{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.07840185655596324,
  "eval_steps": 500,
  "global_step": 2500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 10.651023864746094,
      "epoch": 0.00015680371311192648,
      "grad_norm": 11.5625,
      "learning_rate": 2e-06,
      "loss": 10.8028,
      "mean_token_accuracy": 6.958942394703626e-05,
      "num_tokens": 14082.0,
      "step": 5
    },
    {
      "entropy": 10.651004600524903,
      "epoch": 0.00031360742622385296,
      "grad_norm": 11.375,
      "learning_rate": 4.5e-06,
      "loss": 10.7633,
      "mean_token_accuracy": 6.574622238986194e-05,
      "num_tokens": 29900.0,
      "step": 10
    },
    {
      "entropy": 10.650343418121338,
      "epoch": 0.00047041113933577947,
      "grad_norm": 8.4375,
      "learning_rate": 7e-06,
      "loss": 10.5185,
      "mean_token_accuracy": 0.01316747289383784,
      "num_tokens": 43394.0,
      "step": 15
    },
    {
      "entropy": 10.64456491470337,
      "epoch": 0.0006272148524477059,
      "grad_norm": 5.34375,
      "learning_rate": 9.5e-06,
      "loss": 10.2603,
      "mean_token_accuracy": 0.03233455698937178,
      "num_tokens": 58563.0,
      "step": 20
    },
    {
      "entropy": 10.616307353973388,
      "epoch": 0.0007840185655596325,
      "grad_norm": 3.609375,
      "learning_rate": 1.2e-05,
      "loss": 9.9833,
      "mean_token_accuracy": 0.034109539538621905,
      "num_tokens": 73923.0,
      "step": 25
    },
    {
      "entropy": 10.588720226287842,
      "epoch": 0.0009408222786715589,
      "grad_norm": 2.984375,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 9.8466,
      "mean_token_accuracy": 0.03426761133596301,
      "num_tokens": 87652.0,
      "step": 30
    },
    {
      "entropy": 10.595056247711181,
      "epoch": 0.0010976259917834855,
      "grad_norm": 2.671875,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 9.7741,
      "mean_token_accuracy": 0.031784738413989544,
      "num_tokens": 101369.0,
      "step": 35
    },
    {
      "entropy": 10.59932107925415,
      "epoch": 0.0012544297048954118,
      "grad_norm": 2.515625,
      "learning_rate": 1.95e-05,
      "loss": 9.6922,
      "mean_token_accuracy": 0.03506205677986145,
      "num_tokens": 114428.0,
      "step": 40
    },
    {
      "entropy": 10.581777667999267,
      "epoch": 0.0014112334180073384,
      "grad_norm": 2.3125,
      "learning_rate": 2.2e-05,
      "loss": 9.706,
      "mean_token_accuracy": 0.03386611118912697,
      "num_tokens": 130472.0,
      "step": 45
    },
    {
      "entropy": 10.577643585205077,
      "epoch": 0.001568037131119265,
      "grad_norm": 2.40625,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.6852,
      "mean_token_accuracy": 0.030213401652872562,
      "num_tokens": 144077.0,
      "step": 50
    },
    {
      "entropy": 10.59252758026123,
      "epoch": 0.0017248408442311913,
      "grad_norm": 2.15625,
      "learning_rate": 2.7e-05,
      "loss": 9.6373,
      "mean_token_accuracy": 0.03242910895496607,
      "num_tokens": 158941.0,
      "step": 55
    },
    {
      "entropy": 10.579074954986572,
      "epoch": 0.0018816445573431179,
      "grad_norm": 2.140625,
      "learning_rate": 2.95e-05,
      "loss": 9.563,
      "mean_token_accuracy": 0.02959225047379732,
      "num_tokens": 174362.0,
      "step": 60
    },
    {
      "entropy": 10.570305156707764,
      "epoch": 0.002038448270455044,
      "grad_norm": 2.09375,
      "learning_rate": 3.2e-05,
      "loss": 9.4485,
      "mean_token_accuracy": 0.033058703131973746,
      "num_tokens": 190929.0,
      "step": 65
    },
    {
      "entropy": 10.55489387512207,
      "epoch": 0.002195251983566971,
      "grad_norm": 2.015625,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.3387,
      "mean_token_accuracy": 0.033958965167403224,
      "num_tokens": 203660.0,
      "step": 70
    },
    {
      "entropy": 10.530017757415772,
      "epoch": 0.0023520556966788973,
      "grad_norm": 2.0625,
      "learning_rate": 3.7e-05,
      "loss": 9.2826,
      "mean_token_accuracy": 0.027879416570067407,
      "num_tokens": 218421.0,
      "step": 75
    },
    {
      "entropy": 10.534087371826171,
      "epoch": 0.0025088594097908237,
      "grad_norm": 2.09375,
      "learning_rate": 3.95e-05,
      "loss": 9.1439,
      "mean_token_accuracy": 0.03472804371267557,
      "num_tokens": 233734.0,
      "step": 80
    },
    {
      "entropy": 10.496679878234863,
      "epoch": 0.0026656631229027505,
      "grad_norm": 1.9765625,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.0182,
      "mean_token_accuracy": 0.02818337455391884,
      "num_tokens": 247105.0,
      "step": 85
    },
    {
      "entropy": 10.467080688476562,
      "epoch": 0.002822466836014677,
      "grad_norm": 1.9140625,
      "learning_rate": 4.45e-05,
      "loss": 8.928,
      "mean_token_accuracy": 0.03165087569504976,
      "num_tokens": 262487.0,
      "step": 90
    },
    {
      "entropy": 10.403542804718018,
      "epoch": 0.002979270549126603,
      "grad_norm": 1.9921875,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 8.7777,
      "mean_token_accuracy": 0.036813986487686634,
      "num_tokens": 276705.0,
      "step": 95
    },
    {
      "entropy": 10.354411220550537,
      "epoch": 0.00313607426223853,
      "grad_norm": 1.9765625,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 8.6775,
      "mean_token_accuracy": 0.04005452003329992,
      "num_tokens": 291031.0,
      "step": 100
    },
    {
      "entropy": 10.269740104675293,
      "epoch": 0.0032928779753504563,
      "grad_norm": 1.890625,
      "learning_rate": 5.2e-05,
      "loss": 8.5069,
      "mean_token_accuracy": 0.040437552519142626,
      "num_tokens": 305444.0,
      "step": 105
    },
    {
      "entropy": 10.179180717468261,
      "epoch": 0.0034496816884623826,
      "grad_norm": 1.8671875,
      "learning_rate": 5.45e-05,
      "loss": 8.3811,
      "mean_token_accuracy": 0.04589438661932945,
      "num_tokens": 321168.0,
      "step": 110
    },
    {
      "entropy": 10.070209217071532,
      "epoch": 0.0036064854015743094,
      "grad_norm": 1.7265625,
      "learning_rate": 5.7e-05,
      "loss": 8.1795,
      "mean_token_accuracy": 0.05213871449232101,
      "num_tokens": 334988.0,
      "step": 115
    },
    {
      "entropy": 9.922878646850586,
      "epoch": 0.0037632891146862357,
      "grad_norm": 1.6796875,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.0559,
      "mean_token_accuracy": 0.05347799621522427,
      "num_tokens": 349049.0,
      "step": 120
    },
    {
      "entropy": 9.7627742767334,
      "epoch": 0.0039200928277981625,
      "grad_norm": 1.765625,
      "learning_rate": 6.2e-05,
      "loss": 8.0302,
      "mean_token_accuracy": 0.05416244864463806,
      "num_tokens": 365497.0,
      "step": 125
    },
    {
      "entropy": 9.585789394378661,
      "epoch": 0.004076896540910088,
      "grad_norm": 1.5,
      "learning_rate": 6.450000000000001e-05,
      "loss": 7.8073,
      "mean_token_accuracy": 0.05755908451974392,
      "num_tokens": 380324.0,
      "step": 130
    },
    {
      "entropy": 9.368918418884277,
      "epoch": 0.004233700254022015,
      "grad_norm": 1.59375,
      "learning_rate": 6.7e-05,
      "loss": 7.6844,
      "mean_token_accuracy": 0.06051859185099602,
      "num_tokens": 393766.0,
      "step": 135
    },
    {
      "entropy": 9.158668899536133,
      "epoch": 0.004390503967133942,
      "grad_norm": 1.4296875,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.6272,
      "mean_token_accuracy": 0.06358711272478104,
      "num_tokens": 409369.0,
      "step": 140
    },
    {
      "entropy": 8.945510005950927,
      "epoch": 0.004547307680245868,
      "grad_norm": 1.546875,
      "learning_rate": 7.2e-05,
      "loss": 7.5019,
      "mean_token_accuracy": 0.06252376027405263,
      "num_tokens": 425475.0,
      "step": 145
    },
    {
      "entropy": 8.711659908294678,
      "epoch": 0.004704111393357795,
      "grad_norm": 1.59375,
      "learning_rate": 7.45e-05,
      "loss": 7.3832,
      "mean_token_accuracy": 0.06495344713330269,
      "num_tokens": 440853.0,
      "step": 150
    },
    {
      "entropy": 8.498833179473877,
      "epoch": 0.0048609151064697214,
      "grad_norm": 1.609375,
      "learning_rate": 7.7e-05,
      "loss": 7.3471,
      "mean_token_accuracy": 0.06971911787986755,
      "num_tokens": 457407.0,
      "step": 155
    },
    {
      "entropy": 8.355002212524415,
      "epoch": 0.005017718819581647,
      "grad_norm": 1.09375,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.2971,
      "mean_token_accuracy": 0.06931432262063027,
      "num_tokens": 472813.0,
      "step": 160
    },
    {
      "entropy": 8.208080291748047,
      "epoch": 0.005174522532693574,
      "grad_norm": 1.5,
      "learning_rate": 8.2e-05,
      "loss": 7.3059,
      "mean_token_accuracy": 0.06723905019462109,
      "num_tokens": 489742.0,
      "step": 165
    },
    {
      "entropy": 8.143133544921875,
      "epoch": 0.005331326245805501,
      "grad_norm": 1.4375,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.2399,
      "mean_token_accuracy": 0.06843472644686699,
      "num_tokens": 503493.0,
      "step": 170
    },
    {
      "entropy": 8.036779308319092,
      "epoch": 0.005488129958917427,
      "grad_norm": 1.1796875,
      "learning_rate": 8.7e-05,
      "loss": 7.1275,
      "mean_token_accuracy": 0.07654446959495545,
      "num_tokens": 520143.0,
      "step": 175
    },
    {
      "entropy": 7.934807920455933,
      "epoch": 0.005644933672029354,
      "grad_norm": 1.3203125,
      "learning_rate": 8.95e-05,
      "loss": 7.232,
      "mean_token_accuracy": 0.06283588744699956,
      "num_tokens": 534776.0,
      "step": 180
    },
    {
      "entropy": 7.953633880615234,
      "epoch": 0.00580173738514128,
      "grad_norm": 1.3359375,
      "learning_rate": 9.2e-05,
      "loss": 7.2097,
      "mean_token_accuracy": 0.06803618557751179,
      "num_tokens": 550256.0,
      "step": 185
    },
    {
      "entropy": 7.894899654388428,
      "epoch": 0.005958541098253206,
      "grad_norm": 1.1796875,
      "learning_rate": 9.45e-05,
      "loss": 7.2178,
      "mean_token_accuracy": 0.0686919379979372,
      "num_tokens": 564797.0,
      "step": 190
    },
    {
      "entropy": 7.844494724273682,
      "epoch": 0.006115344811365133,
      "grad_norm": 1.1953125,
      "learning_rate": 9.7e-05,
      "loss": 6.9952,
      "mean_token_accuracy": 0.0732124924659729,
      "num_tokens": 579721.0,
      "step": 195
    },
    {
      "entropy": 7.716418170928955,
      "epoch": 0.00627214852447706,
      "grad_norm": 1.359375,
      "learning_rate": 9.95e-05,
      "loss": 6.9564,
      "mean_token_accuracy": 0.07807120829820632,
      "num_tokens": 593431.0,
      "step": 200
    },
    {
      "entropy": 7.715715551376343,
      "epoch": 0.006428952237588986,
      "grad_norm": 1.6796875,
      "learning_rate": 0.000102,
      "loss": 6.9708,
      "mean_token_accuracy": 0.07868832051753998,
      "num_tokens": 608176.0,
      "step": 205
    },
    {
      "entropy": 7.692095470428467,
      "epoch": 0.0065857559507009125,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010449999999999999,
      "loss": 6.9833,
      "mean_token_accuracy": 0.07725078389048576,
      "num_tokens": 621784.0,
      "step": 210
    },
    {
      "entropy": 7.700586032867432,
      "epoch": 0.006742559663812839,
      "grad_norm": 1.421875,
      "learning_rate": 0.000107,
      "loss": 6.9716,
      "mean_token_accuracy": 0.07797399312257766,
      "num_tokens": 634707.0,
      "step": 215
    },
    {
      "entropy": 7.623541641235351,
      "epoch": 0.006899363376924765,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001095,
      "loss": 7.0494,
      "mean_token_accuracy": 0.0720802016556263,
      "num_tokens": 647835.0,
      "step": 220
    },
    {
      "entropy": 7.631550598144531,
      "epoch": 0.007056167090036692,
      "grad_norm": 1.3125,
      "learning_rate": 0.000112,
      "loss": 6.9591,
      "mean_token_accuracy": 0.07618656158447265,
      "num_tokens": 662587.0,
      "step": 225
    },
    {
      "entropy": 7.684398460388183,
      "epoch": 0.007212970803148619,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001145,
      "loss": 7.0602,
      "mean_token_accuracy": 0.07651265673339366,
      "num_tokens": 676329.0,
      "step": 230
    },
    {
      "entropy": 7.561251878738403,
      "epoch": 0.007369774516260545,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011700000000000001,
      "loss": 6.953,
      "mean_token_accuracy": 0.08274711072444915,
      "num_tokens": 692541.0,
      "step": 235
    },
    {
      "entropy": 7.560929489135742,
      "epoch": 0.0075265782293724715,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.0198,
      "mean_token_accuracy": 0.07723658978939056,
      "num_tokens": 706997.0,
      "step": 240
    },
    {
      "entropy": 7.557625770568848,
      "epoch": 0.007683381942484398,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000122,
      "loss": 6.9217,
      "mean_token_accuracy": 0.08156572207808495,
      "num_tokens": 720719.0,
      "step": 245
    },
    {
      "entropy": 7.592370510101318,
      "epoch": 0.007840185655596325,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001245,
      "loss": 7.0018,
      "mean_token_accuracy": 0.08049147836863994,
      "num_tokens": 734572.0,
      "step": 250
    },
    {
      "entropy": 7.522368669509888,
      "epoch": 0.007996989368708251,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000127,
      "loss": 6.8688,
      "mean_token_accuracy": 0.08029020316898823,
      "num_tokens": 751365.0,
      "step": 255
    },
    {
      "entropy": 7.458198308944702,
      "epoch": 0.008153793081820177,
      "grad_norm": 1.375,
      "learning_rate": 0.0001295,
      "loss": 6.8057,
      "mean_token_accuracy": 0.08540241345763207,
      "num_tokens": 765549.0,
      "step": 260
    },
    {
      "entropy": 7.365204620361328,
      "epoch": 0.008310596794932104,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000132,
      "loss": 6.7974,
      "mean_token_accuracy": 0.0849799670279026,
      "num_tokens": 780071.0,
      "step": 265
    },
    {
      "entropy": 7.453489208221436,
      "epoch": 0.00846740050804403,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013450000000000002,
      "loss": 6.8488,
      "mean_token_accuracy": 0.08373216427862644,
      "num_tokens": 795328.0,
      "step": 270
    },
    {
      "entropy": 7.333805704116822,
      "epoch": 0.008624204221155956,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00013700000000000002,
      "loss": 6.759,
      "mean_token_accuracy": 0.09399082660675048,
      "num_tokens": 808404.0,
      "step": 275
    },
    {
      "entropy": 7.400871515274048,
      "epoch": 0.008781007934267884,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001395,
      "loss": 6.857,
      "mean_token_accuracy": 0.08847371190786361,
      "num_tokens": 823204.0,
      "step": 280
    },
    {
      "entropy": 7.292252063751221,
      "epoch": 0.00893781164737981,
      "grad_norm": 1.5,
      "learning_rate": 0.00014199999999999998,
      "loss": 6.7636,
      "mean_token_accuracy": 0.08776607438921928,
      "num_tokens": 836276.0,
      "step": 285
    },
    {
      "entropy": 7.366798543930054,
      "epoch": 0.009094615360491736,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001445,
      "loss": 6.7562,
      "mean_token_accuracy": 0.09157689064741134,
      "num_tokens": 850090.0,
      "step": 290
    },
    {
      "entropy": 7.297074222564698,
      "epoch": 0.009251419073603663,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000147,
      "loss": 6.7845,
      "mean_token_accuracy": 0.08689201548695565,
      "num_tokens": 865931.0,
      "step": 295
    },
    {
      "entropy": 7.316424083709717,
      "epoch": 0.00940822278671559,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001495,
      "loss": 6.7825,
      "mean_token_accuracy": 0.08415777459740639,
      "num_tokens": 879960.0,
      "step": 300
    },
    {
      "entropy": 7.245685482025147,
      "epoch": 0.009565026499827515,
      "grad_norm": 1.125,
      "learning_rate": 0.000152,
      "loss": 6.7297,
      "mean_token_accuracy": 0.08968461528420449,
      "num_tokens": 894711.0,
      "step": 305
    },
    {
      "entropy": 7.20286979675293,
      "epoch": 0.009721830212939443,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015450000000000001,
      "loss": 6.6488,
      "mean_token_accuracy": 0.08769082948565483,
      "num_tokens": 911461.0,
      "step": 310
    },
    {
      "entropy": 7.2604657173156735,
      "epoch": 0.009878633926051369,
      "grad_norm": 1.515625,
      "learning_rate": 0.000157,
      "loss": 6.8178,
      "mean_token_accuracy": 0.08734595999121667,
      "num_tokens": 926618.0,
      "step": 315
    },
    {
      "entropy": 7.233215236663819,
      "epoch": 0.010035437639163295,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001595,
      "loss": 6.7365,
      "mean_token_accuracy": 0.09149369075894356,
      "num_tokens": 942310.0,
      "step": 320
    },
    {
      "entropy": 7.098740816116333,
      "epoch": 0.010192241352275222,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000162,
      "loss": 6.638,
      "mean_token_accuracy": 0.09136478006839752,
      "num_tokens": 958180.0,
      "step": 325
    },
    {
      "entropy": 7.278954648971558,
      "epoch": 0.010349045065387148,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00016450000000000001,
      "loss": 6.7407,
      "mean_token_accuracy": 0.08464238122105598,
      "num_tokens": 973484.0,
      "step": 330
    },
    {
      "entropy": 7.1389687061309814,
      "epoch": 0.010505848778499074,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00016700000000000002,
      "loss": 6.5988,
      "mean_token_accuracy": 0.09544083774089814,
      "num_tokens": 988712.0,
      "step": 335
    },
    {
      "entropy": 7.1739630699157715,
      "epoch": 0.010662652491611002,
      "grad_norm": 1.203125,
      "learning_rate": 0.00016950000000000003,
      "loss": 6.6637,
      "mean_token_accuracy": 0.0873688019812107,
      "num_tokens": 1005316.0,
      "step": 340
    },
    {
      "entropy": 7.009389734268188,
      "epoch": 0.010819456204722928,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00017199999999999998,
      "loss": 6.5965,
      "mean_token_accuracy": 0.09233827069401741,
      "num_tokens": 1020624.0,
      "step": 345
    },
    {
      "entropy": 7.205262613296509,
      "epoch": 0.010976259917834854,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017449999999999999,
      "loss": 6.6497,
      "mean_token_accuracy": 0.09442537128925324,
      "num_tokens": 1035346.0,
      "step": 350
    },
    {
      "entropy": 7.131465578079224,
      "epoch": 0.011133063630946781,
      "grad_norm": 1.5,
      "learning_rate": 0.000177,
      "loss": 6.6518,
      "mean_token_accuracy": 0.08939319550991058,
      "num_tokens": 1052396.0,
      "step": 355
    },
    {
      "entropy": 7.066566705703735,
      "epoch": 0.011289867344058707,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001795,
      "loss": 6.6706,
      "mean_token_accuracy": 0.08813088536262512,
      "num_tokens": 1068158.0,
      "step": 360
    },
    {
      "entropy": 7.096987533569336,
      "epoch": 0.011446671057170633,
      "grad_norm": 1.21875,
      "learning_rate": 0.000182,
      "loss": 6.532,
      "mean_token_accuracy": 0.10468010231852531,
      "num_tokens": 1081891.0,
      "step": 365
    },
    {
      "entropy": 6.936192178726197,
      "epoch": 0.01160347477028256,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001845,
      "loss": 6.4988,
      "mean_token_accuracy": 0.10371973067522049,
      "num_tokens": 1094953.0,
      "step": 370
    },
    {
      "entropy": 6.988458728790283,
      "epoch": 0.011760278483394487,
      "grad_norm": 1.375,
      "learning_rate": 0.000187,
      "loss": 6.4762,
      "mean_token_accuracy": 0.09902411252260208,
      "num_tokens": 1109923.0,
      "step": 375
    },
    {
      "entropy": 7.017359066009521,
      "epoch": 0.011917082196506413,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001895,
      "loss": 6.5871,
      "mean_token_accuracy": 0.09580302238464355,
      "num_tokens": 1126987.0,
      "step": 380
    },
    {
      "entropy": 6.971819591522217,
      "epoch": 0.01207388590961834,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000192,
      "loss": 6.5812,
      "mean_token_accuracy": 0.09617941230535507,
      "num_tokens": 1139931.0,
      "step": 385
    },
    {
      "entropy": 6.953200578689575,
      "epoch": 0.012230689622730266,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001945,
      "loss": 6.4646,
      "mean_token_accuracy": 0.10358070284128189,
      "num_tokens": 1155721.0,
      "step": 390
    },
    {
      "entropy": 6.9849913120269775,
      "epoch": 0.012387493335842192,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.5469,
      "mean_token_accuracy": 0.10082540512084961,
      "num_tokens": 1171655.0,
      "step": 395
    },
    {
      "entropy": 6.938650178909302,
      "epoch": 0.01254429704895412,
      "grad_norm": 1.171875,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.5281,
      "mean_token_accuracy": 0.10091597810387612,
      "num_tokens": 1186666.0,
      "step": 400
    },
    {
      "entropy": 6.9116593360900875,
      "epoch": 0.012701100762066046,
      "grad_norm": 1.46875,
      "learning_rate": 0.000202,
      "loss": 6.4346,
      "mean_token_accuracy": 0.09778324812650681,
      "num_tokens": 1198283.0,
      "step": 405
    },
    {
      "entropy": 6.876869249343872,
      "epoch": 0.012857904475177971,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.4735,
      "mean_token_accuracy": 0.09719930961728096,
      "num_tokens": 1212613.0,
      "step": 410
    },
    {
      "entropy": 6.932867479324341,
      "epoch": 0.013014708188289899,
      "grad_norm": 1.6171875,
      "learning_rate": 0.000207,
      "loss": 6.4643,
      "mean_token_accuracy": 0.10083647891879081,
      "num_tokens": 1227515.0,
      "step": 415
    },
    {
      "entropy": 6.873914337158203,
      "epoch": 0.013171511901401825,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002095,
      "loss": 6.4762,
      "mean_token_accuracy": 0.1099552720785141,
      "num_tokens": 1239022.0,
      "step": 420
    },
    {
      "entropy": 6.844020032882691,
      "epoch": 0.013328315614513751,
      "grad_norm": 1.1875,
      "learning_rate": 0.000212,
      "loss": 6.4853,
      "mean_token_accuracy": 0.09865991845726967,
      "num_tokens": 1254653.0,
      "step": 425
    },
    {
      "entropy": 6.919657754898071,
      "epoch": 0.013485119327625679,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002145,
      "loss": 6.4894,
      "mean_token_accuracy": 0.09527975097298622,
      "num_tokens": 1270766.0,
      "step": 430
    },
    {
      "entropy": 6.905207824707031,
      "epoch": 0.013641923040737605,
      "grad_norm": 1.25,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.4774,
      "mean_token_accuracy": 0.10371477827429772,
      "num_tokens": 1284742.0,
      "step": 435
    },
    {
      "entropy": 6.912538671493531,
      "epoch": 0.01379872675384953,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002195,
      "loss": 6.4678,
      "mean_token_accuracy": 0.10031345337629319,
      "num_tokens": 1297881.0,
      "step": 440
    },
    {
      "entropy": 6.800807285308838,
      "epoch": 0.013955530466961458,
      "grad_norm": 1.09375,
      "learning_rate": 0.000222,
      "loss": 6.4071,
      "mean_token_accuracy": 0.10895167514681817,
      "num_tokens": 1314727.0,
      "step": 445
    },
    {
      "entropy": 6.85034818649292,
      "epoch": 0.014112334180073384,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002245,
      "loss": 6.4329,
      "mean_token_accuracy": 0.10191441029310226,
      "num_tokens": 1332630.0,
      "step": 450
    },
    {
      "entropy": 6.820253705978393,
      "epoch": 0.01426913789318531,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.5253,
      "mean_token_accuracy": 0.09508129432797433,
      "num_tokens": 1348028.0,
      "step": 455
    },
    {
      "entropy": 6.8252214908599855,
      "epoch": 0.014425941606297238,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.3729,
      "mean_token_accuracy": 0.10276357978582382,
      "num_tokens": 1363723.0,
      "step": 460
    },
    {
      "entropy": 6.869743585586548,
      "epoch": 0.014582745319409163,
      "grad_norm": 1.625,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.4717,
      "mean_token_accuracy": 0.10454175770282745,
      "num_tokens": 1378400.0,
      "step": 465
    },
    {
      "entropy": 6.816213941574096,
      "epoch": 0.01473954903252109,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.3712,
      "mean_token_accuracy": 0.10962107256054879,
      "num_tokens": 1393383.0,
      "step": 470
    },
    {
      "entropy": 6.703394031524658,
      "epoch": 0.014896352745633017,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000237,
      "loss": 6.3905,
      "mean_token_accuracy": 0.10753661692142487,
      "num_tokens": 1409609.0,
      "step": 475
    },
    {
      "entropy": 6.746933174133301,
      "epoch": 0.015053156458744943,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002395,
      "loss": 6.3711,
      "mean_token_accuracy": 0.109779604524374,
      "num_tokens": 1423378.0,
      "step": 480
    },
    {
      "entropy": 6.657280206680298,
      "epoch": 0.015209960171856869,
      "grad_norm": 1.28125,
      "learning_rate": 0.000242,
      "loss": 6.3545,
      "mean_token_accuracy": 0.10648095905780793,
      "num_tokens": 1437722.0,
      "step": 485
    },
    {
      "entropy": 6.752739667892456,
      "epoch": 0.015366763884968796,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002445,
      "loss": 6.4004,
      "mean_token_accuracy": 0.10294785127043724,
      "num_tokens": 1452969.0,
      "step": 490
    },
    {
      "entropy": 6.7490668296813965,
      "epoch": 0.015523567598080722,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000247,
      "loss": 6.406,
      "mean_token_accuracy": 0.10591119825839997,
      "num_tokens": 1466014.0,
      "step": 495
    },
    {
      "entropy": 6.7571159362792965,
      "epoch": 0.01568037131119265,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002495,
      "loss": 6.4293,
      "mean_token_accuracy": 0.1040202483534813,
      "num_tokens": 1482087.0,
      "step": 500
    },
    {
      "entropy": 6.598506212234497,
      "epoch": 0.015837175024304574,
      "grad_norm": 1.3125,
      "learning_rate": 0.000252,
      "loss": 6.2766,
      "mean_token_accuracy": 0.1113006167113781,
      "num_tokens": 1495241.0,
      "step": 505
    },
    {
      "entropy": 6.743651676177978,
      "epoch": 0.015993978737416502,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002545,
      "loss": 6.3714,
      "mean_token_accuracy": 0.1056169517338276,
      "num_tokens": 1509170.0,
      "step": 510
    },
    {
      "entropy": 6.695261478424072,
      "epoch": 0.01615078245052843,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000257,
      "loss": 6.3142,
      "mean_token_accuracy": 0.11198153868317604,
      "num_tokens": 1522215.0,
      "step": 515
    },
    {
      "entropy": 6.772069597244263,
      "epoch": 0.016307586163640354,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002595,
      "loss": 6.3489,
      "mean_token_accuracy": 0.10926218628883362,
      "num_tokens": 1535315.0,
      "step": 520
    },
    {
      "entropy": 6.584142017364502,
      "epoch": 0.01646438987675228,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000262,
      "loss": 6.2308,
      "mean_token_accuracy": 0.11487055495381356,
      "num_tokens": 1550143.0,
      "step": 525
    },
    {
      "entropy": 6.6475685119628904,
      "epoch": 0.01662119358986421,
      "grad_norm": 1.140625,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.3562,
      "mean_token_accuracy": 0.1018921285867691,
      "num_tokens": 1564977.0,
      "step": 530
    },
    {
      "entropy": 6.716810035705566,
      "epoch": 0.016777997302976133,
      "grad_norm": 1.421875,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.3036,
      "mean_token_accuracy": 0.10546633452177048,
      "num_tokens": 1580060.0,
      "step": 535
    },
    {
      "entropy": 6.631731700897217,
      "epoch": 0.01693480101608806,
      "grad_norm": 1.0625,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.2829,
      "mean_token_accuracy": 0.10685937479138374,
      "num_tokens": 1594789.0,
      "step": 540
    },
    {
      "entropy": 6.6427947044372555,
      "epoch": 0.01709160472919999,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.2729,
      "mean_token_accuracy": 0.1120539978146553,
      "num_tokens": 1610957.0,
      "step": 545
    },
    {
      "entropy": 6.588653326034546,
      "epoch": 0.017248408442311913,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002745,
      "loss": 6.2819,
      "mean_token_accuracy": 0.11637941747903824,
      "num_tokens": 1624447.0,
      "step": 550
    },
    {
      "entropy": 6.620056772232056,
      "epoch": 0.01740521215542384,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000277,
      "loss": 6.3078,
      "mean_token_accuracy": 0.10762870758771896,
      "num_tokens": 1639204.0,
      "step": 555
    },
    {
      "entropy": 6.609020042419433,
      "epoch": 0.017562015868535768,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002795,
      "loss": 6.3212,
      "mean_token_accuracy": 0.110382679104805,
      "num_tokens": 1652377.0,
      "step": 560
    },
    {
      "entropy": 6.693290662765503,
      "epoch": 0.017718819581647692,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.2751,
      "mean_token_accuracy": 0.10745934993028641,
      "num_tokens": 1667254.0,
      "step": 565
    },
    {
      "entropy": 6.560057878494263,
      "epoch": 0.01787562329475962,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002845,
      "loss": 6.3225,
      "mean_token_accuracy": 0.11314097568392753,
      "num_tokens": 1682021.0,
      "step": 570
    },
    {
      "entropy": 6.629387378692627,
      "epoch": 0.018032427007871547,
      "grad_norm": 1.375,
      "learning_rate": 0.000287,
      "loss": 6.1881,
      "mean_token_accuracy": 0.11611460223793983,
      "num_tokens": 1697211.0,
      "step": 575
    },
    {
      "entropy": 6.556789445877075,
      "epoch": 0.01818923072098347,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002895,
      "loss": 6.2833,
      "mean_token_accuracy": 0.10666922554373741,
      "num_tokens": 1712561.0,
      "step": 580
    },
    {
      "entropy": 6.522421073913574,
      "epoch": 0.0183460344340954,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000292,
      "loss": 6.211,
      "mean_token_accuracy": 0.11197596043348312,
      "num_tokens": 1728298.0,
      "step": 585
    },
    {
      "entropy": 6.53973650932312,
      "epoch": 0.018502838147207327,
      "grad_norm": 1.375,
      "learning_rate": 0.0002945,
      "loss": 6.2439,
      "mean_token_accuracy": 0.11992142051458358,
      "num_tokens": 1741697.0,
      "step": 590
    },
    {
      "entropy": 6.4527688980102536,
      "epoch": 0.01865964186031925,
      "grad_norm": 1.53125,
      "learning_rate": 0.000297,
      "loss": 6.2042,
      "mean_token_accuracy": 0.1115458443760872,
      "num_tokens": 1756477.0,
      "step": 595
    },
    {
      "entropy": 6.483115386962891,
      "epoch": 0.01881644557343118,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002995,
      "loss": 6.161,
      "mean_token_accuracy": 0.1176509566605091,
      "num_tokens": 1770965.0,
      "step": 600
    },
    {
      "entropy": 6.551352834701538,
      "epoch": 0.018973249286543106,
      "grad_norm": 1.3125,
      "learning_rate": 0.000302,
      "loss": 6.2809,
      "mean_token_accuracy": 0.11110794097185135,
      "num_tokens": 1785184.0,
      "step": 605
    },
    {
      "entropy": 6.579122543334961,
      "epoch": 0.01913005299965503,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003045,
      "loss": 6.3017,
      "mean_token_accuracy": 0.11201615557074547,
      "num_tokens": 1799353.0,
      "step": 610
    },
    {
      "entropy": 6.484968090057373,
      "epoch": 0.019286856712766958,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000307,
      "loss": 6.2272,
      "mean_token_accuracy": 0.11322264298796654,
      "num_tokens": 1813457.0,
      "step": 615
    },
    {
      "entropy": 6.4240114212036135,
      "epoch": 0.019443660425878886,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003095,
      "loss": 6.1879,
      "mean_token_accuracy": 0.11162598803639412,
      "num_tokens": 1829820.0,
      "step": 620
    },
    {
      "entropy": 6.436847734451294,
      "epoch": 0.01960046413899081,
      "grad_norm": 1.09375,
      "learning_rate": 0.000312,
      "loss": 6.1562,
      "mean_token_accuracy": 0.11525302529335021,
      "num_tokens": 1843376.0,
      "step": 625
    },
    {
      "entropy": 6.483410835266113,
      "epoch": 0.019757267852102738,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003145,
      "loss": 6.2293,
      "mean_token_accuracy": 0.1060644507408142,
      "num_tokens": 1858854.0,
      "step": 630
    },
    {
      "entropy": 6.426254606246948,
      "epoch": 0.019914071565214665,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000317,
      "loss": 6.0136,
      "mean_token_accuracy": 0.125533090531826,
      "num_tokens": 1873740.0,
      "step": 635
    },
    {
      "entropy": 6.404020833969116,
      "epoch": 0.02007087527832659,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003195,
      "loss": 6.1716,
      "mean_token_accuracy": 0.11641796976327896,
      "num_tokens": 1888942.0,
      "step": 640
    },
    {
      "entropy": 6.520569324493408,
      "epoch": 0.020227678991438517,
      "grad_norm": 1.171875,
      "learning_rate": 0.000322,
      "loss": 6.2086,
      "mean_token_accuracy": 0.12039469107985497,
      "num_tokens": 1904054.0,
      "step": 645
    },
    {
      "entropy": 6.439871692657471,
      "epoch": 0.020384482704550445,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.1303,
      "mean_token_accuracy": 0.11612799763679504,
      "num_tokens": 1916464.0,
      "step": 650
    },
    {
      "entropy": 6.333846569061279,
      "epoch": 0.02054128641766237,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.0516,
      "mean_token_accuracy": 0.11740432828664779,
      "num_tokens": 1930947.0,
      "step": 655
    },
    {
      "entropy": 6.440066814422607,
      "epoch": 0.020698090130774296,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.1644,
      "mean_token_accuracy": 0.11625355631113052,
      "num_tokens": 1945060.0,
      "step": 660
    },
    {
      "entropy": 6.518965578079223,
      "epoch": 0.020854893843886224,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.3127,
      "mean_token_accuracy": 0.11131328493356704,
      "num_tokens": 1959187.0,
      "step": 665
    },
    {
      "entropy": 6.498908805847168,
      "epoch": 0.02101169755699815,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.2751,
      "mean_token_accuracy": 0.11170570999383926,
      "num_tokens": 1974516.0,
      "step": 670
    },
    {
      "entropy": 6.321500682830811,
      "epoch": 0.021168501270110076,
      "grad_norm": 1.09375,
      "learning_rate": 0.000337,
      "loss": 6.0763,
      "mean_token_accuracy": 0.12224758714437485,
      "num_tokens": 1987459.0,
      "step": 675
    },
    {
      "entropy": 6.427319049835205,
      "epoch": 0.021325304983222004,
      "grad_norm": 1.125,
      "learning_rate": 0.0003395,
      "loss": 6.1014,
      "mean_token_accuracy": 0.11473800018429756,
      "num_tokens": 2004345.0,
      "step": 680
    },
    {
      "entropy": 6.32919487953186,
      "epoch": 0.021482108696333928,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000342,
      "loss": 6.1893,
      "mean_token_accuracy": 0.11112356036901475,
      "num_tokens": 2021128.0,
      "step": 685
    },
    {
      "entropy": 6.442146730422974,
      "epoch": 0.021638912409445855,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.1285,
      "mean_token_accuracy": 0.11685249358415603,
      "num_tokens": 2034534.0,
      "step": 690
    },
    {
      "entropy": 6.332594347000122,
      "epoch": 0.021795716122557783,
      "grad_norm": 1.171875,
      "learning_rate": 0.000347,
      "loss": 5.994,
      "mean_token_accuracy": 0.12094285562634469,
      "num_tokens": 2047522.0,
      "step": 695
    },
    {
      "entropy": 6.32836503982544,
      "epoch": 0.021952519835669707,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003495,
      "loss": 6.1174,
      "mean_token_accuracy": 0.11575590446591377,
      "num_tokens": 2063026.0,
      "step": 700
    },
    {
      "entropy": 6.352781915664673,
      "epoch": 0.022109323548781635,
      "grad_norm": 1.203125,
      "learning_rate": 0.000352,
      "loss": 6.1027,
      "mean_token_accuracy": 0.12405385375022888,
      "num_tokens": 2077313.0,
      "step": 705
    },
    {
      "entropy": 6.297963857650757,
      "epoch": 0.022266127261893563,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003545,
      "loss": 6.0533,
      "mean_token_accuracy": 0.12451517134904862,
      "num_tokens": 2090545.0,
      "step": 710
    },
    {
      "entropy": 6.329249334335327,
      "epoch": 0.022422930975005487,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000357,
      "loss": 6.1159,
      "mean_token_accuracy": 0.11173808798193932,
      "num_tokens": 2104252.0,
      "step": 715
    },
    {
      "entropy": 6.367383718490601,
      "epoch": 0.022579734688117414,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003595,
      "loss": 6.047,
      "mean_token_accuracy": 0.11943526864051819,
      "num_tokens": 2118020.0,
      "step": 720
    },
    {
      "entropy": 6.267186212539673,
      "epoch": 0.022736538401229342,
      "grad_norm": 1.046875,
      "learning_rate": 0.000362,
      "loss": 6.0012,
      "mean_token_accuracy": 0.1282973624765873,
      "num_tokens": 2132429.0,
      "step": 725
    },
    {
      "entropy": 6.323628377914429,
      "epoch": 0.022893342114341266,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003645,
      "loss": 6.1295,
      "mean_token_accuracy": 0.1254815012216568,
      "num_tokens": 2145763.0,
      "step": 730
    },
    {
      "entropy": 6.322244787216187,
      "epoch": 0.023050145827453194,
      "grad_norm": 1.03125,
      "learning_rate": 0.000367,
      "loss": 6.083,
      "mean_token_accuracy": 0.12271661087870597,
      "num_tokens": 2160583.0,
      "step": 735
    },
    {
      "entropy": 6.244843339920044,
      "epoch": 0.02320694954056512,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003695,
      "loss": 6.0526,
      "mean_token_accuracy": 0.1167644202709198,
      "num_tokens": 2176254.0,
      "step": 740
    },
    {
      "entropy": 6.375514793395996,
      "epoch": 0.023363753253677046,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000372,
      "loss": 6.1026,
      "mean_token_accuracy": 0.1191796325147152,
      "num_tokens": 2190456.0,
      "step": 745
    },
    {
      "entropy": 6.307696533203125,
      "epoch": 0.023520556966788973,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003745,
      "loss": 6.0764,
      "mean_token_accuracy": 0.12289957031607628,
      "num_tokens": 2203102.0,
      "step": 750
    },
    {
      "entropy": 6.40611457824707,
      "epoch": 0.0236773606799009,
      "grad_norm": 1.34375,
      "learning_rate": 0.000377,
      "loss": 6.0772,
      "mean_token_accuracy": 0.12269680425524712,
      "num_tokens": 2216866.0,
      "step": 755
    },
    {
      "entropy": 6.236187839508057,
      "epoch": 0.023834164393012825,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003795,
      "loss": 6.0033,
      "mean_token_accuracy": 0.11777962148189544,
      "num_tokens": 2231275.0,
      "step": 760
    },
    {
      "entropy": 6.256259632110596,
      "epoch": 0.023990968106124753,
      "grad_norm": 1.234375,
      "learning_rate": 0.000382,
      "loss": 6.101,
      "mean_token_accuracy": 0.1180748425424099,
      "num_tokens": 2244788.0,
      "step": 765
    },
    {
      "entropy": 6.308438873291015,
      "epoch": 0.02414777181923668,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003845,
      "loss": 6.1113,
      "mean_token_accuracy": 0.1172371581196785,
      "num_tokens": 2260548.0,
      "step": 770
    },
    {
      "entropy": 6.242420339584351,
      "epoch": 0.024304575532348605,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.0436,
      "mean_token_accuracy": 0.1195620097219944,
      "num_tokens": 2277030.0,
      "step": 775
    },
    {
      "entropy": 6.2474593162536625,
      "epoch": 0.024461379245460532,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.038,
      "mean_token_accuracy": 0.11677318438887596,
      "num_tokens": 2291350.0,
      "step": 780
    },
    {
      "entropy": 6.340010404586792,
      "epoch": 0.02461818295857246,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.1619,
      "mean_token_accuracy": 0.11335810571908951,
      "num_tokens": 2306766.0,
      "step": 785
    },
    {
      "entropy": 6.247185087203979,
      "epoch": 0.024774986671684384,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.0213,
      "mean_token_accuracy": 0.12314986139535904,
      "num_tokens": 2322556.0,
      "step": 790
    },
    {
      "entropy": 6.070211410522461,
      "epoch": 0.02493179038479631,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039700000000000005,
      "loss": 5.8597,
      "mean_token_accuracy": 0.12424484714865684,
      "num_tokens": 2334061.0,
      "step": 795
    },
    {
      "entropy": 6.18665885925293,
      "epoch": 0.02508859409790824,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003995,
      "loss": 6.0215,
      "mean_token_accuracy": 0.12226074188947678,
      "num_tokens": 2350082.0,
      "step": 800
    },
    {
      "entropy": 6.2820933818817135,
      "epoch": 0.025245397811020164,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000402,
      "loss": 5.9866,
      "mean_token_accuracy": 0.12534554153680802,
      "num_tokens": 2365799.0,
      "step": 805
    },
    {
      "entropy": 6.109661722183228,
      "epoch": 0.02540220152413209,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004045,
      "loss": 5.9468,
      "mean_token_accuracy": 0.12130692824721337,
      "num_tokens": 2379491.0,
      "step": 810
    },
    {
      "entropy": 6.178446578979492,
      "epoch": 0.02555900523724402,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040699999999999997,
      "loss": 5.9715,
      "mean_token_accuracy": 0.1236006774008274,
      "num_tokens": 2393840.0,
      "step": 815
    },
    {
      "entropy": 6.055525779724121,
      "epoch": 0.025715808950355943,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004095,
      "loss": 5.8116,
      "mean_token_accuracy": 0.13396925255656242,
      "num_tokens": 2408364.0,
      "step": 820
    },
    {
      "entropy": 6.2112726211547855,
      "epoch": 0.02587261266346787,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000412,
      "loss": 5.941,
      "mean_token_accuracy": 0.12613581195473672,
      "num_tokens": 2423471.0,
      "step": 825
    },
    {
      "entropy": 6.06286735534668,
      "epoch": 0.026029416376579798,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004145,
      "loss": 5.9854,
      "mean_token_accuracy": 0.1259685769677162,
      "num_tokens": 2437773.0,
      "step": 830
    },
    {
      "entropy": 6.202285575866699,
      "epoch": 0.026186220089691722,
      "grad_norm": 1.203125,
      "learning_rate": 0.000417,
      "loss": 5.9824,
      "mean_token_accuracy": 0.12420864850282669,
      "num_tokens": 2451210.0,
      "step": 835
    },
    {
      "entropy": 6.162640047073364,
      "epoch": 0.02634302380280365,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004195,
      "loss": 5.9749,
      "mean_token_accuracy": 0.12392436563968659,
      "num_tokens": 2466855.0,
      "step": 840
    },
    {
      "entropy": 6.198370170593262,
      "epoch": 0.026499827515915578,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000422,
      "loss": 6.0288,
      "mean_token_accuracy": 0.12249074876308441,
      "num_tokens": 2483544.0,
      "step": 845
    },
    {
      "entropy": 6.083363008499146,
      "epoch": 0.026656631229027502,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004245,
      "loss": 5.9422,
      "mean_token_accuracy": 0.12914909645915032,
      "num_tokens": 2496906.0,
      "step": 850
    },
    {
      "entropy": 6.218140172958374,
      "epoch": 0.02681343494213943,
      "grad_norm": 1.09375,
      "learning_rate": 0.000427,
      "loss": 5.9327,
      "mean_token_accuracy": 0.12357857301831246,
      "num_tokens": 2510929.0,
      "step": 855
    },
    {
      "entropy": 6.035630893707276,
      "epoch": 0.026970238655251357,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004295,
      "loss": 5.8866,
      "mean_token_accuracy": 0.12890503630042077,
      "num_tokens": 2523088.0,
      "step": 860
    },
    {
      "entropy": 6.202908706665039,
      "epoch": 0.02712704236836328,
      "grad_norm": 1.125,
      "learning_rate": 0.000432,
      "loss": 6.0462,
      "mean_token_accuracy": 0.12496983706951141,
      "num_tokens": 2538069.0,
      "step": 865
    },
    {
      "entropy": 6.155341863632202,
      "epoch": 0.02728384608147521,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004345,
      "loss": 5.9648,
      "mean_token_accuracy": 0.12080245763063431,
      "num_tokens": 2550457.0,
      "step": 870
    },
    {
      "entropy": 6.164808702468872,
      "epoch": 0.027440649794587137,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000437,
      "loss": 6.0062,
      "mean_token_accuracy": 0.12498117238283157,
      "num_tokens": 2564626.0,
      "step": 875
    },
    {
      "entropy": 6.074118041992188,
      "epoch": 0.02759745350769906,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004395,
      "loss": 5.9586,
      "mean_token_accuracy": 0.1210741564631462,
      "num_tokens": 2580773.0,
      "step": 880
    },
    {
      "entropy": 6.084677696228027,
      "epoch": 0.02775425722081099,
      "grad_norm": 1.046875,
      "learning_rate": 0.000442,
      "loss": 5.9504,
      "mean_token_accuracy": 0.12536775544285775,
      "num_tokens": 2595110.0,
      "step": 885
    },
    {
      "entropy": 6.080619478225708,
      "epoch": 0.027911060933922916,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004445,
      "loss": 5.8725,
      "mean_token_accuracy": 0.12975198924541473,
      "num_tokens": 2610259.0,
      "step": 890
    },
    {
      "entropy": 6.119405221939087,
      "epoch": 0.02806786464703484,
      "grad_norm": 1.078125,
      "learning_rate": 0.000447,
      "loss": 5.9856,
      "mean_token_accuracy": 0.1261485055088997,
      "num_tokens": 2625589.0,
      "step": 895
    },
    {
      "entropy": 6.066659688949585,
      "epoch": 0.028224668360146768,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044950000000000003,
      "loss": 5.9433,
      "mean_token_accuracy": 0.12746250554919242,
      "num_tokens": 2640145.0,
      "step": 900
    },
    {
      "entropy": 6.1185914993286135,
      "epoch": 0.028381472073258696,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045200000000000004,
      "loss": 5.9215,
      "mean_token_accuracy": 0.13048022910952567,
      "num_tokens": 2653621.0,
      "step": 905
    },
    {
      "entropy": 6.019134902954102,
      "epoch": 0.02853827578637062,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045450000000000004,
      "loss": 5.997,
      "mean_token_accuracy": 0.12859755456447602,
      "num_tokens": 2668380.0,
      "step": 910
    },
    {
      "entropy": 6.19635124206543,
      "epoch": 0.028695079499482547,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045700000000000005,
      "loss": 5.9644,
      "mean_token_accuracy": 0.1297648772597313,
      "num_tokens": 2682862.0,
      "step": 915
    },
    {
      "entropy": 6.027379941940308,
      "epoch": 0.028851883212594475,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045950000000000006,
      "loss": 5.8723,
      "mean_token_accuracy": 0.13426159247756003,
      "num_tokens": 2696586.0,
      "step": 920
    },
    {
      "entropy": 5.995415496826172,
      "epoch": 0.0290086869257064,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000462,
      "loss": 5.8748,
      "mean_token_accuracy": 0.1332511007785797,
      "num_tokens": 2712941.0,
      "step": 925
    },
    {
      "entropy": 6.09525785446167,
      "epoch": 0.029165490638818327,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004645,
      "loss": 5.9343,
      "mean_token_accuracy": 0.12271871790289879,
      "num_tokens": 2728371.0,
      "step": 930
    },
    {
      "entropy": 5.963246011734009,
      "epoch": 0.029322294351930255,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000467,
      "loss": 5.8632,
      "mean_token_accuracy": 0.13176240101456643,
      "num_tokens": 2741856.0,
      "step": 935
    },
    {
      "entropy": 6.0740649700164795,
      "epoch": 0.02947909806504218,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004695,
      "loss": 5.9106,
      "mean_token_accuracy": 0.12672237455844879,
      "num_tokens": 2757663.0,
      "step": 940
    },
    {
      "entropy": 6.102129697799683,
      "epoch": 0.029635901778154106,
      "grad_norm": 1.09375,
      "learning_rate": 0.000472,
      "loss": 5.8991,
      "mean_token_accuracy": 0.1298680104315281,
      "num_tokens": 2771074.0,
      "step": 945
    },
    {
      "entropy": 5.961727428436279,
      "epoch": 0.029792705491266034,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004745,
      "loss": 5.8354,
      "mean_token_accuracy": 0.13212114572525024,
      "num_tokens": 2784369.0,
      "step": 950
    },
    {
      "entropy": 6.022281265258789,
      "epoch": 0.029949509204377958,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000477,
      "loss": 5.9385,
      "mean_token_accuracy": 0.12027354538440704,
      "num_tokens": 2800233.0,
      "step": 955
    },
    {
      "entropy": 6.097547292709351,
      "epoch": 0.030106312917489886,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004795,
      "loss": 5.9377,
      "mean_token_accuracy": 0.12930759862065316,
      "num_tokens": 2815270.0,
      "step": 960
    },
    {
      "entropy": 6.130005693435669,
      "epoch": 0.030263116630601813,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000482,
      "loss": 6.0192,
      "mean_token_accuracy": 0.12478073239326477,
      "num_tokens": 2830833.0,
      "step": 965
    },
    {
      "entropy": 6.0437352657318115,
      "epoch": 0.030419920343713738,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004845,
      "loss": 5.9442,
      "mean_token_accuracy": 0.13879294395446778,
      "num_tokens": 2845321.0,
      "step": 970
    },
    {
      "entropy": 6.048642206192016,
      "epoch": 0.030576724056825665,
      "grad_norm": 1.109375,
      "learning_rate": 0.000487,
      "loss": 5.9032,
      "mean_token_accuracy": 0.13587410748004913,
      "num_tokens": 2859112.0,
      "step": 975
    },
    {
      "entropy": 6.0042726516723635,
      "epoch": 0.030733527769937593,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004895,
      "loss": 5.8294,
      "mean_token_accuracy": 0.12910271510481836,
      "num_tokens": 2872888.0,
      "step": 980
    },
    {
      "entropy": 6.00495228767395,
      "epoch": 0.030890331483049517,
      "grad_norm": 1.234375,
      "learning_rate": 0.000492,
      "loss": 5.938,
      "mean_token_accuracy": 0.13228848949074745,
      "num_tokens": 2887363.0,
      "step": 985
    },
    {
      "entropy": 6.064482307434082,
      "epoch": 0.031047135196161445,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004945,
      "loss": 5.9331,
      "mean_token_accuracy": 0.1261969804763794,
      "num_tokens": 2902598.0,
      "step": 990
    },
    {
      "entropy": 5.962180280685425,
      "epoch": 0.031203938909273372,
      "grad_norm": 1.09375,
      "learning_rate": 0.000497,
      "loss": 5.9316,
      "mean_token_accuracy": 0.12509682774543762,
      "num_tokens": 2917404.0,
      "step": 995
    },
    {
      "entropy": 6.087647294998169,
      "epoch": 0.0313607426223853,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995,
      "loss": 6.0481,
      "mean_token_accuracy": 0.12149560004472733,
      "num_tokens": 2934527.0,
      "step": 1000
    },
    {
      "entropy": 5.987649393081665,
      "epoch": 0.03151754633549723,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499998026082006,
      "loss": 5.8305,
      "mean_token_accuracy": 0.13421109914779664,
      "num_tokens": 2949222.0,
      "step": 1005
    },
    {
      "entropy": 6.066426849365234,
      "epoch": 0.03167435004860915,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999900070995136,
      "loss": 5.9763,
      "mean_token_accuracy": 0.12479947283864021,
      "num_tokens": 2964168.0,
      "step": 1010
    },
    {
      "entropy": 5.988660573959351,
      "epoch": 0.031831153761721076,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999758199023239,
      "loss": 5.8726,
      "mean_token_accuracy": 0.13721302077174186,
      "num_tokens": 2978217.0,
      "step": 1015
    },
    {
      "entropy": 5.860611200332642,
      "epoch": 0.031987957474833004,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999554648793858,
      "loss": 5.8288,
      "mean_token_accuracy": 0.12761253491044044,
      "num_tokens": 2993577.0,
      "step": 1020
    },
    {
      "entropy": 5.974480104446411,
      "epoch": 0.03214476118794493,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999289425887425,
      "loss": 5.8514,
      "mean_token_accuracy": 0.12751337438821791,
      "num_tokens": 3007649.0,
      "step": 1025
    },
    {
      "entropy": 5.889734697341919,
      "epoch": 0.03230156490105686,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998962537575161,
      "loss": 5.8314,
      "mean_token_accuracy": 0.14106257036328315,
      "num_tokens": 3021619.0,
      "step": 1030
    },
    {
      "entropy": 6.131463050842285,
      "epoch": 0.03245836861416879,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998573992818874,
      "loss": 5.9148,
      "mean_token_accuracy": 0.12463216632604598,
      "num_tokens": 3036726.0,
      "step": 1035
    },
    {
      "entropy": 5.956810188293457,
      "epoch": 0.03261517232728071,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004998123802270715,
      "loss": 5.9448,
      "mean_token_accuracy": 0.12305018752813339,
      "num_tokens": 3049786.0,
      "step": 1040
    },
    {
      "entropy": 5.986479806900024,
      "epoch": 0.032771976040392635,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997611978272886,
      "loss": 5.9525,
      "mean_token_accuracy": 0.1329242967069149,
      "num_tokens": 3063521.0,
      "step": 1045
    },
    {
      "entropy": 6.006673288345337,
      "epoch": 0.03292877975350456,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997038534857298,
      "loss": 5.8735,
      "mean_token_accuracy": 0.12820919156074523,
      "num_tokens": 3077790.0,
      "step": 1050
    },
    {
      "entropy": 5.896439933776856,
      "epoch": 0.03308558346661649,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004996403487745194,
      "loss": 5.848,
      "mean_token_accuracy": 0.1296282596886158,
      "num_tokens": 3095577.0,
      "step": 1055
    },
    {
      "entropy": 5.958034658432007,
      "epoch": 0.03324238717972842,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499570685434671,
      "loss": 5.8715,
      "mean_token_accuracy": 0.12729506567120552,
      "num_tokens": 3111244.0,
      "step": 1060
    },
    {
      "entropy": 5.850221729278564,
      "epoch": 0.033399190892840346,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994948653760405,
      "loss": 5.6822,
      "mean_token_accuracy": 0.14201920107007027,
      "num_tokens": 3125407.0,
      "step": 1065
    },
    {
      "entropy": 5.910600900650024,
      "epoch": 0.033555994605952266,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004994128906772729,
      "loss": 5.7931,
      "mean_token_accuracy": 0.13524388447403907,
      "num_tokens": 3137260.0,
      "step": 1070
    },
    {
      "entropy": 5.954425382614136,
      "epoch": 0.033712798319064194,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000499324763585746,
      "loss": 5.8919,
      "mean_token_accuracy": 0.12603772133588792,
      "num_tokens": 3150418.0,
      "step": 1075
    },
    {
      "entropy": 5.915726375579834,
      "epoch": 0.03386960203217612,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004992304865175085,
      "loss": 5.837,
      "mean_token_accuracy": 0.13079922497272492,
      "num_tokens": 3165385.0,
      "step": 1080
    },
    {
      "entropy": 5.868867015838623,
      "epoch": 0.03402640574528805,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004991300620572138,
      "loss": 5.841,
      "mean_token_accuracy": 0.1295164331793785,
      "num_tokens": 3179144.0,
      "step": 1085
    },
    {
      "entropy": 5.89624342918396,
      "epoch": 0.03418320945839998,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004990234929580494,
      "loss": 5.7452,
      "mean_token_accuracy": 0.13632771372795105,
      "num_tokens": 3192831.0,
      "step": 1090
    },
    {
      "entropy": 5.911666440963745,
      "epoch": 0.034340013171511904,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004989107821416609,
      "loss": 5.8807,
      "mean_token_accuracy": 0.13042519465088845,
      "num_tokens": 3205820.0,
      "step": 1095
    },
    {
      "entropy": 6.020514869689942,
      "epoch": 0.034496816884623825,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004987919326980723,
      "loss": 5.8385,
      "mean_token_accuracy": 0.13410276398062707,
      "num_tokens": 3219618.0,
      "step": 1100
    },
    {
      "entropy": 5.817402315139771,
      "epoch": 0.03465362059773575,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004986669478856011,
      "loss": 5.7457,
      "mean_token_accuracy": 0.13494893535971642,
      "num_tokens": 3233648.0,
      "step": 1105
    },
    {
      "entropy": 5.92337851524353,
      "epoch": 0.03481042431084768,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004985358311307688,
      "loss": 5.8936,
      "mean_token_accuracy": 0.12879264429211618,
      "num_tokens": 3248906.0,
      "step": 1110
    },
    {
      "entropy": 5.852339553833008,
      "epoch": 0.03496722802395961,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004983985860282081,
      "loss": 5.7109,
      "mean_token_accuracy": 0.13579089120030402,
      "num_tokens": 3263212.0,
      "step": 1115
    },
    {
      "entropy": 5.866457986831665,
      "epoch": 0.035124031737071536,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004982552163405623,
      "loss": 5.7612,
      "mean_token_accuracy": 0.14021414518356323,
      "num_tokens": 3277244.0,
      "step": 1120
    },
    {
      "entropy": 5.96242036819458,
      "epoch": 0.03528083545018346,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004981057259983839,
      "loss": 5.9119,
      "mean_token_accuracy": 0.12455229237675666,
      "num_tokens": 3292200.0,
      "step": 1125
    },
    {
      "entropy": 5.904011678695679,
      "epoch": 0.035437639163295384,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004979501191000262,
      "loss": 5.7927,
      "mean_token_accuracy": 0.13359547033905983,
      "num_tokens": 3308999.0,
      "step": 1130
    },
    {
      "entropy": 5.829094743728637,
      "epoch": 0.03559444287640731,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004977883999115311,
      "loss": 5.8299,
      "mean_token_accuracy": 0.13213476166129112,
      "num_tokens": 3325380.0,
      "step": 1135
    },
    {
      "entropy": 5.848357009887695,
      "epoch": 0.03575124658951924,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004976205728665113,
      "loss": 5.7401,
      "mean_token_accuracy": 0.14331733286380768,
      "num_tokens": 3339997.0,
      "step": 1140
    },
    {
      "entropy": 5.814362859725952,
      "epoch": 0.03590805030263117,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004974466425660307,
      "loss": 5.7597,
      "mean_token_accuracy": 0.13575922846794128,
      "num_tokens": 3353388.0,
      "step": 1145
    },
    {
      "entropy": 5.941767978668213,
      "epoch": 0.036064854015743095,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004972666137784759,
      "loss": 5.8378,
      "mean_token_accuracy": 0.13422609940171243,
      "num_tokens": 3368534.0,
      "step": 1150
    },
    {
      "entropy": 5.867988395690918,
      "epoch": 0.03622165772885502,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004970804914394271,
      "loss": 5.7803,
      "mean_token_accuracy": 0.13961924463510514,
      "num_tokens": 3384683.0,
      "step": 1155
    },
    {
      "entropy": 5.836901712417602,
      "epoch": 0.03637846144196694,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004968882806515225,
      "loss": 5.6851,
      "mean_token_accuracy": 0.14083293005824088,
      "num_tokens": 3397823.0,
      "step": 1160
    },
    {
      "entropy": 5.793954658508301,
      "epoch": 0.03653526515507887,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004966899866843177,
      "loss": 5.753,
      "mean_token_accuracy": 0.14006715640425682,
      "num_tokens": 3411862.0,
      "step": 1165
    },
    {
      "entropy": 5.860671329498291,
      "epoch": 0.0366920688681908,
      "grad_norm": 1.015625,
      "learning_rate": 0.000496485614974142,
      "loss": 5.7202,
      "mean_token_accuracy": 0.14168946966528892,
      "num_tokens": 3426306.0,
      "step": 1170
    },
    {
      "entropy": 5.748768091201782,
      "epoch": 0.036848872581302726,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004962751711239492,
      "loss": 5.7055,
      "mean_token_accuracy": 0.137933249771595,
      "num_tokens": 3441906.0,
      "step": 1175
    },
    {
      "entropy": 5.866535282135009,
      "epoch": 0.037005676294414654,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004960586609031636,
      "loss": 5.8129,
      "mean_token_accuracy": 0.12675541639328003,
      "num_tokens": 3456643.0,
      "step": 1180
    },
    {
      "entropy": 5.953323316574097,
      "epoch": 0.03716248000752658,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004958360902475224,
      "loss": 5.8209,
      "mean_token_accuracy": 0.13259705379605294,
      "num_tokens": 3472218.0,
      "step": 1185
    },
    {
      "entropy": 5.888777065277099,
      "epoch": 0.0373192837206385,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004956074652589125,
      "loss": 5.8791,
      "mean_token_accuracy": 0.13352179303765296,
      "num_tokens": 3487173.0,
      "step": 1190
    },
    {
      "entropy": 5.888743495941162,
      "epoch": 0.03747608743375043,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004953727922052035,
      "loss": 5.8697,
      "mean_token_accuracy": 0.1344257652759552,
      "num_tokens": 3502963.0,
      "step": 1195
    },
    {
      "entropy": 5.723510456085205,
      "epoch": 0.03763289114686236,
      "grad_norm": 1.0,
      "learning_rate": 0.0004951320775200756,
      "loss": 5.638,
      "mean_token_accuracy": 0.1404781922698021,
      "num_tokens": 3517807.0,
      "step": 1200
    },
    {
      "entropy": 5.961473083496093,
      "epoch": 0.037789694859974285,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004948853278028436,
      "loss": 5.8022,
      "mean_token_accuracy": 0.1324881412088871,
      "num_tokens": 3531597.0,
      "step": 1205
    },
    {
      "entropy": 5.893240833282471,
      "epoch": 0.03794649857308621,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004946325498182755,
      "loss": 5.878,
      "mean_token_accuracy": 0.12567512765526773,
      "num_tokens": 3549526.0,
      "step": 1210
    },
    {
      "entropy": 5.814656019210815,
      "epoch": 0.03810330228619814,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004943737504964076,
      "loss": 5.7029,
      "mean_token_accuracy": 0.13802678138017654,
      "num_tokens": 3562202.0,
      "step": 1215
    },
    {
      "entropy": 5.823633623123169,
      "epoch": 0.03826010599931006,
      "grad_norm": 1.0,
      "learning_rate": 0.000494108936932354,
      "loss": 5.7657,
      "mean_token_accuracy": 0.13985563889145852,
      "num_tokens": 3576824.0,
      "step": 1220
    },
    {
      "entropy": 5.86934757232666,
      "epoch": 0.03841690971242199,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004938381163861124,
      "loss": 5.7675,
      "mean_token_accuracy": 0.13630534335970879,
      "num_tokens": 3589347.0,
      "step": 1225
    },
    {
      "entropy": 5.739215517044068,
      "epoch": 0.038573713425533916,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004935612962823645,
      "loss": 5.7174,
      "mean_token_accuracy": 0.1372951328754425,
      "num_tokens": 3604280.0,
      "step": 1230
    },
    {
      "entropy": 5.7907041072845455,
      "epoch": 0.038730517138645844,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004932784842102739,
      "loss": 5.6566,
      "mean_token_accuracy": 0.14609543979167938,
      "num_tokens": 3617854.0,
      "step": 1235
    },
    {
      "entropy": 5.782031059265137,
      "epoch": 0.03888732085175777,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004929896879232758,
      "loss": 5.7427,
      "mean_token_accuracy": 0.13488340973854065,
      "num_tokens": 3632627.0,
      "step": 1240
    },
    {
      "entropy": 5.888515567779541,
      "epoch": 0.0390441245648697,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004926949153388668,
      "loss": 5.8072,
      "mean_token_accuracy": 0.13958224803209304,
      "num_tokens": 3649406.0,
      "step": 1245
    },
    {
      "entropy": 5.856204605102539,
      "epoch": 0.03920092827798162,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004923941745383859,
      "loss": 5.8247,
      "mean_token_accuracy": 0.13710009753704072,
      "num_tokens": 3664042.0,
      "step": 1250
    },
    {
      "entropy": 5.801159381866455,
      "epoch": 0.03935773199109355,
      "grad_norm": 1.015625,
      "learning_rate": 0.000492087473766794,
      "loss": 5.7248,
      "mean_token_accuracy": 0.13974272459745407,
      "num_tokens": 3678964.0,
      "step": 1255
    },
    {
      "entropy": 5.812677669525146,
      "epoch": 0.039514535704205475,
      "grad_norm": 1.0625,
      "learning_rate": 0.000491774821432448,
      "loss": 5.7867,
      "mean_token_accuracy": 0.13592285215854644,
      "num_tokens": 3694725.0,
      "step": 1260
    },
    {
      "entropy": 5.827065658569336,
      "epoch": 0.0396713394173174,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004914562261068693,
      "loss": 5.6938,
      "mean_token_accuracy": 0.14478985965251923,
      "num_tokens": 3708942.0,
      "step": 1265
    },
    {
      "entropy": 5.804870748519898,
      "epoch": 0.03982814313042933,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004911316965245098,
      "loss": 5.7437,
      "mean_token_accuracy": 0.1342754565179348,
      "num_tokens": 3727171.0,
      "step": 1270
    },
    {
      "entropy": 5.7694480419158936,
      "epoch": 0.03998494684354126,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000490801241582512,
      "loss": 5.7494,
      "mean_token_accuracy": 0.14143916815519333,
      "num_tokens": 3742249.0,
      "step": 1275
    },
    {
      "entropy": 5.81528754234314,
      "epoch": 0.04014175055665318,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000490464870340465,
      "loss": 5.6568,
      "mean_token_accuracy": 0.13865938186645507,
      "num_tokens": 3756267.0,
      "step": 1280
    },
    {
      "entropy": 5.706241655349731,
      "epoch": 0.040298554269765106,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004901225920201563,
      "loss": 5.6238,
      "mean_token_accuracy": 0.14538088589906692,
      "num_tokens": 3769846.0,
      "step": 1285
    },
    {
      "entropy": 5.818896913528443,
      "epoch": 0.040455357982877034,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000489774416005319,
      "loss": 5.7882,
      "mean_token_accuracy": 0.13648639097809792,
      "num_tokens": 3783489.0,
      "step": 1290
    },
    {
      "entropy": 5.728751993179321,
      "epoch": 0.04061216169598896,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004894203518413742,
      "loss": 5.6705,
      "mean_token_accuracy": 0.14063634797930719,
      "num_tokens": 3800797.0,
      "step": 1295
    },
    {
      "entropy": 5.821841621398926,
      "epoch": 0.04076896540910089,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004890604092351701,
      "loss": 5.7148,
      "mean_token_accuracy": 0.14129465371370314,
      "num_tokens": 3816678.0,
      "step": 1300
    },
    {
      "entropy": 5.754094028472901,
      "epoch": 0.04092576912221282,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000488694598054715,
      "loss": 5.7167,
      "mean_token_accuracy": 0.13984414786100388,
      "num_tokens": 3830435.0,
      "step": 1305
    },
    {
      "entropy": 5.7894409656524655,
      "epoch": 0.04108257283532474,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004883229283289071,
      "loss": 5.688,
      "mean_token_accuracy": 0.14409852996468545,
      "num_tokens": 3844910.0,
      "step": 1310
    },
    {
      "entropy": 5.704872941970825,
      "epoch": 0.041239376548436665,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048794541024725993,
      "loss": 5.7349,
      "mean_token_accuracy": 0.13705151826143264,
      "num_tokens": 3859688.0,
      "step": 1315
    },
    {
      "entropy": 5.772118949890137,
      "epoch": 0.04139618026154859,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004875620541596221,
      "loss": 5.6293,
      "mean_token_accuracy": 0.14292458593845367,
      "num_tokens": 3875604.0,
      "step": 1320
    },
    {
      "entropy": 5.77454195022583,
      "epoch": 0.04155298397466052,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048717287057589454,
      "loss": 5.7073,
      "mean_token_accuracy": 0.13829230964183808,
      "num_tokens": 3891463.0,
      "step": 1325
    },
    {
      "entropy": 5.741397094726563,
      "epoch": 0.04170978768777245,
      "grad_norm": 1.125,
      "learning_rate": 0.0004867778701657417,
      "loss": 5.7261,
      "mean_token_accuracy": 0.14032770767807962,
      "num_tokens": 3905505.0,
      "step": 1330
    },
    {
      "entropy": 5.805401229858399,
      "epoch": 0.041866591400884376,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048637706375829955,
      "loss": 5.6083,
      "mean_token_accuracy": 0.14735801219940187,
      "num_tokens": 3919059.0,
      "step": 1335
    },
    {
      "entropy": 5.642607164382935,
      "epoch": 0.0420233951139963,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000485970462341878,
      "loss": 5.6655,
      "mean_token_accuracy": 0.14223852679133414,
      "num_tokens": 3934554.0,
      "step": 1340
    },
    {
      "entropy": 5.732195806503296,
      "epoch": 0.042180198827108224,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048555807706366044,
      "loss": 5.6321,
      "mean_token_accuracy": 0.14635981246829033,
      "num_tokens": 3948836.0,
      "step": 1345
    },
    {
      "entropy": 5.717943096160889,
      "epoch": 0.04233700254022015,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048513991922939756,
      "loss": 5.6328,
      "mean_token_accuracy": 0.14274636507034302,
      "num_tokens": 3963374.0,
      "step": 1350
    },
    {
      "entropy": 5.73873291015625,
      "epoch": 0.04249380625333208,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00048471600030309744,
      "loss": 5.7642,
      "mean_token_accuracy": 0.1433413989841938,
      "num_tokens": 3978790.0,
      "step": 1355
    },
    {
      "entropy": 5.791727018356323,
      "epoch": 0.04265060996644401,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048428633190671186,
      "loss": 5.7165,
      "mean_token_accuracy": 0.14199780970811843,
      "num_tokens": 3993320.0,
      "step": 1360
    },
    {
      "entropy": 5.743384075164795,
      "epoch": 0.042807413679555935,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004838509258198167,
      "loss": 5.647,
      "mean_token_accuracy": 0.14719821512699127,
      "num_tokens": 4006867.0,
      "step": 1365
    },
    {
      "entropy": 5.710117816925049,
      "epoch": 0.042964217392667856,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048340979397929,
      "loss": 5.5949,
      "mean_token_accuracy": 0.14546966552734375,
      "num_tokens": 4022664.0,
      "step": 1370
    },
    {
      "entropy": 5.678111791610718,
      "epoch": 0.04312102110577978,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048296294847898386,
      "loss": 5.6193,
      "mean_token_accuracy": 0.14935135692358018,
      "num_tokens": 4036213.0,
      "step": 1375
    },
    {
      "entropy": 5.756679725646973,
      "epoch": 0.04327782481889171,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004825104015693934,
      "loss": 5.7018,
      "mean_token_accuracy": 0.14161612838506699,
      "num_tokens": 4050766.0,
      "step": 1380
    },
    {
      "entropy": 5.732282876968384,
      "epoch": 0.04343462853200364,
      "grad_norm": 0.875,
      "learning_rate": 0.0004820521656573208,
      "loss": 5.6864,
      "mean_token_accuracy": 0.148099435120821,
      "num_tokens": 4065983.0,
      "step": 1385
    },
    {
      "entropy": 5.816153573989868,
      "epoch": 0.043591432245115566,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048158825330553505,
      "loss": 5.7293,
      "mean_token_accuracy": 0.14191074445843696,
      "num_tokens": 4081755.0,
      "step": 1390
    },
    {
      "entropy": 5.647218370437622,
      "epoch": 0.043748235958227494,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048111867723242763,
      "loss": 5.6105,
      "mean_token_accuracy": 0.146306511759758,
      "num_tokens": 4095747.0,
      "step": 1395
    },
    {
      "entropy": 5.771636009216309,
      "epoch": 0.043905039671339414,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004806434503116637,
      "loss": 5.6161,
      "mean_token_accuracy": 0.15057915598154067,
      "num_tokens": 4109547.0,
      "step": 1400
    },
    {
      "entropy": 5.725102043151855,
      "epoch": 0.04406184338445134,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004801625855718296,
      "loss": 5.7245,
      "mean_token_accuracy": 0.1395110748708248,
      "num_tokens": 4125984.0,
      "step": 1405
    },
    {
      "entropy": 5.679219055175781,
      "epoch": 0.04421864709756327,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047967609619607477,
      "loss": 5.6238,
      "mean_token_accuracy": 0.14524023979902267,
      "num_tokens": 4140603.0,
      "step": 1410
    },
    {
      "entropy": 5.675964784622193,
      "epoch": 0.0443754508106752,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004791839955217513,
      "loss": 5.5852,
      "mean_token_accuracy": 0.14772498831152917,
      "num_tokens": 4156800.0,
      "step": 1415
    },
    {
      "entropy": 5.734459352493286,
      "epoch": 0.044532254523787125,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047868629704004786,
      "loss": 5.694,
      "mean_token_accuracy": 0.1451198197901249,
      "num_tokens": 4170821.0,
      "step": 1420
    },
    {
      "entropy": 5.695484256744384,
      "epoch": 0.04468905823689905,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00047818301439561965,
      "loss": 5.6216,
      "mean_token_accuracy": 0.14631207138299943,
      "num_tokens": 4185555.0,
      "step": 1425
    },
    {
      "entropy": 5.674433755874634,
      "epoch": 0.04484586195001097,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047767416138621454,
      "loss": 5.6494,
      "mean_token_accuracy": 0.1443865902721882,
      "num_tokens": 4200164.0,
      "step": 1430
    },
    {
      "entropy": 5.7222973823547365,
      "epoch": 0.0450026656631229,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000477159751962295,
      "loss": 5.6172,
      "mean_token_accuracy": 0.1548900417983532,
      "num_tokens": 4213393.0,
      "step": 1435
    },
    {
      "entropy": 5.670509672164917,
      "epoch": 0.04515946937623483,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047663980022665507,
      "loss": 5.5355,
      "mean_token_accuracy": 0.15677547156810762,
      "num_tokens": 4226794.0,
      "step": 1440
    },
    {
      "entropy": 5.806225967407227,
      "epoch": 0.045316273089346756,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047611432043403437,
      "loss": 5.7712,
      "mean_token_accuracy": 0.13765899315476418,
      "num_tokens": 4243575.0,
      "step": 1445
    },
    {
      "entropy": 5.715469264984131,
      "epoch": 0.045473076802458684,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004755833269907267,
      "loss": 5.6882,
      "mean_token_accuracy": 0.14131314307451248,
      "num_tokens": 4257947.0,
      "step": 1450
    },
    {
      "entropy": 5.7218750476837155,
      "epoch": 0.04562988051557061,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004750468344541857,
      "loss": 5.6569,
      "mean_token_accuracy": 0.14194273352622985,
      "num_tokens": 4274140.0,
      "step": 1455
    },
    {
      "entropy": 5.794920873641968,
      "epoch": 0.04578668422868253,
      "grad_norm": 1.0,
      "learning_rate": 0.00047450485753262525,
      "loss": 5.6275,
      "mean_token_accuracy": 0.14596797823905944,
      "num_tokens": 4287415.0,
      "step": 1460
    },
    {
      "entropy": 5.602665996551513,
      "epoch": 0.04594348794179446,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047395741108461633,
      "loss": 5.5512,
      "mean_token_accuracy": 0.15124925673007966,
      "num_tokens": 4301281.0,
      "step": 1465
    },
    {
      "entropy": 5.795319700241089,
      "epoch": 0.04610029165490639,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047340451011867985,
      "loss": 5.6918,
      "mean_token_accuracy": 0.14298617839813232,
      "num_tokens": 4316413.0,
      "step": 1470
    },
    {
      "entropy": 5.6644445896148685,
      "epoch": 0.046257095368018315,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047284616979287515,
      "loss": 5.6087,
      "mean_token_accuracy": 0.14939066469669343,
      "num_tokens": 4330355.0,
      "step": 1475
    },
    {
      "entropy": 5.713419389724732,
      "epoch": 0.04641389908113024,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047228240541438433,
      "loss": 5.5912,
      "mean_token_accuracy": 0.147909464687109,
      "num_tokens": 4344710.0,
      "step": 1480
    },
    {
      "entropy": 5.616871118545532,
      "epoch": 0.04657070279424217,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047171323243909257,
      "loss": 5.57,
      "mean_token_accuracy": 0.1550672844052315,
      "num_tokens": 4357853.0,
      "step": 1485
    },
    {
      "entropy": 5.6620786666870115,
      "epoch": 0.04672750650735409,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047113866647116457,
      "loss": 5.5682,
      "mean_token_accuracy": 0.1481015995144844,
      "num_tokens": 4372473.0,
      "step": 1490
    },
    {
      "entropy": 5.713062810897827,
      "epoch": 0.04688431022046602,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004705587232626164,
      "loss": 5.7357,
      "mean_token_accuracy": 0.13680261969566346,
      "num_tokens": 4387178.0,
      "step": 1495
    },
    {
      "entropy": 5.670087432861328,
      "epoch": 0.04704111393357795,
      "grad_norm": 0.890625,
      "learning_rate": 0.00046997341871288424,
      "loss": 5.6205,
      "mean_token_accuracy": 0.1452601358294487,
      "num_tokens": 4402092.0,
      "step": 1500
    },
    {
      "entropy": 5.699921464920044,
      "epoch": 0.047197917646689874,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004693827688683879,
      "loss": 5.5736,
      "mean_token_accuracy": 0.15034027099609376,
      "num_tokens": 4415190.0,
      "step": 1505
    },
    {
      "entropy": 5.647108316421509,
      "epoch": 0.0473547213598018,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004687867899220914,
      "loss": 5.5482,
      "mean_token_accuracy": 0.14983870908617974,
      "num_tokens": 4430402.0,
      "step": 1510
    },
    {
      "entropy": 5.681677961349488,
      "epoch": 0.04751152507291373,
      "grad_norm": 0.890625,
      "learning_rate": 0.00046818549821305846,
      "loss": 5.6694,
      "mean_token_accuracy": 0.14604835733771324,
      "num_tokens": 4445547.0,
      "step": 1515
    },
    {
      "entropy": 5.612594366073608,
      "epoch": 0.04766832878602565,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046757891022600494,
      "loss": 5.5101,
      "mean_token_accuracy": 0.15138526260852814,
      "num_tokens": 4460175.0,
      "step": 1520
    },
    {
      "entropy": 5.647563648223877,
      "epoch": 0.04782513249913758,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004669670425908471,
      "loss": 5.6422,
      "mean_token_accuracy": 0.1430087998509407,
      "num_tokens": 4475687.0,
      "step": 1525
    },
    {
      "entropy": 5.663858413696289,
      "epoch": 0.047981936212249506,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004663499120822451,
      "loss": 5.6162,
      "mean_token_accuracy": 0.148489810526371,
      "num_tokens": 4492494.0,
      "step": 1530
    },
    {
      "entropy": 5.622017478942871,
      "epoch": 0.04813873992536143,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004657275356191437,
      "loss": 5.4682,
      "mean_token_accuracy": 0.15244067311286927,
      "num_tokens": 4506566.0,
      "step": 1535
    },
    {
      "entropy": 5.626275539398193,
      "epoch": 0.04829554363847336,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00046509993026430804,
      "loss": 5.6286,
      "mean_token_accuracy": 0.14129947870969772,
      "num_tokens": 4522463.0,
      "step": 1540
    },
    {
      "entropy": 5.617114162445068,
      "epoch": 0.04845234735158529,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004644671132238558,
      "loss": 5.5217,
      "mean_token_accuracy": 0.15152865946292876,
      "num_tokens": 4537401.0,
      "step": 1545
    },
    {
      "entropy": 5.60228967666626,
      "epoch": 0.04860915106469721,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046382910184678585,
      "loss": 5.5717,
      "mean_token_accuracy": 0.1512614145874977,
      "num_tokens": 4549712.0,
      "step": 1550
    },
    {
      "entropy": 5.602184009552002,
      "epoch": 0.04876595477780914,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004631859136245025,
      "loss": 5.4804,
      "mean_token_accuracy": 0.1518227055668831,
      "num_tokens": 4563601.0,
      "step": 1555
    },
    {
      "entropy": 5.686416435241699,
      "epoch": 0.048922758490921064,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004625375661903357,
      "loss": 5.5304,
      "mean_token_accuracy": 0.14626965522766114,
      "num_tokens": 4577309.0,
      "step": 1560
    },
    {
      "entropy": 5.540439081192017,
      "epoch": 0.04907956220403299,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046188407731905787,
      "loss": 5.4883,
      "mean_token_accuracy": 0.1563427686691284,
      "num_tokens": 4590389.0,
      "step": 1565
    },
    {
      "entropy": 5.711526823043823,
      "epoch": 0.04923636591714492,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00046122546492639643,
      "loss": 5.6104,
      "mean_token_accuracy": 0.14507472813129424,
      "num_tokens": 4607042.0,
      "step": 1570
    },
    {
      "entropy": 5.665983533859253,
      "epoch": 0.04939316963025685,
      "grad_norm": 0.921875,
      "learning_rate": 0.000460561747068543,
      "loss": 5.5964,
      "mean_token_accuracy": 0.15060784667730331,
      "num_tokens": 4624377.0,
      "step": 1575
    },
    {
      "entropy": 5.604173374176026,
      "epoch": 0.04954997334336877,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004598929419416578,
      "loss": 5.5906,
      "mean_token_accuracy": 0.14650059193372728,
      "num_tokens": 4638816.0,
      "step": 1580
    },
    {
      "entropy": 5.548513317108155,
      "epoch": 0.049706777056480696,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045921906788137123,
      "loss": 5.4659,
      "mean_token_accuracy": 0.1494405135512352,
      "num_tokens": 4653170.0,
      "step": 1585
    },
    {
      "entropy": 5.642472791671753,
      "epoch": 0.04986358076959262,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045854014336228115,
      "loss": 5.5829,
      "mean_token_accuracy": 0.15180073976516723,
      "num_tokens": 4668809.0,
      "step": 1590
    },
    {
      "entropy": 5.599613761901855,
      "epoch": 0.05002038448270455,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045785618699744615,
      "loss": 5.5059,
      "mean_token_accuracy": 0.15332444459199907,
      "num_tokens": 4682695.0,
      "step": 1595
    },
    {
      "entropy": 5.674669170379639,
      "epoch": 0.05017718819581648,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00045716721753787543,
      "loss": 5.5992,
      "mean_token_accuracy": 0.1441897891461849,
      "num_tokens": 4697566.0,
      "step": 1600
    },
    {
      "entropy": 5.615376663208008,
      "epoch": 0.050333991908928406,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004564732538720148,
      "loss": 5.5736,
      "mean_token_accuracy": 0.15400946587324144,
      "num_tokens": 4713419.0,
      "step": 1605
    },
    {
      "entropy": 5.671611595153808,
      "epoch": 0.05049079562204033,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00045577431502522877,
      "loss": 5.6863,
      "mean_token_accuracy": 0.14871388226747512,
      "num_tokens": 4728407.0,
      "step": 1610
    },
    {
      "entropy": 5.560825538635254,
      "epoch": 0.050647599335152255,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004550704201592787,
      "loss": 5.466,
      "mean_token_accuracy": 0.16336653530597686,
      "num_tokens": 4742860.0,
      "step": 1615
    },
    {
      "entropy": 5.664677143096924,
      "epoch": 0.05080440304826418,
      "grad_norm": 1.0,
      "learning_rate": 0.0004543615885717981,
      "loss": 5.5615,
      "mean_token_accuracy": 0.1503000631928444,
      "num_tokens": 4757249.0,
      "step": 1620
    },
    {
      "entropy": 5.638796377182007,
      "epoch": 0.05096120676137611,
      "grad_norm": 0.96875,
      "learning_rate": 0.00045364783969576296,
      "loss": 5.595,
      "mean_token_accuracy": 0.14745519459247589,
      "num_tokens": 4772147.0,
      "step": 1625
    },
    {
      "entropy": 5.648314714431763,
      "epoch": 0.05111801047448804,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004529291930989592,
      "loss": 5.5434,
      "mean_token_accuracy": 0.15235277265310287,
      "num_tokens": 4784627.0,
      "step": 1630
    },
    {
      "entropy": 5.654696750640869,
      "epoch": 0.051274814187599965,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004522056684834464,
      "loss": 5.6175,
      "mean_token_accuracy": 0.14991778507828712,
      "num_tokens": 4799817.0,
      "step": 1635
    },
    {
      "entropy": 5.680637693405151,
      "epoch": 0.051431617900711886,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004514772856850173,
      "loss": 5.5911,
      "mean_token_accuracy": 0.14739327132701874,
      "num_tokens": 4813178.0,
      "step": 1640
    },
    {
      "entropy": 5.5131996154785154,
      "epoch": 0.051588421613823814,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004507440646726542,
      "loss": 5.4405,
      "mean_token_accuracy": 0.16182133853435515,
      "num_tokens": 4827652.0,
      "step": 1645
    },
    {
      "entropy": 5.645104265213012,
      "epoch": 0.05174522532693574,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004500060255479818,
      "loss": 5.552,
      "mean_token_accuracy": 0.14753093272447587,
      "num_tokens": 4839978.0,
      "step": 1650
    },
    {
      "entropy": 5.603615236282349,
      "epoch": 0.05190202904004767,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004492631885447151,
      "loss": 5.5562,
      "mean_token_accuracy": 0.15101770609617232,
      "num_tokens": 4854612.0,
      "step": 1655
    },
    {
      "entropy": 5.616217136383057,
      "epoch": 0.052058832753159597,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00044851557402810616,
      "loss": 5.5119,
      "mean_token_accuracy": 0.14714247435331346,
      "num_tokens": 4869188.0,
      "step": 1660
    },
    {
      "entropy": 5.665006017684936,
      "epoch": 0.052215636466271524,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00044776320249438444,
      "loss": 5.5835,
      "mean_token_accuracy": 0.15070897787809373,
      "num_tokens": 4882938.0,
      "step": 1665
    },
    {
      "entropy": 5.568822383880615,
      "epoch": 0.052372440179383445,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00044700609457019565,
      "loss": 5.4814,
      "mean_token_accuracy": 0.1596212700009346,
      "num_tokens": 4897999.0,
      "step": 1670
    },
    {
      "entropy": 5.672763919830322,
      "epoch": 0.05252924389249537,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004462442710120359,
      "loss": 5.6476,
      "mean_token_accuracy": 0.14675597250461578,
      "num_tokens": 4912664.0,
      "step": 1675
    },
    {
      "entropy": 5.588794136047364,
      "epoch": 0.0526860476056073,
      "grad_norm": 0.890625,
      "learning_rate": 0.000445477752705683,
      "loss": 5.5276,
      "mean_token_accuracy": 0.15189943909645082,
      "num_tokens": 4928610.0,
      "step": 1680
    },
    {
      "entropy": 5.594437313079834,
      "epoch": 0.05284285131871923,
      "grad_norm": 0.984375,
      "learning_rate": 0.00044470656066562336,
      "loss": 5.4753,
      "mean_token_accuracy": 0.16013056188821792,
      "num_tokens": 4942044.0,
      "step": 1685
    },
    {
      "entropy": 5.559233140945435,
      "epoch": 0.052999655031831155,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004439307160344765,
      "loss": 5.5138,
      "mean_token_accuracy": 0.15794520378112792,
      "num_tokens": 4955380.0,
      "step": 1690
    },
    {
      "entropy": 5.6098854541778564,
      "epoch": 0.05315645874494308,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00044315024008241473,
      "loss": 5.503,
      "mean_token_accuracy": 0.15251349955797194,
      "num_tokens": 4968707.0,
      "step": 1695
    },
    {
      "entropy": 5.52094841003418,
      "epoch": 0.053313262458055004,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004423651542065806,
      "loss": 5.583,
      "mean_token_accuracy": 0.15648411214351654,
      "num_tokens": 4982462.0,
      "step": 1700
    },
    {
      "entropy": 5.645457363128662,
      "epoch": 0.05347006617116693,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00044157547993050006,
      "loss": 5.4863,
      "mean_token_accuracy": 0.1566183090209961,
      "num_tokens": 4996556.0,
      "step": 1705
    },
    {
      "entropy": 5.49437985420227,
      "epoch": 0.05362686988427886,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00044078123890349227,
      "loss": 5.3734,
      "mean_token_accuracy": 0.1659584552049637,
      "num_tokens": 5009338.0,
      "step": 1710
    },
    {
      "entropy": 5.506396293640137,
      "epoch": 0.05378367359739079,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043998245290007606,
      "loss": 5.5344,
      "mean_token_accuracy": 0.1501413643360138,
      "num_tokens": 5024586.0,
      "step": 1715
    },
    {
      "entropy": 5.665162134170532,
      "epoch": 0.053940477310502714,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00043917914381937323,
      "loss": 5.4667,
      "mean_token_accuracy": 0.15752882063388823,
      "num_tokens": 5038807.0,
      "step": 1720
    },
    {
      "entropy": 5.454025506973267,
      "epoch": 0.05409728102361464,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043837133368450815,
      "loss": 5.5053,
      "mean_token_accuracy": 0.15899511128664018,
      "num_tokens": 5051883.0,
      "step": 1725
    },
    {
      "entropy": 5.630865240097046,
      "epoch": 0.05425408473672656,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004375590446420037,
      "loss": 5.432,
      "mean_token_accuracy": 0.15725974589586258,
      "num_tokens": 5067601.0,
      "step": 1730
    },
    {
      "entropy": 5.523392963409424,
      "epoch": 0.05441088844983849,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004367422989611743,
      "loss": 5.4978,
      "mean_token_accuracy": 0.15299071222543717,
      "num_tokens": 5083331.0,
      "step": 1735
    },
    {
      "entropy": 5.6384416103363035,
      "epoch": 0.05456769216295042,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004359211190335153,
      "loss": 5.5192,
      "mean_token_accuracy": 0.15755498856306077,
      "num_tokens": 5098192.0,
      "step": 1740
    },
    {
      "entropy": 5.597709608078003,
      "epoch": 0.054724495876062346,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00043509552737208923,
      "loss": 5.4884,
      "mean_token_accuracy": 0.15653315633535386,
      "num_tokens": 5114763.0,
      "step": 1745
    },
    {
      "entropy": 5.558363723754883,
      "epoch": 0.05488129958917427,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043426554661090853,
      "loss": 5.5527,
      "mean_token_accuracy": 0.15365547090768814,
      "num_tokens": 5128419.0,
      "step": 1750
    },
    {
      "entropy": 5.5542590618133545,
      "epoch": 0.0550381033022862,
      "grad_norm": 0.8671875,
      "learning_rate": 0.00043343119950431516,
      "loss": 5.4396,
      "mean_token_accuracy": 0.16456325799226762,
      "num_tokens": 5142470.0,
      "step": 1755
    },
    {
      "entropy": 5.544502449035645,
      "epoch": 0.05519490701539812,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00043259250892635644,
      "loss": 5.4659,
      "mean_token_accuracy": 0.15672120749950408,
      "num_tokens": 5157122.0,
      "step": 1760
    },
    {
      "entropy": 5.574978923797607,
      "epoch": 0.05535171072851005,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004317494978701582,
      "loss": 5.4053,
      "mean_token_accuracy": 0.16559310555458068,
      "num_tokens": 5169717.0,
      "step": 1765
    },
    {
      "entropy": 5.495390892028809,
      "epoch": 0.05550851444162198,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004309021894472943,
      "loss": 5.5142,
      "mean_token_accuracy": 0.15167766213417053,
      "num_tokens": 5185024.0,
      "step": 1770
    },
    {
      "entropy": 5.585932874679566,
      "epoch": 0.055665318154733905,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004300506068871534,
      "loss": 5.4923,
      "mean_token_accuracy": 0.16034476310014725,
      "num_tokens": 5199670.0,
      "step": 1775
    },
    {
      "entropy": 5.567832994461059,
      "epoch": 0.05582212186784583,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00042919477353630135,
      "loss": 5.4805,
      "mean_token_accuracy": 0.1543068364262581,
      "num_tokens": 5213033.0,
      "step": 1780
    },
    {
      "entropy": 5.524223852157593,
      "epoch": 0.05597892558095776,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000428334712857842,
      "loss": 5.4247,
      "mean_token_accuracy": 0.16250368505716323,
      "num_tokens": 5227857.0,
      "step": 1785
    },
    {
      "entropy": 5.496453952789307,
      "epoch": 0.05613572929406968,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00042747044843077304,
      "loss": 5.4351,
      "mean_token_accuracy": 0.15133761167526244,
      "num_tokens": 5244198.0,
      "step": 1790
    },
    {
      "entropy": 5.47076268196106,
      "epoch": 0.05629253300718161,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00042660200394934047,
      "loss": 5.3509,
      "mean_token_accuracy": 0.16515569239854813,
      "num_tokens": 5259603.0,
      "step": 1795
    },
    {
      "entropy": 5.541412210464477,
      "epoch": 0.056449336720293536,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00042572940322238844,
      "loss": 5.4829,
      "mean_token_accuracy": 0.15599187165498735,
      "num_tokens": 5275078.0,
      "step": 1800
    },
    {
      "entropy": 5.515583229064942,
      "epoch": 0.056606140433405464,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042485267017270664,
      "loss": 5.4015,
      "mean_token_accuracy": 0.15718774795532225,
      "num_tokens": 5288614.0,
      "step": 1805
    },
    {
      "entropy": 5.5446430206298825,
      "epoch": 0.05676294414651739,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004239718288363745,
      "loss": 5.4667,
      "mean_token_accuracy": 0.15240791141986848,
      "num_tokens": 5306437.0,
      "step": 1810
    },
    {
      "entropy": 5.567139291763306,
      "epoch": 0.05691974785962932,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004230869033621023,
      "loss": 5.5704,
      "mean_token_accuracy": 0.15007242411375046,
      "num_tokens": 5320750.0,
      "step": 1815
    },
    {
      "entropy": 5.589888191223144,
      "epoch": 0.05707655157274124,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004221979180105688,
      "loss": 5.5042,
      "mean_token_accuracy": 0.15344637930393218,
      "num_tokens": 5334329.0,
      "step": 1820
    },
    {
      "entropy": 5.502320384979248,
      "epoch": 0.05723335528585317,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00042130489715375645,
      "loss": 5.3988,
      "mean_token_accuracy": 0.16416364461183547,
      "num_tokens": 5348480.0,
      "step": 1825
    },
    {
      "entropy": 5.4765801429748535,
      "epoch": 0.057390158998965095,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042040786527428335,
      "loss": 5.3333,
      "mean_token_accuracy": 0.16420571655035018,
      "num_tokens": 5363805.0,
      "step": 1830
    },
    {
      "entropy": 5.436630821228027,
      "epoch": 0.05754696271207702,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004195068469647315,
      "loss": 5.3937,
      "mean_token_accuracy": 0.1599104017019272,
      "num_tokens": 5377536.0,
      "step": 1835
    },
    {
      "entropy": 5.564624500274658,
      "epoch": 0.05770376642518895,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00041860186692697297,
      "loss": 5.4683,
      "mean_token_accuracy": 0.15769407153129578,
      "num_tokens": 5393066.0,
      "step": 1840
    },
    {
      "entropy": 5.546552610397339,
      "epoch": 0.05786057013830088,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00041769294997149264,
      "loss": 5.3845,
      "mean_token_accuracy": 0.16675785779953003,
      "num_tokens": 5406260.0,
      "step": 1845
    },
    {
      "entropy": 5.554831600189209,
      "epoch": 0.0580173738514128,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004167801210167081,
      "loss": 5.4875,
      "mean_token_accuracy": 0.15391432791948317,
      "num_tokens": 5421238.0,
      "step": 1850
    },
    {
      "entropy": 5.410466814041138,
      "epoch": 0.058174177564524726,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004158634050882861,
      "loss": 5.3638,
      "mean_token_accuracy": 0.1567578598856926,
      "num_tokens": 5436752.0,
      "step": 1855
    },
    {
      "entropy": 5.50055193901062,
      "epoch": 0.058330981277636654,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004149428273184569,
      "loss": 5.4708,
      "mean_token_accuracy": 0.15318142026662826,
      "num_tokens": 5453104.0,
      "step": 1860
    },
    {
      "entropy": 5.572831535339356,
      "epoch": 0.05848778499074858,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004140184129453253,
      "loss": 5.4378,
      "mean_token_accuracy": 0.15755112171173097,
      "num_tokens": 5467277.0,
      "step": 1865
    },
    {
      "entropy": 5.463003301620484,
      "epoch": 0.05864458870386051,
      "grad_norm": 1.0,
      "learning_rate": 0.000413090187312178,
      "loss": 5.3706,
      "mean_token_accuracy": 0.15910724699497222,
      "num_tokens": 5482366.0,
      "step": 1870
    },
    {
      "entropy": 5.539995718002319,
      "epoch": 0.05880139241697244,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004121581758667898,
      "loss": 5.4547,
      "mean_token_accuracy": 0.15378581136465072,
      "num_tokens": 5497444.0,
      "step": 1875
    },
    {
      "entropy": 5.549549913406372,
      "epoch": 0.05895819613008436,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00041122240416072533,
      "loss": 5.4401,
      "mean_token_accuracy": 0.158327154815197,
      "num_tokens": 5510313.0,
      "step": 1880
    },
    {
      "entropy": 5.460352230072021,
      "epoch": 0.059114999843196285,
      "grad_norm": 1.0,
      "learning_rate": 0.0004102828978486385,
      "loss": 5.4253,
      "mean_token_accuracy": 0.16176005005836486,
      "num_tokens": 5525084.0,
      "step": 1885
    },
    {
      "entropy": 5.481763887405395,
      "epoch": 0.05927180355630821,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004093396826875695,
      "loss": 5.3245,
      "mean_token_accuracy": 0.16924351155757905,
      "num_tokens": 5538274.0,
      "step": 1890
    },
    {
      "entropy": 5.577309608459473,
      "epoch": 0.05942860726942014,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00040839278453623837,
      "loss": 5.4952,
      "mean_token_accuracy": 0.15003783777356147,
      "num_tokens": 5554272.0,
      "step": 1895
    },
    {
      "entropy": 5.522237920761109,
      "epoch": 0.05958541098253207,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004074422293543363,
      "loss": 5.4811,
      "mean_token_accuracy": 0.16081224530935287,
      "num_tokens": 5569006.0,
      "step": 1900
    },
    {
      "entropy": 5.49424934387207,
      "epoch": 0.059742214695643996,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004064880432018137,
      "loss": 5.4787,
      "mean_token_accuracy": 0.16519856750965117,
      "num_tokens": 5581573.0,
      "step": 1905
    },
    {
      "entropy": 5.609365940093994,
      "epoch": 0.059899018408755916,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040553025223816615,
      "loss": 5.5393,
      "mean_token_accuracy": 0.1521975040435791,
      "num_tokens": 5596483.0,
      "step": 1910
    },
    {
      "entropy": 5.554807329177857,
      "epoch": 0.060055822121867844,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040456888272171653,
      "loss": 5.4246,
      "mean_token_accuracy": 0.1567584291100502,
      "num_tokens": 5609804.0,
      "step": 1915
    },
    {
      "entropy": 5.5278465270996096,
      "epoch": 0.06021262583497977,
      "grad_norm": 0.96875,
      "learning_rate": 0.00040360396100889577,
      "loss": 5.5035,
      "mean_token_accuracy": 0.15543259531259537,
      "num_tokens": 5623651.0,
      "step": 1920
    },
    {
      "entropy": 5.511450386047363,
      "epoch": 0.0603694295480917,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004026355135535202,
      "loss": 5.4059,
      "mean_token_accuracy": 0.16002338379621506,
      "num_tokens": 5637973.0,
      "step": 1925
    },
    {
      "entropy": 5.48571138381958,
      "epoch": 0.06052623326120363,
      "grad_norm": 0.890625,
      "learning_rate": 0.000401663566906066,
      "loss": 5.5188,
      "mean_token_accuracy": 0.15829627513885497,
      "num_tokens": 5653942.0,
      "step": 1930
    },
    {
      "entropy": 5.569460535049439,
      "epoch": 0.060683036974315555,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00040068814771294134,
      "loss": 5.4165,
      "mean_token_accuracy": 0.1597804084420204,
      "num_tokens": 5668894.0,
      "step": 1935
    },
    {
      "entropy": 5.483691549301147,
      "epoch": 0.060839840687427475,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003997092827157562,
      "loss": 5.3076,
      "mean_token_accuracy": 0.16481563746929168,
      "num_tokens": 5683529.0,
      "step": 1940
    },
    {
      "entropy": 5.411105680465698,
      "epoch": 0.0609966444005394,
      "grad_norm": 0.9375,
      "learning_rate": 0.000398726998750589,
      "loss": 5.3979,
      "mean_token_accuracy": 0.16174078732728958,
      "num_tokens": 5698624.0,
      "step": 1945
    },
    {
      "entropy": 5.516795873641968,
      "epoch": 0.06115344811365133,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00039774132274725076,
      "loss": 5.3894,
      "mean_token_accuracy": 0.15667195618152618,
      "num_tokens": 5711924.0,
      "step": 1950
    },
    {
      "entropy": 5.5290087223052975,
      "epoch": 0.06131025182676326,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039675228172854707,
      "loss": 5.4199,
      "mean_token_accuracy": 0.16616879552602767,
      "num_tokens": 5726094.0,
      "step": 1955
    },
    {
      "entropy": 5.541859483718872,
      "epoch": 0.061467055539875186,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003957599028095371,
      "loss": 5.4431,
      "mean_token_accuracy": 0.16165485680103303,
      "num_tokens": 5741372.0,
      "step": 1960
    },
    {
      "entropy": 5.532270002365112,
      "epoch": 0.061623859252987113,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00039476421319679017,
      "loss": 5.4086,
      "mean_token_accuracy": 0.16190743297338486,
      "num_tokens": 5756986.0,
      "step": 1965
    },
    {
      "entropy": 5.437832164764404,
      "epoch": 0.061780662966099034,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00039376524018764,
      "loss": 5.3666,
      "mean_token_accuracy": 0.16611046195030213,
      "num_tokens": 5770209.0,
      "step": 1970
    },
    {
      "entropy": 5.436462259292602,
      "epoch": 0.06193746667921096,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039276301116943616,
      "loss": 5.3436,
      "mean_token_accuracy": 0.16721372902393342,
      "num_tokens": 5784315.0,
      "step": 1975
    },
    {
      "entropy": 5.456621170043945,
      "epoch": 0.06209427039232289,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003917575536187936,
      "loss": 5.3639,
      "mean_token_accuracy": 0.16664596050977706,
      "num_tokens": 5799298.0,
      "step": 1980
    },
    {
      "entropy": 5.469586610794067,
      "epoch": 0.06225107410543482,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00039074889510083894,
      "loss": 5.3858,
      "mean_token_accuracy": 0.16036681234836578,
      "num_tokens": 5814394.0,
      "step": 1985
    },
    {
      "entropy": 5.520676326751709,
      "epoch": 0.062407877818546745,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038973706326845495,
      "loss": 5.4764,
      "mean_token_accuracy": 0.1597631722688675,
      "num_tokens": 5829537.0,
      "step": 1990
    },
    {
      "entropy": 5.514448499679565,
      "epoch": 0.06256468153165867,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0003887220858615225,
      "loss": 5.4459,
      "mean_token_accuracy": 0.15997925251722336,
      "num_tokens": 5843452.0,
      "step": 1995
    },
    {
      "entropy": 5.548208999633789,
      "epoch": 0.0627214852447706,
      "grad_norm": 0.9375,
      "learning_rate": 0.0003877039907061597,
      "loss": 5.4272,
      "mean_token_accuracy": 0.15218425393104554,
      "num_tokens": 5857283.0,
      "step": 2000
    },
    {
      "entropy": 5.423493385314941,
      "epoch": 0.06287828895788253,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003866828057139598,
      "loss": 5.348,
      "mean_token_accuracy": 0.16844379752874375,
      "num_tokens": 5872202.0,
      "step": 2005
    },
    {
      "entropy": 5.522428369522094,
      "epoch": 0.06303509267099446,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00038565855888122503,
      "loss": 5.4183,
      "mean_token_accuracy": 0.15921104848384857,
      "num_tokens": 5885593.0,
      "step": 2010
    },
    {
      "entropy": 5.485387945175171,
      "epoch": 0.06319189638410637,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00038463127828819975,
      "loss": 5.3541,
      "mean_token_accuracy": 0.16191442757844926,
      "num_tokens": 5899446.0,
      "step": 2015
    },
    {
      "entropy": 5.489001369476318,
      "epoch": 0.0633487000972183,
      "grad_norm": 0.90625,
      "learning_rate": 0.00038360099209830043,
      "loss": 5.351,
      "mean_token_accuracy": 0.1628739595413208,
      "num_tokens": 5916710.0,
      "step": 2020
    },
    {
      "entropy": 5.440134334564209,
      "epoch": 0.06350550381033022,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0003825677285573433,
      "loss": 5.432,
      "mean_token_accuracy": 0.1569529414176941,
      "num_tokens": 5933406.0,
      "step": 2025
    },
    {
      "entropy": 5.470552062988281,
      "epoch": 0.06366230752344215,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00038153151599277027,
      "loss": 5.3142,
      "mean_token_accuracy": 0.16807927638292314,
      "num_tokens": 5947305.0,
      "step": 2030
    },
    {
      "entropy": 5.413301801681518,
      "epoch": 0.06381911123655408,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0003804923828128723,
      "loss": 5.4033,
      "mean_token_accuracy": 0.1616388663649559,
      "num_tokens": 5964244.0,
      "step": 2035
    },
    {
      "entropy": 5.560881328582764,
      "epoch": 0.06397591494966601,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003794503575060104,
      "loss": 5.4285,
      "mean_token_accuracy": 0.16694391518831253,
      "num_tokens": 5979126.0,
      "step": 2040
    },
    {
      "entropy": 5.386257219314575,
      "epoch": 0.06413271866277793,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037840546863983484,
      "loss": 5.2873,
      "mean_token_accuracy": 0.17639683783054352,
      "num_tokens": 5991237.0,
      "step": 2045
    },
    {
      "entropy": 5.490569543838501,
      "epoch": 0.06428952237588986,
      "grad_norm": 1.0,
      "learning_rate": 0.0003773577448605015,
      "loss": 5.4583,
      "mean_token_accuracy": 0.1567297637462616,
      "num_tokens": 6005763.0,
      "step": 2050
    },
    {
      "entropy": 5.600405073165893,
      "epoch": 0.06444632608900179,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003763072148918872,
      "loss": 5.4518,
      "mean_token_accuracy": 0.1669362559914589,
      "num_tokens": 6019246.0,
      "step": 2055
    },
    {
      "entropy": 5.475091743469238,
      "epoch": 0.06460312980211372,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003752539075348017,
      "loss": 5.394,
      "mean_token_accuracy": 0.16765405088663102,
      "num_tokens": 6033353.0,
      "step": 2060
    },
    {
      "entropy": 5.4313554763793945,
      "epoch": 0.06475993351522565,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00037419785166619817,
      "loss": 5.3058,
      "mean_token_accuracy": 0.164634045958519,
      "num_tokens": 6047396.0,
      "step": 2065
    },
    {
      "entropy": 5.438332796096802,
      "epoch": 0.06491673722833757,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003731390762383818,
      "loss": 5.3261,
      "mean_token_accuracy": 0.16622865200042725,
      "num_tokens": 6059997.0,
      "step": 2070
    },
    {
      "entropy": 5.47605676651001,
      "epoch": 0.06507354094144949,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0003720776102782158,
      "loss": 5.4121,
      "mean_token_accuracy": 0.1672607406973839,
      "num_tokens": 6076310.0,
      "step": 2075
    },
    {
      "entropy": 5.506420993804932,
      "epoch": 0.06523034465456141,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037101348288632555,
      "loss": 5.4357,
      "mean_token_accuracy": 0.16408559679985046,
      "num_tokens": 6089342.0,
      "step": 2080
    },
    {
      "entropy": 5.500313711166382,
      "epoch": 0.06538714836767334,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0003699467232363012,
      "loss": 5.4016,
      "mean_token_accuracy": 0.16145640909671782,
      "num_tokens": 6104146.0,
      "step": 2085
    },
    {
      "entropy": 5.442271518707275,
      "epoch": 0.06554395208078527,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003688773605738973,
      "loss": 5.2839,
      "mean_token_accuracy": 0.1639725610613823,
      "num_tokens": 6118856.0,
      "step": 2090
    },
    {
      "entropy": 5.418323945999146,
      "epoch": 0.0657007557938972,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00036780542421623134,
      "loss": 5.2861,
      "mean_token_accuracy": 0.16729723662137985,
      "num_tokens": 6132745.0,
      "step": 2095
    },
    {
      "entropy": 5.440960121154785,
      "epoch": 0.06585755950700913,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003667309435509802,
      "loss": 5.4028,
      "mean_token_accuracy": 0.16144924014806747,
      "num_tokens": 6148961.0,
      "step": 2100
    },
    {
      "entropy": 5.454626750946045,
      "epoch": 0.06601436322012105,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003656539480355741,
      "loss": 5.2672,
      "mean_token_accuracy": 0.17965516299009324,
      "num_tokens": 6163253.0,
      "step": 2105
    },
    {
      "entropy": 5.343049764633179,
      "epoch": 0.06617116693323298,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0003645744671963891,
      "loss": 5.3038,
      "mean_token_accuracy": 0.16757653951644896,
      "num_tokens": 6177629.0,
      "step": 2110
    },
    {
      "entropy": 5.498579788208008,
      "epoch": 0.06632797064634491,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003634925306279376,
      "loss": 5.4428,
      "mean_token_accuracy": 0.16230330765247344,
      "num_tokens": 6192102.0,
      "step": 2115
    },
    {
      "entropy": 5.5992106914520265,
      "epoch": 0.06648477435945684,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0003624081679920574,
      "loss": 5.4747,
      "mean_token_accuracy": 0.15547768026590347,
      "num_tokens": 6209005.0,
      "step": 2120
    },
    {
      "entropy": 5.49295563697815,
      "epoch": 0.06664157807256876,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0003613214090170977,
      "loss": 5.3607,
      "mean_token_accuracy": 0.15989728271961212,
      "num_tokens": 6223134.0,
      "step": 2125
    },
    {
      "entropy": 5.478620433807373,
      "epoch": 0.06679838178568069,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003602322834971048,
      "loss": 5.4698,
      "mean_token_accuracy": 0.1545787662267685,
      "num_tokens": 6238474.0,
      "step": 2130
    },
    {
      "entropy": 5.428533124923706,
      "epoch": 0.0669551854987926,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003591408212910051,
      "loss": 5.3,
      "mean_token_accuracy": 0.16924979537725449,
      "num_tokens": 6253296.0,
      "step": 2135
    },
    {
      "entropy": 5.48041467666626,
      "epoch": 0.06711198921190453,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0003580470523217863,
      "loss": 5.3992,
      "mean_token_accuracy": 0.1572578251361847,
      "num_tokens": 6267239.0,
      "step": 2140
    },
    {
      "entropy": 5.5361556053161625,
      "epoch": 0.06726879292501646,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003569510065756771,
      "loss": 5.3284,
      "mean_token_accuracy": 0.16762975305318834,
      "num_tokens": 6281887.0,
      "step": 2145
    },
    {
      "entropy": 5.402921390533447,
      "epoch": 0.06742559663812839,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003558527141013254,
      "loss": 5.2977,
      "mean_token_accuracy": 0.17072063982486724,
      "num_tokens": 6295972.0,
      "step": 2150
    },
    {
      "entropy": 5.447606754302979,
      "epoch": 0.06758240035124032,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003547522050089742,
      "loss": 5.3777,
      "mean_token_accuracy": 0.16163324415683747,
      "num_tokens": 6311624.0,
      "step": 2155
    },
    {
      "entropy": 5.480730867385864,
      "epoch": 0.06773920406435224,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00035364950946963606,
      "loss": 5.3373,
      "mean_token_accuracy": 0.17015375643968583,
      "num_tokens": 6326691.0,
      "step": 2160
    },
    {
      "entropy": 5.453211641311645,
      "epoch": 0.06789600777746417,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0003525446577142663,
      "loss": 5.4,
      "mean_token_accuracy": 0.16562338769435883,
      "num_tokens": 6340656.0,
      "step": 2165
    },
    {
      "entropy": 5.415660095214844,
      "epoch": 0.0680528114905761,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035143768003293395,
      "loss": 5.3165,
      "mean_token_accuracy": 0.16582199484109877,
      "num_tokens": 6354504.0,
      "step": 2170
    },
    {
      "entropy": 5.484099197387695,
      "epoch": 0.06820961520368803,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0003503286067739913,
      "loss": 5.434,
      "mean_token_accuracy": 0.16360560953617095,
      "num_tokens": 6370042.0,
      "step": 2175
    },
    {
      "entropy": 5.540794420242309,
      "epoch": 0.06836641891679995,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00034921746834324193,
      "loss": 5.4698,
      "mean_token_accuracy": 0.16043284088373183,
      "num_tokens": 6385413.0,
      "step": 2180
    },
    {
      "entropy": 5.499316692352295,
      "epoch": 0.06852322262991188,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003481042952031072,
      "loss": 5.3512,
      "mean_token_accuracy": 0.16809642761945726,
      "num_tokens": 6400468.0,
      "step": 2185
    },
    {
      "entropy": 5.459881019592285,
      "epoch": 0.06868002634302381,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003469891178717911,
      "loss": 5.4017,
      "mean_token_accuracy": 0.16188558787107468,
      "num_tokens": 6414720.0,
      "step": 2190
    },
    {
      "entropy": 5.51483473777771,
      "epoch": 0.06883683005613572,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003458719669224436,
      "loss": 5.4079,
      "mean_token_accuracy": 0.1607223004102707,
      "num_tokens": 6429678.0,
      "step": 2195
    },
    {
      "entropy": 5.382878208160401,
      "epoch": 0.06899363376924765,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003447528729823221,
      "loss": 5.2761,
      "mean_token_accuracy": 0.17314398288726807,
      "num_tokens": 6442481.0,
      "step": 2200
    },
    {
      "entropy": 5.489427852630615,
      "epoch": 0.06915043748235958,
      "grad_norm": 0.90625,
      "learning_rate": 0.0003436318667319525,
      "loss": 5.445,
      "mean_token_accuracy": 0.15897778123617173,
      "num_tokens": 6458397.0,
      "step": 2205
    },
    {
      "entropy": 5.430074548721313,
      "epoch": 0.0693072411954715,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00034250897890428716,
      "loss": 5.234,
      "mean_token_accuracy": 0.17394062131643295,
      "num_tokens": 6471874.0,
      "step": 2210
    },
    {
      "entropy": 5.454744720458985,
      "epoch": 0.06946404490858343,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003413842402838633,
      "loss": 5.4338,
      "mean_token_accuracy": 0.16406074166297913,
      "num_tokens": 6485719.0,
      "step": 2215
    },
    {
      "entropy": 5.4392256259918215,
      "epoch": 0.06962084862169536,
      "grad_norm": 1.0,
      "learning_rate": 0.00034025768170595834,
      "loss": 5.2209,
      "mean_token_accuracy": 0.17636980563402177,
      "num_tokens": 6500368.0,
      "step": 2220
    },
    {
      "entropy": 5.4150715351104735,
      "epoch": 0.06977765233480729,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0003391293340557446,
      "loss": 5.2868,
      "mean_token_accuracy": 0.1690964236855507,
      "num_tokens": 6516829.0,
      "step": 2225
    },
    {
      "entropy": 5.421437788009643,
      "epoch": 0.06993445604791922,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0003379992282674431,
      "loss": 5.3989,
      "mean_token_accuracy": 0.16200968623161316,
      "num_tokens": 6532550.0,
      "step": 2230
    },
    {
      "entropy": 5.415826034545899,
      "epoch": 0.07009125976103114,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003368673953234749,
      "loss": 5.3133,
      "mean_token_accuracy": 0.17040503472089769,
      "num_tokens": 6547899.0,
      "step": 2235
    },
    {
      "entropy": 5.430303144454956,
      "epoch": 0.07024806347414307,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00033573386625361176,
      "loss": 5.2606,
      "mean_token_accuracy": 0.17038411349058152,
      "num_tokens": 6564150.0,
      "step": 2240
    },
    {
      "entropy": 5.394917631149292,
      "epoch": 0.070404867187255,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033459867213412567,
      "loss": 5.2915,
      "mean_token_accuracy": 0.17163457870483398,
      "num_tokens": 6579795.0,
      "step": 2245
    },
    {
      "entropy": 5.3988746166229244,
      "epoch": 0.07056167090036693,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000333461844086937,
      "loss": 5.3316,
      "mean_token_accuracy": 0.16311462223529816,
      "num_tokens": 6595974.0,
      "step": 2250
    },
    {
      "entropy": 5.435396146774292,
      "epoch": 0.07071847461347884,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033232341327876097,
      "loss": 5.3441,
      "mean_token_accuracy": 0.17145417332649232,
      "num_tokens": 6610385.0,
      "step": 2255
    },
    {
      "entropy": 5.472650051116943,
      "epoch": 0.07087527832659077,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003311834109202531,
      "loss": 5.3341,
      "mean_token_accuracy": 0.17234935164451598,
      "num_tokens": 6625078.0,
      "step": 2260
    },
    {
      "entropy": 5.461251878738404,
      "epoch": 0.0710320820397027,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033004186826515416,
      "loss": 5.2843,
      "mean_token_accuracy": 0.16021159738302232,
      "num_tokens": 6638390.0,
      "step": 2265
    },
    {
      "entropy": 5.407567644119263,
      "epoch": 0.07118888575281462,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003288988166094324,
      "loss": 5.3173,
      "mean_token_accuracy": 0.1673991709947586,
      "num_tokens": 6651721.0,
      "step": 2270
    },
    {
      "entropy": 5.420019674301147,
      "epoch": 0.07134568946592655,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032775428729042656,
      "loss": 5.3146,
      "mean_token_accuracy": 0.1679161623120308,
      "num_tokens": 6667175.0,
      "step": 2275
    },
    {
      "entropy": 5.4789347648620605,
      "epoch": 0.07150249317903848,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000326608311685986,
      "loss": 5.4204,
      "mean_token_accuracy": 0.16072620302438737,
      "num_tokens": 6682681.0,
      "step": 2280
    },
    {
      "entropy": 5.408848476409912,
      "epoch": 0.0716592968921504,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003254609212136108,
      "loss": 5.2796,
      "mean_token_accuracy": 0.1738677978515625,
      "num_tokens": 6697555.0,
      "step": 2285
    },
    {
      "entropy": 5.338811588287354,
      "epoch": 0.07181610060526233,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032431214732959036,
      "loss": 5.1247,
      "mean_token_accuracy": 0.1783951237797737,
      "num_tokens": 6711298.0,
      "step": 2290
    },
    {
      "entropy": 5.344261217117309,
      "epoch": 0.07197290431837426,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000323162021528141,
      "loss": 5.2803,
      "mean_token_accuracy": 0.17247539311647414,
      "num_tokens": 6726700.0,
      "step": 2295
    },
    {
      "entropy": 5.388544130325317,
      "epoch": 0.07212970803148619,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00032201057534054264,
      "loss": 5.22,
      "mean_token_accuracy": 0.1700620949268341,
      "num_tokens": 6742399.0,
      "step": 2300
    },
    {
      "entropy": 5.38937554359436,
      "epoch": 0.07228651174459812,
      "grad_norm": 0.90625,
      "learning_rate": 0.00032085784033427414,
      "loss": 5.2808,
      "mean_token_accuracy": 0.17642759680747985,
      "num_tokens": 6756718.0,
      "step": 2305
    },
    {
      "entropy": 5.344116592407227,
      "epoch": 0.07244331545771004,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003197038481121478,
      "loss": 5.1643,
      "mean_token_accuracy": 0.17331337034702302,
      "num_tokens": 6770915.0,
      "step": 2310
    },
    {
      "entropy": 5.406969451904297,
      "epoch": 0.07260011917082196,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0003185486303114436,
      "loss": 5.3816,
      "mean_token_accuracy": 0.16489416509866714,
      "num_tokens": 6785250.0,
      "step": 2315
    },
    {
      "entropy": 5.440972566604614,
      "epoch": 0.07275692288393389,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003173922186030409,
      "loss": 5.3405,
      "mean_token_accuracy": 0.16699183732271194,
      "num_tokens": 6800057.0,
      "step": 2320
    },
    {
      "entropy": 5.389354228973389,
      "epoch": 0.07291372659704581,
      "grad_norm": 1.078125,
      "learning_rate": 0.000316234644690551,
      "loss": 5.2416,
      "mean_token_accuracy": 0.1720912516117096,
      "num_tokens": 6814100.0,
      "step": 2325
    },
    {
      "entropy": 5.375819158554077,
      "epoch": 0.07307053031015774,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003150759403094473,
      "loss": 5.19,
      "mean_token_accuracy": 0.18120640218257905,
      "num_tokens": 6826933.0,
      "step": 2330
    },
    {
      "entropy": 5.3471251964569095,
      "epoch": 0.07322733402326967,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00031391613722619587,
      "loss": 5.1391,
      "mean_token_accuracy": 0.1811688706278801,
      "num_tokens": 6840279.0,
      "step": 2335
    },
    {
      "entropy": 5.294892311096191,
      "epoch": 0.0733841377363816,
      "grad_norm": 0.953125,
      "learning_rate": 0.000312755267237384,
      "loss": 5.1874,
      "mean_token_accuracy": 0.17787426859140396,
      "num_tokens": 6854987.0,
      "step": 2340
    },
    {
      "entropy": 5.373378753662109,
      "epoch": 0.07354094144949352,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003115933621688488,
      "loss": 5.2862,
      "mean_token_accuracy": 0.16869913190603256,
      "num_tokens": 6870559.0,
      "step": 2345
    },
    {
      "entropy": 5.493012857437134,
      "epoch": 0.07369774516260545,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00031043045387480487,
      "loss": 5.3371,
      "mean_token_accuracy": 0.1719384267926216,
      "num_tokens": 6884027.0,
      "step": 2350
    },
    {
      "entropy": 5.370590591430664,
      "epoch": 0.07385454887571738,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003092665742369703,
      "loss": 5.327,
      "mean_token_accuracy": 0.1683666244149208,
      "num_tokens": 6897276.0,
      "step": 2355
    },
    {
      "entropy": 5.350070476531982,
      "epoch": 0.07401135258882931,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00030810175516369343,
      "loss": 5.2637,
      "mean_token_accuracy": 0.1680775538086891,
      "num_tokens": 6912054.0,
      "step": 2360
    },
    {
      "entropy": 5.490565586090088,
      "epoch": 0.07416815630194123,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0003069360285890775,
      "loss": 5.3106,
      "mean_token_accuracy": 0.1678343892097473,
      "num_tokens": 6926864.0,
      "step": 2365
    },
    {
      "entropy": 5.410408210754395,
      "epoch": 0.07432496001505316,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030576942647210547,
      "loss": 5.2572,
      "mean_token_accuracy": 0.17173388600349426,
      "num_tokens": 6939946.0,
      "step": 2370
    },
    {
      "entropy": 5.332781887054443,
      "epoch": 0.07448176372816508,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030460198079576355,
      "loss": 5.2163,
      "mean_token_accuracy": 0.1741424486041069,
      "num_tokens": 6953320.0,
      "step": 2375
    },
    {
      "entropy": 5.467446136474609,
      "epoch": 0.074638567441277,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003034337235661648,
      "loss": 5.3685,
      "mean_token_accuracy": 0.16742643415927888,
      "num_tokens": 6967118.0,
      "step": 2380
    },
    {
      "entropy": 5.401697874069214,
      "epoch": 0.07479537115438893,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0003022646868116714,
      "loss": 5.2851,
      "mean_token_accuracy": 0.17228549867868423,
      "num_tokens": 6982949.0,
      "step": 2385
    },
    {
      "entropy": 5.350200128555298,
      "epoch": 0.07495217486750086,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003010949025820163,
      "loss": 5.2321,
      "mean_token_accuracy": 0.1780702739953995,
      "num_tokens": 6997750.0,
      "step": 2390
    },
    {
      "entropy": 5.374381256103516,
      "epoch": 0.07510897858061279,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0002999244029474252,
      "loss": 5.2905,
      "mean_token_accuracy": 0.1750371888279915,
      "num_tokens": 7012581.0,
      "step": 2395
    },
    {
      "entropy": 5.446659708023072,
      "epoch": 0.07526578229372471,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029875321999773684,
      "loss": 5.239,
      "mean_token_accuracy": 0.1738226056098938,
      "num_tokens": 7026192.0,
      "step": 2400
    },
    {
      "entropy": 5.442251110076905,
      "epoch": 0.07542258600683664,
      "grad_norm": 1.0,
      "learning_rate": 0.00029758138584152333,
      "loss": 5.4083,
      "mean_token_accuracy": 0.1644318714737892,
      "num_tokens": 7040496.0,
      "step": 2405
    },
    {
      "entropy": 5.438568544387818,
      "epoch": 0.07557938971994857,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0002964089326052102,
      "loss": 5.2907,
      "mean_token_accuracy": 0.17047504484653472,
      "num_tokens": 7055739.0,
      "step": 2410
    },
    {
      "entropy": 5.405134677886963,
      "epoch": 0.0757361934330605,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0002952358924321949,
      "loss": 5.339,
      "mean_token_accuracy": 0.16206610202789307,
      "num_tokens": 7070071.0,
      "step": 2415
    },
    {
      "entropy": 5.410668659210205,
      "epoch": 0.07589299714617243,
      "grad_norm": 1.03125,
      "learning_rate": 0.00029406229748196657,
      "loss": 5.3378,
      "mean_token_accuracy": 0.16102805286645888,
      "num_tokens": 7085329.0,
      "step": 2420
    },
    {
      "entropy": 5.337376546859741,
      "epoch": 0.07604980085928435,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002928881799292235,
      "loss": 5.1477,
      "mean_token_accuracy": 0.17660113275051117,
      "num_tokens": 7100313.0,
      "step": 2425
    },
    {
      "entropy": 5.369079828262329,
      "epoch": 0.07620660457239628,
      "grad_norm": 1.0625,
      "learning_rate": 0.00029171357196299154,
      "loss": 5.2666,
      "mean_token_accuracy": 0.17177634239196776,
      "num_tokens": 7113781.0,
      "step": 2430
    },
    {
      "entropy": 5.37522292137146,
      "epoch": 0.0763634082855082,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0002905385057857414,
      "loss": 5.2384,
      "mean_token_accuracy": 0.1755241796374321,
      "num_tokens": 7128338.0,
      "step": 2435
    },
    {
      "entropy": 5.393016338348389,
      "epoch": 0.07652021199862012,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002893630136125058,
      "loss": 5.224,
      "mean_token_accuracy": 0.17592531442642212,
      "num_tokens": 7142392.0,
      "step": 2440
    },
    {
      "entropy": 5.360267114639282,
      "epoch": 0.07667701571173205,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0002881871276699967,
      "loss": 5.2994,
      "mean_token_accuracy": 0.1711569085717201,
      "num_tokens": 7158080.0,
      "step": 2445
    },
    {
      "entropy": 5.409264707565308,
      "epoch": 0.07683381942484398,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028701088019572114,
      "loss": 5.253,
      "mean_token_accuracy": 0.1703270971775055,
      "num_tokens": 7172421.0,
      "step": 2450
    },
    {
      "entropy": 5.4173094749450685,
      "epoch": 0.0769906231379559,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0002858343034370977,
      "loss": 5.2463,
      "mean_token_accuracy": 0.16973855420947076,
      "num_tokens": 7188369.0,
      "step": 2455
    },
    {
      "entropy": 5.4142214298248295,
      "epoch": 0.07714742685106783,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00028465742965057267,
      "loss": 5.3641,
      "mean_token_accuracy": 0.1718568906188011,
      "num_tokens": 7203434.0,
      "step": 2460
    },
    {
      "entropy": 5.403614521026611,
      "epoch": 0.07730423056417976,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028348029110073533,
      "loss": 5.2666,
      "mean_token_accuracy": 0.1755361244082451,
      "num_tokens": 7217709.0,
      "step": 2465
    },
    {
      "entropy": 5.4026562690734865,
      "epoch": 0.07746103427729169,
      "grad_norm": 0.8671875,
      "learning_rate": 0.00028230292005943365,
      "loss": 5.2665,
      "mean_token_accuracy": 0.1695285990834236,
      "num_tokens": 7233475.0,
      "step": 2470
    },
    {
      "entropy": 5.397445344924927,
      "epoch": 0.07761783799040362,
      "grad_norm": 0.8515625,
      "learning_rate": 0.00028112534880488945,
      "loss": 5.2636,
      "mean_token_accuracy": 0.16677384972572326,
      "num_tokens": 7250753.0,
      "step": 2475
    },
    {
      "entropy": 5.282281112670899,
      "epoch": 0.07777464170351554,
      "grad_norm": 0.9375,
      "learning_rate": 0.0002799476096208137,
      "loss": 5.1742,
      "mean_token_accuracy": 0.1840490221977234,
      "num_tokens": 7266061.0,
      "step": 2480
    },
    {
      "entropy": 5.370245885848999,
      "epoch": 0.07793144541662747,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00027876973479552087,
      "loss": 5.3002,
      "mean_token_accuracy": 0.17231762111186982,
      "num_tokens": 7281798.0,
      "step": 2485
    },
    {
      "entropy": 5.4180694103240965,
      "epoch": 0.0780882491297394,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00027759175662104424,
      "loss": 5.2373,
      "mean_token_accuracy": 0.1746474653482437,
      "num_tokens": 7295541.0,
      "step": 2490
    },
    {
      "entropy": 5.420738077163696,
      "epoch": 0.07824505284285131,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0002764137073922508,
      "loss": 5.3273,
      "mean_token_accuracy": 0.16565654724836348,
      "num_tokens": 7311099.0,
      "step": 2495
    },
    {
      "entropy": 5.381645679473877,
      "epoch": 0.07840185655596324,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00027523561940595505,
      "loss": 5.176,
      "mean_token_accuracy": 0.18013110309839248,
      "num_tokens": 7323866.0,
      "step": 2500
    }
  ],
  "logging_steps": 5,
  "max_steps": 4000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.2850889213952e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}