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001198122328674867,
      "loss": 4.707,
      "mean_token_accuracy": 0.24594861418008804,
      "num_tokens": 165909655.0,
      "step": 95655
    },
    {
      "entropy": 5.3747947216033936,
      "epoch": 7.442637619140245,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011979216593795502,
      "loss": 4.6219,
      "mean_token_accuracy": 0.2442621573805809,
      "num_tokens": 165918374.0,
      "step": 95660
    },
    {
      "entropy": 5.346795272827149,
      "epoch": 7.443026648511963,
      "grad_norm": 1.375,
      "learning_rate": 0.00011977210136345516,
      "loss": 4.6367,
      "mean_token_accuracy": 0.24686167389154434,
      "num_tokens": 165926883.0,
      "step": 95665
    },
    {
      "entropy": 5.362995672225952,
      "epoch": 7.44341567788368,
      "grad_norm": 1.375,
      "learning_rate": 0.00011975203914429157,
      "loss": 4.5725,
      "mean_token_accuracy": 0.2416232004761696,
      "num_tokens": 165936112.0,
      "step": 95670
    },
    {
      "entropy": 5.39398603439331,
      "epoch": 7.4438047072553974,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011973197928076872,
      "loss": 4.6294,
      "mean_token_accuracy": 0.23947879523038865,
      "num_tokens": 165945044.0,
      "step": 95675
    },
    {
      "entropy": 5.351245164871216,
      "epoch": 7.444193736627115,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011971192177319089,
      "loss": 4.4969,
      "mean_token_accuracy": 0.25778488218784334,
      "num_tokens": 165953513.0,
      "step": 95680
    },
    {
      "entropy": 5.3747358322143555,
      "epoch": 7.444582765998833,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011969186662186256,
      "loss": 4.5913,
      "mean_token_accuracy": 0.2514837920665741,
      "num_tokens": 165962360.0,
      "step": 95685
    },
    {
      "entropy": 5.343138647079468,
      "epoch": 7.4449717953705505,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011967181382708787,
      "loss": 4.6475,
      "mean_token_accuracy": 0.24636254757642745,
      "num_tokens": 165971512.0,
      "step": 95690
    },
    {
      "entropy": 5.365707015991211,
      "epoch": 7.445360824742268,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011965176338917125,
      "loss": 4.6436,
      "mean_token_accuracy": 0.24505784064531327,
      "num_tokens": 165980759.0,
      "step": 95695
    },
    {
      "entropy": 5.387465143203736,
      "epoch": 7.445749854113986,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001196317153084168,
      "loss": 4.5935,
      "mean_token_accuracy": 0.24762510657310485,
      "num_tokens": 165988729.0,
      "step": 95700
    },
    {
      "entropy": 5.36963963508606,
      "epoch": 7.446138883485704,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00011961166958512883,
      "loss": 4.5838,
      "mean_token_accuracy": 0.2451114609837532,
      "num_tokens": 165998229.0,
      "step": 95705
    },
    {
      "entropy": 5.3854756355285645,
      "epoch": 7.44652791285742,
      "grad_norm": 1.375,
      "learning_rate": 0.00011959162621961144,
      "loss": 4.5873,
      "mean_token_accuracy": 0.2519699841737747,
      "num_tokens": 166006636.0,
      "step": 95710
    },
    {
      "entropy": 5.405295658111572,
      "epoch": 7.446916942229138,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011957158521216872,
      "loss": 4.6053,
      "mean_token_accuracy": 0.24768915474414827,
      "num_tokens": 166015052.0,
      "step": 95715
    },
    {
      "entropy": 5.241939401626587,
      "epoch": 7.447305971600856,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001195515465631048,
      "loss": 4.4524,
      "mean_token_accuracy": 0.2540135011076927,
      "num_tokens": 166023388.0,
      "step": 95720
    },
    {
      "entropy": 5.315170383453369,
      "epoch": 7.4476950009725735,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011953151027272372,
      "loss": 4.5529,
      "mean_token_accuracy": 0.2503159955143929,
      "num_tokens": 166032038.0,
      "step": 95725
    },
    {
      "entropy": 5.36104736328125,
      "epoch": 7.448084030344291,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011951147634132961,
      "loss": 4.6369,
      "mean_token_accuracy": 0.24176654368638992,
      "num_tokens": 166040242.0,
      "step": 95730
    },
    {
      "entropy": 5.343714046478271,
      "epoch": 7.448473059716009,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011949144476922628,
      "loss": 4.6016,
      "mean_token_accuracy": 0.23465361148118974,
      "num_tokens": 166048778.0,
      "step": 95735
    },
    {
      "entropy": 5.343772459030151,
      "epoch": 7.448862089087726,
      "grad_norm": 1.296875,
      "learning_rate": 0.00011947141555671783,
      "loss": 4.5574,
      "mean_token_accuracy": 0.24613203704357148,
      "num_tokens": 166058124.0,
      "step": 95740
    },
    {
      "entropy": 5.33361668586731,
      "epoch": 7.449251118459443,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011945138870410809,
      "loss": 4.5354,
      "mean_token_accuracy": 0.2586942493915558,
      "num_tokens": 166065880.0,
      "step": 95745
    },
    {
      "entropy": 5.383159351348877,
      "epoch": 7.449640147831161,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011943136421170091,
      "loss": 4.6304,
      "mean_token_accuracy": 0.2486148178577423,
      "num_tokens": 166074077.0,
      "step": 95750
    },
    {
      "entropy": 5.322824621200562,
      "epoch": 7.450029177202879,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011941134207980018,
      "loss": 4.5373,
      "mean_token_accuracy": 0.25074647814035417,
      "num_tokens": 166082161.0,
      "step": 95755
    },
    {
      "entropy": 5.406968402862549,
      "epoch": 7.4504182065745965,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011939132230870967,
      "loss": 4.6791,
      "mean_token_accuracy": 0.243515844643116,
      "num_tokens": 166091239.0,
      "step": 95760
    },
    {
      "entropy": 5.309253358840943,
      "epoch": 7.450807235946314,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001193713048987332,
      "loss": 4.5222,
      "mean_token_accuracy": 0.2577675715088844,
      "num_tokens": 166101446.0,
      "step": 95765
    },
    {
      "entropy": 5.372639322280884,
      "epoch": 7.451196265318032,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011935128985017442,
      "loss": 4.6099,
      "mean_token_accuracy": 0.2511883169412613,
      "num_tokens": 166110423.0,
      "step": 95770
    },
    {
      "entropy": 5.386543464660645,
      "epoch": 7.451585294689749,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011933127716333712,
      "loss": 4.6641,
      "mean_token_accuracy": 0.24615657180547715,
      "num_tokens": 166118841.0,
      "step": 95775
    },
    {
      "entropy": 5.432804441452026,
      "epoch": 7.451974324061466,
      "grad_norm": 1.328125,
      "learning_rate": 0.00011931126683852489,
      "loss": 4.6706,
      "mean_token_accuracy": 0.242329640686512,
      "num_tokens": 166127592.0,
      "step": 95780
    },
    {
      "entropy": 5.3710548877716064,
      "epoch": 7.452363353433184,
      "grad_norm": 1.265625,
      "learning_rate": 0.00011929125887604137,
      "loss": 4.6138,
      "mean_token_accuracy": 0.2450222685933113,
      "num_tokens": 166136718.0,
      "step": 95785
    },
    {
      "entropy": 5.4355552196502686,
      "epoch": 7.452752382804902,
      "grad_norm": 1.375,
      "learning_rate": 0.0001192712532761902,
      "loss": 4.6659,
      "mean_token_accuracy": 0.23889132887125014,
      "num_tokens": 166145592.0,
      "step": 95790
    },
    {
      "entropy": 5.419160509109497,
      "epoch": 7.4531414121766195,
      "grad_norm": 1.3125,
      "learning_rate": 0.00011925125003927485,
      "loss": 4.6844,
      "mean_token_accuracy": 0.2371038407087326,
      "num_tokens": 166154325.0,
      "step": 95795
    },
    {
      "entropy": 5.3729997158050535,
      "epoch": 7.453530441548337,
      "grad_norm": 1.515625,
      "learning_rate": 0.00011923124916559894,
      "loss": 4.5341,
      "mean_token_accuracy": 0.2546359434723854,
      "num_tokens": 166162993.0,
      "step": 95800
    },
    {
      "entropy": 5.357900762557984,
      "epoch": 7.453919470920054,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011921125065546585,
      "loss": 4.5155,
      "mean_token_accuracy": 0.24916453361511232,
      "num_tokens": 166171720.0,
      "step": 95805
    },
    {
      "entropy": 5.39055438041687,
      "epoch": 7.454308500291772,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011919125450917912,
      "loss": 4.6472,
      "mean_token_accuracy": 0.24852005988359452,
      "num_tokens": 166180161.0,
      "step": 95810
    },
    {
      "entropy": 5.34354453086853,
      "epoch": 7.454697529663489,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000119171260727042,
      "loss": 4.6439,
      "mean_token_accuracy": 0.24071654081344604,
      "num_tokens": 166189237.0,
      "step": 95815
    },
    {
      "entropy": 5.421201992034912,
      "epoch": 7.455086559035207,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011915126930935808,
      "loss": 4.6576,
      "mean_token_accuracy": 0.24766531735658645,
      "num_tokens": 166197772.0,
      "step": 95820
    },
    {
      "entropy": 5.314159202575683,
      "epoch": 7.455475588406925,
      "grad_norm": 1.46875,
      "learning_rate": 0.00011913128025643059,
      "loss": 4.5439,
      "mean_token_accuracy": 0.2533565878868103,
      "num_tokens": 166206170.0,
      "step": 95825
    },
    {
      "entropy": 5.386200428009033,
      "epoch": 7.4558646177786425,
      "grad_norm": 1.46875,
      "learning_rate": 0.00011911129356856276,
      "loss": 4.6654,
      "mean_token_accuracy": 0.24182116389274597,
      "num_tokens": 166214776.0,
      "step": 95830
    },
    {
      "entropy": 5.357251310348511,
      "epoch": 7.45625364715036,
      "grad_norm": 1.265625,
      "learning_rate": 0.00011909130924605799,
      "loss": 4.5627,
      "mean_token_accuracy": 0.24915471524000168,
      "num_tokens": 166224452.0,
      "step": 95835
    },
    {
      "entropy": 5.418583488464355,
      "epoch": 7.456642676522077,
      "grad_norm": 1.5,
      "learning_rate": 0.00011907132728921939,
      "loss": 4.6164,
      "mean_token_accuracy": 0.23772496581077576,
      "num_tokens": 166232553.0,
      "step": 95840
    },
    {
      "entropy": 5.256591653823852,
      "epoch": 7.457031705893795,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011905134769835028,
      "loss": 4.493,
      "mean_token_accuracy": 0.25568588376045226,
      "num_tokens": 166241742.0,
      "step": 95845
    },
    {
      "entropy": 5.4266359329223635,
      "epoch": 7.457420735265512,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011903137047375366,
      "loss": 4.6194,
      "mean_token_accuracy": 0.2555432364344597,
      "num_tokens": 166250563.0,
      "step": 95850
    },
    {
      "entropy": 5.350070953369141,
      "epoch": 7.45780976463723,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011901139561573275,
      "loss": 4.5356,
      "mean_token_accuracy": 0.25546029657125474,
      "num_tokens": 166258775.0,
      "step": 95855
    },
    {
      "entropy": 5.356121587753296,
      "epoch": 7.458198794008948,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011899142312459072,
      "loss": 4.6393,
      "mean_token_accuracy": 0.2511505126953125,
      "num_tokens": 166267197.0,
      "step": 95860
    },
    {
      "entropy": 5.354912519454956,
      "epoch": 7.4585878233806655,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011897145300063045,
      "loss": 4.6155,
      "mean_token_accuracy": 0.2456697478890419,
      "num_tokens": 166276451.0,
      "step": 95865
    },
    {
      "entropy": 5.406034231185913,
      "epoch": 7.458976852752383,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011895148524415509,
      "loss": 4.6829,
      "mean_token_accuracy": 0.24152234196662903,
      "num_tokens": 166285197.0,
      "step": 95870
    },
    {
      "entropy": 5.4375701427459715,
      "epoch": 7.4593658821241,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011893151985546751,
      "loss": 4.6596,
      "mean_token_accuracy": 0.23399144113063813,
      "num_tokens": 166293763.0,
      "step": 95875
    },
    {
      "entropy": 5.389990901947021,
      "epoch": 7.459754911495818,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011891155683487079,
      "loss": 4.632,
      "mean_token_accuracy": 0.24837333112955093,
      "num_tokens": 166302881.0,
      "step": 95880
    },
    {
      "entropy": 5.390180015563965,
      "epoch": 7.460143940867535,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011889159618266768,
      "loss": 4.578,
      "mean_token_accuracy": 0.251559779047966,
      "num_tokens": 166311810.0,
      "step": 95885
    },
    {
      "entropy": 5.385951042175293,
      "epoch": 7.460532970239253,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011887163789916118,
      "loss": 4.5858,
      "mean_token_accuracy": 0.24798345118761062,
      "num_tokens": 166320250.0,
      "step": 95890
    },
    {
      "entropy": 5.355606508255005,
      "epoch": 7.460921999610971,
      "grad_norm": 1.546875,
      "learning_rate": 0.000118851681984654,
      "loss": 4.5814,
      "mean_token_accuracy": 0.2486318051815033,
      "num_tokens": 166328599.0,
      "step": 95895
    },
    {
      "entropy": 5.421585464477539,
      "epoch": 7.4613110289826885,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011883172843944907,
      "loss": 4.6885,
      "mean_token_accuracy": 0.2322828009724617,
      "num_tokens": 166336878.0,
      "step": 95900
    },
    {
      "entropy": 5.4028849601745605,
      "epoch": 7.461700058354406,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011881177726384912,
      "loss": 4.6513,
      "mean_token_accuracy": 0.2369014948606491,
      "num_tokens": 166345563.0,
      "step": 95905
    },
    {
      "entropy": 5.401794147491455,
      "epoch": 7.462089087726123,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011879182845815678,
      "loss": 4.6335,
      "mean_token_accuracy": 0.24470696747303008,
      "num_tokens": 166353395.0,
      "step": 95910
    },
    {
      "entropy": 5.29552321434021,
      "epoch": 7.462478117097841,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011877188202267487,
      "loss": 4.4087,
      "mean_token_accuracy": 0.2603929743170738,
      "num_tokens": 166362245.0,
      "step": 95915
    },
    {
      "entropy": 5.348288106918335,
      "epoch": 7.462867146469558,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011875193795770585,
      "loss": 4.5486,
      "mean_token_accuracy": 0.2499289557337761,
      "num_tokens": 166370883.0,
      "step": 95920
    },
    {
      "entropy": 5.334297466278076,
      "epoch": 7.463256175841276,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011873199626355265,
      "loss": 4.5959,
      "mean_token_accuracy": 0.252534918487072,
      "num_tokens": 166379245.0,
      "step": 95925
    },
    {
      "entropy": 5.3312122344970705,
      "epoch": 7.463645205212994,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011871205694051763,
      "loss": 4.6581,
      "mean_token_accuracy": 0.2459083989262581,
      "num_tokens": 166388692.0,
      "step": 95930
    },
    {
      "entropy": 5.395445680618286,
      "epoch": 7.4640342345847115,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011869211998890334,
      "loss": 4.6579,
      "mean_token_accuracy": 0.24433287978172302,
      "num_tokens": 166397556.0,
      "step": 95935
    },
    {
      "entropy": 5.381791782379151,
      "epoch": 7.464423263956428,
      "grad_norm": 1.5625,
      "learning_rate": 0.00011867218540901242,
      "loss": 4.534,
      "mean_token_accuracy": 0.24882991909980773,
      "num_tokens": 166405683.0,
      "step": 95940
    },
    {
      "entropy": 5.289937114715576,
      "epoch": 7.464812293328146,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011865225320114717,
      "loss": 4.4724,
      "mean_token_accuracy": 0.26282182335853577,
      "num_tokens": 166414806.0,
      "step": 95945
    },
    {
      "entropy": 5.341854000091553,
      "epoch": 7.465201322699864,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001186323233656102,
      "loss": 4.5796,
      "mean_token_accuracy": 0.2488087981939316,
      "num_tokens": 166423147.0,
      "step": 95950
    },
    {
      "entropy": 5.406926393508911,
      "epoch": 7.465590352071581,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011861239590270379,
      "loss": 4.6653,
      "mean_token_accuracy": 0.24202370047569274,
      "num_tokens": 166431429.0,
      "step": 95955
    },
    {
      "entropy": 5.334937524795532,
      "epoch": 7.465979381443299,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001185924708127304,
      "loss": 4.5903,
      "mean_token_accuracy": 0.25411690324544906,
      "num_tokens": 166440900.0,
      "step": 95960
    },
    {
      "entropy": 5.358508014678955,
      "epoch": 7.466368410815017,
      "grad_norm": 1.3125,
      "learning_rate": 0.00011857254809599226,
      "loss": 4.5808,
      "mean_token_accuracy": 0.2512654513120651,
      "num_tokens": 166450483.0,
      "step": 95965
    },
    {
      "entropy": 5.342708778381348,
      "epoch": 7.4667574401867345,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011855262775279178,
      "loss": 4.6462,
      "mean_token_accuracy": 0.24283837676048278,
      "num_tokens": 166459226.0,
      "step": 95970
    },
    {
      "entropy": 5.382832956314087,
      "epoch": 7.467146469558451,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011853270978343111,
      "loss": 4.6037,
      "mean_token_accuracy": 0.24520778954029082,
      "num_tokens": 166467951.0,
      "step": 95975
    },
    {
      "entropy": 5.390552806854248,
      "epoch": 7.467535498930169,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001185127941882126,
      "loss": 4.5447,
      "mean_token_accuracy": 0.24876424819231033,
      "num_tokens": 166476182.0,
      "step": 95980
    },
    {
      "entropy": 5.390538263320923,
      "epoch": 7.467924528301887,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00011849288096743827,
      "loss": 4.5952,
      "mean_token_accuracy": 0.24609564244747162,
      "num_tokens": 166485039.0,
      "step": 95985
    },
    {
      "entropy": 5.273808765411377,
      "epoch": 7.468313557673604,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001184729701214104,
      "loss": 4.5068,
      "mean_token_accuracy": 0.25508918315172197,
      "num_tokens": 166493719.0,
      "step": 95990
    },
    {
      "entropy": 5.3568971157073975,
      "epoch": 7.468702587045322,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001184530616504311,
      "loss": 4.6041,
      "mean_token_accuracy": 0.25048181265592573,
      "num_tokens": 166501696.0,
      "step": 95995
    },
    {
      "entropy": 5.330262279510498,
      "epoch": 7.46909161641704,
      "grad_norm": 1.328125,
      "learning_rate": 0.00011843315555480237,
      "loss": 4.5932,
      "mean_token_accuracy": 0.2517843395471573,
      "num_tokens": 166510551.0,
      "step": 96000
    },
    {
      "epoch": 7.46909161641704,
      "eval_entropy": 5.083234449208528,
      "eval_loss": 4.84763765335083,
      "eval_mean_token_accuracy": 0.23805892816722335,
      "eval_num_tokens": 166510551.0,
      "eval_runtime": 28.8759,
      "eval_samples_per_second": 1439.194,
      "eval_steps_per_second": 179.908,
      "step": 96000
    }
  ],
  "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": 2.25989335575936e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}